The Reactivity of Rare-Earth Metallocenes towards Alkynes: Mechanism and Synthetic Applications

Victor Fidel Quiroga Norambuena

The Reactivity of Rare-Earth Metallocenes towards Alkynes: Mechanism and Synthetic Applications V. F. Quiroga Norambuena

The research project described in this thesis was supported financially by the Netherlands Organization of Scientific Research (NWO).

Cover design: J. F. Quiroga Espejo Printed by Grafisch bedrijf Ponsen en Looijen BV, Wageningen

Copyright 2006 V. F. Quiroga Norambuena

RIJKSUNIVERSITEIT GRONINGEN

The Reactivity of Rare-Earth Metallocenes towards Alkynes: Mechanism and Synthetic Applications

Proefschrift

ter verkrijging van het doctoraat in de Wiskunde en Natuurwetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op vrijdag 17 november 2006 om 13:15 uur

door

Victor Fidel Quiroga Norambuena

geboren op 12 november 1975 te Heemskerk

Promotor: Prof. dr. B. Hessen Beoordelingscommissie: Prof. dr. ir. H. J. Heeres Prof. dr. K. Lammertsma Prof. dr. U. Rosenthal

ISBN: 90-367-2821-5 (Gedrukte versie) 90-367-2822-3 (Elektronische versie)

Aan mijn ouders

Contents
1. 1.1. 1.2. 1.3. 1.4. 1.5. 2. INTRODUCTION ................................................................................................... 1 HISTORICAL BACKGROUND .................................................................................. 1 GENERAL PROPERTIES .......................................................................................... 2 ORGANO RARE-EARTH METAL COMPLEXES IN HOMOGENEOUS CATALYSIS .......... 4 OBJECTIVE AND OVERVIEW.................................................................................. 5 REFERENCES AND NOTES ..................................................................................... 6 RARE-EARTH METALLOCENE PROPARGYL/ALLENYLS ..................... 13 2.1. INTRODUCTION .................................................................................................. 13 2.2. SYNTHESIS OF SUBSTRATES ............................................................................... 14 2.3. SYNTHESIS OF PROPARGYL/ALLENYLS ............................................................... 17 2.3.1. Introduction ................................................................................................ 17 2.3.2. Reactions of alkyls with 1-aryl-1-propynes ................................................ 17 2.3.3. Reactions of hydrides with 1-aryl-1-propynes ............................................ 19 2.4. STRUCTURAL CHARACTERIZATION OF PROPARGYL/ALLENYLS .......................... 23 2.4.1. Introduction ................................................................................................ 23 2.4.2. Spectroscopic properties of the propargyl/allenyls .................................... 24 2.4.3. Ligand and metal ion size effects ................................................................ 26 2.4.4. The molecular structures of Cp*2LnCH2CCPh (Ln = La, Y)..................... 28 2.5. REACTIVITY OF PROPARGYL/ALLENYLS ............................................................. 31 2.5.1. Introduction ................................................................................................ 31 2.5.2. The reactivity towards protic acids............................................................. 32 2.5.3. Thermolysis................................................................................................. 37 2.5.4. The reactivity towards Lewis bases ............................................................ 39 2.5.5. The molecular structure of Cp*2La(3-CH2CCPh)(py)............................... 42 2.5.6. The reactivity towards unsaturated substrates ........................................... 44 2.6. CONCLUSIONS .................................................................................................... 50 2.7. EXPERIMENTAL SECTION ................................................................................... 50 2.8. REFERENCES AND NOTES ................................................................................... 60 3. THE CYCLODIMERIZATION OF 1-METHYLALK-2-YNES CATALYZED BY RARE-EARTH METALLOCENES....................................................................... 71 3.1. INTRODUCTION .................................................................................................. 71 3.2. THE CYCLODIMERIZATION OF PROPYNYLAROMATICS ........................................ 72 3.2.1. Lanthanocene-catalyzed cyclodimerization................................................ 72 3.2.2. Influence of the metal ion size..................................................................... 78 3.2.3. Influence of the ancillary ligation............................................................... 80 3.2.4. Reactions with 2-propynyltoluene............................................................... 82 3.3. THE CYCLODIMERIZATION OF DIPROPYNYLAROMATICS ..................................... 85 3.3.1. Introduction ................................................................................................ 85 3.3.2. 1,4-Dipropynylbenzene ............................................................................... 86

3.3.3. 1,4-Dipropynyl-2,5-di-n-hexylbenzene. ...................................................... 86 3.4. CONCLUSIONS .................................................................................................... 90 3.5. EXPERIMENTAL SECTION ................................................................................... 91 3.6. REFERENCES AND NOTES ................................................................................... 96 4. THE OLIGOMERIZATION OF PHENYLACETYLENE CATALYZED BY RARE-EARTH METALLOCENES ........................................................................... 101 4.1. INTRODUCTION ................................................................................................ 101 4.2. CATALYTIC OLIGOMERIZATION OF PHENYLACETYLENE ................................... 101 4.2.1. Introduction .............................................................................................. 101 4.2.2. Influence of the precatalysts ..................................................................... 102 4.2.3. Product analysis ....................................................................................... 105 4.2.4. Effect of substrate concentration .............................................................. 106 4.2.5. Reaction kinetics ....................................................................................... 107 4.2.6. Reaction intermediates ............................................................................. 111 4.3. STOICHIOMETRIC REACTIONS WITH PHENYLACETYLENE.................................. 113 4.3.1. Introduction .............................................................................................. 113 4.3.2. The hydride derivative .............................................................................. 114 4.3.3. The alkyl derivative................................................................................... 115 4.3.4. The butatrienediyl derivative .................................................................... 116 4.4. MECHANISM OF CATALYTIC OLIGOMERIZATION .............................................. 120 4.4.1. Catalytic dimerization............................................................................... 120 4.4.2. Catalytic trimerization.............................................................................. 122 4.4.3. Catalyst deactivation ................................................................................ 123 4.4.4. The mechanistic interpretation of the observed kinetic behavior ............. 124 4.5. CONCLUSIONS .................................................................................................. 128 4.6. EXPERIMENTAL SECTION ................................................................................. 129 4.7. REFERENCES AND NOTES ................................................................................. 133 5. THE OLIGOMERIZATION OF (HETERO)AROMATIC 1-ALKYNES CATALYZED BY RARE-EARTH METALLOCENES........................................... 141 5.1. INTRODUCTION ................................................................................................ 141 5.2. SYNTHESIS OF (HETERO)AROMATIC 1-ALKYNES .............................................. 142 5.3. CATALYTIC OLIGOMERIZATION OF (HETERO)AROMATIC 1-ALKYNES ............... 143 5.3.1. Introduction .............................................................................................. 143 5.3.2. Substrate effects at relatively low initial substrate concentration ............ 143 5.3.3. Substrate effects at relatively high initial substrate concentration........... 144 5.3.4. Experiments with 2-ethynylthiophene ....................................................... 150 5.3.5. Experiments with 2-ethynylpyridine.......................................................... 153 5.3.6. Experiments with 2-ethynylanisole ........................................................... 155 5.3.7. Experiments with yttrium precatalyst ....................................................... 157 5.4. STOICHIOMETRIC REACTIONS OF (HETERO)AROMATIC 1-ALKYNES .................. 159 5.4.1. The reactivity of alkyl and hydride derivatives ......................................... 159 5.4.2. The reactivity of the dimeric alkynyl derivatives ...................................... 160 5.4.3. The reactivity of amide derivative towards phenylacetylene .................... 161

5.4.4. The reactivity of amide derivatives towards other 1-alkynes.................... 163 5.4.5. The reactivity of Lewis base adducts of monomeric alkynyl derivatives... 166 5.4.6. The molecular structures of butatrienediyl derivatives............................. 167 5.5. DISCUSSION OF SUBSTRATE EFFECTS ............................................................... 172 5.5.1. Introduction .............................................................................................. 172 5.5.2. Substrate effects in the reactivity of the dimeric alkynyl derivatives ........ 173 5.5.3. Substrate effects in the catalytic 1-alkyne oligomerization reactions ....... 175 5.6. CONCLUSIONS .................................................................................................. 181 5.7. EXPERIMENTAL SECTION ................................................................................. 181 APPENDIX. THE ACIDITY OF THE STUDIED 1-ALKYNES.................................................. 193 5.8. REFERENCES AND NOTES ................................................................................. 197 6. THE ORGANOLANTHANIDE-CATALYZED POLYMERIZATION OF DIYNES ......................................................................................................................... 207 6.1. INTRODUCTION ................................................................................................ 207 6.2. SYNTHESIS OF THE (HETERO)AROMATIC DIYNES.............................................. 209 6.3. POLYMERIZATION REACTIONS WITH (HETERO)AROMATIC DIYNES. .................. 212 6.3.1. Introduction .............................................................................................. 212 6.3.2. Polymerization reactions .......................................................................... 213 6.3.3. Polymerization reactions with end-capping agents .................................. 221 6.3.4. The effect of reaction temperature ............................................................ 224 6.3.5. The effect of monomer concentration........................................................ 225 6.4. STRUCTURAL CHARACTERIZATION AND PROPERTIES OF THE POLYMER. .......... 226 6.5. DISCUSSION ..................................................................................................... 230 6.6. CONCLUSIONS .................................................................................................. 235 6.7. EXPERIMENTAL SECTION ................................................................................. 236 6.8. REFERENCES AND NOTES ................................................................................. 239 SUMMARY ................................................................................................................... 249 SAMENVATTING........................................................................................................ 253 DANKWOORD............................................................................................................. 257 LIST OF ABBREVIATIONS....................................................................................... 259

Introduction

1.

Introduction

1.1.

Historical background

The lanthanides (Ln) consist of the 4f metals, ranging from lanthanum (La) to lutetium (Lu).1 The group 3 metals scandium (Sc) and yttrium (Y) exhibit similar chemical behavior as the lanthanide elements. Therefore, the lanthanide and group 3 metals are often considered as part of the same group, termed the rare earth metals.2 The organometallic chemistry of the rare earth metals has been slow to develop relative to that of other metals. The first well-characterized organometallic complexes of the lanthanide and group 3 elements, the tris(cyclopentadienyl) derivatives Cp3Ln, were prepared in 1954 by Wilkinson and Birmingham.3 Despite this early discovery, further development of this new area of organometallic chemistry was hampered by the intrinsic instability of these organometallic compounds towards moisture and oxygen. Another reason for the slow initial development of the organometallic chemistry of the rare earth metals was the belief that these compounds were ionic and represented merely trivalent versions of alkali and alkaline earth metal organometallic species. As a consequence, two decades of relative stagnation followed, until the availability of more sophisticated experimental and analytical techniques made the rigorous exclusion of traces of air and moisture during preparation and characterization possible in the 1970s. Further progress in the late 1970s and early 1980s demonstrated that many group 3 organometallics and organolanthanides display a rich and interesting chemistry, such as catalytic alkene hydrogenation4 and polymerization5 at very high rates6 and reactivity towards the C-H bonds of arenes, alkenes and alkanes.7 It became evident that the lanthanide and group 3 metals had the potential for some unique chemistry, distinct from anything possible with main-group or transition metals. In addition, the rare earth elements offer exceptional opportunities for tuning catalyst properties depending on ionic radii and ancillary ligand architecture. The low cost of the majority of the rare earth metals as compared to most transition metals and the lack of heavy-metal toxicity also contribute to the increasing attraction of catalysts based on group 3 and lanthanidemetals. The last two decades have witnessed a tremendous growth in research activities in this field, establishing that organo rare-earth metal complexes exhibit distinctive structural and physical properties and that many of them are highly active in variety of catalytic processes (Section 1.3).8 Up to the early 1970s organo rare-earth metal chemistry had been limited to -bonded organometallic compounds, such as tris(cyclopentadienyl)3, tris(indenyl)9 and cyclooctatetraene10 complexes. In addition, a variety of homoleptic compounds, such as Li[LnPh4]11 and Sc(CCPh)311a, and some ill-defined carbyls prepared via metal vapor synthesis had been described as well.12 In the mid-1970s, Tsutsui and Ely synthesized a number of alkyl, aryl and alkynyl bis(cyclopentadienyl) derivatives Cp2LnR.13 Since then, bis(cyclopentadienyl) complexes (metallocenes or sandwich complexes) attracted the most attention in the organometallic chemistry of rare earth metals. Cyclopentadienyl ligands are ideally suited to the rare earth metals, as they are capable of ionic bonding and can readily be tuned sterically with a variety of substituents. Unsubstituted cyclopentadienyl ligands provide complexes that are insoluble in hydrocarbon solution and usually display low reactivity.14 The pentamethylcyclopentadienyl ligand (C5Me5, abbreviated by Cp*), on the other hand, can generally provide excellent solubility and stability for the rare earth metal ions and constitutes the most studied ancillary ligand in the organometallic chemistry of the rare earth metals (e.g. 1 and 2, Scheme 1-1).15 A large variety of metallocenes with different subsituents (e.g. 3) has been prepared in the last decades.16 The Cp rings have, furthermore, been connected with a bridging group to give an ansa system (e.g. 4, 5, 7, 8).17 Besides these ansametallocenes, half-metallocenes or half-sandwich complexes bearing mixed cyclopentadienyl-modentate-anionic ligands have also been explored, the most important example being the silylene-linked amido-cyclopentadienyl ligand (e.g. 6).18 Although cyclopentadienyl ligands have proven to be very suitable for preparing and examing welldefined, monomeric organometallic complexes and much elegant chemistry has been uncovered, generally two Cp donors are required in the cyclopentadienyl chemistry of rare earth metals, thereby limiting the chemistry to reactions involving one M-C bond in neutral complexes. Thus, recent work in this field has shown a departure from cyclopentadienyl chemistry, searching not only for different dianionic ligand environments, but also for mono-anionic and neutral ligand environments that allow the preparation of bis- and tris(alkyl) complexes.19 The

Chapter 1

Scheme 1-1. Examples of metallocenes, ansa-metallocenes and half-metallocenes in organo rare-earth metal chemistry.
Achiral SiMe3 Ln E(SiMe3)2 Ln H Ln H Me3Si H3 C C H3 SiMe3 Ln SiMe3

Ln

Si

Ln E(SiMe3)2

(Me3Si)2E

Ln

Si N

Ln E(SiMe3)2

4
Chiral

E = N or CH Si Ln E(SiMe3)2 Si Ln E(SiMe3)2 R* = (-)-methyl, (+)-neomenthyl or (-)-phenylmethyl

R*

R*

latter two compounds are of interest, because of the potential for generating rare earth metal cations, related to the group 4 metal cations, which have been of fundamental importance in the field of olefin polymerization.20 The last 5-10 years has witnessed a large amount of new ligands for use with group 3 and lanthanide metals and many of them have provided complexes that are of interest as (pre)catalysts (Section 1.3).

1.2.

General properties

The lanthanides have the general electronic configuration [Xe]4fn5d16s2 with n = 0 (La) to 14 (Lu). The 4f valence orbitals of the lanthanides do not protrude significantly beyond the filled 5s2 and 5p6 orbitals. As a result, 4f electrons are commonly thought to be unavailable for bonding and ligand field effects are practically absent.21 For this reason, the chemistry of the lanthanides is believed to be predominantly ionic and governed more by electrostatic factors and steric requirements than by filled orbital considerations.22 As a consequence, the chemistry, spectroscopy and magnetism of rare-earth metal ions differ considerably from d transition-metal ions. With no electronic preferences for particular geometries, irregular geometries are common. General principles of d-transition metal ligand bonding, such as -donor/-acceptor interaction and the 18-electron rule are not observed in group 3 and lanthanide chemistry and examples of imido or alkylidene complexes remain scarce.23 Contrary to common belief, a more recent study showed that both covalent and ionic contributions play an important role in determining the molecular structure of rare-earth metal complexes.24 This finding suggested that the bonding description of rare-earth metal complexes lies somwhere on a continuum between purely ionic and covalent bonding. Lanthanides and group 3 metals have one common thermodynamically favorable oxidation state, which is the trivalent positive state (Ln3+). Cerium is unique in that it also possesses an accessible tetravalent oxidation state. Other readily accessible nontrivalent oxidation states are Sm2+ (4f6), Eu2+ (4f7) and Yb2+ (4f14). The inert-gas electronic configuration of the Ln(III) derivatives implies a similar chemical behavior, resembling in this sense broadly the behavior of the early d-block elements in their highest oxidation states. The preferred coordination numbers of the Ln(III) cations are in the range of 8-12; eight-coordination is typical for rare earth

Introduction metal complexes. High coordination numbers are usually accomplished by forming oligomeric structures or highly solvated complexes. Since the reactivity of lanthanide complexes is correlated with the steric saturation at the metal center, both forms are undesirable for the preparation of highly reactive compounds. The relatively high charge-to-ionic-radius ratio results in an electropositive metal center that behaves as a hard Lewis acid, according to Pearsons hard-soft-acid-base (HSAB) principle.25 Based on their relative preferences for pyridine, Lappert suggested a relative Lewis acidity scale: Cp2ScMe > AlMe3 > Cp2YMe Cp2LnMe (here: Ln = large lanthanide elements).26 Accordingly, metal ions of the rare earth metals prefer coordination to hard ligands, such as O donors. Hence, the term oxophilic is often apllied to these ions. Marks series of bond energies for Cp*2Sm-X compounds illustrates that the bonding preferences are not as clear as the HSAB principle would have it: Cl > CCPh > Br > O(t-Bu) > S(n-Pr) > I > H > NMe2 > PEt2.27 Although the LnX bonds are thermodynamically stable, group 3 and lanthanide complexes display high kinetic lability, due to their typical high ligand exchange ability.28 The pronounced oxophilicity makes organometallic compounds of the rare earth metals very reactive towards water and air, resembling early d-block metals in this respect.

(1.1)

L2Ln

L2Ln

(1.2)

L2Ln

L2Ln Y -

X Z +

L2Ln

Due to the absence of energetically accessible oxidation states that allow two-electron redox processes, oxidative addition and reductive elimination reactions are not observed for rare earth metals. Instead, the reactivity of the organometallic complexes having metals in the trivalent oxidation state is dominated by (i) simple Lewis acid/base-type interactions, (ii) insertion of unsaturated moieties into polar Ln-C and Ln-H bonds (Eq. 1.1), (iii) -H and -alkyl elimination and (iv) -bond metathesis processes (Eq. 1.2). The latter is a concerted, bimolecular reaction that can involve both polar (i.e. alkane C-H bonds) and nonpolar (i.e. H2) bonds and proceeds via a highly ordered, polarized four-centered transition state. Only Sm2+/Sm3+, Eu+/Eu2+, Yb+/Yb2+ and Ce3+/Ce4+ have more than one energetically accessible oxidation state and one-electron redox shuttles are observed for these metals. The effective eight-coordinate ionic radius of Ln3+ gradually decreases from 1.16 for La3+ to 0.977 for Lu3+. This particularly feature is known as the lanthanide contraction. The gradual decrease in ionic radius and the limited radial extension of the valence orbitals are manifested in subtle differences in the formation, coordination geometry and reactivity of rare earth metal complexes, having similar chemical environments, but different metals. The intrinsic properties of the rare earth cations, as revealed by their oxophilicity, hard Lewisacid character and large size, govern the reactivity of the compounds of the rare earth metals. Hence, parallels to the chemistry of aluminum, group 2 and group 4 elements and the actinides are often found. The lanthanide ions exhibit three types of electronic transitions. The colors arising from forbidden 4f 4f transitions show little variation with ligand substitutions. This feature complicates separation techniques. Instead, reactions and appropriate reagents must be chosen, so that one (major) product is formed which can then be purified by (fractional) crystallization or sublimation. In addition, most of the lanthanide ions are luminescent, either fluorescent (e.g. PrIII, NdIII, HoIII, ErIII and YbIII) or phosphorescent (e.g. SmIII, EuIII, GdIII, TbIII, DyIII and TmIII). The ions LaIII and LuIII have no f-f transitions and are not luminescent. The paramagnetic rare earth metals have Laporte allowed 4f 5d transitions and have intense, variable colors that change significantly with metal environment, a clear synthetic advantage. The third type of electronic transition displayed by lanthanide ions are charge-transfer transitions and both ligand-to-metal and metal-to-ligand transitions are allowed. Because their energies are high, however, these transitions are less widespread than encountered in d transition-metal chemistry. The technological importance of the lanthanide ions for magnetic and optical materials has long attracted interest in their physical properties and solid-state chemistry.29 NMR spectroscopy is the most important tool for solution structure determination in the organometallic chemistry of rare earth metals. Because the rare earth metals are paramagnetic for all configurations from 4f1 to 4f13, it is not surprising that many researchers have chosen to investigate the chemistry

Chapter 1 of diamagnetic Sc3+, Y3+, La3+, Yb2+ and Lu3+. These nuclei are also accessible to direct observation by heteronuclear NMR spectroscopy.30 For yttrium complexes, the 89Y nucleus (100% abundance) has a spin of 1/2 and the coupling information available is highly informative. 89Y NMR spectroscopy has become a valuable diagnostic tool for the characterization of organoyttrium complexes, but the utility of this technique is confined to concentrated samples, due to long relaxation times and negative nOe effects.31 The 45Sc (I = 7/2, 100%), 175Lu (I = 7/2, 97.4%) and 139La (I = 7/2, 99.91%) nuclei have a significant quadrupole moment. As a result, the carbon atoms bonded directly to the metal are sometimes difficult to detect and -protons in hydrocarbyl derivatives are broadened to varying degree. Although the 139La and 45Sc nuclei have high relative sensitivities and display a wide range of chemical shifts, their relatively large nuclear spins and quadrupole moments give rise to broad resonances. As a result, 139La and 45Sc NMR data are scarce in literature,32 while the use of 175Lu NMR spectroscopy has virtually completely been limited by its large quadrupole moment. 171Yb (I = 1/2, 14.27%) NMR spectroscopy, on the other hand, is well-developed and has become a useful method to study organoytterbium(II) complexes.33 It should also be noted that interpretable and diagnostic, albeit paramagnetically shifted, NMR spectra for Ce3+, Nd3+, Sm3+ and Sm2+ can be obtained.

1.3.

Organo rare-earth metal complexes in homogeneous catalysis

The first indication of catalytic activity of organolanthanides came from the observation that oxides, halides or alkoxides in the presence of co-catalysts, such as lithium alkyls, aluminum alkyls or other hydrocarbyl transfer reagents catalyzed the cracking of hydrocarbons, the oligomerization of alkenes and the polymerization of alkenes and alkynes.34 Alkene and alkyne hydrogenation catalyst were prepared by co-condensing lanthanide metal vapors with unsaturated hydrocarbons.35 Later, when well-defined organolanthanide complexes became available, it was found that some of these complexes are also effective catalysts. For example, polymerization of ethene, propene and butadiene was observed for bis(cyclopentadienyl)lanthanide hydrocarbyls36, (cyclooctatetraenyl)cerium complexes and anionic tetra(allyl)lanthanide complexes.37 Organolanthanide hydrides38 and cyclopentadienyl ytterbium(II) complexes39 catalyzed the hydrogenation of alkenes and alkynes. In the 1980s metallocene complexes of group 3 and lanthanide metals received much attention as alkene polymerization catalysts.40 This can be attributed to the intrinsic high polymerization activity of neutral alkyl and hydrido derivatives towards ethylene and the possibility to study the debated mechanism of ZieglerNatta polymerization without the complication of ill-defined cocatalysts and possible ion pairing effects. Conventional rare-earth metallocene complexes are, however, inefficient catalysts for the polymerization of 1alkenes higher than ethylene, due to rapid and irreversible alllylic C-H activation. Modifications of the ligand environment of these catalysts have been shown to suppress this side reaction.41 Other monomers, such as dienes42, styrene derivatives43, methacrylates44 and lactones45, have also been polymerized by metallocene complexes of the rare earth metals and their derivatives.46 The discovery that some organometallics of the rare earth metals are highly active as polymerization catalysts prompted interest in other catalytic applications, but efforts to apply these catalysts to organic synthesis have only recently begun.47 Most catalytic applications of the organometallic compounds of group 3 and lanthanide metals involve alkene transformations, such as hydrogenation,48 oligomerization,49 polymerization,50 cyclization,51 hydroamination,52 hydrosilylation,53 hydrophosphination54 and hydroboration.55 Moreover, these reactions have shown to proceed in some cases with high regio- and stereoselectivities, in combination with a high tolerance towards a wide variety of organic functional groups. Particularly impressive results have been obtained for the lanthanidocene-catalyzed hydroamination reaction, including regio-/diastereo-/enantioselective intramolecular cycloamination of aminoalkenes,56 aminoalkynes,57 aminoallenes58 and tandem bicyclizations of aminodienes, aminodiynes and aminoenynes.59 Recent research in this area has focused on extending the scope of catalytic reactions and on varying the bis(cyclopentadienyl) ligand coordinations sphere in order to achieve new catalytic applications. Especially variations of the cyclopentadienyl substituents or the use of ring-bridged cyclopentadienyl ligands has led to novel catalytic applications, such as oligomerization of 1-alkenes,60 isospecific polymerization of 1-alkenes,61 hydrocyclization of ,-dienes,62 C-C -bond activation63 and the hydromethylation of 1-alkenes.64 More recently, also the use of non-cyclopentadienyl supporting ligand environments has led to a wealth of catalytic applications, such as alkene polymerization,65 lactide polymerization,66 hydrosilylation67 and hydroamination.68 In some cases, improved catalytic behavior has been noted for cationic rare-earth metal complexes as compared

Introduction to their neutral analogues,69 such as for ethylene polymerization,70 styrene polymerization,71 diene polymerization,72 hydroamination73 and 1-alkyne dimerization.74

1.4.

Objective and overview

Compared to alkenes, catalytic transformations of alkynes, mediated by rare earth metal compounds, have received much less attention. Earlier it was observed that permethyllanthanidocene alkyl derivatives Cp*2LnCH(SiMe3)2 catalyze the unusual cyclodimerization of 1-methylalk-2-ynes CH3CCR (Eq. 1.3).75 However, the scope of substrates was believed to be confined to 1-methylalk-1-ynes with small alkyl groups and mechanistic details of this remarkable transformation were lacking. It was envisioned that the application of mono- and bis(propynyl)arenes in the catalytic cyclodimerization could provide synthetic access to novel and interesting materials. Therefore, a mechanistic study of the permethyllanthanocene-catalyzed cyclodimerization reaction of propynylarenes was initiated to determine the scope and mechanism of the catalytic cyclodimerization reaction of propynylaromatics, searching for selective catalytic pathways to the construction of a rare class of four-membered carbocycles and (cross-)conjugated polymers.
R R + R Ln = La, Ce R = Me, Et, nPr R

(1.3)

Cp*2LnCH(SiMe3)2 R

1-Alkynes are converted by permethylmetallocene alkyl derivatives Cp*2LnCH(SiMe3)2 to mixtures of enyne dimers and higher oligomers (Eq. 1.4).76 From the desire to apply the rare-earth metallocene-catalyzed oligomerization of 1-alkynes to the synthesis of novel conjugated polymers, a mechanistic study was undertaken to assess the scope of variation in the aromatic moiety (e.g. substituents, heteroatoms) and determine the factors that govern the rate and selectivity of this catalytic transformation.
R R

(1.4)
R

Cp*2LnCH(SiMe3)2 + R Ln = La, Ce, Y R = Me, Et, nPr, tBu, Ph, SiMe3 R + higher oligomers

In Chapter 2 the synthesis, structure and reactivity of lanthanidocene aryl-substituted propargyl/allenyl complexes Cp*2LnCH2CCAr (Ln = Y, La; Ar = C6H5, C6H4Me-2, C6H3Me2-2,6, C6H3iPr2-2,6) is described. Lanthanide propargyl/allenyls are presumed to be the active catalysts in the cyclodimerization reaction of 1-methylalk-2-ynes. Evidence for 3-bonding in these species was indicated by several spectroscopic techniques and this bonding description is also reflected by the observed reactivity, as these derivatives were found to be thermally robust, furnish acetylenic and allenylic quenching products upon reaction with protic acids and do not insert 1-alkenes into the metal-carbon bond. Sterically hindered 1-methylalk-2-ynes CH3CCAr undergo transmetalation with Cp*2LnCH2CCAr, while sterically less hindered 1-methylalk-2-ynes undergo catalytic cyclodimerization with Cp*2LaCH2CCAr. The reactivity of the Cp*2LnCH2CCAr complexes towards hard Lewis bases such as THF and pyridine has also been studied. The cyclodimerization of 1-methylalk-2-ynes CH3CCAr catalyzed by Cp*2LaCH2CCAr is treated in Chapter 3. Factors governing the rate and selectivity, such as substrate (i.e. ortho-methyl substitution) and catalyst structure (i.e. ancillary ligation and metal ionic radius), have been investigated. A plausible mechanistic

Chapter 1 scenario is proposed to account for the observed products. Finally, the cyclodimerization reaction was investigated as a preprative route towards (cross-)conjugated polymers of 1,4-di(prop-1-ynyl)benzene and 2,5-nhexyl-1,4-di(prop-1-ynyl)benzene. The rare-earth metallocene-catalyzed oligomerization of phenylacetylene is discussed in Chapter 4. The effect of metal ion size, ancillary ligation and substrate concentration on the rate and selectivity has been investigated. Relatively high substrate-to-catalyst molar ratios were found to promote catalyst deactivation (via Cp* abstraction) and catalytic trimerization. Stoichiometric reactions established the identity of reaction intermediates observed during catalytic substrate conversion. A kinetic study in conjuction with in situ 1H NMR experiments provided evidence for a rapid pre-equilibrium of a monomeric alkynyl derivative with its base adduct of phenylacetylene. A plausible mechanistic scenario has been proposed to account for the observed reaction products and the observed kinetic behaviour. The scope of the rare-earth metallocene-catalyzed oligomerization of aromatic 1-alkynes has been investigated in Chapter 5 on the basis of detailed studies of six representative substituted (hetero)aromatic 1alkynes. The observed behavior in both catalytic and stoichiometric reactions has been rationalized in terms of substrate/product inhibition, electronic/steric 1-alkyne substituent effects, heteroatom-assisted carbon-carbon bond formation and catalyst deactivation. The relative importance of these effects was found to depend on the specific properties of the 1-alkyne. Appropriate 1-alkyne substitution was found to result in high catalytic rates and selectivities for (E)-but-1-en-3-yne formation. The nature of the 1-alkyne substituent effect in the dimerization of monomeric alkynyl derivatives Cp*2LaCCR to dinuclear butatrienediyl derivatives [(Cp*2La)2(n:n-RC4R)] was also addressed. The oligomerization of (hetero)aromatic diynes catalyzed by Cp*2LaCH(SiMe3)2 is discussed in Chapter 6. The effect of the (hetero)arene moiety, with and without solubilizing aliphatic substituents, on the regioregularity of the formed polymers has been investigated. The use of mono-ethynyl substrates to control the molecular weight of the polymers is discussed. The formed conjugated polymers and their physical properties have been studied by a variety of spectroscopic methods.

1.5.
1 2

References and notes

The symbol Ln, not assigned to any particular element, is commonly used to designate the lanthanides as a class. In this text, however, the symbol Ln is used to designate the rare earth metals. The name rare earth metals is deprecated by IUPAC, since these metals are neither rare in abundance, nor earths (which is an obsolete term for oxides). In fact, each is more common in the earth's crust than silver, gold or platinum, while cerium, yttrium, neodymium and lanthanum are more common than lead. Wilkinson, G.; Birmingham, J. M. J. Am. Chem. Soc. 1954, 76, 6210. For examples, see: (a) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1983, 105, 1401. (b) Ye, C.; Qian, C.; Yang, X. J. Organomet. Chem. 1991, 407, 329 and references therein. For examples, see: (a) Ballard, D. G. H.; Courtis, A.; Holton, J.; McMeeking, J.; Pearce, R. J. Chem. Soc., Chem. Commun. 1978, 994. (b) Ballard, D. G. H.; Courtis, A.; Holton, J.; McMeeking, J.; Pearce, R. J. Chem. Soc., Chem. Commun. 1976, 1060. (c) Jeske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091. The highest turnover frequency in the hydrogenation of 1-hexene was found for [Cp*2La(-H)]2 (120000 h-1), comparing favorably to the activities of well-known transition-metal homogeneous catalysts, such as RhCl(PPh3)3 (3000 h-1) and [(COD)Ir(PCy3)]PF6 (6400 h-1) under similar reaction conditions (25 C, 1 atm.), see: (a) Kobayashi, T; Sakakura, T.; Hayashi, T.; Yumura, M.; Tanaka, M. Chem.Lett. 1992, 1158. Even higher activities have been reported for silylene-bridged metallocene hydrides [{Me2Si(C5Me4)2}Ln(-H)], see: (b) Jeske, G.; Schock, L. E.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8103. For reviews, see: (a) Watson, P. L. In Selective Hydrocarbon Activation. Principles and Progress; Davies, J. A., Watson, P. L., Greenberg, A., Liebman, J. F., Eds.; VCH Publishers, Weinheim; 1990, Chapter 4, p. 78. (b) Rothwell, I. P. In Selective Hydrocarbon Activation. Principles and Progress; Davies, J. A., Watson, P. L., Greenberg, A., Liebman, J. F., Eds.; VCH Publishers, Weinheim; 1990, Chapter 3, p. 43. (c) Ref. 5b.

3 4

Introduction

9 10 11

12

13 14

15

16

17

18

19

For general reviews on organolanthanide chemistry, see: (a) Marks, T. J.; Ernst, R. D. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1982; Chapter 21. (b) Evans, W. J. Adv. Organomet. Chem. 1985, 24, 131. (c) Schaverien, C. J. Adv. Organomet. Chem. 1994, 36, 283. (d) Edelmann, F. T. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1995; Vol. 4, Chapter 2. (e) Schumann, H.; Meese-Marktscheffel, J. A.; Esser, L. Chem. Rev. 1995, 95, 865. (f) Anwander, R.; Herrmann, W. A. Top. Curr. Chem. 1996, 179, 1. (g) Edelmann, F. T. Top. Curr. Chem. 1996, 179, 247. (h) Edelmann, F. T. In Metallocenes; Togni, A., Halterman, Ronald L., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 1, Chapter 2. (i) Chirik, P. J.; Bercaw, J. E. In Metallocenes; Togni, A., Halterman, Ronald L., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 1, Chapter 3. (j) Edelmann, F. T.; Freckmann, D. M. M.; Schumann, H, Chem. Rev. 2002, 102, 1851. (k) Arndt, S.; Okuda, J. Chem. Rev. 2002, 102, 1953. (l) Shibasaki, M.; Yoshikawa, N. Chem. Rev. 2002, 102, 2187. (m) Inanaga, J.; Furuno, H.; Hayano, T. Chem. Rev. 2002, 102, 2211. Tsutsui, M.; Gysling, H. J. J. Am. Chem. Soc. 1969, 91, 3175. Hayes, R. G.; Thomas, J. L. J. Am. Chem. Soc. 1969, 91, 6880. (a) Hart, F. A.; Saran, M. S. Chem. Commun. 1968, 1614. (b) Hart, F. A.; Massey, A. G.; Saran, M. S. J. Organomet. Chem. 1970, 21, 147. (c) Cotton, S. A.; Hart, F. A.; Hursthouse, M. B.; Welch, A. J. J. Chem. Soc., Chem. Commun. 1972, 1225. For early reviews on organolanthanide chemistry, see: (a) Tsutsui, M.; Ely, N.; Dubois, R. Acc. Chem. Res. 1976, 9, 216. (b) Schumann, H, Angew. Chem. Int. Ed. Engl. 1984, 23, 474. (c) Evans, W. J. Adv. Organometallic Chem. 1985, 24, 131. (d) Watson, P. L.; Parshall Acc. Chem. Res 1985, 18, 51. (a) Tsutsui, M.; Ely, N. J. Am. Chem. Soc. 1974, 96, 4042. (b) Tsutsui, M.; Ely, N. J. Am. Chem. Soc. 1975, 97, 3551. (c) Ely, M.; Tsutsui, M. J. Am. Chem. Soc. 1975, 97, 1280. (a) Maginn, R. E.; Manastryrskyj, S.; Dubeck, M. J. Am. Chem. Soc. 1963, 85, 672. (b) Namy, J. L.; Girard, P.; Kagan, H. B. Caro, P. E. Nouv. J. Chim. 1981, 5, 479. (c) Evans, W. J.; Meadows, J. H.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1984, 106, 1291. For Cp*2LnCH(SiMe3)2 complexes, see: (a) Sc: St. Clair, M.; Santarsiero, B. D.; Bercaw, J. E. Organometallics 1989, 8, 17. (b) Y: den Haan, K. H.; de Boer, J. L.; Teuben, J. H.; Spek, A. L.; Kojic,-Prodic, B.; Hays, G. R.; Huis, R. Organometallics 1986, 5, 1726. (c) La, Sm, Nd, Lu: Ref. 41a. (d) Ce: Heeres, H. J.; Renkema, J.; Booij, M.; Meetsma, A.; Teuben, J. H. Organometallics 1988, 7, 2495. For [Cp*2Ln(-H)]2 complexes, see: (h) Sc: Ref. 40b. (i) Y: den Haan, K. H.; Wielstra, Y.; Teuben, J. H. Organometallics 1987, 6, 2053. (j) La, Nd, Lu: Ref. 41a. (k) Ce: Ref. 15d. (l) Sm: Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1983, 105, 1401. For [(C5H4SiMe3)2Ln(-CH3)]2 complexes, see: (a) Y, Er: Ref. 5a. (b) Lu: Voskoboynikov, A. Z.; Parshina, I. N.; Shetstakova,A. K.; Butin, K. P.; Beletskaya, I. P.; Kuzmina, L. G.; Howard, J. A. K. Organometallics 1997, 16, 4041. (c) Nd: Molander, G. A.; Dowdy, E. D.; Noll, B. C. Organometallics 1998, 17, 3734. For Me2SiCp2LnCH(SiMe3)2 complexes with Cp = 5-C5Me4, see: (a) Sc: Ref. 49b. (b) Y: Coughlin, E. B.; Henling, L. M.; Bercaw, J. E. Inorg. Chim. Acta 1996, 242, 205. (c) Nd, Sm, Lu: Ref. 6b. For bis(indenyl)ethane-derived lanthanocenes, see: (d) Gilbert, A. T.; Davis, B. L.; Emge, T. J.; Broene, R. D. Organometallics 1999, 18, 2125. For Me2Si(Cp)(CpR*)LnE(SiMe3)2 complexes with R* = (+)-neomenthyl, (-)-menthyl and E = N, CH, see: (d) Y, Nd, Sm, La, Lu: Giardello, M. A.; Conticello, V. P.; Brard, L.; Sabat, M.; Rheingold, A. L.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10212. For Me2Si(OHF)(CpR*) complexes with OHF = 5-octahydrofluorenyl and R* = ()-menthyl, see: (d) Y, Sm, Lu: Douglass, M. R.; Ogasawara, M.; Hong, S.; Metz, M. V.; Marks, T. J. Organometallics 2002, 21, 283. For Me2SiCp(tBuN)LnE(SiMe3)2 complexes, see: (a) Ln = Sc, E = CH: Ref. 41h. (b) Ln = Sm, Nd, Yb, Lu, E = N, CH: Tian, S.; Arredondo,V. M.; Stern, C. L.; Marks, T. J. Organometallics 1999, 18, 2568. For reviews of non-cyclopentadienyl organometallic complexes of lanthanide and group 3 metals, see: (a) Edelmann, F. T.; Freckmann, D. M. M.; Schumann, H. Chem. Rev. 2002, 102, 1851. (b) Piers, W. E.; Emslie, D. J. H. Coord. Chem. Rev. 2002, 233-234, 131. (c) Edelmann, F. T.; Angew. Chem., Int. Ed. Engl. 1995, 34, 2466.

Chapter 1

20

21 22

23

24 25

26 27 28

29

30

31

32

For reviews, see: (a) Kaminsky, W.; Sinn, H. Transition Metals and Organometallics as Catalysts for Olefin Polymerization; Springer-Verlag: Berlin, 1988. (b) Keii, T.; Soga, K. Catalytic Olefin Polymerization; Elsevier: Amsterdam, 1990. (c) Guram, S. A.; Jordan, R. F. In Comprehensive Organometallic Chemistry II, Elsevier, Oxford; 1995, Vol. 4.; p. 589. (d) Britzinger, H. H.; Fischer, D.; Mlhaupt, R.; Rieger, B.; Waymouth, R. M. Angew. Chem. Int. Ed. Engl. 1995, 34, 1143. (e) Britovsek, G. J. P.; Gibson, V. C.; Wass, D. F. Angew. Chem. Int. Ed. 1999, 38, 428. (f) Coates, G. W. Chem. Rev. 2000, 100, 1223. For example, typical ligand field effects are a few hundred cm-1 for rare-earth metal complexes, whereas d transition metal complexes normally display values of 5 000-30 000 cm-1.29h (a) Raymond, K. N.; Eigenbrot, C. W. Acc. Chem. Res. 1980, 13, 276. (b) Moeller, T. In Comprehensive Inorganic Chemistry, Trotman-Dickenson, A. F. et al. (Eds.); Pergamon Press: Oxford, 1973; Chapter 44. For a review, see: (a) Giesbrecht, G. R.; Gordon, J. C. Dalton Trans. 2004, 2387. For examples, see: (b) Aparna, K.; Ferguson, M.; Cavell, R. G. J. Am. Chem. Soc. 2000, 122, 726. (c) Cantat, T.; Jaroschik, F.; Nief, F.; Ricard, L.; Mzailles, N.; Le Floch, P. Chem Commun. 2005, 41, 5178. (d) Cantat, T.; Jaroschik, F.; Ricard, L.; Le Floch, P.; Nief, F.; Mzailles, N. Organometallics 2006, 25, 1329. (e) Dietrich, H. M.; Trnroos, K. W.; Anwander, R. J. Am. Chem. Soc. 2006, 128, 9298. Brady, E. D.; Clark, D. L.; Gordon, J. C.; Hay, P. J.; Keogh, D. W.; Poli, R.; Scott, B. L.; Watkin, J. G. Inorg. Chem. 2003, 42, 6682. Lanthanide cations are located between Sr(II) and Ti(IV), see: (a) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533. (b) Hard and Soft Acids and Bases; Pearson, R. G. (ed.); Dowden, Hutchinson and Ross, Stroudsberg, PA; 1973. (b) Fleming, I. Frontier Orbitals and Organic Chemical Reactions, Wiley-Interscience, London; 1976. Holton, J.; Lappert, M. F.; Ballard, D. G. H.; Pearce, R.; Atwood, J. L.; Hunter, W. E. J. Chem. Soc., Dalton Trans. 1979, 54. Nolan, S. P.; Stern, D.; Marks, T. J. J. Am. Chem. Soc. 1989, 111, 7844. Even the thermodynamically very stable Ln-OR bond undergoes rapid ligand exchange reactions, see: (a) Hubert-Pfalzgraf, L. G. New. J. Chem. 1987, 11, 163. (b) Hubert-Pfalzgraf, L. G. New J. Chem. 1995, 19. 727. For reviews, see: (a) Carnall, W. T. In Handbook on the Physics and Chemistry of Rare Earths, Gschneidner, Jr., K. A., Eyring, L. (Eds.); North Holland Publishing Company: Amsterdam, 1979; Vol. 3, Ch. 24. (b) G. Blasse In Handbook on the Physics and Chemistry of Rare Earths, Gschneidner, Jr., K. A., Eyring, L. (Eds.); North Holland Publishing Company: Amsterdam, 1979; Vol. 4, Ch. 34. (c) Shionoya, S.; Yen, W. M. In Phosphor Handbook; Shionoya, S., Yen, W. M. (Eds.); CRC Press, Inc.: Boca Raton, Florida, 1999; Chapter 3. (d) Parker, D.; Williams, J. A. G. In Metal Ions in Biological Systems, Sigel, A; Sigel, H. (Eds.); Marcel Dekker Inc.: New York, 2003;Vol. 40. (e) Dssing, A. Eur. J. Inorg Chem. 2005, 1425. (f) Bnzli, J.-C. G.; Piguet, C. Chem. Soc. Rev. 2005, 34, 1048-1077. (g) Faulkner, S.; Pope, S. J. A.; Burton-Pye, B. P. Appl. Spectrosc. Rev. 2005, 40, 1. (h) Bnzli, J.-C. Acc. Chem. Res. 2006, 39, 53. Solid state spectra have been obtained on paramagnetic compounds using nuclei with relatively low quadrupole moments. For general references, see: (a) Harris, R. K.; Mann, B. E. NMR and the Periodic Table, Academic Press, 1978. (b) Brevard, C.; Granger, P. Handbook of High Resolution Multinuclear NMR, Wiley, New York; 1981. (c) Fischer, R. D. In Fundamental and Technological Aspects of Organo-f-Element Chemistry; NATO ASI Series; Marks, T.J., Fragal, I. L., Eds.; D. Reidel, Boston; 1985. For examples, see: (a) Evans, W. J.; Meadows, J. H.; Kostka, A. G.; Closs, G. L. Organometallics 1985, 4, 324. (b) Schaverien, C. J.; Frijns, J. H. G.; Heeres, H. J.; van den Hende, J. R.; Teuben, J. H.; Spek, A. L. J. Chem. Soc., Chem. Commun. 1991, 642. (c) Evans, W. J.; Gonzales, S. L.; Ziller, J. W. J. Am. Chem. Soc. 1994, 116, 2600. For examples of 139La NMR spectroscopy, see: (a) Eggers, S. H.; Fischer, R. D. J. Organomet. Chem. 1986, 315, C61. (b) Eggers S. H.; Adam, M.; Haupt, E. T. K.; Fischer, R. D. Inorg. Chim. Acta 1987, 139, 315. (c) Adam, M.; Haupt, E. T. K.; Fischer, R. D Bull. Magn. Res. 1990, 12, 101. (d) Windisch, H.; Scholz, J.; Taube, R.; Wrackmeyer, B. J. Organomet. Chem. 1996, 520, 23. For examples of 45Sc NMR spectroscopy, see: (d) Mancini, M. Inorg. Chem. 1984, 23, 1072. (e) Pougeard, P.; Mancini, M.; Sayer, B. G.; McGlinchey, M. J. Inorg. Chem. 1985, 24, 93.

Introduction

33

34

35 36 37

38 39 40

41

42

43

44

For examples, see: (a) Avent, A. G.; Edelman, M. A.; Lappert, M. F.; Lawless, G. A. J. Am. Chem. Soc. 1989, 111, 3423. (b) Recknagel, A.; Edelman, F. T. Angew. Chem., Int. Ed. Engl. 1991, 30, 693. (c) van den Hende, J. R.; Hitchcock, P. B.; Lappert, M. F.; Nile, T. A. J. Organomet. Chem. 1994, 472, 79. For examples, see: (a) Shen, Z.; Farona, M. F. Polym. Bulletin 1983, 10, 298. (b) Shen, Z; Farona, M. F. J. Polym. Sci.: Polym. Chem. Ed. 1984, 22, 1009. (c) Zhiquan, S.; Mujie, Y.; Mingxiao, S.; Yiping, C. J. Polym. Sci.: Polym. Lett. Ed. 1982, 20, 417. For reviews, see: (a) Ref. 12a. (b) Bruzzone, M. In Fundamental and Technological Aspects of Organo-f-Element Chemistry; NATO ASI Series; Marks, T.J., Fragal, I. L., Eds.; D. Reidel, Boston; 1985. (a) Evans, W. J.; Coleson, K. M.; Engerer, S. C. Inorg. Chem. 1981, 20, 4320. (b) Evans, W. J.; Engerer, S. C.; Coleson, K. M. J. Am. Chem. Soc. 1981, 103, 6672. (a) Ballard, D. G. H.; Courtis, A.; Holton, J.; McMeeking, J.; Pearce, R. J. Chem. Soc., Chem. Commun. 1978, 994. (b) Ballard, D. G. H.; Pearce, R. J. Chem. Soc., Chem. Commun. 1975, 621. For examples, see: (a) Chigir, N. N.; Guzman, I. Sh.; Sharaev, O. K.; Tiniakova, E. I.; Dolgoplosk, B. A. Dokl. Akad. Nauk USSR, 1981, 263, 375. For reviews, see: (b) Mazzei, A. In Organometallics of the f-Elements; Marks, T. J, Fischer, R. D. (Eds.); D. Reidel Publishers, Boston; 1979. (a) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1983, 105, 1401. (b) Evans, W. J. J. Organomet. Chem. 1983, 250, 217. Zinnen, H. A.; Pluth, J. J.; Evans, W. J. J. Chem. Soc., Chem. Commun. 1980, 810. For examples, see: (a) Watson, P. L. J. Am. Chem. Soc. 1982, 104, 337. (b) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1987, 109, 203. (c) Burger, B. J.; Thompson, M. E.; Cotter, W. D.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112, 1566. For reviews, see: (d) Burger, B. J.; Cotter, W. D.; Coughlin, E. B.; Chacon, S. T.; Hajela, S. H.; Herzog, T. A.; Kohn, R.; Mitchell, J.; Piers, W. E.; Shapiro, P. J.; Bercaw, J. E. In Ziegler Catalysts; Fink, G.; Mlhaupt R.; Brintzinger, H. H., Eds.; Springer Verlag: Berlin, 1995. (e) Gromada, J.; Carpentier, J.-F.; Motreux, S. Coord. Chem. Rev. 2004, 248, 397. For examples, see: (a) Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091. (b) Jeske, G.; Laurel, E.; Schock, P.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8103. (c) Marks, T. J.; Mauermann, H. WO Patent 8605788 (Northwestern University, USA), 1986. (d) Wiesenfeldt, H.; Reinmuth, A.; Barsties, E.; Evertz, K.; Brintzinger, H.-H. J. Organomet. Chem. 1989, 369, 359. (e) Roll, W.; Brintzinger, H.H.; Rieger, B.; Zolk, R. Angew. Chem. Int. Ed. Engl. 1990, 29, 279. (f) Shapiro, P. J.; Bunel, E. E.; Schaefer, W. P.; Bercaw, J. E. Organometallics 1990, 9, 867. (g) Stern, D.; Sabat, M.; Marks, T. J. J. Am. Chem. Soc. 1990, 112, 9558. (h) Shapiro, P. J.; Cotter, W. D.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1994, 116, 4623. (i) Hajela, S.; Bercaw, J. E. Organometallics 1994, 13, 1147. (j) Mitchell, J. P.; Hajela, S.; Brookhart, K.; Hardcastle, K. I.; Henling, H. M.; Bercaw, J. E. J. Am. Chem. Soc. 1996, 118, 1045. (k) Gilchrist, J. H.; Bercaw, J. E. J. Am. Chem. Soc. 1996, 118, 12021. (l) Ihara, E.; Nodono, M.; Katsura, K.; Adachi, Y.; Yasuda, M.; Yamagashira, H.; Hashimoto, H.; Kanehisa, N.; Kai, Y. Organometallics 1998, 17, 3945. (m) Ihara, E.; Yoshioka, S.; Furo, M.; Katsura, K.; Yasuda, M.; Mohri, S.; Kanehisa, M.; Kai, Y. Organometallics 2001, 20, 1752. (a) Chen, W.; Xiao, S.; Wang, Y.; Yu, G. Kexue Tongbao 1984, 29, 892. (b) Qian, H.; Yu, G.; Chen, W. Gaofenzi Tongxun 1984, 3, 226. (c) Yu, G.; Chen. W.; Wang, Y. Kexue Tongbao 1984, 29, 421. (d) Taube, R.; Windisch, H. J. Organomet. Chem. 1994, 472, 71. (e) Barbier-Baudry, D.; Andre, N.; Dormond, A.; Pardes, C.; Richard, P.; Visseaux, M.; Zhu, C. J. Eur. J. Inorg. Chem. 1998, 1721. (e) Bonnet, F.; Barbier-Baudry, D.; Dormond, A.; Visseaux, M. Polymer Int. 2002, 51, 986. (f) BarbierBaudry, D.; Bonnet, F.; Domenichini, B.; Dormond, A.; Visseaux, M. J. Organometl. Chem. 2002, 647, 167. (a) Bogaert, S.; Carpentier, J.-F.; Chenal, T.; Mortreux, A.; Ricart, G.; Macromol. Chem. Phys. 2000, 201, 1813. (b) Kirillov, E.; Lehmann, C. W.; Razavi, A.; Carpentier, J.-F. J. Am. Chem. Soc. 2004, 126, 12240. (a) Yasuda, H.; Furo, M.; Yamaoto, H. Macromolecules 1992, 25, 5115. (b) Jiang, T.; Shen, Q.; Lin, Y.; Jns, S. J. Organomet. Chem. 1993, 450, 121. (c) Yasuda, H.; Yamamoto, H.; Yamashita, M.; Yokota, K.; Nakamura, A, Miyake, S.; Kai, Y. Kanchisa, N. Macromolecules 1993, 26, 7134. (d) Yasuda, H.; Yamamoto, H.; Yokota, K.; Miyake, S.; Nakamura, A. J. Am. Chem. Soc. 1992, 114, 4908. (e) Simpson, C. K.; White, R. E.; Carlson, C. N.; Wrobleski, D. A.; Kuehl, C. J.; Croce, T. A.;

Chapter 1

45 46 47

48

49

50

51 52 53

54 55 56

57 58 59 60 61

62 63 64

Steele, I. M.; Scott, B. L.; Young Jr., V. G.; Hanusa, T. P.; Sattelberger, A. P.; John, K. D. Organometallics 2005, 24, 3685. (a) Yasuda, H. Chem Abstr. 1994, 120, 135475a. (b) Evans, W. J.; Katsumata, H. Macromolecules 1994, 27, 2330. (c) Onozawa, S.; Sakakura, T.; Tanaka, M. Chem. Lett. 1994, 531. For reviews, see: (a) Yasuda, H. Prog. Polym. Sci. 2000, 25, 573. (b) Hou, Z.; Wakatsuki, Y. Coord. Chem. Rev. 2002, 231, 1. (c) Yasuda, H. J. J. Organomet. Chem. 2002, 647, 128. For reviews, see: (a) Molander, G. A. Chemtracts: Org.Chem. 1998, 18, 237. (b) Molander, G. A.; Dowdy, E. D. Top. Organomet. Chem. 2000, 2, 119. (c) Molander, G. A.; Romero, J. A. C. Chem. Rev. 2002, 102, 2161. (a) Molander, G. A.; Hoberg, J. O. J. Am. Chem. Soc. 1992, 114, 3123. (b) Molander, G. A.; Hoberg, J. O. J. Org. Chem. 1992, 57, 3266. (c) Giardello, M. A.; Conticello, V. P.; Brard, L.; Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10241. (d) Haar, C. M.; Stern, C. L.; Marks, T. J. Organometalllics 1996, 15, 1765. (e) Roesky, P. W.; Denninger, U.; Stern, C. L.; Marks, T. J. Organometallics. 1997, 16, 4486. (f) Obora, Y.; Ohta, T.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1997, 119, 3745. (a) Watson, P. L. J. Am. Chem. Soc. 1982, 104, 337. (b) Piers, W. E.; Shapiro, P. J.; Bunel, E. E. Synlett 1990, 74. (c) Kretschmer, W. P.; Troyanov, S. I.; Meetsma, A.; Hessen, B.; Teuben, J. H. Organometallics 1998, 17, 284. (a) Olonde, X.; Mortreux, A.; Petit, F.; Bujadoux, K. J. Mol. Catal.1993, 82, 75. (b) Mauermann, J.; Swepston, P. N.; Marks, T. J. J. Am. Chem. Soc. 1985, 4, 200. (c) Jeske, G. ; Lauke, H. ; Mauermann, H. ; Swepston, P. N. ; Schumann, H. ; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091. (d) Watson, P. L. J. Chem. Soc., Chem. Commun. 1983, 276. (e) Watson, P. L. J. Am. Chem. Soc. 1983, 105, 6491. (f) Evans, W. J. ; Bloom, I. ; Hunter, W. E. ; Atwood, J. L. J. Am. Chem. Soc. 1983, 105, 1401. (a) Molander, G. A.; Hoberg, J. O. J. Am. Chem. Soc. 1992, 14, 3123. (b) Piers, W. E.; Shapiro, P. J.; Bunel, E. E. Synlett 1990, 74. For a review of the organolanthanide-catalyzed hydroamination reaction, see: Hong, S.; Marks, T. J. Acc. Chem. Res. 2004, 37, 673. (a) Sakakura, T.; Lautenschlager, H.; Tanaka, M. J. Chem. Soc., Chem. Commun. 1991, 40. (b) Molander, G. A.; Julius, M. J. Org. Chem. 1992, 57, 6347. (c) Fu, P.-F.; Brard, L.; Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 7157. (d) Molander, G. A.; Corrette, C. P. Organometallics 1998, 17, 5504. (e) Schumann, H.; Keitsch, M. R.; Winterfeld, J.; Mhle, S.; Molander, G. A. J. Organomet. Chem. 1998, 559, 181. (a) Douglass, M. R.; Marks, T. J. J. Am. Chem. Soc. 2000, 122, 1824. (b) Douglass, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 2001, 123, 10221. (a) Harrison, K. N. ; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 9220. (b) Bijpost, E.; Duchateau, R.; Teuben, J. H. J. Nol. Catal. 1995, 95, 121. (c) Molander, G. A.; Pfeiffer, D. Org. Lett. 2001, 3, 361. (a) Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1989, 111, 4108. (b) Gagn, M. R.; Nolan, S. P.; Marks, T. J. Organometallics 1990, 9, 1716 (c) Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (d) Giardello, M. A.; Conticello, V. P.; Brard, L. ; Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10241. (a) Li, Y.; Fu, P.-F.; Marks, T. J. Organometallics 1994, 13, 439. (b) Li, Y.; Marks, T. J. Organometallics 1996, 118, 9295. (a) Arredondo, V. M.; McDonald, F. E. Marks, T. J. J. Am. Chem. Soc. 1998, 120, 4871. (b) Arredondo, V. M.; McDonald, F. E.; Marks, T. J. Organometallics 1999, 18, 1949. (a) Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1996, 118, 707. (b) Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 1757. (c) Hong, S.; Kawaoka, A. M.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 15878. Shapiro, P. J.; Bunel, E.; Schaefer, W. P.; Bercaw, J. E. Organometallics 1990, 9, 867. (a) Gagn, M. R.; Brard, L.; Conticello, V. P.; Giardello, M. A.; Stern, C. L.; Marks, T. J. Organometallics 1992, 11, 2003. (b) Coughlin, E. B.; Bercaw, J. E. J. Am. Chem. Soc. 1992, 114, 7606. (c) Mitchell, J. P.; Hajela, S.; Brookhart, S. K.; Hardcastle, K. I. Henling, L. M. Bercaw, J. E. J. Am. Chem. Soc. 1996, 118, 1045. Piers, W. E.; Shapiro, P. J.; Bunel, E. E. Synlett 1990, 74. Bunel, E. E.; Burger, B. J.; Bercaw, J. E. J. Am. Chem. Soc. 1988, 110, 976. Fontaine, F.-G.; Tilley, T. D. Organometallics 2005, 24, 4340.

10

Introduction

65

66 67 68

69

70

71 72

73 74

75 76

(a) Piers, W. E.; Hayes, P. G.; McDonald, R. J. Am. Chem. Soc. 2002, 124, 2132. (b) Hajela, S.; Schaefer, W. P.; Bercaw, J. E. J. Organomet. Chem. 1997, 532, 45. (c) Long, D. B.; Bianchoni, P. A. J. Am. Chem. Soc. 1996, 118, 12453. (d) Bambirra, S.; van Leusen, D.; Meetsma, A.; Hessen, B.; Teuben, J. H. Chem. Commun. 2001, 637. (e) Trifonov, A. A.; Federova, E. A.; Fukin, G. K.; Bochkarev, M. N. Eur. J. Inorg. Chem. 2004, 4396. Giesbrecht, G. R.; Whitener, G. D.; Arnold, J. J. Chem. Soc., Dalton Trans. 2001, 923. Gountchev, T. I.; Tilley, T. D. Organometallics 1999, 18, 5661. (a) Hultsch, K. C.; Hampel, F.; Wagner, Th. Organometallics 2004, 23, 2601. (b) Brgstein, M. R.; Berberich, H.; Roesky, P. W. Organometallics 1998, 17, 1452. (c) Kim, Y. K.; Livingstone, T.; Bercaw, J. E. Tetrahedron 2001, 42, 2933. (d) Brgstein, M. R.; Berberich, H.; Roesky, P. W.; Chem. Eur. J. 2001, 7, 3078. (e) OShaughnessy, P. N.; Knight, P. D.; Morton, C.; Gillespie, K. M.; Sott, P. Chem. Commun. 2003, 2330. (f) Gribkov, D. V.; Hultsch, K. C.; Hampel, F. Chem. Eur. J. 2003, 9, 4796. (g) Kim, Y. K.; Livinghouse, T.; Horino, Y. J. Am. Chem. Soc. 2003, 125, 9560. (h) Hong, S.; Tian, S.; Metz, M. V.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 14768. (i) Collin, J.; Daran, J.-D.; Schulz, E.; Trifonov, A. Chem. Commun. 2003, 3048. (j) O Shaugnessy, P. N.; Scott, P. Tetrahedron Asymmetry 2003, 14, 1979. (k) Gribkov, D. V.; Hultsch, K. C. Chem. Commun. 2004, 730. For a recent review of cationic organo rare-earth metal complexes, see: (a) Zeimentz, P. M.; Arndt, S.; Elvidge, B. R.; Okuda. J. Chem. Rev. 2006, 106, 2404. (b) Hou, Z.; Luo, Y.; Li, X. J. Organomet. Chem. 2006, 691, 3114. (c) Arndt, S.; Okuda, J. Adv. Synth. Catal. 2005, 347, 339. (a) Bambirra, S.; van Leusen, D.; Meetsma, A.; Hessen, B.; Teuben, J. H. Chem. Commun. 2003, 522. (b) Bambirra, S.; Bouwkamp, M. W.; Meetsma, A.; Hessen, B. J. Am. Chem. Soc. 2004, 126, 9182. (c) Hayes, P. G.; Piers, W. E.; McDonald, R. J. Am. Chem. Soc. 2002, 124, 2132. (d) Bambirra, S.; van Leusen, D.; Meetsma, A.; Hessen, B.; Teuben, J. H. Chem. Commun. 2001, 637. (e) Bambirra, S.; Boot, S. J.; van Leusen, D.; Meetsma, A.; Hessen, B. Organometallics 2004, 23, 1891. (f) Hessen, B.; Bambirra, S. WO2002032909. (g) Christopher, J. N.; Squire, L. R.; Canich, J. A. M.; Schaffer, T. D. WO0018808. (h) Arndt, S.; Spaniol, T. P.; Okuda, J. Angew. Chem., Int. Ed. 2003, 42, 5075. (i) Okuda, J.; Arndt, S.; Matsui, S. JP2005015579. (j) Ward, B. D.; Bellemin-Laponnaz, S.; Gade, L. H. Angew. Chem., Int. Ed. 2005, 44, 1668. (k) Lawrence, S. C.; Ward, B. D.; Dubberley, S. R.; Kozak, C. M.; Mountford, P. Chem. Commun. 2003, 2880. (l) Hajela, S.; Schaefer, W. P.; Bercaw, J. E. J. Organomet. Chem. 1997, 532, 45. (m) Tredget, C. S.; Bonnet, F.; Cowley, A. R.; Mountford, P. Chem. Commun. 2005, 3301. (n) Ward, B. D.; Bellemin-Laponnaz, S.; Gade, L. H. Angew. Chem. Int. Ed. 2005, 44, 1668. Luo, Y.; Baldamus, J.; Hou, Z. J. Am. Chem. Soc. 2004, 126, 13910. (a) Kaita, S.; Hou, Z. M.; Nishiura, M.; Doi, Y.; Kurazumi, J.; Horiuchi, A. C.; Wakatsuki, Y. Macromol. Rapid Commun. 2003, 24, 179. (b) Taube, R.; Maiwald, S.; Sieler, J. J. Organomet. Chem. 2001, 621, 327. (c) Kaita, S.; Hou, Z.; Wakatsuki, Y. US2002/0119889A1. (d) Zhang, L.; Luo, Y.; Hou, Z. J. Am. Chem. Soc. 2005, 127, 14562. Lauterwasser, F.; Hayes, P. G.; Brse, S.; Piers, W. E.; Schafer, L. L. Organometallics 2004, 23, 2234. (a) Tazelaar, C. G. J.; Bambirra, S.; van Leusen, D.; Meetsma, A.; Hessen, B.; Teuben, J. H. Organometallics 2004, 23, 1891. (b) Nishiura, M.; Hou, Z.; Wakatsuki, Y.; Yamaki, T.; Miyamoto, T. J. Am. Chem. Soc. 2003, 125, 1184. Heeres, H. J.; Heeres, A.; Teuben, J. H. Organometallics 1990, 10, 1508 (a) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1987, 109, 203. (b) den Haan, K. H.; Wielstra, Y.; Teuben, J. H. Organometallics 1987, 6, 2053. (c) Heeres, H. J.; Teuben J. H. Organometallics 1991, 11, 1980.

11

Rare-earth metallocene propargyl/allenyls

2.

Rare-earth metallocene propargyl/allenyls

2.1.

Introduction

Organo rare-earth metal chemistry has witnessed a tremendous growth during the last two decades.1 This spectacular growth in research activities can be attributed to that the fact that many of these compounds exhibit not only unique structural and physical properties, but are highly active catalysts in a variety of catalytic processes as well. Among the different classes of organo rare-earth metal complexes, the hydride and alkyl derivatives are the most studied, as they often display activities dramatically higher than that of comparable transition metal complexes. Other classes, such as alkynyl, aryl and allyl derivatives, have been investigated less extensively, but their preparation, molecular structure and properties are well-documented. Scheme 2-1. Structural isomerism in allenyl/propargyl metal complexes.

R R M . H R

M R H R

R H

II

III

In contrast to allenyl and propargyl derivatives of transition metals, alkali metals and alkaline earth metals, which play an important role in organic synthesis,2 relatively little is known about the corresponding rare-earth metal congeners. 1-Allenyl (I) and -propargyl (III) metal complexes are known to exhibit interesting structural features and an unusual reactivity (Scheme 2-1).3 An aspect that has received much recent attention is their tautomeric behavior. Metal 1-propargyls have been shown to undergo both reversible and irreversible tautomerization into metal 1-allenyls. In addition, well-characterized delocalized -bonded 3-allenyl/propargyl complexes (II) are known as well.4 To the best of our knowledge, the first report of rare-earth metal allenyl and propargyl derivatives is the reaction of Cp*2LuMe with allene and 2-butyne in the early 1980s.5 A facile 1,3-metal shift of the propargyl to yield the observed allenyl was put forward in a review article (Scheme 2-2), but no experimental data were given in the references. In 1990 Heeres et al. reported a series of spectroscopically characterized alkyl- and silylsubstituted lanthanidocene propargyls Cp*2LnCH2CCR (Ln = La, Ce).6 The reactivity of these derivatives remained largely unexplored and the exact nature of the bonding was unknown, as structural data were lacking.

Scheme 2-2. Early examples of propargyl and allenyl lanthanides.

Lu CH3

Lu

CH4

Lu CH3

Lu

Lu

13

Chapter 2

Scheme 2-3. Other examples of propargyl and allenyl lanthanides.

H N N Y Me Me N . N Y Me Me N N Me3Si Si Sm I I Li SiMe3 N N N B N Yb Me3Si SiMe3 N

N N N

SiPh3

A'

Since then, only three studies involving well-defined allenyl/propargyl lanthanide species have appeared in literature, including the spectroscopically characterized yttrium 1-allenyl species A, which is reported to be in a rapid equilibrium with its 1-propargyl isomer A,7 the structurally characterized 3-propargyl samarium half-sandwhich complexes (e.g. B), incorporating a chelating allenyl/propargyl ligand system,8 and a divalent ytterbium 3-propargyl (C) which has also been structurally characterized (Scheme 2-3).9 The only other reports regarding propargyl and/or allenyl rare-earth metal complexes are reactive intermediates in synthetic organic reactions, prepared in situ via transmetalation reactions of organopalladium intermediates10 or Grignard reagents.11 Scheme 2-4. The catalytic cyclodimerization of 1-methylalk-1-ynes.
R Cp*2LnCH(SiMe3)2 R R Ln = La, Ce R = Me, Et,nPr + R R

Lanthanide propargyl derivatives Cp*2LnCH2CCR have been proposed to be the active catalyst in the catalytic cyclodimerization of alkyl- and silyl-substituted 1-methylalk-2-ynes CH3CCR in the presence of Cp*2LnCH(SiMe3)2 (Ln = La, Ce) (Scheme 2-4).6 Motivated by the possibility to apply this reaction to bifunctional aromatic substrates and thereby prepare a novel class of polymers (Chapter 3), a study of the synthesis, structure and reactivity of rare-earth metallocene aryl-substituted propargyls was initiated. The investigations described in this chapter represent the first systematic study of allenyl/propargyl derivatives of group 3 and lanthanide metals and include the first crystallographic structure determination of these derivatives having the common trivalent oxidation state and non-chelating allenyl/propargyl ligands.

2.2.

Synthesis of substrates

Introduction In order to explore the scope of propargylic C-H activation of 1-methylalk-1-ynes by rare-earth metallocene derivatives, a series of 1-aryl-1-propynes was prepared. Ortho-substituted phenyl derivatives were prepared to study steric effects and a pentafluorophenyl derivative was prepared to address electronic effects. 1-(2-methylphenyl)-1-propyne (2) The synthesis of 2-propynyltoluene (2) was straightforward and accomplished by palladiumcatalyzed cross-coupling of the corresponding iodoarene and propynylzinc bromide (the Negishi reaction, Scheme 2-5).12 The zinc reagent was prepared in situ from commercially available propynylmagnesium bromide, propyne gas and anhydrous zinc bromide.

14

Rare-earth metallocene propargyl/allenyls

Scheme 2-5. The synthesis of propynylarenes 2 and 3.

(ii) 85% 2

NH2

(i) 87%

(ii) 78% 3

Reaction conditions: (i): CH3CCMgBr (1.2 equiv), ZnBr2 (1.2 equiv), Pd(PPh3)4 (5 mol%), THF, 50 C. (ii): (1) NaNO2, HCl; (2) KI. 1-(2,6-dimethylphenyl)-1-propyne (3) The preparation of 2,6-dimethylpropynylbenzene (3) was achieved analogously to 2-propynyltoluene, but started with the aqueous iododediazotization of 2,6-dimethylaniline, according to a published procedure.13 When 2,6-dimethyliodobenzene was cross-coupled via the Negishi protocol, the corresponding 1-aryl-1-propyne was obtained in high yield and purity. 1-(2,6-diisopropylphenyl)-1-propyne (4) Unfortunately, the application of a similar strategy in the preparation of 1-(2,6-diisopropylphenyl)-1propyne (4) was not successful. When 2,6-diisopropylaniline was subjected to the standard iododediazotization methodology, a mixture was obtained which contained 2,6-diisopropylchlorobenzene (41%, GC), 2,6diisopropylphenol (7%), 2,6-diisopropyliodobenzene (21%) and an unknown compound (m/z 352, 11%), according to GC-MS analysis.14 It was believed that the low solubility of 2,6-diisopropylaniline in aqueous solution allowed for the occurrence of competing reactions.15 The crude yield of the iodo compound (47%, GC) was increased by the addition of ethanol and the use of aqueous hydrogen iodide solution. Surprisingly, the use of copper(I) iodide or copper bronze was also found to increase the crude yield of the desired iodo compound (60-65%, GC).16 Subsequent column chromatography (silica, hexanes) and vacuum distillation (120-140 C, 5 mmHg) furnished samples of desired aryliodide in reasonable purity (95%, GC). Nonetheless, samples of the corresponding propynylarene (97% pure, GC), prepared via cross-coupling with propynylcopper (vide infra), were unsuitable for the present reactivity studies involving organo rare-earth metallocenes, presumably due to inseparable, unidentified impurities of the iododiazotization reaction. It should be noted that the synthesis of 2,6-diisopropyliodobenzene (by aqueous iododiazotization of 2,6-diisopropylaniline) has been reported after preparation of this manuscript. In all of these cases, however, experimental details are lacking17 and/or the isolated yields are quite low (35%).18 The synthesis of 2,6diisopropylbromobenzene is also documented in literature (albeit in low yields and purity either via the Scheme 2-6. The synthesis of 2,6-diisopropyliodobenzene.
NH2 (i) N2BF4 (ii) 97% I

NH2

(iii) 78%

Reaction conditions: (i): HBF4, n-BuONO, 0 C, EtOH/H2O; (ii): KI, I2, copper bronze, DMSO, 50 C; (iii) n-BuONO, KI, I2, copper bronze, DMSO, 50-60 C.

15

Chapter 2

Scheme 2-7. The synthesis of 1-(2,6-diisopropylbenzene)-1-propyne 4.

CuSO4 + CH3C CH (i) CuC CCH3

I + CuC CCH3

(ii) 76% 4

Reaction conditions: (i): NH2OHHCl, NH3, RT (aq); (ii): pyridine, 110-120 C. Sandmeyer reaction19 or via aprotic diazotization20), but was not considered in this study, as iodides are generally more reactive than bromides and triflates in transition metal-catalyzed cross-coupling reactions with organometallics.21 In view of the low solubility of 2,6-diisopropylaniline in aqueous hydrogen halide solutions, continued efforts to prepare 2,6-diisopropyliodobenzene were directed along the lines of aprotic diazotization.22 Several reports involving the in situ aprotic diazotization of 2,6-diisopropylanilines (followed by chlorination23a or cyanation23b) and the preparation of substituted iodobenzenes24 (by in situ aprotic diazotization of the corresponding anilines in the presence of iodine) suggested that in situ aprotic iododiazotization of 2,6diisopropylaniline should, in principle, be feasible, but initial attempts based on these protocols were unsuccessful.25 Based on experiments with 2,6-diisopropylbenzenediazonium tetrafluoroborate,26 it was possible to develop a convenient procedure in which 2,6-diisopropylaniline could directly be transformed into the corresponding iodide by treatment with n-butylnitrite in the presence of iodine, potassium iodide and copper bronze under an inert atmosphere and using dry DMSO as a solvent (Scheme 2-6). The only by-product was 1,3diisopropylbenzene (<1%, GC/GC-MS) which could completely be separated by column chromatography (silica, hexanes) or vacuum distillation (120-140 C, 5 mmHg). In contrast to 2,6-dimethyliodobenzene, its diisopropyl analogue did not react with propynylmagnesium bromide and zinc bromide in the presence of Pd(PPh3)4, not even after a prolonged reaction time or after increasing the reaction temperature and catalyst loading.27 This lack of reactivity can plausibly be ascribed to the steric hindrance of the two ortho-isopropyl groups.28 Other attempts involving the cross-coupling of 2,6-diisopropyliodobenzene with a propynylboronic ester29 (the Suzuki-Miyaura reaction30) and the use of 2,6diisopropylphenyl triflate were equally disappointing.31 An early report in which the substitution of 2,4,6-triisopropyliodobenzene was achieved by a copper(I) acetylide32, albeit in low yield, prompted us to explore this reaction (the Castro reaction) in the present context.33 Motivated by the well-recognized highly explosive nature of copper(I) acetylides,34 attempts were undertaken to develop a convenient one-pot procedure.35 These attempts failed, however, and the Castro reaction had to be performed with isolated propynylcopper (prepared in ammonia from propyne gas, according to a modified published procedure36). The potentially explosive copper compound was kept moist with ether after isolation and added directly in a large excess to a solution of the 2,6-diisopropyliodobenzene in dry pyridine. Heating the suspension at reflux under an inert nitrogen atmosphere for several days afforded the desired propynylarene (4) in good yield (Scheme 2-7). The only contaminants in the crude product mixture were 1,3diisopropylbenzene (4%, GC/GC-MS) and 2,6-diisopropyliodobenzene (5%) which could be removed to a large extent after repeated column chromatography (silica, hexanes).37 The final sample of 4 contained small amounts of 1,3-diisopropylbenzene (<2%, 1H NMR). As this compound was considered to be relatively inert towards the rare-earth metallocene derivatives in subsequent reactivity studies, no further attempts were undertaken to obtain a sample of higher purity. 1-Pentafluorophenyl-1-propyne (5) In order to investigate electronic effects in the propynylaromatic substrates as well, it was decided to prepare pentafluoro-propynylbenzene (5). According to a modification of a general procedure, pentafluorophenylpropyne (5) was prepared in a low yield (24%) from the reaction of propynyllithium with excess hexafluorobenzene in THF at low temperature (Scheme 2-8).38 Propynyllithium was generated in situ from propyne gas and n-butyllithium in THF at -70 C. The product mixture was evaporated to dryness after

16

Rare-earth metallocene propargyl/allenyls

Scheme 2-8. The synthesis of propynylarene 5.

F F F F F F (i) + LiC CCH3 24% F 5 F F F F

Reaction conditions: (i): THF, -60 C, 2 h.

which the product was conveniently isolated by sublimation. The major by-product of the reaction was identified as 1,4-dipropynylhexa-fluorobenzene.

2.3.
2.3.1.

Synthesis of propargyl/allenyls
Introduction

As synthetic routes in organo rare-earth metal chemistry starting from halide derivatives are known to be plagued by the undesired incorporation of alkali metal salts and solvents, the most straightforward synthetic strategy towards Cp*2LnCH2CCR complexes via salt metathesis reaction of a halide precursor and a propargylic metal salt was a priori considered to be problematic.39 Owing to this tendency to form ate complexes, Lewisbase free complexes are typically prepared from alkyl or hydride precursors in the organometallic chemistry of rare-earth metals. The alkyl derivative of the type Cp*2LnCH(SiMe3)2 represents the most commonly used and wellcharacterized entry into the synthetic and reactive chemistry of Lewis-base free, 14-electron lanthanidocenes.40 Attempts to prepare Lewis-base free, monomeric propargyl/allenyl rare-earth metal complexes started therefore with reactivity studies of Cp*2LnCH(SiMe3)2 (6) and [Cp*2Ln(-H)]2 (7) (Ln = La, Y) towards 1-methylalk-1ynes. 2.3.2. Reactions of alkyls with 1-aryl-1-propynes

1-Phenyl-1-propyne (1) and ortho-substituted 1-phenyl-1-propynes (2-4) When an equimolar amount of 1-phenyl-1-propyne (1) and Cp*2LaCH(SiMe3)2 (6a) was allowed to react in benzene-d6, no reaction was observed by 1H NMR spectroscopy after several days. The clean, but slow formation of the desired propargyl species Cp*2LaCH2CCPh (7a) was apparent upon warming the reaction mixture to 80 C. For example, 2% of 6a was converted into 7a and CH2(SiMe3)2 after 3 hours at 80 C (Scheme 2-9). The rate of propargylic metalation was only moderately increased by increasing the reaction temperature. For example, 14% of 6a was converted into 7a and CH2(SiMe3)2 after 20 hours at 100 C. The use of an excess amount of substrate, on the other hand, resulted in a significant acceleration of propargylic metalation, accompanied by the formation of several organic products, as indicated by 1H NMR spectroscopy and GC-MS analysis after quenching the reaction mixture with methanol-d4 (Section 2.5.6). For example, the use of a 10-fold molar excess of substrate led to the complete conversion of 6a into 7a, CH2(SiMe3)2 and several organic products after 2 hours at 100 C. Analogous reactions of Cp*2YCH(SiMe3)2 (6b) with 1-phenyl-1-propyne (1) and Cp*2LaCH(SiMe3)2 (6a) with CH3CCC6H4Me-2 (2), CH3CCC6H4Me2-2,6 (3) and CH3CCC6H4iPr2-2,6 (4) gave similar results, but the corresponding propargyl species Cp*2LnCH2CCAr (7-10) was formed at a considerably lower rate (Scheme 2-9). For example, a stoichiometric mixture of 6b and 1 gave 5% conversion of the alkyl derivative (6b) into the propargyl derivative (7b) after 2 days at 100 C. These reactions allowed for the spectroscopic characterization of the corresponding propargyls (7-10) by means of multinuclear 1D and 2D NMR spectroscopy. In all cases, the Cp* ligands were magnetically

17

Chapter 2

Scheme 2-9. The reactions of 6 with 1-5.

Cp*2LnCH(SiMe3)2 6 Ln = La (a), Y(b) + CH3C CAr 1-5 Cp*2LnCH2C CAr 7-11 + CH2(SiMe3)2

F F

Ar
F F

CH3CCAr Cp*2LnCH2CCAr

1 7

2 8

3 9

4 10

5 11

equivalent and one CH2 signal was observed in the 1H NMR spectrum, suggesting that the propargyls Cp*2Ln(3-CH2CCAr) in solution adopt either a static 3-propargyl structure or exist as a rapid equilibrium mixture of the 1-propargyl and 1-allenyl tautomers. Variable-temperature 1H and 13C NMR studies of Cp*2LaCH2CCPh (7a), Cp*2YCH2CCPh (7b) and Cp*2LaCCC6H3Me2-2,6 (9a) in toluene-d8 down to -80 C and up to 120 C indicated that these complexes are nonfluxional on the 1H and 13C NMR time scale in this temperature range. The details of NMR and infrared spectroscopic parameters are discussed in Section 2.4. These initial reactivity studies demonstrated that the reactions of the alkyl derivatives Cp*2LnCH(SiMe3)2 (6) with 1-aryl-1-propynes (1-4) proceed via slow and selective propargylic C-H activation of the 1-aryl-1-propynes, even at elevated temperatures. The slow rate of propargylic metalation renders the preparation of propargyl derivatives Cp*2LnCH2CCAr (7-10) via the reactions of 6 with 1-aryl-1-propynes inconvenient. Although the rate of reaction was considerably increased by the use of an excess amount of substrate, the concomitantly formed organic products impeded the isolation of 7 by fractional crystallization and extraction. As a consequence, the preparation of Cp*2LnCH2CCAr (7-10) from the reaction of 6 and 1-4 were found to be impractical and other routes towards these propargyls were explored. 1-Pentafluorophenyl-1-propyne (5) In marked contrast to the reactions of Cp*2LnCH(SiMe3)2 6 (Ln = La, Y) with phenylacetylene and ortho-substituted analogues, such as 2-ethynyltoluene and (2,6-dimethylphenyl)acetylene, that proceed via rapid acetylenic metalation, even at low temperatures (Chapter 4), the present reactions of 6 with 1-phenyl-1-propyne and ortho-substituted analogues were found to proceed via slow propargylic metalation. It seems natural to ascribe the lower rate of propargylic C-H activation relative to acetylenic C-H activation to the lower (kinetic) acidity of the propargylic C-H bond. A study of the lithiation reactions of a series of 1-aryl-1-propynes revealed that the rate of propargylic lithiation of 1-methyl-1-alkynes increases with the electron-withdrawing capacity of the aromatic group.41 Accordingly, the reaction of 6 and 1-pentafluorophenyl-1-propyne (5) was conducted to investigate whether propargylic metalation by Cp*2LnCH(SiMe3)2 (6) could be increased upon increasing the acidity of the propargylic C-H bond in this system.42 The electron-withdrawing and carbanion-stabilizing capacity of the pentafluorophenyl group is well established.43 When a mixture of Cp*2LaCH(SiMe3)2 (6a) and a stoichiometric amount of CH3CCC6F5 (5) was heated in benzene-d6 to 50 C for 1 day, no reaction was observed by 1H NMR spectroscopy. The slow formation of the propargyl Cp*2LaCH2CCC6F5 (11a) and CH2(SiMe3)2 was observed after heating the reaction mixture for several hours at 80 C. For example, 12% of 6a was converted after 52 hours at 80 C. Complete conversion of 6a into 11a and CH2(SiMe3)2 was observed after heating the reaction mixture for 13 days at 80 C, albeit less selectively as observed previously for 1-phenyl-1-propyne and its ortho-substituted analogues (vide supra). Although the reaction of Cp*2LaCH(SiMe3)2 (6a) and CH3CCC6F5 (5) yielded 11a as the major organometallic product, based on the presence of one major Cp* 1H NMR resonance at 1.80 ppm, the yield was only 47%, based on integration versus CH2(SiMe3)2.

18

Rare-earth metallocene propargyl/allenyls Table 2-1. NMR and IR spectral data of the 1-aryl-1-propynes.a

H3C C C Ar
C-2 C-3 H-1 C-1 C C (2JCH) (3JCH) (1JCH) (cm-1) 1 Ph 3.92 80.46 86.08 1.64 2250 (1) (131.4) (4.8) (10.2) 2 C6H3Me2-2,6 4.18 78.08 94.20 1.74 2243 (3) (131.0) (4.5) (10.5) 3 C6H3iPr2-2,6 4.26 77.19 93.59 1.75 2243 (4) (131.0) (4.7) (10.5) 4 C 6 F5 4.02 64.34 99.84 1.50 2254 (5) (133.0) (4.6) (10.8) a NMR measurements performed in C6D6 at 25 C under a nitrogen atmosphere. Chemical shifts are reported in ppm and coupling constants in Hz. Repeated experiments suggested that 13C chemical shifts were accurate to 0.02 ppm, 1H chemical shifts to 0.01 ppm and 1JCH coupling constants to 0.2 Hz. IR measurements as liquid films or nujol mulls. Entry Ar Multinuclear NMR spectroscopy and GC/GC-MS analysis of the reaction mixture after quenching with H2O and D2O plausibly ruled out the occurrence of H/D scrambling at the Cp* methyl groups and C-F activation of 5, but failed to shed light on the identities of the formed organometallic by-products. Because quantitative 1H NMR analysis indicated that 0.90 equiv. of 5 was converted relative to CH2(SiMe3)2 after complete conversion of 6a, it is believed that the low yield of Cp*2LaCH2CCC6F5 (11a) is the result of the limited thermal stability of 11a. Additional support for this view is provided by the observed lower yield of 11a (25%), when the analogous reaction was conducted at 100 C. The validity of the linear relationship between the rate of proton transfer from a carbon acid and the first-order carbon-hydrogen coupling constant of the corresponding carbon (1JCH) has been demonstrated by many studies.44 The rationale for this behavior is found in the relative amount of s-character of the hybrid orbital on the carbon atom. As the s-character increases, the hybrid orbital is more tightly bound to the carbon atom and the C-H bond becomes more polar, thereby increasing the acidity of the hydrogen atom. Conversely, the formed anion is the most stable in the orbital with the highest s-character. Hence, the following relative order of increasing kinetic acidity was found: 3 4 < 1 < 5, on the basis of the first-order carbon-hydrogen coupling constant of the propargylic methyl group (1JCH).45 It seems that the anticipated rate increase of propargylic C-H activation is only modest upon increasing the kinetic acidity of the propargylic methyl group by substituting the phenyl group in 1-pheny-1propyne with a pentafluorophenyl group. In addition, experminents indicated that the corresponding propargyl Cp*2LaCH2CCC6F5 (11a) was thermally less stable than its non-fluorinated analogues (7-10), decomposing into unidentified compounds after prolonged times at relatively high temperatures. Thus, both the relative slow rate and thermal instability limit the utility of the present reaction as a convenient preparative route towards propargyl derivatives of rare-earth metallocenes. 2.3.3. Reactions of hydrides with 1-aryl-1-propynes

1-Phenyl-1-propyne (1) Because the formation of the corresponding propargyl was slow from the reaction of alkyl derivatives Cp*2LnCH(SiMe3)2 (6) with 1-aryl-1-propynes (1-5), analogous reactions of [Cp*2Ln(-H)]2 (12) were studied as a more convenient preparative route towards propargyl derivatives of rare-earth metallocenes Cp*2LnCH2CCAr (7-10). When 0.5 equiv. of 1-phenyl-1-propyne was added to a solution of [Cp*2La(-H)]2 (12a) in benzene-d6, a complex reaction mixture formed with a time-dependent composition, as indicated by in situ 1H NMR spectroscopy. After 5 min at room temperature, the hydride was converted completely into three major Cp*2La species (on the basis of Cp* 1H NMR resonances at 1.92, 1.90 and 1.88 ppm) of which the propargyl species Cp*2LaCH2CCPh ( 1.92 ppm) constituted only 31% of the total intensity of the Cp* 1H NMR

19

Chapter 2 resonances present. In addition, cis-1-phenylprop-1-ene was identified in the reaction mixture and its concentration increased slowly in time.

[Cp*2La(-H)]2 + 2 CH3C CPh

(2.1)

12a Ph Cp*2LaCH2C CPh 7a + Cp*2La 13a + 14a LaCp*2 Ph + Ph

Interestingly, the propargyl derivative became the major organometallic product present (75%, 1H NMR) after standing for 1 day at room temperature. GC/GC-MS and 1H NMR analysis of the major organic products in the reaction mixture after addition of methanol-d4 indicated the presence of cis-1-phenylprop-1-ene, cis-1-phenylprop-1-ene-d1, phenylallene-d1 and 1-phenyl-1-propyne-d1. The latter two compounds are formed from deuterolyis of Cp*2LaCH2CCPh (7a) (Section 2.5.2). The use of toluene-d8 as solvent gave similar results, thereby indicating that the reactions described above do not involve solvent molecules, as previously observed for lanthanide hydrides and alkyls in aromatic solvents.46 These results can plausibly be rationalized by proposing three separate pathways by which a monomeric hydride derivative reacts with 1-phenyl-1-propyne (Scheme 2-10). Although no spectroscopic evidence for Cp*2LaH was obtained in this study, ample evidence exist in literature that hydride derivatives of rare-earth metallocenes [Cp*2Ln(-H)]2 (12) are present in solution as an equilibrium between dimer and monomer.47 The formed monomeric hydride derivative may, in principle, undergo both alkyne insertion into the La-H bond and propargylic C-H activation. Propargylic metalation of 1-phenyl-1-propyne by Cp*2LaH yields 7a and dihydrogen. The reversible nature of propargylic metalation was demonstrated by the independent reaction of 7a with excess dihydrogen (1 atm.). In this reaction, 7a was consumed within several hours, giving rise to phenylpropane and [Cp*2La(-H)2] (12a) over the course of 1 day at room temperature. Although the relatively low rate of the reaction of 7a with excess H2 may suggest that reversible metalation of Cp*2LaH by 1-phenyl-1propyne is only of minor importance in the present system, efficient hydrogen capture is not without precedent.48 Insertion of the carbon-carbon triple bond into the La-H bond of the monomeric hydride derivative may take place both in a 2,1- and 1,2-manner, affording the corresponding 1-phenylprop-1-en-1-yl 12a and 1phenylprop-1-en-2-yl derivatives 13a, respectively. Clearly, the ratio of these insertion products is determined by Scheme 2-10. The proposed reaction sequences in the reaction of 12a with 1.
H H

Cp*2La 12a

LaCp*2

Ph Cp*2La 13a

CH3 H H2

CH3C CPh Cp*2La H Ph CH3C CPh H2

CH3C CPh

CH3 Cp*2La H2 14a

Ph H

Ph CH3C CPh H2

Cp*2LaCH2C CPh 7a CH3C CPh Ph Ph CH3C CPh

20

Rare-earth metallocene propargyl/allenyls

Scheme 2-11. The preparation of Cp*2LaC(Ph)=C(Ph)H (15a) and its reaction with 1-phenyl-1-propyne.
Ph [Cp*2La(-H)]2 12a Ph Cp*2La Ph + H CH3C CPh Cp*2LaCH2C CPh 7a + + Ph Ph Cp*2La 15a Ph Ph Ph H

the electronic and steric properties of the substituents at the carbon-carbon triple bond and those of the metal complex and is difficult to predict a priori. 1,2-Insertion may be favored under steric control,49 but 2,1-insertion of 1-arylalk-1-ynes and 1-arylalk-1-enes is commonly observed in rare-earth metal chemistry (Chapter 4).50 It should also be noted that no evidence for reversible alkyne insertion was obtained in the reaction of [Cp*2Ln(H)]2 with diphenylacetylene.51 The transient alkenyl species Cp*LaC(Ph)=C(CH3)H (13a) and Cp*LaC(CH3)=C(Ph)H (14a) were inferred from the presence of cis-1-phenylprop-1-ene and the formation of cis-1-phenylprop-1-ene-d1 upon quenching the reaction mixture with methanol-d4. However, the relative preference for 2,1- or 1,2-insertion in this system could not be determined by means of NMR spectroscopy or GC-MS analysis. Attempts to spectroscopically identify the proposed 1-phenylprop-1-en-1-yl and 1-phenylprop-1-en-2-yl derivatives were hampered by the complexity of the 1H NMR spectrum and overlapping signals. The reaction of the formed propenyl derivatives with 1-phenyl-1-propyne may give rise to the observed cis-1-phenylprop-1-ene via propargylic C-H activation. Evidence for this hypothesis was provided by the reaction of Cp*2LaC(Ph)=C(Ph)H (15a) with 1-phenyl-1-propyne, affording Cp*2LaCH2CCPh (7a) and cisdiphenylethene (Scheme 2-11). This alkenyl derivative was conveniently prepared from ([Cp*2La(-H)]2 and diphenylethyne. Another route towards the observed cis-1-phenylprop-1-ene may plausibly involve the reaction of the propenyl derivatives with dihydrogen formed from propargylic metalation of 1-phenyl-1-propyne, as implicated by the reaction of Cp*2LaCH2CCPh with excess dihydrogen, giving rise to phenylpropane and [Cp*2La(-H)]2.52 Because the reaction of the hydride [Cp*2La(-H)]2 (12a) and 1-phenyl-1-propyne (1) was not found to be clean at room temperature, it was decided to investigate whether the proposed competition between propargylic C-H activation and alkyne insertion could be influenced in the favor of propargylic C-H activation by lowering the reaction temperature. Thus, the reaction of 12a and 1 (2 equiv.) in toluene-d8 at -60 C was monitored with 1H NMR spectroscopy. The 1H NMR spectrum of the reaction mixture was more complex, however, exhibiting nine Cp* resonances of which the propargyl represented only 13% of the total Cp* 1H NMR resonance intensities. No significant changes in the composition were observed upon slowly warming to room temperature and further standing. The presence of cis-1-phenylprop-1-ene-d1, phenylallene-d1, 1-phenyl-1propyne-d1 and several unidentified compounds was indicated by NMR spectroscopy and GC-MS analysis after quenching with methanol-d4. The occurrence of insertion reactions of unsaturated substrates into La-C bonds is ruled out by the observed masses of the reaction products. These considerations suggest that alternative C-H activation sequences compete more effectively with propargylic C-H activation at lower reaction temperatures. Examples of lanthanidocene derivatives undergoing allylic, vinylic and aromatic C-H activation are welldocumented.53 The above reaction sequences imply that alkyne insertion is competitive with propargylic C-H activation due to reversible propargylic metalation in a closed atmosphere. Following this line of reasoning, it was put forward that the removal of dihydrogen gas may render propargylic metalation irreversible, thereby suppressing alkyne insertion under the above reaction conditions. This notion was confirmed by performing the reactions under dynamic vacuum, affording the propargyl in high yields (~90%, 1H NMR). Even so, the presence of cis-1-phenyl-1-propene indicated that alkyne insertion was still competitive with propargylic C-H activation under these reaction conditions. The desired propargyl Cp*2LaCH2CCPh (7a) was successfully prepared by performing the reaction of the hydride 12a in an aliphatic (e.g. pentane, hexane) or aromatic solvent at room temperature under dynamic vacuum. Also, the use of an excess amount of substrate was found to be necessary to compensate for the seemingly unavoidable hydrogenation of the substrate. A preparative more convenient method that avoids the

21

Chapter 2

Scheme 2-12. The preparation of Cp*2LnCH2CCPh (7) (Ln = Y b, La a) from the reaction of the hydride precursor, generated via in situ hydrogenolysis, and 1-phenyl-1-propyne.
H2 -CH2(SiMe3)2 CH3C CPh vacuum Cp*2LnCH2C CPh + H2

Cp*2LnCH(SiMe3)2

Ln = La (91%), Y (92%)

handling of the highly reactive hydride involves the in situ formation of the hydride [Cp*2Ln(-H)]2 from the alkyl Cp*2LnCH(SiMe3)2 (Scheme 2-12). Cp*2LaCH2CCPh (7a) is extremely soluble in alkane and aromatic solvents and was initially isolated as a red oil which did not easily form crystals.54,55 A crystalline solid was obtained after complete in vacuo removal of the organic compounds in the crude product mixture (e.g. cis-1phenyl-1-propene, CH2(SiMe3)2, unreacted 1-phenyl-1-propyne) and prolonged cooling at -80 C. Recrystallization from toluene afforded single-crystals suitable for an X-ray crystallographic study in which 7a was identified as an 3-propargyl/allenyl complex in the solid state. Structural data is discussed in more detail in Section 2.4.4. A similar methodology also afforded the yttrium analogue Cp*2YCH2CCPh (7b) in good isolated yields and single-crystals suitable for X-ray analysis were obtained analogously. Ortho-substituted 1-phenyl-1-propynes (3-4) When [Cp*2La(-H)]2 (12a) was allowed to react in benzene-d6 at room temperature with 2 equiv. of CH3CCC6H4Me2-2,6 (3), an instantaneous reaction was observed with 1H NMR spectroscopy. Although propargylic C-H activation was found to be the major pathway, forming the propargyl Cp*2LaCH2CCC6H4Me22,6 (9a) quite selectively (>90%, 1H NMR), hydrogenation was still operative, as indicated by the presence of cis-1-(2,6-dimethylphenyl)prop-1-ene (1H NMR, GC-MS) in the reaction mixture. When the reaction mixture was allowed to stand at room temperature for several days, no significant changes were observed with 1H NMR spectroscopy. The complex Cp*2LaCH2CCC6H4Me2-2,6 (9a) was isolated on a preparative scale by employing the above methodology consisting of the in situ preparation of the hydride derivative and the application of dynamic vacuum. However, the isolated yield was low (22%), most likely due incomplete removal of organic products from the reaction mixture, thereby hampering the crystallization of the propargyl. Unfortunately, attempts to obtain single-crystals suitable for X-ray analysis were unsuccessful. The reaction of [Cp*2La(-H)]2 (12a) with 0.5 equiv of CH3CCC6H4iPr2-2,6 (4) in benzene-d6 gave two major Cp* resonances ( 2.22, 1.92) in the 1H NMR spectrum after 1 h at room temperature of which the propargyl was the major species ( 1.92, 58%). Again, the substrate was partially hydrogenated, as indicated by the presence of cis-1-(2,6-diisopropylphenyl)-1-propene (1H NMR, GC-MS). After 2 days at room temperature both the concentration of the propargyl (74%) and hydrogenated substrate increased. Unfortunately, no crystalline product could be isolated at a preparative scale from the reaction mixture of Cp*2LaCH(SiMe3)2 and 4, following the above procedure. The organic products were difficult to remove in vacuo and unidentified organometallic reaction products could not be separated by extraction with appropriate solvents. 1-Pentafluorophenyl-1-propyne (5) When the prop-1-ynylpentafluorobenzene (5) (1.0 equiv per La) was added to a solution of 12a in benzene-d6 at rom temperature, a complex time-dependent reaction mixture formed, as indicated by 1H NMR spectroscopy. After 10 min the 1H NMR resonances of the hydride and substrate had completely disappeared and four major Cp* 1H NMR resonances were observed ( 1.88, 1.84, 1.83 and 1.80 ppm) of which the propargyl Cp*2LaCH2CCC6F5 (11a: 1.88) represented only 7% of the total Cp* 1H NMR resonance intensities. Attempts to identify the major organometallic products by means of NMR spectroscopy failed. After two days standing at room temperature, the concentration of 11a increased to 14%, while more Cp* 1H NMR resonances appeared concomitantly. 19F NMR spectroscopy indicated the presence of at least four different pentafluorophenyl groups in a 0.4:1.0:0.3:0.3 ratio. When the reaction mixture was quenched with methanol-d4 after 2 days at room temperature, NMR and GC/GC-MS analysis indicated the presence of two major organic products that were plausibly identified as 1-pentafluorophenyl-1-propene-d1 and the monodeuterated coupling product C6F5C(D)=C(Me)C(C6F5)=CHMe in a ratio of 1.00:0.33, respectively. The geometry of the double bonds in the latter double insertion product could not be determined unequivocally and the proposed structure is based on the

22

Rare-earth metallocene propargyl/allenyls

Scheme 2-13. The proposed reaction sequences of the reaction of 12a with 5.
C6F5 [Cp*2La(-H)]2 + 2 CH3C CC6F5 C 6D 6 25 C Cp*2La C6F5 H 16a + H Cp*2La C 6F 5 17a CD3OD CD3OD C6F5 D C6F5 H D C6F5 H

well-recognized tendency of organo rare-earth metal complexes to give 2,1-insertion products of 1-aryl-1alkenes and 1-aryl-1-alkynes into metal-carbon and metal-hydrogen bonds ( Scheme 2-13).50 Based on the inferred presence of the propenyl derivative, Cp*2LaC(C6F5)=CH(CH3) (16a) could be characterized with 1H and 19F NMR spectroscopy ( 7.67, q, J = 6.7 Hz, CH, 1 H; 1.84, s, Cp*, 30 H; 1.51, d, J = 6.7 Hz, CH3, 3 H). This assignment is further supported by the close resemblance of its 1H NMR spectral data with that of Cp*2LaC(Ph)=CH(Ph) (i.e. the vinylic proton resonates at 7.69 ppm and the Cp* ligands appear as one singlet at 1.84 ppm) (vide supra). The complexity of the 1H NMR spectrum thwarted attempts to characterize Cp*2LaC(C6F5)=C(CH3)C(C6F5)=CH(CH3) (17a) unambiguously.

2.4.
2.4.1.

Structural characterization of propargyl/allenyls

Introduction

The Cp*2LnCH2CCAr complexes exhibit equivalent Cp* and CH2 resonances in the 1H and 13C NMR spectrum indicative of either a static 3-propargyl structure or a rapid equilibrium mixture of the 1-propargyl and 1-allenyl species. Because no 1H and 13C NMR resonances attributable to distinct 1-propargyl and 1allenyl derivatives of Cp*2LaCH2CCPh (7a), Cp*2YCH2CCPh (7b) and Cp*2LaCCC6H3Me2-2,6 (9a) were observed at -80 C, the former picture is presently favored. To investigate this issue further, these compounds were also characterized by 13C NMR and infrared spectroscopy and, in some cases, by X-ray crystallography. Scheme 2-14. The possible limiting structures of Cp*2LnCH2CCAr.
R Ln H H D R
Ln

R
Ln

R . H

H H E

Furthermore, spectroscopic parameters are likely to reflect structural changes in the present Cp*2LnCH2CCAr complexes resulting from differences in the steric and electronic properties of the arylsubstituted ligands employed in this study. If the Cp*2LnCH2CCAr complexes are static 3-propargyls (E), it is believed that substitution at the ortho-positions of the phenyl ring will shift the 3-propargyl structure closer to a more 1-propargyl-like structure (D) in the continuum of canonical structures, due to the larger steric hindrance between the Cp* ligands and the ortho-substituent in the 1-allenyl-like structure (F) (Scheme 2-14). Alternatively, if the Cp*2LnCH2CCAr complexes are engaged in a rapid equilibrium mixture between the 1-

23

Chapter 2 Table 2-2. Spectroscopic data of the Cp*2LnCH2CCAr complexes.a Entry Ln, Ar CH2 C-2 C-3 Cp* C-1 C C (1JYC) (1JCH, 1JYC) (cm-1) 1 Y, Ph 2.82 1.93 49.33 155.09 106.79 1920 (7b) (159.1, 5.2) (11.9) 2 La, Ph 2.82 1.91 54.00 152.19 112.75 1944 (7a) (158.3) 3 La, C6H3Me2-2,6 2.71 1.92 49.00 149.28 105.79 1975 (9a) (156.9) c 4 La, C6H3iPr2-2,6 2.68 1.92 47.69 145.61 100.31 (10a) (156.7) c 3.08 1.88 55.77 165.24 108.76 5 La, C6F5 (11a) (162.5) 6 CH3CCPh 24.9 79.9 85.9 2251 7b CH2CCHPh 78.9 209.9 94.1 1940 a NMR measurements in C6D6 at 25 C under a nitrogen atmosphere. The chemical shifts are reported in ppm and coupling constants in Hz. IR measurements under a nitrogen atmosphere. NMR measurements in CDCl3 at 25 C for entries 6 and 7. b Data taken from Ref. 67. c Compounds not isolated. propargylic (D) and 1-allenylic tautomer (F), the formation of F is likely to be disfavored upon orthosubstitution and more D will be present in the equilibrium mixture. 2.4.2. Spectroscopic properties of the propargyl/allenyls

The 1H and 13C NMR resonances of the rare-earth metallocene propargyl derivatives Cp*2LnCH2CCAr are shown in Table 2-2. The CH2 proton resonances are observed between 3.08 and 2.71 ppm. These values suggest a considerable contribution from the propargylic structure, as 1-bound allenylic methylene protons typically resonate within the region of 3.9-4.3 ppm and those associated with the propargylic tautomer lie at higher field between 1.75 and 3.0 ppm (Table 2-3).3c The 13C NMR spectra exhibit downfield shifted CH2 propargyl carbon resonances (C-1, 47.7-57.8 ppm) relative to the neutral 1-phenyl-1-propyne ligand ( 3.92 ppm, see Table 2-1). The chemical shifts are in the range of values observed for other substituted, sp3 -carbons in rare-earth metallocenes, such as Cp*2Y[3CH(Ph)CH2Ph]56 ( 56.40 ppm, 1JCH = 128 Hz), Cp*2La[3-CH(Ph)CH2Ph]56 ( 49.20 ppm, 1JCH = 131 Hz) and Cp*2YCH2C6H3Me2-3,5 ( 47.35 ppm, 1JCH = 133.4 Hz).57 The first-order carbon-hydrogen coupling constants 1 JCH (156.7-162.5 Hz), on the other hand, are significantly larger than those observed in the above complexes and reveal that the C-H bonds of the propargylic carbon have considerably more s character. The values are close to that of a sp2 carbon (typically 156.4 Hz) and may be interpreted as to discard 1-propargyl bonding (sp3 hybridization is expected for the propargylic carbon in an 1-propargyl, Scheme 2-16). However, such an unambiguous interpretation is complicated by the fact that inductive/field effects originating from polar substituents influence 1JCH values to such an extent that the 1JCH values mostly do not reflect the true hybridization state of the carbon.58 The CH2CC central carbon resonances (C-2, 145.6-165.2 ppm) are closer to the central carbon atom of the neutral phenylallene ( 209.9 ppm) than that of 1-phenyl-1-propyne ( 79.9 ppm), thereby suggesting a high degree of allenylic character for the central carbon. Also, the CH2CC carbon resonances (C-3, 100.3-112.8 ppm) are closer to that of the corresponding carbon atom of phenylallene than to that of 1-phenyl-1-propyne ( 85.9 ppm). Extensive studies in organolithium chemistry have shown that the central carbon (C-2, 165-180 ppm for allenyl) and the propargyl carbon (C-1, 90-100 ppm) are very sensitive indicators of allenylic and propargylic structures.59 Carbon NMR resonances and associated 1JCH values have also been used to differentiate between 1-allenyl and 1-propargyl structures for other metal complexes, but it is presently difficult to make generalizations based on the limited data reported in literature.3c-d Nevertheless, when the NMR spectral data of the present rare-earth metal complexes are compared to that of reported (phenyl-substituted) allenyl and propargyl metal complexes, it can be seen that the observed spectral parameters point to an 3-allenyl/propargyl

24

Rare-earth metallocene propargyl/allenyls Table 2-3. Selected NMR data for allenyl and propargyl complexes.a Entry Complex CH2 C-1 (1JCH) 1 [(PPh3)2Pt(3-CH2CCPh)]+ 2.74 48.3 (170) 2 [(PPh3)2Pd(3-CH2CCPh)]+ 3.19 50.9 3 Cp2ZrMe(3-CH2CCPh) 3.37 55.5 (167) 4 CpMo(3-CH2CCPh)(2-CH3CCPh) 3.90 41.6 (158, 165) 5 Cp2Zr(CH2CCPh)2 2.80 30.7 (151) 6 Cp*(TBM)Zr(3-CH2CCMe) 2.51 51.0 7 [Cp*2Zr(3-CH2CCMe)]+ 2.93 61.7 (158) 8 [Pt(CO)(PPh3)2(1-CH2CCPh)]+ 1.98 7.1 9 [Pt(CO)(PPh3)2(1-CPhCCH2)]+ 3.65 72.9 a Chemical shifts in ppm and coupling constants in Hz.

C-2 97.3 94.2 120.5 142.3 128.4 99.9 128.4 92.2 203.1

C-3 102.1 105.4 114.1 124.8 103.5 91.7 103.5 86.1 101.5

Ref. [4f,e] [4g] [4b] [63b] [4b] [4h] [88] [4e] [4e]

structure (Table 2-3). The observation that the 13C NMR chemical shifts are intermediate of those observed for 1-CH2CCPh and 1-C(Ph)=C=CH2 complexes is in accord with this view (Entries 8 and 9). Infrared spectroscopy has been widely used as a diagnostic tool in determining the relative contribution of allenylic and propargylic structures.2,59 In the 2250-1750 cm-1 region of the infrared spectrum, strong absorptions at 1944 and 1920 cm-1 are observed for Cp*2LaCH2CCPh (7a) and Cp*2YCH2CCPh (7b), respectively (Figure 2-8). Comparison with reported data indicates that these values are intermediate of those typically observed for -propargyl (2100-2215 cm-1) and -allenyl metal complexes (1800-1920 cm-1),60 thereby providing additional support to the view that the present rare-earth metallocene complexes adopt an 3propargyl/allenyl structure. Remarkably, comparison with IR data of reported 3-propargyl/allenyl complexes is difficult, because IR data of these compounds are rare in literature.3d-f One of the few literature examples involves the crystallographically characterized Cp2ZrMe(3-CH2CCPh), which is reported to display two medium absorptions at 2176 and 1924 cm-1.4b,61 These vibrations were not assigned and their understanding is not clear-cut, because a static 3-bonding would be expected to lead to a single absorption in this region. The observation of two absorptions in the 1800-1920 and 2100-2215 cm-1 range has, in fact, typically been interpreted in terms of an equilibrium between 1-allenylic and 1-propargylic organometallic species.2 However, no fluxional behavior was observed in solution Cp2ZrMe(3-CH2CCPh) down to -40 C and the authors did not comment on this ambiguity. The titanocene derivative Cp*2Ti(CH2CCPh) is reported to exhibit only one strong absorption at 1902 cm-1 in this region and was assigned the 3-hapticity by comparison with Cp2ZrMe(3-CH2CCPh).62 Following this line of reasoning, it seems that both NMR and IR spectroscopy are consistent with an 3-allenyl/propargyl structure for the present rare-earth metal complexes. Yttrium has a nucleus with I = in a natural abundance of 100% and a coupling interaction in Cp*2YCH2CCPh (7b) is observed with the terminal carbons of the CH2CC ligand (Table 2-2). These couplings are in agreement with the nature of metal CH2CC bonding. According to several molecular orbital studies, the CH2CCR ligand binds mainly through its terminal carbons, while the -bond of the carbon-carbon triple bond does not interact significantly with the metal fragment.63 The observed yttrium-carbon coupling of the propargyl carbon (5.9 Hz) is notably smaller than that of other yttrocene-bonded methylene carbons, such as Cp*2Y(CH2C6H3Me2-3,5)64 (1JYC = 23.2 Hz) and Cp*2Y[3-CH(Ph)CH2Ph]56 (1JYC = 18.7 Hz), but is comparable to that of the terminal allyl carbon in Cp*2Y(3-CH2CH=CH2)40 (1JYC = 3.8 Hz) and the smallest coupling Scheme 2-15. Reported first-order yttrium-carbon coupling constants in [Cp2Y(-1:1-C,N-HC=NCtBu)]2 and [(Cp*2Y)2(-O,C-OCH=CHC C)].

N Cp2Y N 35 Hz YCp2

5 Hz Cp*2Y

O 12 Hz

53 Hz YCp*2

25

Chapter 2

200 Carbon chemical shift (ppm) 180 160 140 120 100 80 60 40 20 130.5 131 131.5
1

C-1

C-2

C-3

132 J CH (Hz)

132.5

133

133.5

Figure 2-1. Plot of the carbon chemical shifts of the propargylic (C-1), internal (C-2) and terminal (C-3) carbon atoms of the CH2CC group in 7a, 9a, 10a and 11a (Table 2.2) versus the first-order carbonhydrogen coupling constant (1JCH) of the propargylic methyl group of the neutral ligand CH3CCAr (Table 2.1). constant in the asymmetrically bridging formimidoyl carbon in [Cp2Y(-1:1-N,C-HC=NCtBu)]2 (1JYC = 5 Hz, 1 JYC = 35 Hz) (Scheme 2-15).65 The yttrium-carbon coupling constant of the terminal carbon (11.9 Hz) in 7b is similar to that observed for the terminal acetylide carbon with the distant Cp*2Y moiety in [(Cp*2Y)2(-O,COCH=CHC C)] (1JYC = 12 Hz, 1JYC = 53 Hz) (Scheme 2-15).66 These findings reveal that the observed yttrium-carbon coupling constants are consistent with the proposed delocalized metal 3-CH2CCPh bonding. 2.4.3. Ligand and metal ion size effects

The observed NMR spectral parameters of the propargyl derivatives Cp*2LnCH2CCAr clearly indicate that the chemical shift of the propargylic CH2 protons varies with the nature of the aromatic group ( 3.08-2.71 ppm), while that of the Cp* ligand changes only little ( 1.93-1.88 ppm) (Table 2-2). In fact, the CH2 proton resonances move upfield with increasing ortho-substitution in the Cp*2LaCH2CCC6H3R2-2,6 (R = H 7a, Me 9a, iPr 10a) series. In view of the relatively high field proton resonance of the allenylic CH2 group in phenylallene67 ( 4.85 ppm) and 1-allenyl metal complexes (e.g. entry 8, Table 2-3) as compared to the the relatively low field proton resonance of the propargylic CH2 group in 1-phenyl-1-propyne ( 1.65 ppm) and 1propargyl metal complexes (e.g. entry 9, Table 2-3), this finding suggests a shift to either a more propargyl-like 3-propargyl structure or a shift towards the formation of the 1-propargyl in an equilibrium mixture between a 1-propargyl and a 1-allenyl derivative.3c Replacing the phenyl group by a pentafluorophenyl group seems to have the opposite effect, favoring either a more allenyl-like 3-propargyl structure or the formation of the 1-allenyl in the equilibrium mixture.68 It seems reasonable to ascribe this change in character to the -carbanion-stabilizing property of the pentafluorophenyl group.43 Electronic substituent effects in propargyl-allenyl equilibria of organoindium derivatives have been reported.69 For these compounds, both experimental and theoretical data revealed that the electron-donating nature of a methyl group at the terminal carbon destabilizes the sp2 carbanion in the allenylic structure relative to the propargylic structure, thereby favoring the formation of the 1-propargyl species. The 13C NMR spectral data within the Cp*2LaCH2CCAr series are in agreement with the trend observed with 1H NMR spectroscopy (Table 2-2). When the steric bulk at the ortho-positions of Cp*2LaCH2CCAr is increased, the CH2 and CH2CC resonances shift upfield and CH2CC resonances shift downfield, consistent with the view that an increase of steric bulk at the ortho-positions increases the propargylic character of Cp*2LaCH2CCAr. When the phenyl group in Cp*2LaCH2CCPh is replaced by a pentafluorophenyl group, the opposite change in character is observed with 13C NMR spectroscopy (i.e. the CH2 and CH2CC resonances shift downfield and the CH2CC shift upfield) which is in agreement with the proposed increase in allenylic character.

26

Rare-earth metallocene propargyl/allenyls

3.4 Proton chemical shift (ppm) 3.2 3 2.8 2.6 2.4 130.5

131

131.5
1

132 J CH (Hz)

132.5

133

133.5

Figure 2-2. Plot of the proton chemical shift of the CH2 group of 7a, 9a, 10a and 11a (Table 2.2) versus the first-order carbon-hydrogen coupling constant (1JCH) of the propargylic methyl group of the neutral ligand CH3CCAr (Table 2.1) Another interesting feature of the NMR spectral data concerns the first-order carbon-hydrogen couplings of the propargylic carbon (C-1) in the Cp*2LaCH2CCAr series. It can be seen that these 1JCH values (in Hz) decrease in the same order as the order of increasing propargylic character based on the carbon chemical shifts, i.e. C6F5 (162.5) > C6H5 (158.3) > C6H3Me2-2,6 (156.9) > C6H3iPr2-2,6 (156.7). Although 1JCH couplings are well-known to correlate with the hybridization of the carbon atom in hydrocarbons, polar substituents have a much greater effect on 1JCH couplings and their presence renders these couplings unsuitable for estimating the s character of the carbon bonding orbital in some cases.58 Assuming that the polar effects of the metal center are constant in the series Cp*2LaCCAr, the decrease in 1JCH for the propargylic carbon arguably reflects a change in hybridization which can be interpreted as a change from a sp2-type carbon in H and I to a more sp3-type carbon in G (Scheme 2-16).70 Additional evidence for the proposed changes in the bonding of the present Cp*2LaCH2CCAr complexes as a function of the steric and electronic properties of the ligand is supplied by comparing the spectral properties of the complexes with those of the corresponding neutral ligand. Based on both NMR spectroscopy, the relative -electron-withdrawing ability of the ligand CH3CCAr was found to increase in the following order: 3 (C6H3Me2-2,6, 131.0) 4 (C6H3iPr2-2,6, 131.0) < 1 (Ph, 131.4) < 5 (C6F5, 133.0) (Table 2-1). It can be seen that several structurally diagnostic spectral parameters of the Cp*2LaCH2CCAr complexes, such as the proton chemical shift of the CH2 group (Figure 2-1), the carbon chemical shifts of the C3 ligand (Figure 2-2) and the first-order carbon-hydrogen coupling constant of the CH2 group (Figure 2-3), correlate in a similar manner with the -electron withdrawing ability of the ligand. These spectral parameters indicate also that the propargylic character of Cp*2LaCH2CCC6H3Me2-2,6 is somewhat higher than that of Cp*2LaCH2CCC6H3iPr2-2,6, despite the fact that the -electron withdrawing ability of their aromatic substituents is similar. This minor discrepancy may plausibly be ascribed to the increased steric bulk of the 2,6-diisopropylphenyl group relative to the 2,6dimethylphenyl group. In summary, these findings strongly suggest that the bonding of Cp*2LaCH2CCAr becomes increasingly more propargylic in character upon increasing the steric bulk of the aromatic substituent Scheme 2-16. The hybridization of the terminal carbon atoms in the limiting structures of the prop-2ynyl/allenyl anion.70

CH2 C CH C C C H H sp3 sp

CH2 C CH H C C C H H sp
2

CH2 C CH H C C C H sp
2

H
2

sp

sp

27

Chapter 2

170 165 J CH (Hz)

1

160 155 150 130.5

131

131.5
1

132 J CH (Hz)

132.5

133

133.5

Figure 2-3. Plot of the first-order carbon-hydrogen coupling constant (1JCH) of the propargylic CH2 group in 7a, 9a, 10a and 11a (Table 2.1) versus the first-order carbon-hydrogen coupling constant (1JCH) of the propargylic methyl group of the neutral ligand CH3CCAr (Table 2.2) and more allenylic in character upon an increase in the -electron-withdrawing ability of the aromatic substituent. When the lanthanum metal center is substituted by the smaller yttrium,71 the relative position of the 3 -allenyl/propargyl ligand in either an 3-allenyl/propargyl continuum or an 1-allenyl/1-propargyl equilibrium mixture is difficult to ascertain spectroscopically. A clear-cut interpretation of the observed spectral parameters is complicated by a difference in inductive/field effects of the metals.72 Also, the solid-state molecular structures of 7a and 7b did not provide additional insight into the influence of the metal ion size on the nature of bonding of the 3-allenyl/propargyl ligand in 7a and 7b (Section 2.4.4). 2.4.4. The molecular structures of Cp*2LnCH2CCPh (Ln = La, Y).

In order to shed more light on the nature of the bonding in the present propargyl derivatives Cp*2LnCH2CCAr, X-ray crystallographic studies were performed on Cp*2LaCH2CCPh (7a) and Cp*2YCH2CCPh (7b). In both cases, single crystals were grown from toluene solutions at low temperature. Complexes 7a and 7b crystallize in the orthorhombic space groups Pna21 with Z = 4 and Pca2a1 with Z = 4, respectively. The crystal structures of 7a and 7b are shown in Figures 2-3 and 2-4, respectively, and selected bond distances and angles are given in Table 2-4. The bent metallocene in 7a is similar to that in other lanthanum73 and lanthanide metallocenes.74 The 137.1 Ct1-La-Ct2 angle and the average 2.54 La-Ct1 and 2.51 La-Ct2 distances are normal. The C5Me5 are in the usual staggered conformation with a twist angle of 28.9 compared to 36 for a perfectly staggered arrangement.75 The phenyl-substituted propargyl/allenyl ligand is clearly bound in an 3-mode. The La-C(propargyl) distances for La-C21, La-C22 and La-C23 are 2.811(6), 2.695(7) and 2.740(7) , respectively. At first glance, the main interaction of the metal center with the C3 ligand center would seem to take place on C22. However, theoretical studies have shown that the metal bonding with an 3-propargyl/allenyl ligand occurs through the terminal carbon atoms.63 In agreement with delocalized 3-bonding, the observed La-C21 and La-C22 distances are longer than that of reported La-C -bond lengths, such as 2.537(5) and 2.588(4) in Cp*La[CH(SiMe3)2]2 and 2.651(8) and 2.627(10) in Cp*La[CH(SiMe3)2]2(THF).76 The backbone of the 3-propargyl/allenyl ligand is virtually coplanar with La (atom displacements from the least squares plane: La 0.000, C21 0.001, C22 0.003, C23 0.002 ). Such coplanar rearrangements have also been reported for other 3-propargyl/allenyl complexes and this structural feature is believed to be characteristic of 3-propargyl/allenyl complexes.3d-f The atoms La, C21, C22 and C23 all lie in the mirror plane that reflects the two Cp* ligands of 7a. This symmetry, obviously, results in the observed magnetic equivalence of the two hydrogen atoms bonded to C21 and the Cp* ligands in the 1H and 13C NMR spectrum of 7a (Section 2.4.2). The crystallographic C21-C22 and C22-C23 bond distances of 1.36(1) and 1.23(1) , respectively, are

28

Rare-earth metallocene propargyl/allenyls

Figure 2-4. The molecular structure of 7a with thermal ellipsoids drawn at the 50% probability level. The hydrogen atoms are omitted for clarity.

Figure 2-5. The molecular structure of 7b with thermal ellipsoids drawn at the 50% probability level. The hydrogen atoms are omitted for clarity.

intermediate between those generally accepted for C-C single (1.45 ) and double (1.31 ) and double and triple (1.20 ) bonds, respectively, and the C3 ligand backbone is thus consistent with resonance between propargyl and allenyl structures. The relatively large difference of 0.13(1) between the two C-C bond lengths suggests that the propargylic contribution to the bonding description of 7a is relatively more important than the allenylic contribution. Although both of the resonance forms of the 3-propargyl/allenyl ligand would be expected to lead to a linear geometry at the central carbon atom, the bond angle C21-C22-C23 of159.3(7) indicates a considerable deviation from linearity. This structural feature appears to be another characteristic of 3propargyl/allenyl complexes, as angles of 145-155 have been reported for other (phenyl-substituted) 3propargyl/allenyl complexes (Table 2-4).3d-f The overall geometry of Cp*2YCH2CCPh (7b) is similar to that of its lanthanum congener 7a. The YC bond distances are smaller and reflect the differences in Y3+ (1.019 ) and La3+ (1.160 ) ionic radii for eightcoordination.71 The metrical parameters corresponding to the Cp*2Y fragment of 7b are not exceptional, as

29

Chapter 2 Table 2-5. Selected bond lengths () and angles () in Cp*2LnCH2CCPh (Ln = La, Y) complexes.a Ln = La (7a) Ln = Y (7b) Bond lengths C21-C22 1.362(10) C21-C22 1.366(4) C22-C23 1.234(10) C22-C23 1.268(4) C23-C24 1.465(9) C23-C24 1.462(4) La-C21 2.811(6) Y-C21 2.653(3) La-C22 2.695(7) Y-C22 2.529(3) La-C23 2.740(7) Y-C23 2.560(3) av. La-C(Cp*) 2.806(16) av. Y-C(Cp*) 2.654(4) 2.780(16) 2.652(4) La-Ct1 2.535(3) Y-Ct1 2.362 La-Ct2 2.509(4) Y-Ct2 2.362 Bond angles Ct1-La-Ct2 137.1(1) Ct1-Y-Ct2 139.3 C21-C22-C23 159.3(7) C21-C22-C23 155.9(3) C22-C23-C24 141.2(7) C22-C23-C24 141.5(3) La-C21-C22 71.0(4) Y-C21-C22 70.2(2) La-C23-C22 74.9(4) Y-C23-C22 74.7(2) Torsion angles C22-C23-C24-C29 77.9(14) C22-C23-C24-C25 174.6(4) C21-C22-C23-C24 177.0(16) C21-C22-C23-C24 172.9(6) a Cp*1 = C1-C5, Cp*2 = C11-C15, Ct = centroid. values for the Ct1-Y-Ct2 angle (139.3) and the average Y-Ct distances (2.36 ) are normal.77 The C5Me5 ligands are in a staggered conformation with a twist angle of 24.1. The Y-C(propargyl) bond lengths of 2.653(3), 2.529(3) and 2.560(3) are longer than that of reported Y-C -bond lengths, such as 2.44(2) in Cp*2YMe(THF)78a, 2.468(7) in Cp*2YCH(SiMe3)2.78b It is interesting to note that the smaller bond distance of Y-C23 (2.560(3) ) relative to Y-C21 (2.653(3) ) is reflected by the smaller yttrium-carbon coupling constant of the propargylic carbon (1JCH = 5.2 Hz) relative to that of the terminal carbon (1JCH = 11.9 Hz). Similar to Cp*2LaCH2CCPh (7a), Cp*2YCH2CCPh (7b) exhibits a nearly coplanar rearrangement of the C3 propargyl carbons and the metal center, albeit slightly more distorted (atom displacements from the least squares plane: Y 0.001, C21 0.005, C22 0.009, C23 0.005 ). Most of the observed differences corresponding to the propargyl/allenyl ligand of 7a and 7b are within experimental error, but there are two exceptions. Firstly, the C21-C22-C23 angle of 155.9(3) indicates that the C3 ligand is more bent the yttrium derivative. It seems natural to ascribe this to the smaller size of yttrium. Secondly, the phenyl group is almost coplanar with the plane defined by the C3 ligand and the metal in 7b, whereas the phenyl group is virtually perpendicular to the same plane in 7a. This difference may also be the result of the smaller size of the metal center, as the more proximate Cp* ligands are likely to force the phenyl group more into the equatorial girdle, but other effects resulting from Table 2-4. Selected bond distances () and angles () for 7a and 7b and related compounds CpZr(Me)(3CH2CCPh)4b (D), Cp*Zr(TBM)(3-CH2CCMe)4h (E), CpMo(2-MeCCPh)(3-CH2CCPh)63b (F) and [(PPh3)2Pt(3-CH2CCPh)]OTf 4f (G).

H H

M
Complex C1-C2 C2-C3 M-C1 M-C2 M-C3 7a 1.362(10) 1.234(10) 2.811(6) 2.695(7) 2.740(7) 159.3(7) 7b 1.366(4) 1.268(4) 2.653(3) 2.529(3) 2.560(3) 155.9(3) D 1.344(5) 1.259(4) 2.658(4) 2.438(3) 2.361(3) 155.4(3) E 1.376(7) 1.218(7) 2.498(3) 2.408(4) 2.444(5) F 1.404(9) 1.281(8) 2.278(6) 2.164(5) 2.105(5) 146.1(6) G 1.395(14) 1.227(13) 2.186(11) 2.150(8) 2.273(9) 152.2(9)

30

Rare-earth metallocene propargyl/allenyls conjugation and solid-state packing cannot be excluded. When comparing the structural data of 7a and 7b with that of other (phenyl-substituted) propargyl/allenyl complexes reported in literature (Table 2-4), it can be seen that they have similar structures. In all cases, the metal propargylic carbon (C1) distance is larger than the metal internal carbon (C2) distance. The M-C1 distance is larger than the M-C3 distance for CpZr(Me)(3-CH2CCPh), Cp*Zr(TBM)(3-CH2CCMe), CpMo(2-MeCCPh)(3-CH2CCPh) and smaller for [(PPh3)2Pt(3-CH2CCPh)]OTf. This observation may be related to the relative contribution of the allenyl or propargyl resonance structure in the 3-propargyl/allenyl complexes, but classification of the present 3-propargyl/allenyl rare-earth metal complexes in either of these two groups is hampered by the observed experimental error. The carbon-carbon distances in the 3propargyl/allenyl ligand are similar for all complexes, the C1-C2 distance being larger than the C2-C3 distance in all complexes, and the angle varies from 145 to 160, seemingly both as a function of metal radius and the coordination environment.

2.5.
2.5.1.

Reactivity of propargyl/allenyls
Introduction

Spectroscopic and structural analysis of the lanthanidocene propargyls has provided evidence for static 3-propargyl/allenyl structures. Further insight into the properties of this largely unexplored class of rareearth metallocenes was obtained by exploring their reactivity towards a variety of substrates. The documented reactivity of 3-propargyl/allenyl transition-metal complexes is dominated by the addition of nucleophiles to cationic complexes and the reported reactions may be classified as nucleophilic additions to the metal or the nucleophilic, regiospecific additions to the central carbon of the 3-propargyl/allenyl ligand (Scheme 2-17).3d-f When the nucleophile adds to the metal, 1-propargyl or 1-allenyl complexes are formed, depending on the nature of the nucleophile. These 1-propargyl and 1-allenyl complexes may display further reactivity, such as insertion of unsaturated reagents into the M-CH2 bond and the addition of electrophiles to the ligand.3a-c If the 1-tautomers are kinetically accessible, the reactive chemistry of present 3-propargyl/allenyl Scheme 2-17. Typical reactions of 3-propargyl/allenyl metal complexes.
X+ X [M] YH [M] X [M] X + [M] X .

[M]

complexes will most likely involve reactions of 1-propargyl and 1-allenyl derivatives, depending on their relative importance. It seems reasonable to expect that the 1-propargylic and 1-allenylic tautomer exhibit each distinct C-H activation and insertion reaction sequences, in analogy to rare-earth metalllocene 1-alkyl and 1-hydride derivatives. Examples of rare-earth metal complexes containing ligands undergoing a conversion from n- to 1-bonding include allyl Cp*2Ln(3-CH2CH=CHR) and tris(pentamethylcyclo-pentadienyl) derivatives Cp*3Ln. Permethyllanthanidocene allyls undergo insertion reactions with ethylene, CO and CO2 and the bonding of the allyl ligand is fluxional in solution.79,80 Although these findings indicate that access to an 1 form is facile in solution, the crystallographic studies obtained so far have shown that the allyl ligands adopt 3structures in the solid state. Similarly, the observed reactivity of Cp*3Sm with ethylene, H2, CO, PhCN and PhCNO is in agreement with that of an 1-alkyl, but no spectroscopic or structural evidence for a Cp*2Sm(1C5Me5) species has been observed in solution or in the solid state.81

31

Chapter 2 Scheme 2-18. The reactions of the lanthanidocene 3-propargyl/allenyl complexes with methanol-d4.
Ar Cp*2Ln CD3OD D Ar + Ar D

Ln = La, Ar = Ph (7a), C6H3Me2-2,6 (8a) Ln = Y, Ar = Ph (7b)

2.5.2. Methanol

The reactivity towards protic acids

When Cp*2YCH2CCPh (7b), Cp*2LaCH2CCPh (7a) and Cp*2LaCH2CCC6H3Me2-2,6 (8a) were allowed to reacted with a an excess of methanol-d4 (5-10-fold molar excess) in benzene-d6, the orange solutions turned yellow instantaneously. On the basis of NMR and GC-MS analysis, the organic reaction products were identified as the corresponding acetylenic ArCCCH2D and allenylic ArCD=C=CH2 deuterolysis products (Ar = Ph, C6H3Me2-2,6) (Scheme 2-18). Because the relative amount of allenylic and acetylenic quenchings products is expected to reflect the relative importance of the 1-propargylic and 1-allenylic tautomers in the present 3allenyl/propargyl complexes, analogous reactions with methanol were conducted for Cp*2LnCH2CCAr. The ratios of acetylenic and allenylic products were determined by in situ 1H NMR spectroscopy (using appropriate long pulse delays to avoid signal saturation under the present anaerobic conditions and long experiment times to allow for reliable signal-to-noise ratios) and are shown in Table 2-6. NMR spectroscopy indicated the following order of increasing propargylic character within the lanthanum series: Cp*2LaCH2CCC6F5 (10a) < Cp*2LaCH2CCPh (7a) < Cp*2LaCH2CCC6H3Me2-2,6 (8a) < Cp*2LaCH2-CCC6H3iPr2-2,6 (9a) (Section 2.4.3). Surprisingly, the relative amount of acetylenic quenching product did not increase with the degree of propargylic bonding in the above 3-allenyl/propargyl complexes Cp*2LnCH2CCAr. Instead, the relative amount of acetylenic quenching product was found to decrease with the steric size of the substituent of the 3-propargyl/alleny ligand. Moreover, the higher relative amount of acetylenic quenching product for 7a as compared to 7b suggests that the formation of acetylenic quenching products increases with the coordination space around the metal center under the assumption that the bonding in Cp*2LaCH2CCPh (7a) and Cp*2YCH2CCPh (7b) have a similar degree of propargylic character (Section 2.4.3). These results can plausibly be rationalized as proceeding via intial Lewis base coordination of methanol (Scheme 2-19). Nucleophilic attack at the metal center may take place at the substituted side or at the unsubstituted side of the 3-propargyl/allenyl ligand. Obviously, the latter mode of nucleophilic attack is sterically favored. The formation of the Lewis-base adducts is followed by the electrophilic attack of the

Table 2-6. Allene/alkyne ratios for the reaction of Cp*2LnCH2CCAr with methanol.a
Ar Cp*2Ln CH3OH Ar Ar +
.

Entry 1 2 3 4 5 a Reactions in spectroscopy.

Ln, Ar Y, C6H5 (7b) La, C6H5 (7a) La, C6H5Me2-2,6 (8a) La, C6H5iPr2-2,6 (9a) La, C6F5 (10a) benzene-d6 at room temperature. The ratios were

alkyne:allene 1.00:1.00 1.11:1.00 1.06:1.00 0.58:1.00 1.17:1.00 determined by in situ 1H NMR

32

Rare-earth metallocene propargyl/allenyls Scheme 2-19. The proposed mechanism for the reactions of the Cp*2Ln(3-CH2CCAr) complexes with methanol-d4.
CD3 O D Cp*2Ln CD3OD Ar J Ar CD3 O D Cp*2Ln C H H Ar Cp*2LnOCD3 + Ar D

Ln

Cp*2Ln O D3 C K D Ar

Ar C D Cp*2Ln O CD3

. +

Ar D

Cp*2LnOCD3

polarized proton on the nearest carbon of the 3-propargyl/allenyl ligand. As a result, the 1-propargyl-like Lewis base adduct J, formed from nucleophilic base attack at the unsubstituted side of the 3-propargyl/allenyl ligand, will give rise to the acetylenic quenching product upon protonolysis and the 1-allenyl-like Lewis base adduct K, formed from nucleophilic base attack at the substituted side of the 3-propargyl/allenyl ligand, will give rise to the allenylic quenching product. The formation of Lewis bases is well-documented for transition-metal 3propargyl/allenyl complexes,3d-f while initial Lewis base coordination has previously been proposed for the reactions of rare-earth metallocene alkyl derivatives Cp*2LnR (Ln = Sc102a, R = Me; Ln = Y82a, La82b, Ce82b, R= CH(SiMe3)2) with alcohols ROH.

Ph Cp*2La + PhC CH Ph LaCp*2 Cp*2La Ph 18a + Cp*2La Ph 19a

(2.2)

7a Ph Ph Ph + Ph + . + Ph

Phenylacetylene The above results revealed that Lewis base coordination of methanol at the unsubstituted side is only moderately favored over Lewis base coordination at the phenyl-substituted side of the 3-propargyl/allenyl ligand in Cp*2LaCH2CCPh (7a) and Cp*2YCH2CCPh (7b). This finding suggests that electronic effects are more important than steric effects in determining the side of nucleophilic attack at the metal center in the propargyl complexes Cp*2LnCH2CCPh. In this light, it was decided to study the effect of another Brnsted acid having different Lewis basic properties. The choice of phenylacetylene was based on the relatively high acidity of the acetylenic proton and the soft Lewis basic character of both the carbon-carbon triple bond and the phenyl group.83 Methanol, in contrast, is typically classified as a hard Lewis base, according to Pearsons hard-soft-acidbase (HSAB) principle.84 Reactions of Cp*2La(3-CH2CCPh) (7a) with phenylacetylene were performed in benzene-d6 at room temperature and their progress was monitored by in situ 1H NMR spectroscopy, using hexamethyldisiloxane (HMDSO) as an internal standard. When a stoichiometric amount of phenylacetylene was added to a benzene-d6 solution of 7a at room temperature, the light-orange solution turned immediately dark red and only 57% of the propargyl was converted upon complete consumption of phenylacetylene. The propargyl derivative (7a) was

33

Chapter 2

Scheme 2-20. The proposed reaction sequences for the reaction of 7a with phenylacetylene at room temperature.
Ph Cp*2La 7a . Cp*2LaC CPh Ph Ph Ph LaCp*2 18a Ph Ph Cp*2La Ph 19a Ph

Ph Ph + Ph Ph Ph

Ph

Cp*2La Ph

LaCp*2 20a

Ph Cp*2La Ph

converted into two new organometallic products that were identified as the butatrienediyl 18a (24%) and the but1-en-3-yn-1-yl derivative 19a (9%), on the basis of NMR spectroscopy (Chapter 4). The major organic products in the reaction mixture were identified as the protonated 3-propargyl/allenyl ligand (phenylallene and 1-phenyl1-propyne in a 1.00:0.20 ratio, respectively) and trans-1,4-diphenylbut-1-en-3-yne (eq. 2.2). Upon standing at room temperature the amount of phenylallene and 19a decreased under the formation of two unidentified lanthanocene species having Cp* 1H NMR resonances at 1.97 and 1.92 ppm, while the amount of 18a and 7a did not change significantly. When 7a was allowed to react with a two- and tenfold molar excess of phenylacetylene, 80% and 88% of 7a was converted, respectively, upon complete consumption of phenylacetylene. In both cases, phenylacetylene was converted into phenylallene, 1-phenyl-1propyne, (E)-1,4-diphenylbut-1-en-3-yne and traces of 2,4-diphenylbut-1-en-3-yne, according to NMR and GCMS analysis. The formation of phenylacetylene dimers and the observation of lanthanocene derivatives, previously observed in the lanthanocene-catalyzed oligomerization reaction of phenylacetylene, strongly suggest that the reaction of 7a with phenylacetylene gives rise to the monomeric, alkynyl derivative Cp*2LaCCPh which has been implicated as the active catalyst in the permethyllanthanocene-catalyzed oligomerization reaction of phenylacetylene (Chapter 4). Monomeric, permethyllanthanidocene alkynyl derivatives Cp*2LnCCR are unstable and dimerize into dinuclear, -alkynyl derivatives [Cp*2Ln(-CCR)]2. In some cases, these dimeric, alkynyl derivatives undergo subsequent C-C coupling to form butatrienediyl derivatives [(Cp*2Ln)2(RC=C=C=CR)].85 Examples of the latter two types of dinuclear compounds have been isolated and characterized, but no spectroscopic or structural evidence for monomeric, lanthanidocene alkynyl derivatives have been reported in literature. The present results reveal that Cp*2LaCCPh reacts faster with phenylacetylene than Cp*2La(3-CH2CCPh) (7a), thereby accounting for the incomplete conversion of 7a in the reaction with excess phenylacetylene. Another feature of the reaction of 7a with phenylacetylene concerns the reactivity of Cp*2LaC(Ph)=C(H)CCPh (19a) with phenylallene. The reactivity of allenes and alkenyl metal derivatives is relatively unexplored in organo rare-earth metal chemistry.86 Even so, both C-H activation and insertion reactions are plausible processes, in analogy to the well-documented reactive chemistry of rare-earth metallocene 1-alkyl and 1-hydride derivatives.8 Reactions of 19a with phenylallene, proceeding via insertion into the La-C bond, yield organic compounds C25H20 after protonolysis. GC-MS analysis did not supply evidence for the presence of such organic compounds after quenching reaction mixtures with H2O, thereby excluding the possibility that 19a reacts with phenylallene via insertion reactions.87 Nonregioselective C-H activation of phenylallene by Cp*2LaC(Ph)=C(H)CCPh (19a) yields, in principle, two new allenyl lanthanocene species and Cp*2La(3-CH2CCPh) (Scheme 2-21). This scenario accounts for the observation of two new lanthanocene species, having Cp* 1H NMR resonances at 1.97 and 1.92 ppm (vide supra). Unfortunately, the complexity of the reaction mixtures thwarted attempts to identify these

34

Rare-earth metallocene propargyl/allenyls

Scheme 2-21. The possible C-H activation reaction sequences of Cp*2LaC(Ph)=C(H)CCPh with phenylallene.
Ph Ph Cp*2La Ph Ph Ph Cp*2La Ph Ph Ph Cp*2La Ph + H . Cp*2La + . H Ph + Ph Cp*2La Ph + Ph Ph + . H Cp*2La Ph + Ph Ph Ph

proposed allenyl derivatives unequivocally and quenching experiments with H2O and D2O did provide experimental evidence to support this hypothesis. For the sake of convenience, these allenylic derivatives will be referred to as Cp*2LaX ( 1.92 ppm) and Cp*2LaY ( 1.97 ppm) in the proceeding discussion. Nnonselective reactivity of allenylic substrates towards alkyl d0 metal complexes has also been observed for isolectronic group 4 metallocene cations.88 In an effort to obtain additional evidence for the above proposed reaction sequences, reactions of Cp*2La(3-CH2CCPh) (7a) with phenylacetylene were performed in toluene-d8 at low temperatures and monitored by normalized, in situ 1H NMR spectroscopy. Phenylacetylene (1.1 equiv.) was condensed onto a toluene-d8 solution of 7a at -196 C and allowed to warm up to -60 C in the probe of the spectrometer. 1H NMR spectroscopy indicated quantitative conversion of 7a into the dimeric, bridged alkynyl derivative [Cp*2La(CCPh)]2 (20a) (one singlet for Cp* at 2.21 ppm, see Figure 2-6) and Cp*2LaC(Ph)=C(H)CCPh (19a) (one singlet for Cp* at 2.01 ppm), accompanied by the formation of phenylallene and 1-phenyl-1-propyne (in a 1.00:0.08 ratio, respectively). Slowly warming the reaction mixture to room temperature resulted in the conversion of the dimeric alkynyl derivative [Cp*2La(-CCPh)]2 (20a) into the butatrienediyl species [(Cp*2La)2(-3:3-PhC=C=C=CPh)] (18a) (one singlet for Cp* at 2.04 ppm which overlaps at low temperatures with the solvent, see Figure 2-6)

(a) (b) (c) (d) (e) (f) (g) (h) (i) (j) 2.20 2.10 2.00 1.90 ppm

Figure 2-6. Variable-temperature 500 MHz 1H NMR spectra of the Cp* region during the reaction of 7a and phenylacetylene at low temperature in C7D8 under the following conditions: (a, upper line) 7a at -60 C, (b) addition of phenylacetylene at -60 C, (c) -50 C, (d) -40 C, (e) -30C , (f) -20 C, (g) -10 C, (h) 0 C, (i) 10 C, (j, lower line) after 14 h at 25 C. See text for details.

35

Chapter 2

35 30

phenylallene 20a 18a Cp*2LaX

7a 19a Cp*2LaY

Concentration (mM)

25 20 15 10 5 0

2 -60 C

3 -50 C

4 -40 C

5 -30 C

6 -20 C

7 -10 C

8 0 C

9 10 C

10 25 C

11

Figure 2-7. Concentration profile of the reaction of 7a and phenylacetylene at low temperature in C7D8, followed in time during slowly warming to room temperature. Reaction conditions: (1) 7a at -60 C, 10 min; (2) 10 min after addition of phenylacetylene at -60 C; (3) after 10 min at -50 C; (4) after 10 min at 40 C; (5) after 10 min at -30 C; (6) after 10 min at -20 C; (7) after 10 min at -10 C, (8) after 10 min at 0 C, (9) after 10 min at 10 C, (10) after 10 min at 25 C; (11) after 14 h at 25 C. The lines connecting the data points are drawn as guides for the eye. All species were monitored by normalized, 500 MHz 1H NMR spectroscopy. and the reformation of 7a (one singlet for Cp* at 1.90 ppm). The C-C coupling reaction of 20a to yield 17a has been studied in more detail for reactions of Cp*2LaCH(SiMe3)2 and [Cp*2La(-H)]2 with phenylacetylene (Chapter 4). At higher temperature, the complete conversion of phenylallene was observed, accompanied by the conversion of 19a (one singlet for Cp* at 1.97 ppm) into Cp*2LaX ( 1.92 ppm) and Cp*2LaY ( 1.97 ppm). Finally, the slow formation of a new lanthanocene derivative having a Cp* 1H NMR resonance at 2.01 ppm was also observed. This derivative was also formed as a minor product in the low-temperature reactions of Cp*2LaCH(SiMe3)2 and [Cp*2La(-H)]2 with phenylacetylene and was tentatively assigned to an alkynyl derivative of unknown structure (Chapter 4). The rapid and quantitative conversion of Cp*2La(3-CH2CCPh) (7a) at low temperature followed by its reformation at higher temperature was unexpected and studied in more detail by a quantitative analysis. Scheme 2-22. The reaction sequences proposed to account for the observed reactivity of phenylallene during the reaction of 7a with phenylacetylene at low temperature and subsequent warming to room temperature.
Ph . Ph Cp*2La 7a Ph Cp*2LaC CPh + Ph Cp*2La Ph 19a Ph + . Ph Cp*2La 7a + Cp*2LaX + Cp*2LaY + Ph Ph Ph + Cp*2LaC CPh + Ph Ph LaCp*2 +

Cp*2La 20a

Ph Cp*2La Ph 19a

36

Rare-earth metallocene propargyl/allenyls Kinetic profiles of the reaction products were obtained by monitoring their concentration in time by means of normalized, in situ 1H NMR spectroscopy using an internal reference (hexamethyldisiloxane) and appropriate long pulse delays (Figure 2-7). The rapid and quantitative protonolysis of Cp*2La(3-CH2CCPh) (7a) by phenylacetylene affords Cp*2LaCCPh which dimerizes into 20a. The experimental data indicate that the formed dimeric bridged alkynyl derivative 20a reacts via two distinct processes at -30 C. The reaction of 20a with phenylallene competes with intramolecular C-C coupling of 20a to yield 17a. The kinetic profiles indicate also that the reaction of 20a with phenylallene affords 7a, selectively. The concomitant increase in concentration of 19a undubitately originates from the reaction of Cp*2LaCCPh and phenylacetylene, both formed from the reaction of of 20a with phenylallene (Scheme 2-22). In accord with previous results, the present results also indicated that Cp*2LaC(Ph)=C(H)CCPh (19a) reacts with phenylallene at 0 C, forming mainly Cp*2La(3-CH2CCPh) (7a), but at higher temperature also Cp*2LaX ( 1.92 ppm). At room temperature, both 19a and phenylallene were completely consumed within several hours, to yield among others trans-1,4-diphenylbut-1-en-yne and Cp*2LaY ( 1.97 ppm). Concluding remarks The reactions of Cp*2La(3-CH2CCPh) (7a) with phenylacetylene and methanol revealed that the relative amounts of propargylic and allenylic quenching products depend on the nature of the Lewis basic, Brnsted acid (i.e. methanol gave an allene-to-alkyne ratio of 1.00:1.11, while phenylacetylene resulted in a ratio of 1.00:0.20). Under the assumption that the reaction of 7a with phenylacetylene proceeds via initial alkyne coordination, the present data provide evidence that the tendency for Lewis base coordination at the sterically most hindered side of the 3-propargyl/allenyl ligand in 7a increases with the softness of the Lewis base. Circumstantial evidence for initial Lewis base coordination in the reactions of 7a with phenylacetylene is supplied by the observation of the Lewis base adduct Cp*2LaCCPh(PhCCH) in the permethyllanthanocenecatalyzed oligomerization of phenylacetylene (Chapter 4), coordination equilibria of alkynes and divalent lanthanidocenes Cp*2Ln (Ln = Sm, Eu, Yb) in benzene-d6 solution89 and alkene coordination prior to insertion into Ln-alkyl bonds.90 Based on the linear, rod-shape structure of phenylacetylene, it seems reasonable to assume that the steric requirements for Lewis base coordination of phenylacetylene are either similar or larger than that of methanol in 7a. In this light, the higher relative amount of allenylic quenching product for the reaction of 7a with phenylacetylene as compared to the reaction of 7a with methanol suggests that the side of nucleophilic attack is mainly determined by electronic factors rather than steric factors in the reaction of 7a with nucleophiles. The above results indicate that the relative amount of propargylic and allenylic quenching products is influenced by the reaction temperature as well (i.e. at room temperature the reaction of 7a with phenylacetylene produced an allene-to-alkyne ratio of 1.00:0.20, but at -60 C a ratio of 1.00:0.08 was found). If alkyne coordination takes place prior to protonolysis, it seems that the formation of the sterically most hindered Lewis base adduct, the 1-allenyl-like derivative of Cp*2La(3-CH2CCPh) (7a), is favored by relatively low reaction temperatures. The 1-allenyl-like derivative of 7a may therefore represent the kinetic product, while the 1propargyl-like derivative of 7a represents the thermodynamic product of the reaction of 7a with phenylacetylene. 2.5.3. Thermolysis

Prior to further reactivity studies, it was considered insightful to investigate the thermal stability of the present Cp*2LnCH2CCAr complexes. Thus, a toluene-d8 solution of Cp*2LaCH2CCPh (7a) was heated to 120 C. After two days, small signals appeared in the Cp* ( 2.05, 2.01, 1.99, 1.83 ppm) and aliphatic region ( 0.59, 0.45 ppm) which increased in intensity upon further heating. Also, a color change from light yellow to bright orange was observed after heating several days at 120 C. The propargyl NMR signals disappeared almost completely after 10 days. Attempts to identify the thermolysis products were unsuccesfull due to the high reactivity and instability of the products. The reaction mixture was quenched with methanol-d4 after complete conversion of 7a. Subsequent GC/GC-MS analysis indicated the presence of Cp*H-dn and two oligodeuterated isomers of C18H16, corresponding to the dimers of the phenyl-substituted propargyl ligand. The above results revealed that 7a is thermally quite robust. It seems natural to ascribe the observed thermal stability to the delocalized 3-bonding of the propargyl/allenyl ligand, providing the electrophilic metal center with both steric and electronic saturation. Several studies of the thermal stability of Cp*2LaCH(SiMe3)2 (6a) have been reported in literature. The decomposition of 6a in cyclohexane-d12 at 120 C was found to be relatively rapid (quantitative within 2 h) at 120 C via intramolecular methyl C-H activation, yielding highly

37

Chapter 2

Scheme 2-23. Proposed mechanism for the thermolysis of 7a.

H2 C H La CH2C CPh 120 C Ph 7a H2 C H CH2 La C C Ph La + Ph . La + Ph

La

dn La dn + C7D8-n dn

dn D La d5 D

reactive and unstable products which defied chracterization.91 The decomposition of 6a in toluene-d8 at 112 C produced the benzyl derivative Cp*2LaCD2C6D5 and CH2(SiMe3)2 within several hours (t1/2 = 3 h), both selectively and quantitatively.92 Evidence was presented that solvent metalation is preceded by the intramolecular C-H activation of the Cp* methyl group, affording the fulvene derivative Cp*(C5Me4CH25:2)La. The reversible nature of inter- and intramolecular C-H activation was put forward to account for the observed H/D exchange between solvent and Cp* ligand. Accordingly, thermal decomposition of 7a can be envisioned to occur by intramolecular metalation, forming the fulvene derivative Cp*(C5Me4CH2-5:2)La and the protonated 3-propargyl/allenyl ligand, possibly both 1-phenyl-1-propyne and phenylallene (Scheme 2-23). Subsequent, reversible solvent metalation accounts for the observed deuteration of the Cp* ligand. The formation of the observed dimers of the 3-propargyl/allenyl ligand can be explained by insertion of 1-phenyl-1-propyne and/or phenylallene into the La-C bond of 7a. The observation of only two isomers points to the regiorandom insertion of 1-phenyl-1-propyne into the La-CH2 bond of the 3-propargyl/allenyl derivative, however. Evidence for this view is supplied by the observed tendency of phenylallene to undergo C-H activation rather than insertion reactions with Cp*2LnR complexes (Section 2.5.2) and catalytic reactions of 7a with excess 1-phenyl-1-propyne. A detailed study of the latter process demonstrated that insertion of 1-phenyl-1-propyne into the La-CH2 bond of 7a represents the major reactive pathway under the present reaction conditions (Chapter 3).93 The fact that the dimers of the 3propargyl/allenyl ligand are oligodeuterated implies that the protonated 3-propargyl/allenyl ligand undergoes reversible transmetalation with the benzyl derivative Cp*2LaCD2C6D5 before insertion. Due to the complexity of the 1H NMR spectrum, only Cp*2La(3-CD2C6D5) was identified spectroscopically, based on its reported Cp* 1H NMR resonance at 1.82 ppm.92a,b The observation of 1H NMR resonances at 0.59 and 0.45 ppm are, furthermore, consistent with the presence of -metalated methyl groups.94 Heating mixtures of 7a and 7b for several days to 120 C in benzene-d6 gave similar organic products as observed for 7a in toluene (i.e. only the presence of oligodeuterated Cp*H and isomeric dimers of the protonated 3-propargyl ligand was indicated by GC-MS analysis after treating the reaction mixture with methanol). Quantitative 1H NMR analysis of the decrease of the propargylic CH2 proton resonance normalized against hexamethyldisiloxane (4.5 mM) in benzene-d6 solution indicated that the thermolysis of 7a was firstorder in 7a for at least 4 half-lives, as was evident from a linear plot of ln[7a]0/[7a]t versus time (R2 = 0.9889, t1/2 = 47.1 h).

38

Rare-earth metallocene propargyl/allenyls 2.5.4. The reactivity towards Lewis bases

Tetrahydrofuran The reactions of Lewis base with 3-propargyl/allenyl transition-metal complexes are known to give the Lewis base adduct of 1-propargyl or 1-allenyl derivatives, depending on the nature of the Lewis base.3d-f To investigate whether the present Cp*2LnCH2CCAr complexes exhibit an analogous reactivity, the reaction of Cp*2LaCH2CCPh (7a) with an equimolar amount of tetrahydrofuran (THF) was conducted in benzene-d6 at room temperature. No color change was observed and 1H NMR spectroscopy revealed a rapid and clean reaction. The upfield shifts of the THF 1H and 13C NMR resonances (e.g. -0.55 and -0.04 ppm for the - and -carbon, respectively, in the 13C NMR spectrum) are consistent with the coordination to the electrophilic metal center. The interpretation of the shifted propargyl NMR resonances is somewhat ambiguous, however. An increase in allenylic character is suggested by the downfield shift of the CH2 proton resonance ( +0.13 ppm), whereas the upfield shift of its CH2 carbon resonance ( -1.37 ppm) points to an increase in propargylic character (Table 2-7). In spite of this contradiction, an 3-propargyl/allenyl structure is presently favored, based on the similarity of the NMR spectral parameters with the crystallographically characterized [Cp*2La(3-CH2CCPh)(C5H5N)] (7apy) (Section 2.5.5). The complex 7aTHF is stable in benzene-d6 solution at room temperature, as no changes were observed in the 1H NMR spectrum after standing for two days. When the reaction mixture was heated to 50 C, the intensity of the NMR signals of the base adduct 7aTHF slowly decreased, giving rise to small vinylic 1H NMR resonances and several unidentified proton resonances in the Cp* region. After 24 h, no resonances attributable to 7aTHF were observed and the reaction mixture was quenched with methanol-d4. Subsequent GCMS analysis indicated the presence of Cp*D, THF and one unidentified dimer C18H16 of the protonated phenylsubstituted propargyl ligand (identical to C18H16 found in the thermolysis reaction of 7a, Section 2.5.3). The thermal lability of 7aTHF relative to the parent compound 7a is striking. The exact reason for this difference in stability is unknown at present. One possibility is that the additional ligand provides extra steric and electronic saturation to the metal center, thereby stabilizing the formation of an 1-propargyl (or 1-allenyl) structure in 7a. Addition of THF has, in fact, been found to facilitate the rapid 3-C3H5 to 1-C3H5 interconversion in Cp*2Nd(C3H5) and Cp*2Sm(C3H5).40 If a transient 1-propargyl or 1-allenyl species is formed, it may undergo intramolecular C-H activation with the C-H bond to the oxygen atom, with or without the intermediacy of Cp*FvLa(THF), forming Cp*2La(1-2-C4H4O) and CH3CCPh. The former process has been observed for Cp*2YMe(THF)57 and several ether adducts of lanthanidocene hydrides, Cp*2LnH(ROR) (Ln = Lu95, Y96, Ce96, La96), while the reaction of the samarium derivative Cp*2SmMe(THF) with benzene-d6 and toluene-d8 is reported to proceed via intramolecular Cp* C-H activation.97 Although a temperature effect cannot be ruled out, this rationalization also accounts for the observation of only one of the two previously observed dimers C18H16 of the 3-propargyl/allenyl ligand in the thermolysis reaction of 7a (Section 2.5.3).98 Apparently, the formation of one tautomer is kinetically much facile. The observed mixture of products is probably the result of several types of C-O activation processes, such as reported for Cp*2YH(THF)96 and (C5H4R)2YCH2SiMe3(THF) (R = H, Me).99 Nonetheless, no spectroscopic evidence for typical C-O cleavage products such as Cp*2LaOnBu or [Cp*2La(-OCH=CH2)]n was obtained by comparison with reported NMR data.96 Scheme 2-24. The reactions of 7a and 7b with Lewis bases.
O Cp*2La Ph 7a N Cp*2Ln Ph Ln = La (7a), Y (7b) + Cp*2Ln N Ph + Cp*2La O Ph

39

Chapter 2 Table 2-7. IR and NMR spectral parameters of Lewis base adducts.a 7aTHF 7apy 7a
1

7b 2.82 1.93 49.33 (159.1, 5.2) 155.09 106.79 (11.9)

7bpy 2.69 1.93 47.22 (157.2, 7.7) 151.58 104.88 (9.7)

H NMR 3.16 1.94 C NMR

CH2 Cp* CH2CC CH2CC CH2CC

2.82 1.91 54.00 (158.3) 152.19 112.75

2.95 1.97 52.63 152.90

b

13

51.07 (159.5) 152.61

b

IR 1944 1973 1923 2148 C C NMR spectra measured in C6D6 at 25 C. Chemical shifts in ppm and 1JCH and 1JYC coupling constants in Hz. IR spectra measured as Nujol mulls. b Not observed. c Complex not isolated.
a c

Pyridine A clean and quantitative reaction was also observed for the reaction of Cp*2LaCH2CCPh (7a) with 1 equivalent of pyridine. The formation of the corresponding adduct (7apy) is indicated by the upfield shifts of the pyridine proton ( -0.12, -0.17 and -0.19 ppm for -, -, and -CH, respectively) and carbon resonances ( 0.55, -0.30 and +1.77 ppm for -, -, and -CH, respectively) upon metal coordination. The changes of the propargyl 1H and 13C NMR resonances are similar as observed previously for 7aTHF, albeit slightly larger (Table 2-7). Again, the downfield shift of the CH2 proton resonance ( +0.34 ppm) suggests an increase in allenylic character, while the upfield shift of its carbon resonance ( -2.93 ppm) and the increase in 1JCH imply, conversely, an increase in propargylic character. An increase in propargylic character is also indicated by IR spectroscopy, revealing a shift of the symmetrical C C stretch from 1944 cm-1 to 1973 cm-1 (Figure 2-8). A single-crystal X-ray analysis study of 7apy established unambiguously that the bonding of the propargyl/allenyl ligand in 7apy is still tri-hapto, despite pyridine coordination (Section 2.5.5). The Lewis base adduct Cp*2LaCH2CCPh(C5H5N) (7apy) was obtained on a preparative scale by the reaction of Cp*2LaCH2CCPh (7a) with a small excess of pyridine (1.1 equiv) in a pentane solution at room temperature. Cooling a pentane solution of 7apy afforded yellow crystals in a high isolated yield (83%). The complex 7apy was found to be indefinately stable in the solid state at room temperature, while 1H NMR spectroscopy indicated that benzene-d6 solutions of 7apy remained unchanged at room temperature for at least seven days, but slowly underwent decomposition at 50 C. Thermal decomposition of 7apy in benzene-d6 solution at 50 C was accompanied by a color change from light-yellow to brown. The intensity of the pyridine and Cp* proton resonances decreased significantly within seven days at 50 C, but only one new signal at 3.20 ppm formed concomitantly. Subsequent heating to 80 C led to the formation of small vinylic signals and the complete disappearance of NMR signals due to 7apy within two days. Addition of methanol-d4, followed by GC/GC-MS analysis indicated the presence of 2,2bipyridine as the major organic product. Compounds with m/z 197-198 (a mixture of C14H15N and C14H14ND), 234 (two isomers of C18H14D2) and 333 were formed as minor products. The above results indicate that thermal decomposition of 7apy is complex and that the nature of the quenched decomposition products differ from that observed for the analogous reactions of 7a and 7aTHF. The masses of the observed quenching products provide evidence for the occurence of variety of C-C coupling processes. The broad 1H NMR resonance at 3.20 ppm and the presence of 2,2-bipyridine point to the formation of the paramagnetic species Cp*2La(2-2,2-bipyridyl) which is reported to be formed from Cp*2La(1-2-C5H4N)(C5H4N) upon heating at 50 C.100 However, the complexes Cp*2La(1-2-C5H4N) or Cp*2La(1-2-C5H4N)(C5H4N) were not observed spectroscopically by comparison with reported 1H NMR data. To investigate the effect of metal size on the bonding mode of the 3- propargyl/allenyl ligand upon Lewis base coordination, the reaction of Cp*2YCH2CCPh (7b) with pyridine in benzene-d6 was also studied. NMR spectroscopy indicated that the addition of 1 equiv. of pyridine to 7b led to the rapid and clean formation of the corresponding Lewis base adduct 7bpy. The upfield shifts of the CH2CC carbon resonances are larger than previously observed for 7apy (Table 2-7). The observed yttrium-carbon coupling of the propargyl and terminal carbon resonances indicate that the propargyl/allenyl ligand is still bound in an 3-fashion. Remarkably,

40

Rare-earth metallocene propargyl/allenyls

100

80

60

7a 7 a p y

40

20

0 3500 3000 2500 2000 1500

-1

1000

500

(c m )

100

80

60

7b 7 b p y

40

20

0 3500 3000 2500 2000 1500

-1

1000

500

(c m

Figure 2-8. Infrared spectra of 7a (upper spectrum, black line), 7apy (upper, grey), 7b (lower, black) and 7bpy (lower, grey) as nujol mulls. the CH2 1H NMR resonance moved upfield ( -0.19 ppm) rather than downfield as observed for 7a upon coordination of THF and pyridine. IR spectroscopy indicates a substantial increase of propargylic character, as evidenced by a shift of the C C stretch from 1923 cm-1 to 2148 cm-1 (Figure 2-8). The Lewis base adduct Cp*2YCH2CCPh(C5H5N) (7bpy) was prepared analogously to Cp*2LaCH2CCPh(C5H5N) and isolated in reasonable yields (63%) as an off-white thermolabile solid. Attempts to obtain single-crystals of 7bpy suitable for X-ray analysis failed. Freshly prepared 7bpy turned dark red within hours upon standing at room temperature. NMR and GC/GC-MS analysis indicated the formation of a multitude of decomposition products that defied characterization. In a closed atmosphere, 7bpy was found to be stable in benzene-d6 solution at room temperature for 7 days. Heating to 50 C, however, resulted in slow decomposition in a manner similar to that observed for 7apy. The same organic compounds were found with GC-MS analysis after quenching the reaction mixture with methanol-d4, while the formation a broad signal in the 1 H NMR spectrum at 4.20 ppm, accompanied by slowly disappearing signals due to 7apy, are consistent with the formation of the paramagnetic derivative Cp*2Y(2-2,2-bipyridyl).101 Many examples are reported in literature involving the reaction of rare-earth metallocene derivatives Cp*2LnR (e.g. Ln = Lu95, R = H, Me; Ln = Sc102a, R = Me, CH2C6H5, C6H5; Ln = Y102b, R = CH(SiMe3)2; Ln = Y66a, R = 1-2-C4H3O, 1-2-C4H3S) with pyridine. These reactions proceed via the initial formation of the corresponding Lewis base adduct, followed by transmetalation, forming Cp*2Ln(1-2-C5H4N) and RH. However, no spectroscopic evidence for the (transient) formation of Cp*2Y(1-2-C5H4N) or Cp*2Y(1-2-C5H4N)(C5H4N)

Transmittance (%)

Transmittance (%)

41

Chapter 2

Table 2-8. Selected bond lengths Cp*2LaCH2CCPh(C5H5N) (7apy). 7a C21-C22 C22-C23 C23-C24 La-C21 La-C22 La-C23 av. La-C(Cp*) av. La-C(Cp*) La-Ct1 La-Ct2 Ct1-La-Ct2 C21-C22-C23 C22-C23-C24 La-C21-C22 La-C23-C22 1.362(10) 1.234(10) 1.465(9) 2.811(6) 2.695(7) 2.740(7) 2.806(16) 2.780(16) 2.535(3) 2.509(4) 137.1(1) 159.3(7) 141.2(7) 71.0(4) 74.9(4)

()

and

angles

()

in

Cp*2LaCH2CCPh 7apy

(7a)

and

Bond lengths C26-C27 C27-C28 C28-C29 La-C26 La-C27 La-C28 av. La-C(Cp*) av. La-C(Cp*) La-Ct1 La-Ct2 La-N Bond angles Ct1-La-Ct2 C26-C27-C28 C27-C28-C29 La-C26-C27 La-C28-C27

1.356(4) 1.248(4) 1.454(4) 2.822(3) 2.726(3) 2.823(3) 2.841(7) 2.852(7) 2.571 2.584 2.743(2) 134.2 161.0(3) 142.8(3) 71.9(2) 72.7(2)

was found by comparison with reported 1H NMR data.103 The formation of unidentified material, accompanied by gradual coloration of the reaction mixture to deep red, has previously been noted for rare-earth metal pyridyl and phenyl complexes.104 2.5.5. The molecular structure of Cp*2La(3-CH2CCPh)(py)

Complex 7apy crystallized in the triclinic space group P1,2 with Z = 4. The solid-state molecular structure is depicted in Figure 2-9 and selected bond distances and angles are given in Table 2-8. The Cp*2La moiety of 7apy is similar to that of reported bent permethyllanthanocenes Cp*2La(X)(Y) complexes, having a Ct1-La-Ct2 angle (134.2) and average La-Ct1 (2.57 ) and La-Ct2 (2.58 ) distances.73 Crystallographically characterized formally nine-coordinated bent lanthanidocenes (C5R5)2LnL3 are relatively scarce in literature and the only other example is -to the best of our knowledge- represented by [Cp*2La(NCCH3)(DME)][BPh4].105 Although the geometry of 7apy may seem reminiscent of a bent metallocene with three equatorial ligands or a highly distorted trigonal bipyramid, neither description is entirely accurate. In fact, irregular geometries are common in the structural chemistry of lanthanide complexes, as ligand field effects are practically absent in lanthanide metals.1 The Cp* ligands are in an unusual eclipsed conformation with a twist angle of 2.85 . It has been suggested that the presence of eclipsed Cp* ligands is the result of steric crowding,106 but examples of structures with eclipsed rings in which steric crowding is not so obvious have been reported as well.107 The steric crowding in 7apy is evidenced by the La-N distance of 2.743(2) which is considerably longer than those reported for [Cp*2La(NCCH3)2(DME][BPh4]105 (2.60(2) ), Cp*2LaNHMe(NH2Me)108a (2.70(1) ) and [1,2,4(Me3Si)3C5H2]2LaI(C5H5N)108b (2.643(1) ). Steric crowding in [Cp*2La(NCCH3)2(DME)][BPh4] was indicated by nonequivalent La-N distances. The observed La-N distance of in 7apy is not exceptional, however, as it falls in the 2.70-2.80 range, reported for nonmetallocene lanthanum complexes.109 Comparing the crystal structures of 7a and 7apy reveals several changes upon pyridine coordination. Firstly, a distortion of the coplanar rearrangement of the metal center with the 3-propargyl/allenyl carbon backbone is observed (atom displacements from the least squares plane relative to 7apy: La +0.002, C26 +0.010, C27 +0.018, C28 +0.010 ). An apparent shift towards a more propargyl-like structure is also observed, as evidenced by the increased distances between lanthanum and the terminal carbon ( 0.08(1) ). This change is also reflected by the upfield shift of the CH2CC 13C NMR resonance ( -2.93 ppm), but is in contradiction with the observed downfield shift of the CH2CC 1H NMR resonance ( +0.34 ppm). These findings suggest that the propargylic carbon resonance is a more reliable structural indicator for the relative contribution of propargyl- and allenyl structures than the corresponding proton resonance. In order to accommodate pyridine at the metal center, the average La-Ct1 ( 0.04(2) ) and La-Ct2 distances ( 0.07(2) )

42

Rare-earth metallocene propargyl/allenyls

Figure 2-9. The molecular structure of 7apy with thermal ellipsoids drawn at the 50% probability level. The hydrogen atoms are omitted for clarity.

have increased relative to 7a, together with the Ct1-La-Ct2 angle ( 2.9(1)), as is expected for an increase in coordination number and concomitant steric crowding. Another interesting feature of the crystallographic structure of 7apy is the position of the pyridine ligand which is cis relative to the phenyl group and approximately coplanar with lanthanum and the carbons of the C3 backbone (atom displacements from the least squares plane: La -0.044, N -0.032, C26 -0.066, C27 0.054 ). Such a coplanar and cis rearrangement seems sterically unfavorable and may point to an electronic origin. In principle, quantum chemical investigations lend themselves as a valuable complement to experiments in order to understand the structure and reactivity of metal complexes.110 However, the presence of the large number of electrons in s, p, and f shells renders the quantum chemical calculations of lanthanide metal complexes less straightforward. Even so, several theoretical studies have shown the atomic 4f shell of the lanthanide atom is strongly stabilized and does not contribute significantly to the chemical bonding.111 The [Cp2Ln]+ frontier orbitals may therefore assumed to involve mostly 5d metal orbitals. The electronic structure of Cp2MLn complexes has been thoroughly analyzed in terms of interaction of the bent Cp2M moiety with the n ligands L.112 In C2v symmetry, the Cp2M fragment has three low-lying orbitals 1a1, b2 and 2a1 and two high-lying orbitals b1 and a2. The b1 and a2 orbitals are destabilized by strong interaction with the orbitals of the Cp rings. Only the three low-lying orbitals are capable of bonding with additional ligands L. The a1 orbital is mainly s, z and z2 in character, 1b2 mainly yz and y and 2a1 mainly y2-z2, but the exact contribution of the atomic orbitals to the Cp2M molecular orbitals varies with the metal and the Cp(centroid)-M-Cp(centroid) angle. Interestingly, Lauher and Hoffmann predicted coordination of carbon monoxide along the line perpendicular to the Cp(centroid)-M-Cp(centroid) plane for d0 Cp2MR2 complexes,

Scheme 2-25. A schematic presentation of the frontier orbitals of the bent metallocene Cp2Ln moiety and the LUMO of d0 Cp2MR2.
2a1 x Ln y 1a1 z 1b2 Cp Cp LUMO

R R

43

Chapter 2

Figure 2-10. The calculated LUMO (left) and SLUMO (right) of 7b. because this lateral approach was considered to provide good overlap between the lowest unoccupied molecular orbital (LUMO), approximately of a z2 type, and the carbon lone pair of CO (Scheme 2-25).112a Nucleophilic attack from lateral positions in the carbonylation reaction of zirconocenes was later corroborated experimentally.113 To determine the possible coordination site(s) of the Cp*2Ln(3-CH2CCPh) complexes, a molecular orbital calculation was performed on the model complex Cp2Y(3-CH2CCPh) in a geometry modeled after the crystal structure of Cp*2Y(3-CH2CCPh) (7b). Computational simplifications involving the replacement of lanthanide metals with yttrium and permethylated Cp* ligands with unsubstituted cyclopentadienyl ligands are well-established.114,112 The former simplification is justified by a large body of experimental work that demonstrates that yttrium behaves very similarly to most lanthanides.1 The electronic properties of the ligand, on the other hand, are believed to be only of importance in reactions involving a change of oxidation state.115 The calculations were carried out with the Turbomole program at the bp86/RIDFT level using the Turbomole SV(P) basisset on all atoms (small-core pseudopotential on Y).116 Electron donating reactants are expected to interact with the unoccupied molecular orbitals of 7b. The calculated lowest and second lowest unoccupied molecular orbital (SLUMO) are depicted in Figure 2-10. The position of the coordinated pyridine in the solid-state molecular structure of 7apy corresponds well with the spatial extension of the calculated LUMO and reveals that Lewis bases may coordinate in a trans (to the phenyl group) and coplanar rearrangement as well. The latter coordination mode seems actually less sterically hindered. These results provide additional support for the formation of 3-propargyl-like and 3allenyl-like Lewis bases, the proposed intermediates in the reaction of 7a with methanol and phenylacetylene (Section 2.5.2). It can be anticipated that Lewis base coordination of these protic acids at the position cis to the phenyl group results in the formation of the propargylic quenching product, while coordination trans to the phenyl group leads to the allenylic quenching product (Scheme 2-19). Following this line of reasoning, the regioselectivity of the protonolysis reaction of 7a appears to be related to the preferential nucleophilic attack at either the cis or trans lobe of the LUMO, but more experimental data is needed to support this hypothesis. Interestingly, a similar picture was also put forward to rationalize the reactions of [(PPh3)2Pt(3-CH2CCPh)]+ with nucleophiles.63c In this case, a theoretical study indicated that the nucleophilic addition reactions of simple donor nucleophiles to the platinum metal center proceed via a frontier-orbital controlled nucleophilic attack that utilizes a LUMO which is largely composed of the Pt 6pz atomic orbital. 2.5.6. Ethylene Attempted ethylene polymerization reactions with Cp*2La(3-CH2CCPh) (7a) and Cp*2Y(3-CH2CCPh) (7b) in toluene (1.6 mM) indicated that these complexes did not exhibit any reactivity towards ethylene (7.5 atm) after 0.2 h at 80 C. The reaction mixtures remained solutions without precipitation of polymer and only the expected quenching products (i.e. Cp*H, phenylallene and 1-phenyl-1-propyne) were observed with GC-MS after addition of methanol. The reactivity towards unsaturated substrates

44

Rare-earth metallocene propargyl/allenyls

Scheme 2-26. The proposed route towards the formation of Cp*2LaH as observed in the reaction of 7a with 1-hexene.
Ph Ph + Cp*2LaH

Cp*2La

Cp*2La 7a Ph

120C

Ph Ph Cp*2La + Cp*2LaH

1-Hexene In accord with above results involving ethylene, no reaction was observed with 1H NMR spectroscopy after heating a reaction mixture of 7a and 7b (20 mM) and excess 1-hexene (100-150 equiv.) for 4 days at 100 C in benzene-d6. When the reaction temperature was increased to 120 C, slow substrate conversion ws observed for 7a, but not for 7b. The reaction mixture of 7a and 1-hexene (145 equiv.) was monitored with 1H NMR spectroscopy for 74 days at which point 64% of the substrate was converted and quenched with methanol. GC-MS analysis indicated the presence of cis- and trans-hex-2-ene, hexane and derivatives of 1-hexene dimers (e.g. methyl-substituted undecenes and undecanes), trimers (m/z 250) and tetramers (m/z 334). Unidentified organic compounds having masses consistent with derivatives from the cross-coupled product between 1-phenyl1-propyne and 1-hexene were also found. The absence of substituted derivatives of the pentamethylcyclopentadiene ligand argues against the involvement of fulvene derivatives, while the formation of 1-hexene isomers, hexane and both saturated and unsaturated 1-hexene oligomers strongly suggest that a lanthanum hydride species is formed. The monomeric hydride derivative Cp*2LaH is a well-known and highly active catalyst for alkene hydrogenation.53b,d Because these reactions are kinetically very facile, the generation of such a hydride derivative in the present reaction mixture must be very slow. A plausible route towards Cp*2LaH involves -H elimination after 1,2- or 2,1insertion of 1-hexene into the metal-carbon bond of the 3-allenyl/propargyl derivative. Unfortunately, the present experimental data do not allow differentiation between insertion into the propargylic La-CH2 bond or the allenylic La-CPh bond of 7a. Because the insertion of 1-phenyl-1-propyne into the propargylic La-CH2 bond represents the major pathway in the reaction of 7a with 1-phenyl-1-propyne (vide infra), the former insertion reaction is presently favored (Scheme 2-26). In view of the well-documented high reactivity of Cp*2LaH and the observed slow consumption of 1-hexene, either insertion of 1-hexene or -H elimination must be rate-limiting. As NMR resonances of 7a were still present after several days and no major organometallic intermediates were observed with NMR spectroscopy, slow insertion is presently favored. In spite of the higher degree of propargylic bonding in Cp*2La(3-CH2CCC6H3Me2-2,6) (9a) as compared to Cp*2LaCH2CCPh (7a) (Section 2.4.3), the behavior of 9a towards 1-hexene was similar to that of 7a. Slow conversion of 1-hexene into analogous compounds was observed after heating a benzene-d6 solution of 9a in the presence of an excess of 1-hexene for several days at 100 C. 1-Phenyl-1-propyne When a reaction mixture of Cp*2La(3-CH2CCPh) (7a) was heated to 80 C in the presence of an excess of 1-phenyl-1-propyne (10-25 equiv) slow consumption of 1-phenyl-1-propyne was observed. NMR analysis of the reaction mixture pointed to the catalytic cyclodimerization reaction of 1-phenyl-1-propyne into mainly (E)-3-benzylidene-2-methyl-1-phenylcyclobutene (eq. 2.3). Catalytic cyclodimerization of alkyl- and silyl-substituted methylacetylenes was previously observed for Cp*2LnCH(SiMe3)2 (Ln = La, Ce), while a catalytic cycle based on a kinetic and mechanistic study is given in Chapter 3. The reaction mixtures obtained with 7a were identical to those observed for the reaction of Cp*2LaCH(SiMe3)2 (6a) and [Cp*2La(-H)]2 (12a) with excess 1-phenyl-1-propyne (Section 2.3.2). Interestingly, the major products were found to originate from

45

Chapter 2 substrate insertion into the propargylic La-CH2 bond of 7a, while the minor products originated from substrate insertion into the allenylic La-CH(Ph) bond.

Cp*2LaR 5 mol%

Ph +

Ph Ph + Ph Ph

(2.3)

Ph

C6D6 100 C

Ph

R = CH(SiMe3)2, CH2CCPh, H

Aspects concerning the influence of the reaction temperature, the lanthanide metal and the catalyst structure on the rate and selectivity of the catalytic cyclodimerization reaction are discussed in Chapter 3. It should be noted that no catalytic cyclodimerization of 1-phenyl-1-propyne was observed with Cp*2YCH2CCPh (7b).

(2.4)

Cp*2LaCH2CCPh 5 mol% + C 6D 6 100 C +

unidentified products

1-(2-Methylphenyl)-1-propyne Catalytic cyclodimerization was also observed for 1-(2-methylphenyl)-1-propyne, when a mixture of Cp*2LaCH2CCPh (7a) was heated to 100 C in the presence of an excess of 1-(2-methylphenyl)-1-propyne (1025 equiv) (eq. 2.4). The rate of substrate consumption was considerably lower and the reaction products were also formed less selectively as observed for the reactions of 7a with 1-phenyl-1-propyne, however. Further details are discussed in Chapter 3.

Ph

(2.5)

Cp*2La 7a

+ 3

Cp*2La 9a

+ Ph

1-(2,6-dimethylphenyl)-1-propyne In contrast to the reaction of Cp*2La(3-CH2CCC6H5) (7a) with 1-(2-methylphenyl)-1-propyne, NMR spectroscopy did not indicate the formation of oligomers of 1-(2,6-dimethylphenyl)-1-propyne (3) upon heating a benzene-d6 solution of 7a to 100 C in the presence of 3 (20 equiv.). Instead, transmetalation of 7a with 3 took place, as evidenced by the formation of Cp*2La(3-CH2CCC6H3Me2-2,6) (9a) and 1-phenyl-1-propyne (Eq. 2.5). Upon further heating, both 1-phenyl-1-propyne and 7a were completely converted into (E)-3-benzylidene-2phenyl-1-methylcyclobutene and 9a, respectively. The reaction mixture was quenched after 10 days with methanol-d4 and GC-MS analysis indicated the presence of trace amounts of cross-coupling products of 3 and 1-

46

Rare-earth metallocene propargyl/allenyls

Scheme 2-27. The reaction of 9a with 3 and 1.

Cp*2La 9a

+ 3 Ph

Cp*2La 9a

+ 3

Ph 1

Cp*2La 9a

+ Ph

phenyl-1-propyne (m/z 260) and coupling products of 3 (m/z 288), besides the expected organic compounds formed upon deuterolysis (i.e. Cp*D, 1-2,6-dimethylphenyl-1-propyne-dn, 2,6-dimethylphenyl-propadiene-d1 and isomers of (E)-3-benzylidene-2-phenyl-1-methylcyclobutene). The formation of (E)-3-benzylidene-2-phenyl-1-methylcyclobutene reveals that insertion of 1-phenyl1-propyne into the propargylic La-CH2 bond of 7a took place, forming the cyclodimer after subsequent intramolecular alkyne insertion and protonolysis by 1-phenyl-1-propyne or 1-(2,6-dimethylphenyl)-1-propyne (3), as previously proposed for the lanthanocene-catalyzed cyclodimerization reaction of 1-phenyl-1-propyne (Chapter 3). Because the electronic properties of 1-phenyl-1-propyne and 3 are comparable (Section 2.3.2), the above results suggest that the insertion of 3 into 7a is inhibited for sterical reasons. The large amount of (E)-3benzylidene-2-phenyl-1-methylcyclobutene relative to the cross-coupling product of 1-phenyl-1-propyne and 3 reveals, in addition, that the insertion of 1-phenyl-1-propyne into 7a is rapid relative to the insertion of 1-phenyl1-propyne into 9a. This observation may be attributed to slow transmetalation of 7a with 3, producing only a small amount of 9a relative to 7a, and/or slow insertion of 1-phenyl-1-propyne into 9a. To discriminate between both possibilities, analogous reactions with 9a were conducted. NMR spectroscopy provided evidence that the bonding in Cp*2LaCH2CC(C6H3Me2-2,6) (9a) displayed a higher degree of propargylic character than that in Cp*2LaCH2CCPh (7a). To determine whether an increased propargylic character results in a higher tendency to undergo insertion reactions for the present arylsubstituted 3-propargyl/allenyl complexes Cp*2LaCH2CCAr, the reactivity of 9a was investigated towards 1phenyl-1-propyne and 1-(2,6-dimethylphenyl)-1-propyne (3). When a benzene-d6 solution of 9a was allowed to react in the presence of a 20-fold molar excess of 3, no reactivity was observed after heating several days at 100 C by 1H NMR spectroscopy. The addition of 1-phenyl-1-propyne (1 equiv.), followed by subsequent heating to 100 C, led to its slow conversion into (E)-3-benzylidene-2-phenyl-1-methylcyclobutene (Scheme 2-27). Upon complete conversion of 1-phenyl-1-propyne, the reaction mixture was quenched with methanol-d4. Subsequent GC-MS analysis indicated among others the presence of trace amounts of homo- and cross-coupling products of 3 (m/z 288 and 260). These findings indicate that that insertion of both 1-phenyl-1-propyne and 1-(2,6-dimethylphenyl)-1propyne (3) into Cp*2LaCH2CC(C6H3Me2-2,6) (9a) does not take place. Apparently, the steric constraints posed by the additional 2,6-dimethyl groups in the phenyl-substituted propargyl impede insertion. Similarly, 2,6dimethyl substitution at the phenyl group of the substrate seems to hinder substrate insertion into 7a as well. Thus, catalytic cyclodimerization of 1-arylalk-2-ynes is found to be quite sensitive towards ortho substitution of the simple 1-phenyl-1-propyne system. No insertion chemistry was observed upon substituting the phenyl group of either the 3-propargyl/allenyl ligand or the substrate with two ortho-methyl groups. 1-Pentafluorophenyl-1-propyne When Cp*2La(3-CH2CCPh) 7a was heated to 80 C in benzene-d6 in the presence of a 20-fold molar excess of 1-pentafluorophenyl-1-propyne (5) and monitored with 1H NMR spectroscopy, no significant consumption of 5 was observed after 20 days. Instead, slow conversion of Cp*2La(3-CH2CCPh) into Cp*2La(3-CH2CCC6F5) (11a) took place. This conversion was complete within 48 h at 80 C during which new Cp* proton resonances appeared, accompanied by the formation of several vinylic and aliphatic proton resonances. Further heating led to the decomposition of 11a, as observed previously for the reaction of Cp*2LaCH(SiMe)2 (6a) and 5 (Section 2.3.2). Quenching the mixture with methanol-d4, followed by GC-MS analysis, indicated the presence of 1-pentafluorophenyl-1-propyne-dn, pentafluorophenylpropadiene-d1, Cp*D, 1-

47

Chapter 2

Scheme 2-28. The proposed reaction sequences to account for the products observed for the reaction of 7a with 5.
Cp*2La 7a Ph + C6F5 5 Cp*2La 11a C6F5 C6F5 Cp*2La + C6F5 5 Cp*2La C6F5 1 1 C6F5 C6F5 Cp*2La Ph C6F5 CD3OD C6F5 C6F5 Cp*2La Ph CD3OD C6F5 C6F5 Cp*2La Ph CD3OD C6F5 D Ph C6F5 D Ph C6F5 D Ph D Ph C6F5 Cp*2La Ph CD3OD C6F5 C6F5 C6F5 + Ph 1

C6F5 + Cp*2La C6F5

11a C6F5

C6F5

C6F5

phenyl-1-propyne and small amounts of several organic compounds of which the major products were identified as C18H5DF10 (m/z 413), consistent with the mass of a dimer of 5, and four isomers of C27H13DF10 (m/z 529), consistent with a trimer composed of two 5 units and one 1 unit. Heating a similar mixture to 120 C resulted in a more rapid decomposition of 11a, while C18H5DF10 was the only coupling product observed with GC-MS. These results can be rationalized by the reaction sequences depicted in Scheme 2-28. Rather than insertion into the La-C bond of 7a as observed for 1-phenyl-1-propyne (1), 1-pentafluorophenyl-1-propyne (5) undergoes transmetalation with 7a. The four observed isomers of C27H13DF10 can be explained by regiorandom insertion of 5 and 1 into 11a. The observation of only one isomer of C18H5DF10 suggests that insertion of 5 into 11a is more rapid than insertion of 1-phenyl-1-propyne into 11a. This observation implies as well that one of the two regioisomeric products formed upon insertion of 5 into 11a is more reactive towards 1-phenyl-1-propyne than the other and that the more reactive regioisomeric product is completely converted by 1-phenyl-1-propyne. Because the structure of the products could not be determined unequivocally by NMR spectroscopy, it is not known which of the two regioisomeric insertion products of 5 into 11a is the least reactive towards 1-phenyl-1propyne and produces C18H5DF10 upon deuterolysis. The exact reason for the reluctance of 5 to undergo insertion reactions with 7a as compared to 1 is unknown at present. The similar steric requirements of the pentafluorophenyl group as compared to the phenyl group argue for electronic reasons, however.117 Also, the absence of intramolecular alkyne insertion of the insertion product of 11a and 5 is not understood presently, mainly because the precise effect of the increased electron-withdrawing nature of the alkyne substituent (C6F5 versus C6H5) on the stability of the alk-1-ene-4-yn-1yl species is difficult to assess a priori.118 Diphenylethyne As 1-phenyl-1-propyne was found to undergo insertion into the metal-carbon bond of Cp*2LaCH2CCPh (7a), it was considered interesting to explore the reactivity of 7a towards diphenylethyne as well. Because the steric size of the methyl group is smaller than that of the phenyl group,49 the steric requirements for insertion of diphenylethyne into 7a are somewhat higher than that of 1-phenyl-1-propyne, though.

48

Rare-earth metallocene propargyl/allenyls

Scheme 2-29. The observed reactivity of 7a towards CO and diphenylethyne, in the presence and absence of 1-phenyl-1-propyne.
no reaction Ph Ph Cp*2La 7a Ph Ph
CO
CO

no reaction

Ph

Ph

Ph Ph

catalyst decomposition Ph

A benzene-d6 solution of 7a and diphenylethyne (1.2 equiv) was heated for several days to 100 C and the progress of reaction was monitored with 1H NMR spectroscopy. As the resonances of 7a slowly disappeared, resonances previously observed in the thermolysis of 7a in benzene-d6 solution appeared, while the total amount of proton intensities in the aliphatic and vinylic region decreased as well. Analogous to the thermolysis reaction, the light-yellow solution changed into red-orange during heating. The reaction mixture was quenched with methanol after 20 days and GC/GC-MS analysis indicated the presence of small amounts of 1phenylprop-1-ene-dn and phenylallene-dn, pentamethylcyclopentadiene-dn and several isomeric and oligodeuterated compounds (e.g. m/z 232, 261 and 429) of which trans-diphenylethene and an unknown compound (m/z 429) were identified as the major products. Although the details of the formation of the observed products are not known presently, the masses of the observed organic products are inconsistent with products originating from insertion of diphenylethyne into 7a. It is believed that the observed products are the result of reactions involving a fulvene Cp*(C5Me4CH25:2)La or phenyl derivative Cp*2LaC6D5 both formed upon thermolysis of 7a (Section 2.5.3). Evidence for this view comes from the multitudinous incorporation of deuterium atoms in the observed quenching products, indicating extensive H/D scrambling. To investigate the possibility of a cross-coupling reaction between 1-phenyl-1-propyne and diphenylethyne, a reaction mixture of 7a with a 20-fold molar excess of both diphenylethyne and 1-phenyl-1propyne was heated for several days at 100 C. NMR spectroscopy indicated that diphenylethyne was not consumed and that products from catalytic cyclodimerization of 1-phenyl-1-propyne formed exclusively. Carbon monoxide Many examples of facile carbon monoxide insertion into the metal-carbon bond of rare-earth metallocene 1-alkyl derivatives exist in literature.119 The fact that CO insertion was also observed for n-bound (n > 1) rare-earth metal complexes, such as allyl Cp*2Ln(3-CH2CHCH2),79 ortho-pyridyl Cp*2Y(2-C,NC5H4N)120 and tris(pentamethyl-cyclopentadienyl) derivatives Cp*3Ln,81 encouraged us to explore the reactivity of the present 3-propargyl/allenyl derivatives Cp*2Ln(3-CH2CCAr) towards carbon monoxide as well. When an excess of carbon monoxide (1 atm.) was applied to a benzene-d6 solution of Cp*2La(3CH2CCPh) (7a), no reactivity was observed with 1H NMR spectroscopy after 5 days at room temperature. A multitude of Cp* 1H NMR resonances formed slowly upon heating to 100 C, accompanied by a color change from light-yellow to red-orange. After 7 days at 100 C, only ~5% of 7a was converted. No evidence for CO insertion into 7a was obtained by means of NMR spectroscopy and GC/GC-MS after quenching the mixture with methanol, however. Similar results were also obtained for analogous reactions of Cp*2Y(3-CH2CCPh) (7b). It appears that the 3-propargyl/allenyl bonding of both 7a and 7b is too strong to allow insertion of CO into the metal-carbon bond at room temperature. Nonselective reactivity was observed upon increasing the reaction temperature, but the details of these reactions are unknown at present. As 7a was found to react only slowly with excess CO, even at high temperatures, the possibility of trapping an intermediate of the catalytic cyclodimerization reaction of 1-phenyl-1-propyne was investigated. The intermediates proposed are alkenyl derivatives (Chapter 3) whose (reported) reactivity is relatively unexplored in organolanthanide chemistry.121 The application of CO (1 atm) to a reaction mixture containing 7a and 1-phenyl1-propyne (20 equiv.) led after heating to 100 C to the decomposition of 7a into several unidentified species, as evidenced by the decrease of the 7a proton resonances and the concomitant increase of numerous proton

49

Chapter 2 resonances in the Cp* region. Complete conversion of 7a was observed within 3 days at 100 C. No evidence for CO insertion was found with 1H/13C NMR and GC/GC-MS after quenching the mixture with methanol.

2.6.

Conclusions

Novel rare-earth metallocene 3-propargyl/allenyl derivatives have been prepared via the reactions of the alkyl derivatives Cp*2LnCH(SiMe3)2 with the corresponding 1-aryl-1-propynes CH3CCAr. The synthetic utility of this route towards the Cp*2LnCH2CCAr complexes is limited by the slow rate of propargylic metalation. The reactions of the hydride derivatives [Cp*2Ln(-H)]2 derivatives with 1-aryl-1-propynes CH3CCAr also afforded the desired Cp*2LnCH2CCAr complexes in most cases. However, these reactions were less selectively and the high solubility of Cp*2LnCH2CCAr complexes hindered their isolation from the reaction mixtures in some cases. Spectral and structural analysis provided evidence for a static 3-bonding description of the propargylic/allenylic ligand in Cp*2LnCH2CCAr (Ln = La, Y), but a shift towards more 1-propargylic and more 1-allenylic bonding was observed for Cp*2LaCH2CCAr complexes, depending on the aromatic substituent of the 3-propargyl/allenyl ligand. The former shift was favored by more sterically hindered substituents, while the latter shift was promoted by more -electron-withdrawing substituents. The reactivity of the present 3-propargyl/allenyl complexes Cp*2LnCH2CCAr was found to involve the formation of Lewis base adducts, C-H activation reactions with Brnsted acids and insertion reactions of unsaturated substrates. The reactions of Cp*2LnCH2CCPh (Ln = La, Y) with Lewis bases, such as THF and pyridine, yielded the corresponding Lewis base adduct, both rapidly and selectively, in which the propargyl/allenyl ligand was still bound in a 3-fashion. Reactions of the Cp*2LnCH2CCAr complexes with Brnsted acids furnished both acetylenic (CH3CCAr) and allenylic (CH2=C=CHAr) protonolysis products. The insertion reactivity of Cp*2LnCH2CCPh towards unsaturated substrates was severely limited by the strong 3bonding of the propargylic/allenylic ligand. Even so, insertion of 1-phenyl-1-propyne into Cp*2LaCH2CCAr was implicated by its catalytic conversion into cyclodimers, as observed for the reaction of Cp*2LaCH2CCAr (or its catalyst precursors) with 1-phenyl-1-propyne. Catalytic cyclodimerization seems to be confined to reactions of Cp*2LnR (R = CH(SiMe3)2, H, 3-CH2CCPh) complexes having relatively large metal centers (Ln = La, Ce) in combination with substrates having sterically relatively unhindered carbon-carbon triple bonds (Ar = Ph, C6H4Me-2).

2.7.

Experimental Section

General considerations. All reactions and manipulations of air and moisture sensitive compounds were performed under a nitrogen atmosphere using standard Schlenk, vacuum line and glovebox techniques. Deuterated solvents were dried over Na/K alloy prior to use. Other solvents were dried by percolation over columns of aluminum oxide, BASF R3-11 supported Cu oxygen scavenger, and molecular sieves (4) (pentane), or by distillation from Na/K alloy (THF, cyclohexane). The compounds Cp*2LnCH(SiMe3)2 (Ln = La122, Y123), [(Cp*2La(-H)]2122, 2,6-dimethyliodobenzene13, n-butyl nitrite124, copper bronze125, copper(I) iodide126, propynyllithium127 and Pd(PPh3)2Cl2128 were prepared according to literature procedures. Reagents were purchased from Aldrich, Acros Chimica, Strem, Merck or Fluka and were used as received unless stated otherwise. 1-Phenyl-1-propyne (Aldrich) was dried over CaH2 before use (vide infra). Diphenylethyne (Aldrich) was sublimed before use. Dimethyl sulfoxide was dried at least three times on freshly activated 4- molecular sieves.129 Physical and analytical measurements. NMR spectra were recorded on a Varian VXR-300 (FT, 300 MHz, 1H; 75 MHz, 13C), Varian XL-400 (FT, 400 MHz, 1H; 100 MHz, 13C) or a Varian Inova 500 (FT, 500 MHz, 1H; 125.7 MHz, 13C) spectrometers. NMR experiments on air-sensitive samples were conducted in flamesealable tubes or tubes equipped with a Teflon valve (Young). The 1H NMR spectra were referenced to resonances of residual protons in deuterated solvents and reported in ppm relative to tetramethylsilane ( 0.00 ppm). The 13C NMR spectra were referenced to carbon resonances of deuterated solvents. 19F NMR spectra were referenced externally to hexafluorobenzene in CDCl3 ( -163.0 ppm). IR spectra of pure compounds, KBr pellets or nujol solutions of the samples were recorded on a Mattson 4020 Galaxy FT-IR spectrophotometer. The

50

Rare-earth metallocene propargyl/allenyls elemental analyses were performed by H. Kolbe, Mikroanalytisches Laboratorium, Mlheim an der Ruhr, and the Microanalytical Department of the University of Groningen. GC analyses were performed on a HP 6890 instrument with a HP-1 dimethylpolysiloxane column (19095 Z-123). GC-MS spectra were recorded at 70 eV using a HP 5973 mass-selective detector attached to a HP 6890 GC as described above. General drying procedure for 1-methylalk-2-ynes. The liquids obtained after synthesis or received after purchase were brought in a flask containing freshly ground CaH2 and allowed to react for at least one day at 50 C under nitrogen. Vacuum transfer afforded colorless oils which were stored at -30 C and under nitrogen. 2-Propynyltoluene (2). A 1-L, three-neck, round-bottom flask, equipped with stir bar and gas-inlet adaptor, was charged with THF (200 mL) and degassed by freeze-thaw-pump cycles. A 3-L flask was connected to a lecture bottle of propyne and filled with an appropriate amount of propyne gas (~122 mmol) by adjusting the pressure. From this flask, the gas was allowed to condense into the reaction vessel which was cooled to -100C with a liquid nitrogen-ethanol bath. The 1-L flask was brought under nitrogen, connected to a cooler and drop funnel, while still maintaining the temperature of the bath at ca. -100C. Then n-BuLi (42.0 mL, 105 mmol, 2.5 M in hexanes) was added dropwise under stirring over a period of 1 h after which the temperature of the bath was allowed to warm up to 78 C. A solution of ZnBr2 (22.6 g, 100 mmol, flame-dried in vacuo) in 100 ml of THF was added dropwise to the reaction mixture at -78 C during 30 min while strirring. After addition the cooling bath was removed and the reaction mixture was allowed to warm up to room temperature. Then, 2iodotoluene (8.6 ml, 67 mmol) and Pd(PPh3)Cl2 (2.3 g, 3.3 mmol) were added and the suspension was stirred for 14 h at room temperature. The reaction mixture was quenched with 200 mL of brine and filtered. The organic layer was extracted with petroleum ether (40-60 C) and dried over MgSO4. Rotatory evaporation afforded an orange oil which was purified by means of column chromatography (silica, 230-400 mesh, 60 ) with petroleum ether. The volatiles were removed by means of rotatory evaporation and vacuum distillation using a 20-cm Vigreux column. Drying over CaH2 and subsequent vacuum transfer afforded a light yellow liquid. Yield: 7.34 g (85%). 1 H NMR (300 MHz, CDCl3, 25 C): 2.15 (s, CCCH3, 3 H), 2.43 (s, CH3, 6 H), 6.9-7.2 (m, CH, 3 H). 13C NMR (75 MHz, CDCl3, 25 C): 4.46 (CCCH3), 21.03 (CCH3), 77.24 (CH3C), 94.03 (ArC), 123.74 (iC, Ph), 126.48 (m-C, Ph), 126.82 (p-C, Ph), 140.04 (o-C, Ph). GC-MS, m/z (relative intensity): 131 (11), 130 (M+, 100), 129 (M+ - H, 61), 128 (62), 127 (28), 116 (5), 115 (M+ - CH3, 78), 102 (8), 89 (6), 77 (8), 75 (4), 74 (5), 64 (8), 63 (11), 62 (4), 51 (11), 50 (5), 39 (5). IR (neat, [cm-1]): 3066 (m), 3022 (m) 2917 (m), 2853 (m), 2734 (w), 2249 (w), 1600 (w), 1487 (m), 1456 (m), 1377 (m), 116 (m), 1045 (m), 765 (s), 493 (s). Anal. Calcd. for C10H10 (130.19): C, 92.26%; H, 7.74%. Found: C, 92.12%; H, 7.63%. 1,3-Dimethyl-2-(prop-1-ynyl)benzene (3). A 250-mL Schlenk flask was charged with 13.0 mL (32.5 mmol) of n-BuLi (2.5 M in hexanes) and 30 mL of THF. A 3-L flask was connected to a lecture bottle of propyne and filled with an appropriate amount of propyne gas (40 mmol) by adjusting the pressure. From this flask, propyne was allowed to condense into the Schlenk flask which was cooled to -100C with a liquid nitrogen-ethanol bath. After 15 min of stirring a solution of 7.70 g (34.2 mmol) of anhydrous ZnBr2 in 20 mL of THF was added. The mixture was stirred for 10 min at 78 C. Then 5.0 g (22 mmol) of 2,6dimethyliodobenzene13 and a solution of 0.78 g (1.1 mmol) of Pd(PPh3)Cl2 and 0.90 mL (2.2 mmol) of n-BuLi in 10 mL of THF are added and the mixture is allowed to warm up to room temperature while stirring. The reaction mixture was stirred at room temperature for 10 days and quenched with aqueous solutions of ammonium chloride and ammonium carbonate. Addition of ether, extraction of the organic layer, drying over MgSO4 and rotatory evaporation afforded a yellow oil which was purified by means of column chromatography (silica, 230400 mesh, 60 ) with hexanes. Yield: 2.41 g (78%) 1 H NMR (300 MHz, CDCl3, 25 C): 2.15 (s, CCCH3, 3 H), 2.43 (s, CCH3, 6 H), 6.9-7.2 (m, CH, 3 13 H). C NMR (125.7 MHz, CDCl3, 25 C): 4.50 (q, 1JCH = 131.5 Hz, CCCH3), 21.28 (q, 1JCH = 126.8 Hz, CH3), 77.20 (m, CCCH3), 94.20 (q, 3JCH = 10.6 Hz, CCCH3), 123.72 (m, i-C), 126.47 (dqd, , 1JCH = 158.5 Hz, 2JCH = 5.2 Hz, 3JCH = 8.6 Hz, m-CH), 126.83 (d, 1JCH = 159.2 Hz , p-CH), 140.36 (q, 2JCH = 6.4 Hz, 2JCH = 6.4 Hz, o-C). IR (neat, [cm-1]): 3067 (m), 3022 (m), 2917 (s), 2852 (m), 2733 (m), 2242 (m), 2047 (w), 1928 (m), 1579 (m), 1467 (s), 1378 (m), 1263 (m), 1164 (m), 1093 (m), 1033 (m), 9676 (m), 920 (m), 769 (s) 483 (s). GC-MS, m/z (relative intensity): 144 (M+, 90), 143 (M+ - H, 13), 142 (3), 141 (10), 130 (1), 129 (M+ - CH3, 90), 128 (100), 127 (31), 115 (22), 102 (6), 89 (4), 77 (8), 75 (4), 65 (4), 63 (9), 39 (7). Anal. Calcd. for C11H12 (144.22): C, 91.61%; H, 8.39%. Found: C, 91.85%; H, 8.36%. 2,6-Diisopropylbenzenediazonium tetrafluoroborate. According to a general procedure for dediazotization of amines.14 Aqueous ethanol (30 mL, 96%) and HBF4 (14.4 mL, 110 mmol, 48 wt. % solution) were slowly added to 2,6-diisopropylaminobenzene (3.60 g, 20.3 mmol) in an Erlenmeyer and stirred at 0 C. After addition of an excess of n-butylnitrite124 (5.85 g, 56.7 mmol), the mixture was stirred for 30 min at 0 C.

51

Chapter 2 Addition of cold diethyl ether and storage at -30 C led to the precipitation of a white powder. The powder was subsequently washed on a glass filter with cold diethyl ether and light petroleum ether and kept moist prior to use. The explosive nature upon drying impeded the determination of the (crude) yield. 1 H NMR (300 MHz, CDCl3, 25 C): 1.23 (d, 3JHH = 6.8 Hz, CH3, 12 H), 3.23 (sept, 3JHH = 6.8 Hz, 2 H), 7.0-7.1 (m, CH, 3 H). 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 22.75 (CH3), 27.06 (d, J = 3.7 Hz, CH), 117.12 (m, i-C), 123.98 (m, m-CH), 135.02 (d, J = 15.87 Hz, p-CH), 158.30 (d, J = 244.2 Hz, CCHMe2). 2,6-Diisopropyliodobenzene. A 500-mL, three-neck, round-bottomed flask, equipped with a gasinlet adaptor, drop funnel and a Teflon-coated stir bar, was charged with 2 g of freshly prepared copper bronze125 and placed in a nitrogen atmosphere. 16.52 g (99.5 mmol) of KI, 8.30 g (32.7 mmol) of iodine and 50 mL of dry DMSO were added and the resulting mixture was stirred and heated to 60 C. After addition of 20 g of freshly prepared n-butylnitrite124 a solution of 12.1 g (68.2 mmol) of 2,6-diisopropylaniline in 30 mL of DMSO was added dropwise. The reaction mixture was stirred for 2 h at 60 C. The reaction was stopped by adding aqueous solutions of sodium chloride and sodium bisulfite. Filtration, separation, drying over MgSO4 and rotatory evaporation gave a yellow oil which was purified with column chromatography (silica, hexanes). Yield: 15.2 g (78 %) of a light-yellow viscous liquid. 1 H NMR (300 MHz, CDCl3, 25 C): 7.20 (m, CH, 1 H), 7.04 (m, CH, 2 H), 3.37 (septet, 3JHH = 6.8 Hz, 2 H), 1.20 (d, 3JHH = 6.8 Hz, 6 H). 13C NMR (125.7 MHz, CDCl3, 25 C): 23.37 (q, 1JCH = 126.8 Hz, CH3), 39.41 (d, 1JCH = 128.2 Hz, CH), 109.14 (m, CI), 123.79 (d, 1JCH = 156.4 Hz, CH), 128.30 (d, 1JCH = 160.2 Hz, CH), 151.12 (m, C). IR (neat, [cm-1]): 2950 (m), 2915 (m), 2850(m), 2230 (w), 1907 (w), 1665 (w), 1490 (m), 1463 (m), 1365 (m), 1105 (w), 1020 (m), 910 (m), 872 (m), 835 (s), 790 (m), 728 (s). GC-MS, m/z (relative intensity): 290 (1), 289 (9), 288 (M+, 70), 274 (13), 273 (M+ - CH3, 100), 146 (10), 145 (8), 133 (9), 131 (33), 128 (14), 117 (21), 115 (23), 105 (10), 91 (28). Anal. calcd. for C12H17I (288.17): C, 50.02%; H, 5.95%. Found: C, 50.15%; H, 6.07%. Propynyl copper. According to a modification of reported procedures for the preparation of alkynyl copper compounds.36 25.0 g (100 mmol) of CuSO45H2O was dissolved in 100 mL of concentrated aqueous ammonia in a 500-mL Schlenk flask, equipped with a Teflon-coated stir bar. After stirring at 0 C for 15 min 200 ml of water was added, followed by the slow addition of 13.9 g (200 mmol) of HONH2HCl. Subsequently, propyne gas was bubbled through the solution for 24 h during which a yellow solid precipitated. The suspension was filtered under vacuum and the yellow powder was washed with water, ethanol and ether yellow powder. CAUTION: The yellow powder is explosive when dry and was kept moist with ether. Crude (wet) yield: 8.6 g (84%). 1,3-Diisopropyl-2-(prop-1-ynyl)benzene (4). 8.6 g (84 mmol) of freshly prepared copper(I) propynyl is brought in a 400-mL Schlenk flask, equipped with a Teflon-coated stir bar, and evacuated. After addition of 240 mL of dry pyridine and 3.0 g (10 mmol) of 2,6-diisopropyliodobenzene the reaction mixture was heated on reflux at 120 C for 10 days. The mixture was quenched with water. Addition of petroleum ether (4060), filtration, extraction of the organic layer with petroleum ether (40-60), drying over MgSO4 and rotatory evaporation afforded a yellow oil which was purified by means of column chromatography (neutral alumina) with petroleum ether. Yield: 1.58 g (76%) of a colorless oil. 1 H NMR (500 MHz, CDCl3, 25 C): 1.24 (d, 3JHH = 6.9 Hz, CH(CH3)2, 6 H), 2.13 (s, ArCCH3, 3 H), 3.52 (septet, 3JHH = 6.9 Hz, CHMe2, 2 H), 7.18 (d, 3JHH = 7.8 Hz, m-CH, 2 H), 7.30 (t, 3JHH = 7.8 Hz, p-CH, 1 H). 13 C NMR (125.7 MHz, CDCl3, 25 C): 4.55 (q, 1JCH = 131.4 Hz, CCCH3), 23.61 (qdq, 1JCH = 125.8 Hz, 2JCH = 5.1 Hz, 3JCH = 5.1 Hz, CH3), 31.92 (dsept, 1JCH = 128.7 Hz, 2JCH = 4.0 Hz, CHMe2), 76.73 (q, 2JCH = 4.6 Hz, CH3C), 93.76 (q, 3JCH = 10.7 Hz, ArC), 121.64 (m, i-C), 122.32 (ddd, , 1JCH = 158.5 Hz, 2JCH = 4.8 Hz, 3JCH = 7.9 Hz, m-CH), 127.55 (d, 1JCH = 159.9 Hz, p-CH), 150.52 (m, o-C). IR (neat, [cm-1]): 2962 (s), 2870 (m), 2243 (w), 1575 (w), 1463 (m), 1362 (m), 1253 (w), 1179 (w), 1107 (w), 1060 (w), 935 (w), 801 (m), 754 (m), 483 (s). GCMS, m/z (relative intensity): 201 (8), 200 (M+, 51), 185 (M+ - CH3, 33), 158 (12), 157 (67), 156 (11), 155 (15), 154 (9), 153 (19), 152 (16), 151 (4), 144 (14), 143 (100), 141 (37), 129 (38), 128 (70), 115 (36), 91 (9). HR-MS: C15H20, calc.: 200.15650, found: 200.15734. Pentafluoro(prop-1-ynyl)benzene (5). The following procedure represents a modified version of previously published ones.36 Propyne gas (ca. 60 mmol) was condensed in vacuo in THF (80 mL) which was cooled at -196 C in a single-bulbed, 250-mL Schlenk flask, equipped with stir bar. The flask was brought under a nitrogen atmosphere and allowed to warm up to -80 C. Dropwise addition of n-butyl lithium (21.0 mL, 52.5 mol, 2.5 M in hexanes) under stirring after which the solution was allowed to warm up to -30 C, forming a white suspension. The mixture was cooled to -60 C and a solution of hexafluorobenzene (8.74 g, 47.0 mmol) in THF (50 mL) was slowly added to the reaction mixture under vigorous stirring. After addition the reaction mixture was allowed to slowly warm up to room temperature and stirred overnight. Addition of water (100 mL),

52

Rare-earth metallocene propargyl/allenyls extraction with diethyl ether, drying over MgSO4 and rotatory evaporation (50 C, 100 mbar) afforded a whitish oil. Sublimation at 80 C under vacuum (~ 1 mmHg) provided a white crystalline solid. Yield: 2.32 g (24%). 1 H NMR (400 MHz, CDCl3, 25 C): 2.13 (s, CH3, 3 H). 13C{1H} NMR (100 MHz, CDCl3, 25 C): 1 147.60 (m, JCF = 253 Hz, C5F5), 140.97 (d, 1JCF = 259 Hz, C5F5), 137.55 (d, 1JCF = 251 Hz, C5F5), 100.59 (t, 2JCF = 18 Hz, C5F5C), 99.53 (m, C C), 63.81 (m, C C), 4.821 (s, CH3). 19F NMR (376.5 MHz, CDCl3, 25 C): 139.34 (dd, 3JFF = 21.2 Hz, 4JFF = 7.0 Hz, o-F, 2 F), -156.27 (t, 3JFF = 20.5 Hz, p-F, 1 F), -164.39 (ddd, 3JFF = 21.2 Hz, 3JFF = 21.6 Hz, 4JFF = 7.5 Hz, m-F, 2 F). IR (nujol, [cm-1]): 2926 (s), 2855 (s), 2721 (w), 2254 (w), 2213 (w), 1463 (m), 1377 (m), 1261 (w), 1103 (w), 1020 (w), 974 (w), 835 (m), 489 (s). GC-MS, m/z (relative intensity): 207 (10), 206 (M+, 100), 205 (M+ - H, 73), 188 (7), 187 (69), 186 (5), 180 (6), 179 (9), 167 (7), 161 (8), 156 (28), 155 (5), 117 (7), 105 (6), 104 (4), 103 (4), 93 (8). Anal. Calcd. for C9H3F5: C, 52.45%; H, 1.47%. Found: C, 52.25%; H, 1.38%. Cp*2YCH2CCPh (7b). Hydrogen at a pressure of 1 bar was supplied to a solution of Cp*2YCH(SiMe3)2 (0.58 g, 1.12 mmol) in hexanes (5 mL). Stirring for 3 h at room temperature produced a lightyellow suspension. The hydrogen atmosphere was replaced by nitrogen and 1-phenyl-1-propyne (0.2 mL, 1.60 mmol) was added. Immediately, a red suspension formed. The suspension was stirred for 30 min at room temperature during which it was degassed several times. The solid was separated by decantation and filtration. Subsequent in vacuo removal of the solvent yielded a red crystalline material (0.49 g, 92% yield). Crystals suitable for X-ray analysis were obtained by recrystallization in toluene at low temperature. 1 H NMR (500 MHz, C6D6, 25 C): 1.93 (s, Cp*, 30 H), 2.82 (s, CH2, 2 H), 6.98 (t, 3JHH = 7.5 Hz, pPh, 1 H), 7.09 (t, 3JHH = 7.5 Hz, m-Ph, 2 H), 7.19 (d, 3JHH = 7.8 Hz, o-Ph, 2 H). 13C NMR (125.7 MHz, C6D6, 25 C): 11.15 (q, 1JCH = 125.6 Hz, C5(CH3)5,), 49.33 (dt, 1JCH = 159.1 Hz, 1JYC = 5.2 Hz, CH2C C), 106.79 (d, 1 JYC = 11.9 Hz, CH2C C), 117.78 (m, C5Me5), 126.69 (dt, 1JCH = 160.8, 3JCH = 7.7 Hz, p-CH), 128.60 (dd, 1JCH = 159.4 Hz, 3JCH = 7.6 Hz, m-CH), 132.40 (dtd, 1JCH = 159.1 Hz, 3JCH = 7.1 Hz, 4JCH = 1.4 Hz, Hz, o-CH), 132.77 (tdd, 2JCH = 3.7 Hz, 3JCH = 8.1 Hz, 4JCH = 2.0 Hz, i-C), 155.09 (t, 2JCH = 2.5 Hz, CH2C C). IR (nujol, [cm-1]): 2924 (m), 2856 (m), 2727 (m), 1923 (m), 1592 (m), 1544 (m), 1455 (m), 1378 (m), 1269 (m), 1154 (m), 1065 (m), 1021 (m), 916 (w), 846 (w), 768 (m), 698 (m), 627 (m), 482 (s). Anal. Calcd. for C29H37Y (474.52): C, 73.41; H, 7.86. Found: C, 73.24; H, 7.89. Cp*2LaCH2CCPh (7a). Hydrogen at a pressure of 1 bar was supplied to a solution of Cp*2LaCH(SiMe3)2 (0.82 g, 1.42 mmol) in pentane (5 mL). Stirring for 2 h at room temperature produced a yellow suspension. The hydrogen atmosphere was replaced by nitrogen and 1-phenyl-1-propyne (0.4 mL, 3.20 mmol) was added. Immediately, a deep red suspension forms. The suspension is stirred for 1 h at room temperature during which it was degassed several times. The solid was separated by decantation and filtration. Subsequent in vacuo removal of the solvent yielded a red crystalline material (0.68 g, 91% yield). Crystals suitable for X-ray analysis were obtained by recrystallization in toluene at low temperature. 1 H NMR (300 MHz, C6D6, 25 C): 1.92 (s, Cp*, 30H), 2.82 (s, CH2, 2H), 6.99 (t, 3JHH = 7.3 Hz, pH, 1 H), 7.12 (dd, 3JHH = 7.2 Hz, 3JHH = 7.3 Hz, m-H, 2 H), 7.24 (d, 3JHH = 7.2 Hz, o-H, 2 H). 13C NMR (125.7 MHz, C6D6, 25 C): 10.61 (q, 1JCH = 124.5 Hz, C5(CH3)5), 54.02 (t, 1JCH = 158.3 Hz, CH2C C), 112.76 (m, CH2C C), 119.46 (m, C5Me5), 126.19 (dt, 1JCH = 161.8 Hz, 3JCH = 7.4 Hz, p-CH), 128.71 (dd, 1JCH = 153.0 Hz, 3 JCH = 8.0, m-CH), 130.84 (dtd, 1JCH = 159.9 Hz, 3JCH = 7.0 Hz, 2JCH = 1.4 Hz, o-CH), 132.39 (t, 3JCH = 7.1 Hz, iC), 152.19 (s, CH2C C). IR (nujol, [cm-1]): 2920 (m), 2850 (m), 1944 (s), 1587 (m), 1550 (w), 1455 (s), 1380 (m), 1260 (w), 1090 (w), 1065 (w), 1020 (m), 900 (w), 760 (s), 690 (m), 600 (s), 450 (m). Anal. Calcd for C29H37La (524.52): C, 66.41; H, 7.11. Found: C, 66.24; H, 7.00. Cp*2LaCH2CCC6H3Me2-2,6 (9a). Hydrogen (1 bar) was applied to a solution of Cp*2LaCH(SiMe3)2 (0.56 g, 0.98 mmol) in hexanes (5 mL). The yellow suspension is stirred for 2 h at room temperature. The hydrogen atmosphere was replaced by nitrogen and 1-(2,6-dimethylphenyl)-1-propyne (0.21 g, 1.46 mmol) was added. The suspension was stirred overnight at room temperature and degassed several times. After overnight stirring, the suspension turned deep red. The reaction mixture was extracted with hexanes and evaporated to dryness forming a deep red oil. Crystallization from hexanes at low temperature yielded red crystalline material (0.12 g, 22% yield). Crystals suitable for X-ray analysis were obtained by recrystallization in toluene and cooling the solution. 1 H NMR (300 MHz, C6D6, 25 C): 1.92 (s, Cp*, 30 H), 2.18 (s, CH3, 6 H), 2.71 (s, CH2, 2 H), 6.94 (s, CH, 3 H). 13C NMR (125.7 MHz, C6D6, 25 C): 11.10 (q, 3JCH = 124.8 Hz, C5(CH3)5), 22.36 (q, 3JCH = 125.5 Hz, CH3), 49.12 (t, 1JCH = 156.9 Hz, CH2C C), 105.81 (t, 3JCH = 7.0 Hz, CH2C C), 119.04 (m, C5Me5), 127.14 (d, 1JCH = 159.0, p-CH), 128.31 (d, 1JCH = 156.9, m-CH), 130.96 (m, i-CH), 140.74 (m, o-CH), 149.28 (t, 2JCH = 2.6 Hz, CH2C C). IR (nujol, [cm-1]): 2925 (s), 2853 (s), 2725 (w), 1975 (m), 1786 (w), 1566 (w), 1465 (s),

53

Chapter 2 1377 (m), 1260 (w), 1168 (w), 1093 (w), 1020 (m), 779 (m), 720 (m), 500 (s). Anal. Calcd for C31H41La (552.57): C, 67.38%; H, 7.48%. Found: C, 67.29%; H, 7.54%. Reaction of Cp*2LaCH(SiMe3)2 with CH3CCPh. NMR Scale. Cp*2LaCH(SiMe3)2 (12.3 mg, 21.4 mol) was dissolved in C6D6 and CH3CCPh (3.0 L, 24 mol) was added with a microsyringe in the glovebox. No changes were observed with 1H NMR spectroscopy after 24 h at room temperature. Also, when the sample is heated to 50 C for 12 h, no changes were observed. After heating the sample to 80 C for 3 h, the formation of Cp*2LaCH2CCPh was observed and the yellow solution turned slowly deep red upon further heating. After 17 days at 80 C the reaction mixture was quenched with CD3OD. The presence of CH2(SiMe3)2, Cp*D, 1-phenyl1-propyne-d1 and phenylallene-d1 was indicated by 1H NMR and GC-MS. 1-Phenyl-1-propyne-d1: 1H NMR (C6D6, 25 C, 300 MHz): 1.65 (s, CH2D, 2H).130 GC-MS, m/z (relative intensity): 118 (6), 117 (72), 116 (100), 115 (15), 90 (8), 89 (6), 64 (6), 63 (8), 51 (7). Phenylpropa-1,2-diene-d1: 1H NMR (C6D6, 25 C, 300 MHz): 4.85 (s, CH2, 2 H).131 GC-MS, m/z (relative intensity): 118 (7), 117 (76), 116 (100), 90 (6), 63 (9). Analogous reactions of Cp*2LaCH(SiMe3)2 with CH3CCC6H4Me-2, CH3CCC6H3Me2-2,6 and CH3CCC6H3iPr2-2,6 and Cp*2YCH(SiMe3)2 with CH3CCPh in C6D6 gave similar results. Reaction of Cp*2LaCH(SiMe3)2 with CH3CCC6F5. NMR Scale. Cp*2LaCH(SiMe3)2 (15 mg, 26 mol) was dissolved in C6D6 and CH3CCC6F5 (5.5 mg, 27 mol) was added in the glovebox. No changes were observed with 1H NMR spectroscopy after 24 h at 50 C. After heating the sample to 80 C for 3 h, the formation of Cp*2LaCH2CCPh and CH2(SiMe3)2 was observed. Complete conversion of the alkyl into the propargyl was observed after 13 days at 80 C. The reaction mixture was subsequently quenched with CD3OD, producing a mixture of CH2(SiMe3)2, Cp*D, 1-pentafluorophenyl-1-propyne-d1 and pentafluorophenylpropadiene-d1 as indicated by 1H and 19F NMR and GC-MS. Cp*2LaCH2CCC6F5: 1H NMR (C6D6, 25 C, 400 MHz): 3.08 (s, CH2, 2 H), 1.88 (s, Cp*, 30 H). 19F NMR (C6D6, 25 C, 376 MHz): -140.32 (dd, J = 7.0, 21.2, o-F, 2 F), -156.98 (t, J = 20.5, p-F, 1 F), -164.94 (ddd, J = 8.1, 21.5, 21.6, m-F, 2 F). 13C{1H} NMR (C6D6, 25 C, 125.7 MHz): 10.41 (s, C5Me5), 55.77 (s, CH2), 108.76 (m, CH2CC), 119.70 (s, C5Me5), 137.86 (d, 1JCF = 249.5, m/o-CF), 141.21 (m, i-C), 146.97 (d, 1JCF = 249.5, m/o-CF), 147.70 (d, 1JCF = 249.5, p-CF), 165.24 (m, CH2CC). 1-Pentafluorophenyl-1-propyne-d1: 1H NMR (C6D6, 25 C, 400 MHz): 1.47 (s, CH2D, 2H). 19F NMR (C6D6, 25 C, 376 MHz): -147.46 (dd, J = 7.6, 21.4, o-F, 2 F), -157.00 (t, J = 20.5, p-F, 1 F), -164.96 (ddd, J = 8.1, 21.5, 21.6, m-F, 2 F). GC-MS, m/z (relative intensity): 118 (6), 117 (72), 116 (100), 115 (15), 90 (8), 89 (6), 64 (6), 63 (8), 51 (7). Pentafluorophenylpropadiene-d1: 1H NMR (C6D6, 25 C, 400 MHz): 4.72 (s, CH2, 2H). 19F NMR (C6D6, 25 C, 376 MHz): -145.00 (dd, J = 7.6, 21.4, o-F, 2 F), -159.73 (t, J = 21.4, p-F, 1 F), -165.69 (m, m-F, 2 F). GC-MS, m/z (relative intensity): 208 (2), 207 (M+, 26), 206 (M+ - H, 100), 205 (M+ - 2H or M+ - D, 67), 188 (M+ - F, 21), 187 (M+ - F - H, 75), 186 (M+ - F - 2H or M+ - F - D, 6), 157 (10), 156 (37), 155 (8), 117 (14), 105 (11). GC-MS, m/z (calc., found): 209 (0.4, 0.5), 208 (9.6, 8.7), 207 (100, 100). Reaction of [Cp*2La(-H)]2 with CH3CCPh. NMR Scale. [Cp*2La(-H)]2 (5.2 mg, 6.3 mol) was dissolved in C6D6 and CH3CCPh (1.6 L, 13 mol) was added with a microsyringe in the glovebox. The lightyellow solution turned immediately deep red. A complex mixture is observed with 1H NMR spectroscopy. Within 5 min at room temperature signals attributable to the hydride and substrate have disappeared and three major Cp* signals ( 1.92, 1.90, 1.88 ppm) are observed of which the propargyl ( 1.92 ppm) consititutes only 30% of the total Cp* 1H NMR resonance intensities as determined by line-shape analysis. Besides signals corresponding to cis-1-phenylpropene, unidentified 1H NMR resonances in the vinylic region ( 6.5 - 2.5 ppm) were observed. After 1 day at room temperature the reaction mixture was quenched with CD3OD forming Cp*D, cis-1-phenylprop-1-ene, 1-phenyl-1-propyne-d1 and phenylallene-d1 as indicated by 1H NMR and GC-MS. Cis-1-phenylprop-1-ene: 1H NMR (C6D6, 25 C, 500 MHz): 6.41 (dq, 4JHH = 1.7 Hz, 3JHH = 11.5 Hz, =CH, 1 H), 5.63 (dq, 3JHH = 11.5 Hz, 3JHH = 7.3 Hz, =CH, 1 H), 1.69 (dd, 4JHH = 1.7 Hz, CH3, 3 H).132 GCMS, m/z (relative intensity): 119 (9), 118 (75), 117 (100), 116 (10), 115 (43), 91 (29). The analogous reaction of [Cp*2La(-H)]2 with CH3CCPh in C7D8 gave similar results. Reaction of [Cp*2La(-H)]2 with CH3CCPh at low temperature. NMR Scale. A Teflon-capped NMR tube was charged with [Cp*2La(-H)]2 (6.8 mg, 8.3 mol). On the vacuum line, a solution of CH3CCPh (2.1 L, 17 mol) in 0.45 mL of toluene-d8 was condensed onto the evacuated NMR tube at -196 C. The reaction was monitored with 1H NMR spectroscopy as the temperature was increased gradually. Five Cp* 1H resonances (i.e. 2.05, 2.02, 1.98, 1.91, 1.84) were observed after 15 min at -60 C of which the hydride ( 2.05, 13%) and the propargyl ( 1.91, 12%) represented only a small amount, as determined by line-shape analysis. In addition, several other unidentified vinylic signals were observed. After reaching room temperature nine Cp* 1H

54

Rare-earth metallocene propargyl/allenyls resonances were observed (i.e. 2.04, 2.01, 1.99, 1.97, 1.96, 1.91, 1.88, 1.87, 1.84 ppm) of which the propargyl ( 1.91 ppm, 13%) constituted only a small amount. No changes were observed with 1H NMR spectroscopy after 24 h at room temperature and the reaction mixture was quenched with CD3OD producing a mixture containing Cp*D, cis-1-phenylprop-1-ene-d1, 1-phenyl-1-propyne-d1 and phenylallene-d1, as indicated by 1H NMR and GCMS. Cis-1-phenylprop-1-ene-d1: 1H NMR (C6D6, 25 C, 500 MHz): 6.41 (dq, 4JHH = 1.7 Hz, 3JHH = 11.5 Hz, =CH, 1 H), 5.63 (dq, 3JHH = 11.5 Hz, 3JHH = 7.3 Hz, =CH, 1 H), 1.69 (dd, 4JHH = 1.7 Hz, CH3, 3 H). Reaction of [Cp*2La(-H)]2 with PhCCPh and CH3CCPh. NMR Scale. [Cp*2La(-H)]2 (5.0 mg, 6.1 mol) was dissolved in C6D6 and PhCCPh (2.4 mg, 14 mol) was added in the glovebox. The light-yellow solution turned immediately orange. A clean reaction forming the alkenyl derivative Cp*2LaC(Ph)=CH(Ph) was observed (vide infra). No changes in the 1H NMR spectrum were observed after standing for 5 days at room temperature. Upon addition of 1-phenyl-1-propyne (3.5 mg, 30 mol) the alkenyl derivative was completely converted into the propargyl derivative Cp*2LaCH2CCPh within 24 h at room temperature, giving rise to cisdiphenylethene. The reaction mixture was quenched with CD3OD. GC/GC-MS and 1H NMR analysis indicated the presence of Cp*D, 1-phenyl-1-propyne-d1, 1-phenyl-1-propyne, phenylallene-d1 and cis-diphenylethene. Scheme 2-30. Numbering scheme of Cp*2LaC(Ph)=CH(Ph) (15a).
A B
2 1 10

G
9

4 5 6

Cp*2La

HD

Cp*2LaC(Ph)=CH(Ph) (15a): 1H NMR (C6D6, 25 C, 500 MHz): 7.69 (s, CH, 1 H), 7.29 (m, CH, 2 H), 7.12 (m, CH, 2 H), 7.04 (m, CH, 1 H), 6.99 (m, CH, 5 H), 1.84 (s, Cp*, 30 H). 13C NMR (125.7 MHz, C6D6, 25 C): 10.95 (q, 1JCH = 125.2 Hz, C5Me5), 120.74 (m, C5Me5), 123.44 (dt, 1JCH = 152.6 Hz, 3JCH = 7.1 Hz, oCH), 124.70 (dtd, 1JCH = 159.4 Hz, 2JCH = 1.2 Hz, 3JCH = 7.6 Hz, m-CH), 125.57 (dm, 1JCH = 154.1 Hz, o-CH), 126.59 (dt, 1JCH = 163.2 Hz, 3JCH = 7.1 Hz, m-CH), 129.89 (dtd, 1JCH = 153.7 Hz, 1JCH = 1.4 Hz, 1JCH = 7.9 Hz, pCH), 130.83 (dd, 1JCH = 152.6 Hz, 1JCH = 152.6 Hz, p-CH), 134.28 (dt, 1JCH = 147.2 Hz, 3JCH = 4.0 Hz, =C(Ph)H), 146.25 (s, i-C), 151.08 (s, i-C). The signal corresponding to LaC was not observed. Cis-diphenylethene: 1H NMR (C6D6, 25 C, 500 MHz): 6.46 (s, =CH, 2 H). The aromatic proton signals overlapped with others. GC-MS, m/z (relative intensity): 180 (99), 179 (100), 178 (66), 177 (9), 176 (12), 166 (9), 165 (51), 152 (15), 151 (8), 102 (9), 89 (19), 76 (14), 63 (7), 51 (9). GC-MS, m/z (calc., found): 182 (1.1, 1.7), 181 (18.4, 15.3), 180 (100.0, 100.0). Cp*2LaC(Ph)=C(Ph)H (15a). To a suspension of [Cp*2La(-H)]2 (35.0 mg, 42.6 mol) in cyclohexane (5 mL) was added diphenylacetylene (15.2 mg, 85.3 mol). The light-yellow suspension turned into an orange solution immediately. Removal of the solvent in vacuo gave a yellow solid in quantitative yield. Isolated yield: 49.2 mg (98%). 1 H NMR (500 MHz, C7D14, 25 C): 7.40 (s, D), 7.22 (t, 3JHH = 7.7 Hz, G), 7.20 (t, 3JHH = 7.7 Hz, A), 7.07 (tq, 3JHH = 7.1 Hz, 4JHH = 7.4 Hz, F), 6.95 (tq, 3JHH = 7.4 Hz, , 4JHH = 1.1 Hz, B), 6.91 (d, 3JHH = 7.1 Hz, E), 6.73 (dq, 3JHH = 7.1 Hz, 4JHH = 7.1 Hz, C), 1.77 (s, Cp*, 30 H). 13C NMR (125.7 MHz, C7D14, 25 C): 10.51 (q, 1JCH = 125.4 Hz, C5Me5), 120.60 (s, C5Me5), 122.97 (dt, 1JCH = 148.6 Hz, 3JCH = 7.1 Hz, 3), 124.25 (d, overlap hampered determination of coupling constants, 2), 125.53 (ddd, 1JCH = 153.5 Hz, 3JCH = 6.2 Hz, 3JCH = 6.2 Hz, 8), 126.18 (dt, 1JCH = 162.2 Hz, 3JCH = 7.3 Hz, 9), 129.58 (dd, 1JCH = 156.1 Hz, 3JCH = 7.9 Hz, 1), 130.83 (dd, 1JCH = 158.6 Hz, 3JCH = 7.9 Hz, 10), 133.72 (dt, 1JCH = 147.5 Hz, 3JCH = 4.1 Hz, 6), 146.25 (q, 3JCH = 6.6, 7), 150.83 (q, 3JCH = 7.5 Hz, 4), 217.47 (s, 5). 1H-13C gHSQC (500.0 MHz, 125.7 MHz, C7D14, 25 C): A1, B2, C3, D6, E8, F8, G10. 1H-13C gHMBC (500.1 MHz, 125.7 MHz, C7D14, 25 C): A2-4, B1,3, C1-5, D3-8, E6-9, F8,10, G7-9. Anal. Calcd. for C34H41La (588.61): C, 69.38%; H, 7.02%. Found: C, 69.51%; H, 6.90%. Reaction of Cp*2LaCH2CCPh with excess of H2. NMR Scale. Hydrogen gas was applied (1 atm.) onto a solution of Cp*2LaCH2CCPh (9.6 mg, 18.2 mol) in 0.50 mL of benzene-d6. Within several minutes the yellow solution turned red and became darker in color upon standing. 1H NMR spectroscopy indicated the instantaneous formation of [(Cp*2La)(-H)]2. Two other unidentified lanthanocene derivatives formed concomitantly, based on the presence of new Cp* 1H NMR resonances at 1.95 amnd 1.71 ppm. The reaction mixture was allowed to stand at room temperature for 1 day during which the progress of reaction was monitored

55

Chapter 2 periodically with 1H NMR spectroscopy. Upon standing, more [(Cp*2La)(-H)]2 formed at the expense of Cp*2LaCH2CCPh. After 12 h, Cp*2LaCH2CCPh was completely consumed, while the intensities of the Cp* 1H NMR resonances at 1.95 and 1.71 ppm started to decrease. After 18 h, Cp*2LaCH2CCPh and n-propylbenzene were the only compounds present in the product mixture, as indicated by NMR spectroscopy and GC/GC-MS analysis upon quenching with methanol. Reaction of [Cp*2La(-H)]2 with CH3CCC6H3Me2-2,6. NMR Scale. [Cp*2La(-H)]2 (7.8 mg, 9.5 mol) was dissolved in C6D6 (0.50 mL) and CH3CCC6H3Me2-2,6 (3 mg, 20.8 mol) was added with a microsyringe in the glovebox. The light-yellow solution turned darker and became light-orange within several minutes at room temperature. Within 10 min 1H NMR resonances of the hydride and substrate disappeared and Cp*2LaCH2CCC6H3Me2-2,6 and cis-1-(2,6-dimethylphenyl)prop-1-ene were observed. After 1 day at room temperature the reaction mixture did not change and was quenched with CD3OD forming Cp*D, cis-1-(2,6dimethylphenyl)prop-1-ene, 1-(2,6-dimethylphenyl)-1-propyne-d1 and (2,6-dimethylphenyl)allene-d1 as indicated by 1H NMR and GC-MS. Cis-1-(2,6-dimethylphenyl)prop-1-ene: 1H NMR (C6D6, 25 C, 300 MHz): 6.21 (dm, 3JHH = 11.4 Hz, =CH, 1 H), 5.64 (dq, 3JHH = 11.4 Hz, 3JHH = 6.9 Hz, =CH, 1 H), 2.15 (s, CH3, 6 H), 1.32 (dd, 3JHH = 11.4 Hz, 4 JHH = 1.7 Hz, CH3, 3 H). Minor contamination with its monodeuterated derivative led to unreliable intensities in the mass spectrum. 1-(2,6-Dimethylphenyl)-1-propyne-d1: GC-MS, m/z (relative intensity): 147 (1), 146 (11), 145 (M+, 98), 131(7), 130 (M+ - CH3, 65), 129 (M+ - CH3 - H, 100), 128 (M+ - CH3 - 2 H or M+ - CH3 - D, 62), 116 (14), 115 (10), 89 (3), 77 (C6H6+, 7), 51 (C4H3+, 8). GC-MS, m/z (calc., found): 147 (0.5, 0.7), 146 (12.0, 11.8), 145 (100, 100). (2,6-Dimethylphenyl)propa-1,2-diene-d1: GC-MS, m/z (relative intensity): 146 (4), 145 (M+, 32), 144 + (M - H, 14), 143 (M+ - 2 H or M+ - D, 4), 142 (M+ - 3 H or M+ - H - D, 8), 131(12), 130 (M+ - CH3, 65), 129 (M+ - CH3 - H, 100), 128 (M+ - CH3 - 2 H or M+ - CH3 - D, 62), 115 (8), 103 (5), 102 (3), 77 (C6H6+, 7), 63 (7), 51 (C4H3+, 9). GC-MS, m/z (calc., found): 147 (0.5, 0.9), 146 (12.0, 13.0), 145 (100, 100). Reaction of [Cp*2La(-H)]2 with CH3CCC6H3iPr2-2,6. NMR Scale. [Cp*2La(-H)]2 (6.6 mg, 8.0 mol) was dissolved in C6D6 and CH3CCC6H3iPr2-2,6 (3.5 L, 17 mol) was added with a microsyringe in the glovebox. No clear color change was observed. After 1 h at room temperature two major Cp* 1H NMR resonances are observed ( 2.22, 1.92) of which the propargyl is the major species (58% according to line-shape analysis). Also cis-1-(2,6-diisopropyl-phenyl)-1-propene was observed. After 2 days at room temperature, the relative concentration of the propargyl (74%) had increased. The reaction mixture was subsequently quenched with methanol forming a mixture containing Cp*H, cis-1-(2,6-diisopropylphenyl)prop-1-ene, 1-(2,6diisopropylphenyl)-1-propyne and (2,6-diisopropylphenyl)propa-1,2-diene as indicated by 1H NMR and GC-MS. Cp*2LaCH2CCC6H3iPr2-2,6: 1H NMR (500 MHz, C6D6, 25 C): 1.19 (d, 3JHH = 7.0 Hz, CH3, 12 H), 1.92 (s, Cp*, 30 H), 2.68 (s, CH2, 2 H), 3.34 (sept, 3JHH = 11.1 Hz, CHMe2, 2 H), 7.04-7.18 (m, CH). 13C NMR (125.7 MHz, C6D6, 25 C): 11.00 (q, 3JCH = 125.1 Hz, C5(CH3)5), 23.37 (q, 3JCH = 125.5 Hz, CH3), 32.39 (d, CH), 47.69 (t, 1JCH = 156.7 Hz, CH2C C), 100.31 (t, 3JCH = 7.0 Hz, CH2C C), 119.24 (m, C5Me5), 126.87 (d, 1 JCH = 159.4, p-CH), 122.87 (d, 1JCH = 155.6, m-CH), 128.70 (m, i-CH), 148.00 (m, o-CH), 145.61 (t, 2JCH = 2.4 Hz, CH2C C). Cis-1-(2,6-diisopropylphenyl)prop-1-ene: 1H NMR (C6D6, 25 C, 300 MHz): 6.37 (dq, 4JHH = 1.7 Hz, 3JHH = 11.1 Hz, =CH, 1 H), 5.71 (dq, 3JHH = 11.1 Hz, 3JHH = 6.8 Hz, =CH, 1 H), 3.19 (sept, 3JHH = 7.0 Hz, CHMe2, 2 H), 1.37 (dd, 4JHH = 1.7 Hz, 3JHH = 6.8 Hz, CH3, 3 H), 1.18 (d, 3JHH = 7.0 Hz, CH(CH3)2, 6 H). (2,6-Diisopropylphenyl)propa-1,2-diene: 1H NMR (C6D6, 25 C, 500 MHz): 6.16 (t, 4JHH = 7.0 Hz, CH, 1 H), 4.62 (d, 4JHH = 7.0 Hz, CH2, 2 H), 3.39 (sept, 3JHH = 7.0 Hz, CH, 2 H), 1.18 (d, 3JHH = 7.0 Hz, CH3, 12 H). Reaction of [Cp*2La(-H)]2 with CH3CCC6F5. NMR Scale. [Cp*2La(-H)]2 (10.7 mg, 13.0 mol) was added to a solution of CH3CCC6F5 (6.2 mg, 3.0 mol) in C6D6 in the glovebox. Several Cp* 1H NMR resonances are observed with 1H NMR spectroscopy and the reaction mixture changed slowly in time at RT. After 20 min at RT the 1H NMR resonances of the hydride and substrate have completely disappeared and four major Cp* 1H NMR resonances are observed ( 1.88, 1.84, 1.83 and 1.80) of which the propargyl ( 1.88) represents only 7% of the total Cp* 1H NMR resonance intensities according to line-shape analysis. An unknown species resonating at 1.80 represents the major species (71%). After 3 days more Cp* signals have appeared and the unknown species ( 1.80, 37%) has been consumed in the favor of those resonating at 1.93 (unknown, 12%), 1.88 (propargyl, 14%) and 1.84 (assigned to the alkenyl species Cp*2LaC(C6F5)=C(CH3)H, 12%). 19F NMR analysis points to at least four major different pentafluorophenyl moieties (in a 0.4:1.0:0.3:0.3 ratio). No changes were observed with 1H NMR spectroscopy after 2 days at room temperature and CD3OD is added. NMR

56

Rare-earth metallocene propargyl/allenyls and GC/GC-MS analysis pointed to the presence of 1-(pentafluorophenyl)prop-1-ene-d1 and 2,4bis(pentafluorophenyl)-3-methylhexa-2,4-diene-d1 in a 1.00:0.33 ratio, respectively. Cp*2LaC(C6F5)=C(CH3)H: 1H NMR (C6D6, 25 C, 400 MHz): 7.67 (q, 3JHH = 6.7 Hz, CH, 1 H), 1.84 (s, Cp*, 30 H), 1.51 (d, 3JHH = 6.7 Hz, CH3, 3 H). 1-(Pentafluorophenyl)prop-1-ene-d1: 1H NMR (C6D6, 25 C, 400 MHz): 6.24 (m, CH, 1 H), 1.48 3 (dm, JHH = 6.7 Hz, , 4JHD = 1.8 Hz, CH3, 3 H). GC-MS, m/z (relative intensity): 210 (4), 209 (M+, 45), 208 (M+ H, 84), 190 (M+ - F, 13), 189 (M+ - H - F, 26), 188 (M+ - F - 2H or M+ - F - D, 24), 187 (43), 181 (100), 169 (27), 161 (21), 158 (24). GC-MS, m/z (calc., found): 211 (0.4, 0.3), 210 (9.6, 8.7), 209 (100, 100). 2,4-Bis(pentafluorophenyl)-3-methylhexa-2,4-diene-d1: 1H NMR (C6D6, 25 C, 400 MHz): 5.45 (m, CH, 1 H), 1.57 (s, CH3, 3 H), 1.48 (d, 3JHH = 6.8 Hz, CH3, 3 H). GC-MS, m/z (relative intensity): 417 (1), 416 (10), 415 (M+, 50), 401 (18), 400 (M+ - CH3, 100), 385 (M+ - 2 CH3, 11), 350 (17), 205 (17), 188 (16), 187 (23), 182 (32), 181 (59). GC-MS, m/z (calc., found): 417 (2.2, 1.8), 416 (19.0, 19.6), 415 (100, 100). Reaction of Cp*2LaCH2CCPh with phenylacetylene. NMR Scale. Phenylacetylene (2.2 L, 20 mol) was added with a microsyringe to a solution of Cp*2LaCH2CCPh (9.6 mg, 18.2 mol) in 0.50 mL of benzene-d6 (containing 4.36 M hexa-methyldisiloxane as an internal standard). 1H NMR spectroscopy indicated the formation of [(Cp*2La)2(-2:2-PhC4Ph)], Cp*2LaC(Ph)=C(H)-CCPh and trans-1,4-diphenylbut-1-en-3-yne. The reaction mixture was allowed to stand at room temperature for 12 h after which two consecutive portions (4.4 and 21.0 L) of phenylacetylene were added. After 1 day and 13C NMR analysis, the mixture was quenched with methanol-d4. GC/GC-MS analysis indicated the presence of 1-phenyl-1-propyne, phenylpropadiene, trans1,4-diphenylbut-1-en-3-yne, 1,3-diphenylbut-1-en-3-yne, a trimer of phenylacetylene (m/z 306) and three unidentified dimers C18H16 (one of which was also found in the thermolysis reaction of Cp*2La(3-CH2CCPh) and its THF adduct). The spectral parameters corresponding to [(Cp*2La)2(-2:2-PhC4Ph)], Cp*2LaC(Ph)=C(H)CCPh and the phenylacetylene oligomers were identical to those reported in Chapter 4. C18H16: GC-MS, m/z (relative intensity): 234 (2), 233 (18), 232 (M+, 79), 217 (M+ - CH3, 21), 216 (20), 215 (33), 202 (M+ - 2CH3, 19), 153 (25), 141 (70), 128 (45), 115 (100), 91 (78). GC-MS, m/z (calc., found): 235 (0.1, 0.3), 234 (1.8, 2.3), 233 (19.6, 22.3), 232 (100.0, 100.0). C18H16: GC-MS, m/z (relative intensity): 234 (2), 233 (21), 232 (M+, 93), 217 (M+ - CH3, 24), 216 (23), 215 (37), 202 (M+ - 2CH3, 22), 153 (27), 141 (74), 128 (45), 115 (100, 91 (76). GC-MS, m/z (calc., found): 235 (0.1, 0.1), 234 (1.8, 2.3), 233 (19.6, 21.6), 232 (100.0, 100.0). Thermolysis of Cp*2LaCH2CCPh. NMR Scale. Cp*2LaCH2CCPh (15.0 mg, 28.6 mol) was dissolved in toluene-d8 and heated in a sealed NMR tube for 12 days at 100 C and then for 27 days at 120 C. The progress of the reaction was monitored with 1H NMR spectroscopy. The mixture was quenched with methanol-d4. GC/GC-MS analysis indicated the presence of Cp*H-d5, Cp*H-d6 and two compounds of m/z 233235 and 232-234, respectively. The latter two have been identified as oligodeuterated isomers of C18H16, based upon the identical retention times with nondeuterated C18H16 isomers found in other reactions. The mixture of deuterated isomers rendered the quantitative evaluation of the fragmentation pattern for each compound unreliable. Thermolysis of Cp*2YCH2CCPh. NMR Scale. Cp*2YCH2CCPh (15.0 mg, 31.6 mol) was dissolved in benzene-d6 and heated in a sealed NMR tube for 30 days at 120 C. The progress of the reaction was monitored with 1H NMR spectroscopy. After completion the mixture was quenched with methanol and GC/GCMS analysis indicated the presence of Cp*H-dn (n = 4,5) and three deuterated isomers of C18H16. C18H14D2: GC-MS, m/z (relative intensity): 235 (7), 234 (M+, 36), 220 (19), 219 (100), 205 (17), 204 (53), 157 (14), 141 (23), 128 (22), 115 (47). GC-MS, m/z (calc., found): 237 (0.1, 0.0), 236 (1.8, 1.1), 235 (19.6, 18.0), 234 (100.0, 100.0). C18H14D2: GC-MS, m/z (relative intensity): 236 (1), 235 (6), 234 (M+, 33), 220 (19), 219 (100), 205 (13), 204 (51), 203 (23), 202 (17), 157 (15), 141 (11), 142 (23), 141 (25), 128 (24), 115 (37), 91 (24), 77 (11). GC-MS, m/z (calc., found): 237 (0.1, 0.0), 236 (1.8, 1.6), 235 (19.6, 19.4), 234 (100.0, 100.0). C18H15D: GC-MS, m/z (relative intensity): 235 (1), 234 (6), 233 (M+, 33), 232 (100), 218 (24), 217 (86). 216 (31), 215 (49), 203 (202 (32), 155 (19), 154 (10), 153 (14), 152 (14), 141 (10), 139 (10), 128 (12), 116 (10), 115 (27), 108 (17), 101 (15), 91 (10). GC-MS, m/z (calc., found): 236 (0.1, 0.0), 235 (1.8, 2.6), 234 (19.6, 18.7), 233 (100.0, 100.0). Reaction of Cp*2LaCH2CCPh with THF. NMR Scale. THF (2.3 L, 29 mol) was added with a microsyringe to a solution of Cp*2LaCH2CCPh (15.0 mg, 29.6 mol) in benzene-d6 in the glove box. 1H NMR spectroscopy indicated the clean formation of Cp*2La(3-CH2CCPh)(THF). The reaction mixture was monitored with 1H NMR for 2 days at room temperature, before heating to 50 C. After 1 day at 50 C, the mixture was

57

Chapter 2 quenched with methanol-d4 and GC/GC-MS analysis indicated only the presence of an unidentified dimer C18H16, besides the expected quenching products Cp*D and THF. Cp*2La(3-CH2CCPh)(THF): 1H NMR (300 MHz, C6D6, 25 C): 1.37 (m, CH2, -CH2, 4 H), 1.97 (s, Cp*, 30 H), 2.95 (s, CH2, 2 H), 3.55 (m, OCH2, -CH2, 4 H), 6.95 (t, 3JHH = 7.4 Hz, p-H, 1 H), 7.11 (dd, 3JHH = 7.4 Hz, 3JHH = 7.4 Hz, m-H, 2 H), 7.24 (d, 3JHH = 7.4 Hz, 3JHH = 7.4 Hz, o-H, 2 H). 13C NMR (75 MHz, C6D6, 25 C): 11.02 (C5(CH3)5), 52.63(CH2CC), 25.68 (-CH2), 68.35 (-CH2), 118.86 (C5Me5), 124.91 (p-CH), 128.84 (i-C), 128.55 (m-CH), 129.05 (o-CH). Signals corresponding to CH2CC and CH2CC not observed. C18H16: GC-MS, m/z (relative intensity): 234 (2), 233 (17), 232 (M+, 84), 218 (21), 217 (M+ - CH3, 100), 216 (43), 215 (67), 205 (17), 202 (M+ - 2CH3, 60), 189 (23), 115 (55). GC-MS, m/z (calc., found): 235 (0.1, 0.2), 234 (1.8, 1.9), 233 (19.6, 18.6), 232 (100, 100). Cp*2La(3-CH2CCPh)(py) (7apy). Pyridine (8.0 L, 98 mol) was added with a microsyringe to a solution of Cp*2LaCH2CCPh (50 mg, 95 mmol) in hexanes (2 mL) in the glove box. Stirring for 2 h at room temperature produced a clear light-yellow solution. Cooling and concentrating the solution in vacuo afforded yellow crystals suitable for X-ray analysis. Isolated yield: 52 mg (90%). 1 H NMR (300 MHz, C6D6, 25 C): 1.94 (s, Cp*, 30H), 3.16 (s, CH2, 2H), 6.51 (ddd, 3JHH = 7.7 Hz, 3 JHH = 4.5 Hz, 4JHH = 1.5 Hz, -H, 2 H), 6.81 (tt, 3JHH = 7.6 Hz , 4JHH = 1.9 Hz, -H, 1 H), 6.96 (tt, 3JHH = 7.4 Hz, 4 JHH = 1.2 Hz, p-H, 1 H), 7.13 (d, 3JHH = 7.4 Hz, m-H, 2 H), 7.24 (d, 3JHH = 7.9 Hz, o-H, 2 H), 8.40 (dd, 3JHH = 4.5 Hz, 3JHH = 1.9 Hz, -H, 2 H). 13C NMR (125.7 MHz, C6D6, 25 C): 11.55 (q, 1JCH = 125.1 Hz, C5(CH3)5), 51.07 (t, 1JCH = 159.5 Hz, CH2CC), 117.96 (m, C5Me5), 123.72 (d, 1JCH = 153.6 Hz, p-CH), 123.79 (d, 1JCH = 165.6 Hz, -CH), 128.58 (m- and o-CH), 136.43 (m, i-C), 137.04 (d, 1JCH = 163.3 Hz, -CH), 152.16 (s, CH2C C). The signal corresponding to CH2CC was not observed. IR (nujol, [cm-1]): 2923 (s), 2853 (s), 2713 (m), 1973 (m), 1741 (w), 1653 (w), 1591 (m), 1459 (s), 1377 (m), 1215 (m), 1150 (m), 1067 (m), 1021 (m), 898 (m), 838 (m), 748 (m), 694 (m), 686 (m), 477 (s). Anal. Calcd. for C34H42LaN (603.62): C, 67.65%; H, 7.01%. Found: C, 67.51%; H, 6.90%. Cp*2Y(3-CH2CCPh)(py) (7bpy). Pyridine (20.5 L, 253 mol) was added with a microsyringe to a stirred solution of Cp*2YCH2CCPh (120 mg, 253 mol) in hexanes (4 mL) in the glove box. Immediately the orange solution turned pale yellow. Upon standing at room temperature, the yellow solution gradually turned red and a dark red suspension formed after 1 day. NMR and GC/GC-MS analysis suggested decomposition of the initially formed base adduct into several unidentified compounds. Cooling a freshly prepared solution afforded off-white crystals. Isolated yield: 88.3 mg (63%). After isolation the crystals turned red within several hours at room temperature. Cp*2Y(CH2CCPh)(py): 1H NMR (400 MHz, C6D6, 25 C): 1.93 (s, Cp*, 30 H), 2.71 (s, CH2, 2 H), 3 6.65 (dd, JHH = 5.4 Hz, 3JHH = 5.4 Hz, -CH, 2 H), 6.94 (m , -CH, 1 H), 6.97 (dd, 3JHH = 7.6 Hz, 3JHH = 7.6 Hz, p-CH, 1 H), 7.09 (dd, 3JHH = 7.7 Hz, 3JHH = 7.7 Hz, m-CH, 2 H), 7.22 (dd, 3JHH = 7.8 Hz, 3JHH = 7.8 Hz, o-CH, 2 H), 8.53 (m, -CH, 2 H). 13C NMR (125.7 MHz, C6D6, 25 C): 11.21 (q, 1JCH = 125.8 Hz, C5(CH3)5,), 47.22 (dt, 1JCH = 157.2 Hz, 1JYC = 7.7 Hz, CH2CC), 104.88 (d, 1JYC = 9.7 Hz, CH2CC), 117.65 (m, C5Me5), 123.46 (d, 1 JCH = 163.0 Hz, -CH), 126.39 (dt, 1JCH = 160.9 Hz, 3JCH = 7.4 Hz, p-CH), 128.57 (dd, 1JCH = 168.4 Hz, 3JCH = 7.6 Hz, m-CH), 132.06 (dtd, 1JCH = 159.3 Hz, 3JCH = 6.3 Hz, 4JCH = 1.9 Hz, Hz, o-CH), 132.68 (m, i-C), 135.49 (d, 1JCH = 160.8 Hz, -CH), 150.27 (d, 1JCH = 179.4 Hz, -CH), 151.58 (s, CH2CC). IR (nujol, [cm-1]): 2925 (s), 2854 (m), 2727 (w), 2148 (m), 1920 (w), 1588 (w), 1462 (m), 1377 (m), 1238 (m), 1021 (m), 855 (w), 800 (w), 759 (w), 705 (m), 520 (s). Reaction of Cp*2LaCH2CCPh with 1-phenyl-1-propyne. NMR Scale. Cp*2La(3-CH2CCPh) (10.0 mg, 19.1 mol) was dissolved in 0.50 mL of a benzene-d6 solution of hexamethyldisiloxane (1.4 mM). After addition of 1-phenyl-1-propyne (47.7 L, 381 mol) with a microsyringe the mixture was heated to 120 C and monitored with 1H NMR spectroscopy. After 7 day at 120 C the substrate was completely consumed and the mixture was quenched with methanol-d4. GC/GC-MS analysis indicated the presence of 1-phenylprop-1-ene, Cp*D, phenylpropadiene-d1, 1-phenyl-1-propyne-dn and at least 15 isomers of C18H16. The three major C18H16 isomers were assigned to the three major products which were characterized by 1D and 2D NMR spectroscopy (Chapter 3). Reaction of Cp*2LaCH2CCPh with 1-pentafluorophenyl-1-propyne. NMR Scale. Cp*2LaCH2CCPh (10.0 mg, 19.7 mol) was added to a solution 1-pentafluorophenyl-1-propyne (78.6 mg, 381 mol) in benzene-d6 (0.50 mL). 1H NMR spectroscopy indicated the formation of Cp*2La(3-CH2CCC6F5). The reaction mixture was monitored with 1H and 19F NMR for 20 days at 80 C, quenching with methanol-d4. GC/GC-MS analysis indicated the presence of 1-pentafluorophenyl-1-propyne-dn, pentafluorophenylpropadiened1, Cp*D, 1-phenyl-1-propyne and several organic compounds of which the major products were identified as C18H5DF10 (m/z 413) and four isomers of C27H13DF10 (m/z 529) on the basis of their fragmentation pattern.

58

Rare-earth metallocene propargyl/allenyls Table 2-9: Summary of crystallographic data of the Cp*2La(3-CH2CCPh) (7a), Cp*2Y(3-CH2CCPh) (7b) and [Cp*2La(3-CH2CCPh)(NC5H5)] (7apy) complexes. 7aC5H5N 7a 7b Molecular formula C29H37La C29H37Y C34H42LaN FW 524.52 474.52 603.62 T, K Crystal system Space group a, b, c, () () () V, 3 Z calc, gcm-1 F(000) (Mo K), cm-1 range () wR(F2) refined reflections refined parameters R(F) for Fo > 4.0 (Fo) GooF Weighting (a, b) 100(1) orthorhombic Pna21 20.064(1) 10.051(2) 12.233(3) 2466.9(8) 4 1.412 1072 17.44 2.27, 27.48 0.0990 5124 276 0.0463 1.025 0.0, 0.0 100(1) orthorhombic Pca2a1 16.4646(9) 10.5009(6) 14.7315(8) 2547.0(2) 4 1.237 1000 23.02 2.30, 28.28 0.0685 6224 419 0.0297 1.030 0.0393, 0.0 100(1) triclinic P1,2 8.7176(4) 9.6510(4) 18.4634(9) 78.405(1) 87.337(1) 72.103(1) 1447.86(11) 2 1.385 620 14.97 2.26, 28.28 0.0666 6986 493 0.0307 1.028 0.0316, 0.0

C18H5DF10: GC-MS, m/z (relative intensity): 413 (M+, 49), 412 (100), 398 (21), 397 (M+ - CH3, 47), 378 (87), 328 (21), 245 (25), 232 (10), 231 (37), 205 (82), 193 (41), 181 (96), 161 (26). GC-MS, m/z (calc., found): 416 (0.1, -), 415 (1.8, 1.5), 414 (19.5, 17.6), 413 (100.0, 100.0). C27H13DF10: GC-MS, m/z (relative intensity): 531 (1), 530 (11), 529 (M+, 38), 514 (M+ - CH3, 17), 407 (12), 379 (17), 347 (28), 181 (39), 123 (84), 122 (100), 121 (38), 108 (37), 107 (27), 106 (39), 92 (26), 91 (20). GC-MS, m/z (calc., found): 532 (0.4, -), 531 (4.0, 3.5), 530 (29.3, 30.3), 519 (100.0, 100.0). C27H13DF10: GC-MS, m/z (relative intensity): 531 (2), 530 (12), 529 (M+, 44), 514 (M+ - CH3, 22), 407 (12), 379 (19), 347 (23), 213 (17), 205 (36), 187 (19), 181 (37), 123 (84), 122 (100), 121 (40), 92 (26), 91 (24). GC-MS, m/z (calc., found): 532 (0.4, -), 531 (4.0, 4.6), 530 (29.3, 27.7), 519 (100.0, 100.0). Reaction of Cp*2LaCH2CCPh with an excess of 1-hexene. NMR Scale. A solution of Cp*2LaCH2CCPh (5.0 mg, 9.5 mol) in 0.50 mL of benzene-d6 containing hexamethyldisiloxane (4.5 mM) was prepared and transferred into a NMR tube. After addition of 1-hexene (173 L, 1.38 mmol, 145 equiv.) the tube was sealed off and heated to 100 C. The progress of the reaction was monitored with 1H NMR spectroscopy. After 3.59 days at 100 C no changes were observed and the reaction mixture was heated to 120 C for 74 days. The reaction mixture was quenched with methanol-d4. GC and GC-MS analysis indicated the presence of 1hexene, hexane, cis- and trans-2-hexene, cis- and trans-7-methyl-4-undecene and 3-dodecene. The unidentified organic compounds with m/z 170 (C12H26), 166 (two isomers of C12H22), 167 (C12H23), 195 (C15H15), 197 (C15H17), 200 (cross-coupling product, C15H20), 250 (two isomers of a 1-hexene trimer, C18H34), 281 (three isomers of C21H29) and 334 (1-hexene tetramer, C24H46) were assigned to their elemental composition, based on the assumption that they represented derivatives of homo- or cross-coupling products of 1-hexene and 1-phenyl1-propyne. C12H26: GC-MS, m/z (calc., found): 168 (0.8, 1.0), 167 (13.3, 14.3), 166 (100.0, 100.0). Two isomers of C12H22: GC-MS, m/z (calc., found): 168 (0.8, 1.0), 167 (13.3, 14.3), 166 (100.0, 100.0); m/z (calc., found): 168 (0.8, 1.0), 167 (13.3, 14.7), 166 (100.0, 100.0). C15H20: GC-MS, m/z (calc., found): 202 (1.1, 2.0), 201 (16.5, 19.9), 200 (100.0, 100.0). C15H17: GC-MS, m/z (calc., found): 199 (1.1, 2.1), 198 (16.5, 23.0), 197 (100.0, 100.0). C15H15: GC-MS, m/z (calc., found): 197 (1.1, 2.3), 196 (16.5, 17.5), 195 (100.0, 100.0)

59

Chapter 2 Two isomers of C18H34: GC-MS, m/z (calc., found): 253 (0.1, 0.0), 252 (1.8, 2.8), 251 (20.0, 24.5), 250 (100.0, 100.0); m/z (calc., found): 253 (0.1, 1.4), 252 (1.8, 8.1), 251 (20.0, 20.5), 250 (100.0, 100.0). Three isomers of C21H29: GC-MS, m/z (calc., found): 284 (0.2, 0.0), 283 (2.6, 5.9), 282 (23.1, 36.1), 281 (100.0, 100.0); m/z (calc., found): 253 (0.1, 1.4), 252 (1.8, 8.1), 251 (20.0, 20.5), 250 (100.0, 100.0). C24H46: GC-MS, m/z (calc., found): 337 (0.2, 0.0), 336 (3.3, 3.1), 335 (26.3, 26.3), 334 (100.0, 100.0). X-ray crystallography. In the glovebox, the crystals were transferred from the reaction vessels to a Petri dish filled with a small amount of light mineral oil. A suitable crystal was chosen by examination under a microscope. This crystal was attached to glass fiber which was mounted in a cold nitrogen stream on a Bruker133 SMART APEX CCD diffractometer for data collection. Data collection and structure solution were conducted at the University of Groningen. The structures were solved by Patterson methods and extension of the model was accomplished by direct methods applied to difference structure factors using the program DIRDIF.134 Relevant crystal and data collection parameters are given in Table 2-9.

2.8.
1

References and notes

For general reviews of organo rare-earth metal chemsitry, see: (a) Marks, T. J.; Ernst, R. D. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1982; Chapter 21. (b) Evans, W. J. Adv. Organomet. Chem. 1985, 24, 131-177. (c) Schaverien, C. J. Adv. Organomet. Chem. 1994, 36, 283-362. (d) Edelmann, F. T. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1995; Vol. 4, Chapter 2. (e) Schumann, H.; Meese-Marktscheffel, J. A.; Esser, L. Chem. Rev. 1995, 95, 865-986. (f) Anwander, R.; Herrmann, W. A. Top. Curr. Chem. 1996, 179, 1. (g) Edelmann, F. T. Top. Curr. Chem. 1996, 179, 247. (h) Molander, G. A. Chemtracts: Org. Chem. 1998, 18, 237-263. (i) Edelmann, F. T. In Metallocenes; Togni, A., Halterman, R. L., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 1, Chapter 2. (j) Chirik, P. J., Bercaw, J. E. In Metallocenes; Togni, A., Halterman, R. L., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 1, Chapter 3. (k) Edelmann, F. T.; Freckmann, D. M. M.; Schumann, H, Chem. Rev. 2002, 102, 1851-1896. (l) Arndt, S.; Okuda, J. Chem. Rev. 2002, 102, 1953-1976 (m) Shibasaki, M.; Yoshikawa, N. Chem. Rev. 2002, 102, 2187-2210. (n) Inanaga, J.; Furuno, H.; Hayano, T. Chem. Rev. 2002, 102, 2211-2226. (a) Klein, J. In The Chemistry of the Carbon-Carbon Triple Bond; Patai, S., Ed.; Wiley: New York, 1978; Part 1; Chapter 9, p. 343. (b) Moreau, J. In The Chemistry of Ketenes, Allenes and Related Compounds; Patai, S., Ed.; Wiley: New York, 1980; Part 1; p. 363. (c) Brandsma, L.; Verkruijsse, H. D. Synthesis of Acetylenes, Allenes and Cumulenes. A Laboratory Manual; Elsevier: Amsterdam; 1981. (d) Bielmann, J.-F., Ducep, J.-B. Org. React. 1982, 27, 1. (e) Landor, P. D. In The Chemistry of the Allenes; Landor, R. S., Ed.; Academic Press: London, 1982; p. 19. (f) Epsztein, R. In Comprehensive Carbanion Chemistry; Buncel, E., Durst, T., Eds.; Elsevier: Amsterdam, 1984; Part B; p. 107. (g) Schuster, H. F., Coppola, G. M. Allenes in Organic Chemistry; Wiley: New York, 1984; Chapter 10. (h) Yamamoto, H. In Comprehensive Organic Synthesis; Trost, B. M.; Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 2; p. 81. (i) Tsuji, J.; Mandai, T. Angew. Chem. Int. Ed. Engl. 1995, 34, 2589. For reviews, see: (a) Wojcicki, A.; Shuchart, C. E. Coord. Chem. Rev. 1990, 105, 35. (b) Welker, M. E. Chem. Rev. 1992, 92, 97. (c) Doherty, S.; Corrigan, J. F.; Carty, A. J.; Sappa, E. Adv. Organomet. Chem. 1995, 37, 39. (d) Wojcicki, A. New J. Chem. 1994, 18, 61. (e) Chen, J.-T. Coord. Chem. Rev. 1999, 190-192, 1143. (f) Wojcicki, A. Inorg. Chem. Comm. 2002, 5, 82. For examples, see: (a) Mo: Krivykh, V. V.; Taits, E. S.; Petrovskii, P. V.; Struchkov, Y. T.; Yanovski, A. I. Mendeleev Commun. 1991, 103. (b) Zr: Blosser, P. W.; Gallucci,J. C.; Wojcicki, A. J. Am. Chem. Soc. 1993, 115, 2994. (c) Re: Casey, C. P.; Yi, C. S. J. Am. Chem. Soc. 1992, 114, 6597. (d) Pt: Huang, T.-M.; Chen, J.-T.; Lee, G. H.; Wang, Y. J. Am. Chem. Soc. 1993, 115, 1170. (e) Pt: Blosser, P. W.; Schimpff, D. G.; Gallucci, J. C.; Wojcicki, A. Organometallics 1993, 12, 1993. (f) Pt: Stang, P. J.; Critell, C. M.; Arif, A. M. Organometallics 1993, 12, 4799. (g) Pd: Baize, M. W.; Blosser, P. W.; Plantevin, V.; Schimpff, Gallucci, J. C.; Wojcicki, A. Organometallics 1996, 15, 164. (h) Zr: Rodriguez, G.; Baza, G. C. J. Am. Chem. Soc. 1997, 119, 343. Watson, P. L. In Selective Hydrogen Carbon Activation. Principles and Progress; Davies, J. A.; Watson, P. L.; Greenberg, A.; Liebman, J. F. (Eds.); VCH: Weinheim, 1990, Chapter 4, p.78.

60

Rare-earth metallocene propargyl/allenyls

6 7 8 9 10

11 12

13 14

15

16

17

18

19 20

(a) Heeres, H. J.; Heeres, A.; Teuben, J. H. Organometallics 1990, 9, 1508. (b) Heeres, H. J. Ph. D. Thesis, University of Groningen, 1990; Chapter 5. Hajela, S.; Schaefer, W. P.; Bercaw, J. E. J. Organomet. Chem. 1997, 532, 45. Ihara, E.; Tanaka, M.; Yasuda, H.; Kanehisa, N.; Maruo, T.; Kai, Y. J. Organomet. Chem. 2000, 613, 26. Ferrence, G. M.; McDonald, R.; Takats, J. Angew. Chem. Int. Ed. 1999, 38, 2233. For examples, see: (a) Tabuchi, T.; Inanag, J.; Yamaguchi, M. Tetrahedron Lett. 1986, 27, 5237. (b) Tabuchi, T.; Inanag, J.; Yamaguchi, M. Chem. Lett. 1987, 2275. (c) Makioka, Y.; Koyama, K.; Nishiyama, T.; Takaki, K.; Taniguchi, Y.; Fujiwara, Y. Tetrahedron Lett. 1995, 36, 6283. (d) Mikami, K.; Yoshida, A.; Matsumoto, S.; Feng, F.; Matsumoto, Y.; Sugino, A.; Hanamoto, T.; Inanag, J. Tetrahedron Lett. 1995, 36, 907. (e) Makioka, Y.; Saiki, A.; Takaki, K.; Taniguchi, Y.; Kitamura, T.; Fujiwara, Y. Chem. Lett. 1997, 27. (a) Eckenberg, P.; Groth, U.; Kohler, T. Liebigs Ann. Chem. 1994, 673. (b) Ruan, R.; Zhang, J.; Zhou, X.; Cai, R. Chem. Commun. 2002, 538. Negishi et al. published the synthesis of 2-ethynyl-p-xylene and mesitylacetylene via palladiumcatalyzed cross-coupling with ethynylzinc bromide, see: (a) Negishi, E.; Kotora, M.; Xu, Caiding J. Org. Chem. 1997, 62, 8957. For a review on palladium-catalyzed alkynylation, see: (b) Negishi, E.; Anastasia, L. Chem. Rev. 2003, 103, 1979. Ohno, A.; Tsutsumi, A.; Yamazaki, N.; Okamura, M.; Mikata, Y.; Fujii, M. Bull. Chem. Soc. Jpn. 1996, 69, 1679. Subsequent attempts to purify the crude product mixture by means of column chromatography and vacuum distillation failed. Variation in reaction parameters, such as solvent, reaction time and temperature, did not produce the high crude yields, as typically found for aqueous dediazotization reactions of arylamines. For general remarks on the iododediazotization reaction, see: (a) March, J. Advanced Organic Chemistry, 4th ed.; Wiley: New York, 1992. (b) Vogel, A. I.; Furniss, B. S. Vogel's Textbook of Practical Organic Chemistry, 5th ed.; Longman: London, 1991. Although the synthetic utility of the iododediazotization reaction has been exploited for a several decades, its mechanism has not yet been satisfactorily elucidated, despite considerable study. For reviews, see: (a) March, J. Advanced Organic Chemistry, 4th ed.; Wiley: New York, 1992. (b) Vogel, A. I.; Furniss, B. S. Vogel's Textbook of Practical Organic Chemistry, 5th ed.; Longman: London, 1991. For examples of mechanistic studies of the iododediazotization reaction, see: (b) Abeywickrema, A. N.; Beckwith, A. L. J. J. Org. Chem. 1987, 52, 2568. (c) Packer, J. E.; Taylor, R. E. R. Aust. J. Chem. 1985, 38, 991. (d) Carey, J. G.; Jones, G.; Millar, I. T. Chem. Ind. 1960, 97. (b) Hodgson, H. H.; Birtwell, S.; Walker, J. J. Chem. Soc. 1941, 770. (e) Waters, W. A. J. Chem. Soc. 1942, 266. (f) Galli, C. J. Chem. Soc., Perkin Trans. 2 1981, 1459. (g) Kumar, R.; Singh, P. R. Tetrahedron Lett. 1972, 613. (h) Beckwith, A. L. J.; Meijs, G. F. J. Chem. Soc., Chem. Commun. 1981, 136. In contrast to bromo- and chlorodediazotization reactions, the use of copper is generally assumed not to be required in iododediazotization reactions.14 The replacement of the diazonium group by bromide or chloride in the presence of the appropriate copper(I) salts represents the most widely used method to prepare arylbromides and -chlorides (the Sandmeyer reaction). The Sandmeyer reaction is believed to occur via the reduction of diazonium salts by copper(I) species to give aryl radicals that subsequently abstract halogen from copper(II) halides. The use of finely divided (i.e. freshly precipitated) copper or copper bronze is known to act catalytically in the decomposition of diazonium salts, but the yields in the preparation of arylhalides (Gatterman reaction) are usually not as high as those obtained by the Sandmeyer reaction. (a) Hunter, C. A.; Jones, P. S.; Tiger, P. M. N.; Tomas, S. Chem. Commun. 2003, 1642. (b) Motti, E.; Rossetti, M.; Bocelli, G.; Catellani, M. J. Organomet. Chem. 2004, 689, 3741. (c) Alcalde, E.; Dinars, Rodrguez, S.; Garcia de Miguel, C. Eur. J. Org. Chem. 2005, 1637. Li, Z.; Dong, Y.; Mi, B.; Tang, Y.; Hussler, M.; Tong, H.; Dong, Y.; Lam, J. W. Y.; Ren, Y.; Sung, H. H. Y.; Wong, K. S.; Gao, P.; Williams, I. D.; Kwok, H. S.; Tang, B. Z. J. Phys. Chem. 2005, 109, 10061. (a) van Zanten, B.; Nauta, W. Th. Recl. Trav. Chim. Pays-Bas 1960, 79, 1211. Knorr, R.; Ruhdorfer, J.; Bhrer, P.; Bronberger, H.; Rpple, E. Liebigs Ann. Chem. 1994, 433.

61

Chapter 2

21 22

23 24 25

26

27

28

29

30

31 32 33

(a) Metal-Catalyzed Cross-Coupling Reactions, Diederich, F., Stang, P. J. (Eds.); Wiley: New York, 1997. (b) Tsuji, J. Palladium Reagents and Catalysts, Wiley: New York, 1995. Lithiation or stannylation of 2,6-diisopropylbromobenzene followed by iodinolysis was not explored in this study as a synthetic route towards the desired iodo compound. For examples of this approach in the preparation of 2,6-dimethyliodo-benzene, see: (a) Sasaki, Y.; Hirabuki, M.; Ambo, A.; Ouchi, H.; Yamamoto, Y. Bioorg. Med. Chem. Lett. 2001, 11, 327. For a review on the synthesis of iodoarenes, see: (b) Merkushev, E. B. Synthesis 1988, 923. (a) Giumanini, A. G.; Verardo, G.; Gorassini, F.; Strazzolini, P. Recl. Trav. Chim. Pays-Bas 1995, 114, 311. (b) Giumanini, A. G.; Verardo, G.; Geatti, P.; Strazzolini, P. Tetrahedron 1996, 52, 7137. Friedman, L.; Chlebowski, J. F. J. Org. Chem. 1968, 33, 1636. For example, reactions of 2,6-diisopropylaniline with n-butyl nitrite and several iodinating agents (i.e. iodine, potassium and copper iodide, freshly precipitated copper(I) iodide and methyl iodide), under a nitrogen atmosphere and in air, in the presence and absence of a crown ether, in a variety of solvents (i.e. chloroform, dichloromethane, tetrachloromethane, benzene and THF mixtures of the aforementioned solvents) and at different reaction temperatures yielded mixtures containing the desired iodo compound only in low crude yields (2-20%, GC/GC-MS). 2,6-Diisopropylbenzenediazonium tetrafluoroborate was prepared successfully by the general method of Roe and Hawkins from n-butyl nitrite which was added to a solution of the corresponding amine in 40% tetrafluoroboric acid and ethanol. This compound could be isolated as a white solid and was found to be stable, when it was kept moist with ether below room temperature, but decomposed spontaneously in the dry state. See: Roe, A.; Hawkins, G. F. J. Am. Chem. Soc. 1947, 69, 2443. It has been reported that the Negishi reaction tolerates sterically congested substrates better than the Sonogashira reaction and other palladium-catalyzed cross-coupling protocols based on organomagnesium and -tin reagents (the Stille reaction).12b Performing cross-coupling reactions with sterically hindered substrates represents, in fact, one of the major challenges in this area of research and efforts aimed at developing catalyst systems that are able to bring about this type of reactions are ongoing. Some progress has recently been achieved, but most examples involve either catalysts that are not readily available or the cross-coupling of two sp2 carbons in the synthesis of hindered biaryl compounds. Only recently, efficient catalysts for the Sonohashira coupling of sterically hindered aryl halides with 1-alkynes have been described. For examples, see: (a) Hundertmark, T.; Littke, A. F.; Buchwald, S. L.; Fy, G. C. Org. Lett. 2000, 2, 1729. (b) Kllhofer, A.; Pullmann, T.; Plenio, H. Angew. Chem. Int. Ed. 2003, 42, 1056. (c) Feuerstein, M.; Berthiol, F.; Doucet, H.; Santanelli, M. Synthesis 2004, 1281 and references therein. The propynyl derivative could successfully be prepared from propynylmagnesium bromide and triisopropoxyborane, according to a general procedure for the synthesis of alkynyldiisopropoxyboranes, see: (a) Brown, H. C.; Bhat, N. G.; Srebnik, M. Tetrahedron Lett. 1988, 29, 2631. (b) Brown, H. C.; Cole, T. E. Organometallics 1983, 2, 1317. (c) Pelter, A.; Smith, K.; Brown, H. C. In Borane Reagents; Academic Press: London, 1988; Chapter 2. (d) Chong, J. M.; Shen, L.; Taylor, N. J. J. Am. Chem. Soc. 2000, 122, 1822. (e) Brown, C. D.; Chog, J. M.; Shen, L. Tetrahedron 1999, 55, 14233. For reviews on the Suzuki-Miyaura reaction, see: (a) a) Suzuki, A.; Miyaura, N. Chem. Rev. 1995, 95, 2457. (b) Suzuki, A.; J. Organomet. Chem. 1999, 576, 147. For examples involving the SuzukiMiyaura cross-coupling of di-ortho-substituted aryl halides, see: (c) Watanabe, T.; Miyaura, N.; Suzuki, A. Synlett 1992, 207. (d) Rocca, P.; Marsais, F.; Godard, A.; Quguiner, G. Tetrahedron Lett. 1993, 34, 2937. (e) Hoye, T.; Chen, M. J. Org. Chem. 1996, 61, 7940. (f) Feuerstein, M.; Doucet, H.; Santelli, M. Tetrahedron Lett. 2001, 42, 6667. (g) Yin, J.; Rainka, M. P.; Zhang, X.-X.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 1162. (h) Bedford, R. B.; Hazelwood, S. L.; Limmert, M. E. Chem. Commun. 2002, 2610. (i) Altenhoff, G.; Goddard, R.; Lehmann, C. W.; Glorius, F. Angew. Chem. Int. Ed. 2003, 42, 3690. (j) Navarro, O.; Kelly, R. A.; Nolan, S. P. J. Am. Chem. Soc. 2003, 125, 16194. (k) Altenhoff, G.; Goddard, R.; Lehmann, C. W.; Glorius, F. J. Am. Chem. Soc. 2004, 126, 15195. For the synthesis of 2,6-diisopropylphenyl triflate, see: (a) Netscher, T.; Bohrer, P. Tetrahedron Lett. 1996, 37, 8359. (b) Stang, P. J.; Hanack, M.; Subramanian, L. R. Synthesis 1982, 85. Zimmerman, H. E.; Dodd, J. R. J. Am. Chem. Soc. 1970, 92, 6507. For examples, see: (a) Stephens, R. D.; Castro, C. E. J. Org. Chem. 1963, 28, 3313. (b) Castro, C. E.; Stephens, R. D. J. Org. Chem. 1963, 28, 2163. (c) Castro, C. E.; Gaughan, E. J.; Owsley, D. C. J.

62

Rare-earth metallocene propargyl/allenyls

34

35

36

37 38 39

40

41

42

Org. Chem. 1966, 31, 4071. (d) Owsley, D. C.; Castro, C. E. Org. Synth. Coll. Vol. 1988, 6, 916. For reviews, see: (e) Lindley, J. Tetrahedron 1984, 40, 1433. (f) Normant, J. F. Synthesis 1972, 63. (g) Jukes, A. E. Adv. Organomet. Chem. 1974, 12, 215. (h) Posner, G. H. Org. React. 1975, 22, 253. (i) Sladkov, A. M.; Golding, I. R. Russ. Chem. Rev. 1979, 48, 868. Copper(I) acetylides are dangerous, because they undergo hazardous explosions, when shocked or exposed to heat. Because of their high sensitivity to shock, friction and heat, copper acetylides must be handled with extreme care. They must be kept cool and, if they are to be stored, should be kept wet. See: (a) Bretherick, L. Handbook of Reactive Chemical Hazards, 2nd edition, Butterworths: London, 1979. (b) Sax, N. I. Dangerous Properties of Industrial Materials, 5th edition, Reinhold: New York, 1979. (c) Brameld, V. F.; Clark, M. T.; Seyfang, A. P. J. Soc. Chem. Ind. 1947, 66, 346. According to a general procedure in which copper(I) acetylides are prepared in situ from organolithium or -magnesium reagents in ether or THF, propynylmagnesium bromide was added to a suspension containing copper(I) iodide in ether at 0 C. However, replacing ether in vacuo by solvents with higher boiling points such as THF, di-n-butyl ether (bp 141 C), dimethylformamide (bp 153 C), and pyridine (bp 115 C) and the subsequent addition of the iodoarene did not result in the formation of the desired substitution product after prolonged reaction times and at refluxing temperatures. Also the use of THF and freshly precipitated copper(I) iodide in the preparation of the copper(I) acetylide did not change this outcome. Analogous reactions using propynyllithium as a starting material (formed in situ from propyne gas and n-butyl lithium in THF according to reported procedures127) failed also to give the desired propynylarene. For general synthetic procedures towards copper(I) acetylide, see: (a) Marino, J. P.; Floyd, D. M. J. Am. Chem. Soc. 1974, 96, 7138. (b) House, H. O.; Umen, M. J. J. Org. Chem. 1973, 38, 3893. (c) Brandsma, L.; Verkruijsse, H. D. Preparative Polar Organometallic Chemistry, Springer: Berlin, Vol. 1., p. 42. (d) Taylor, R. J. K. In Organocopper Reagents: A Practical Approach, Taylor, R. J. K. (ed.); Chapters 1 and 2. For the preparation of propynyllithium, see: (e) Midland, M. J. Org. Chem. 1975, 40, 2250. (f) Taschner, M. J.; Rosen, T.; Heathcock, C. H. Org. Synth. Coll. Vol. 7, 226. (g) Stang, P. J.; Boeshar, M.; Wingert, H.; Kitamura, T. J. Am .Chem. Soc. 1988, 110, 3272. (h) Marshall, J. A.; Wang, X. J. Org. Chem. 1991, 56, 960. For propynylcopper, see: (a) Atkinson, R. E.; Curtis, R. F. J. Chem. Soc. (C) 1967, 578. (b) Bossenbroek, B.; Sanders, D. C.; Curry, H. M.; Shechter, H. J. Am. Chem. Soc. 1969, 91, 371. For general procedures, see: (c) Ref. 33d. (d) Owsley, D. C.; Castro, C. E. Org. Synth. 1972, 52, 128. Remarkably, further additions of freshly prepared propynyl copper did not result in the complete consumption of the iodoarene. (a) Wiles, M. R.; Massey, A. G. Chem. Ind. 1967, 663. (b) Coe, P. L.; Plevey, R. G.; Tatlow, J. C. J. Chem. Soc. (C) 1966, 597. (c) Coe, P. L.; Tatlow, J. C.; Terrell, R. C. J. Chem. Soc. (C) 1967, 2626. For example, attempts to prepare salt- and solvent-free allyls from reaction of Cp*2YCl with 2methyl-2-propenylmagnesium chloride led instead to the formation of Cp*2Y(1CH2CMe=CH2)[MgCl22THF], see: den Haan, K. H.; Wielstra, Y.; Eshuis, J. J. W.; Teuben, J. H. J. Organomet. Chem. 1987, 323, 181. Only recently, allyl derivatives, Cp*2Ln(3-CH2CH=CH2), have been recognized as a convenient starting material in organo rare-earth metal chemistry, as they exhibit greater stability in terms of storage and handling than alkyls or hydrides, see: Evans, W. J.; Kozimor, S. A.; Brady, J. C.; Davis, B. L.; Nyce, G. W.; Seibel, C. A.; Ziller, J. W.; Doedens, R. J. Organometallics 2005, 24, 2269 and references therein. (a) Becker, J. Y. J. Organomet. Chem. 1977, 127, 1. (b) Ref. 2a. This finding is consistent with the increased observed kinetic acidity of the 1-alkynes upon substituting the carbon-carbon triple bond with -electron-withdrawing groups, see for examples: (c) Drenth, W.; Loewenstein, A. Recl. Trav. Chim. Pays-Bas 1962, 81, 635. (d) harman, H. B.; Vinard, D. R.; Kreevoy, M. M. J. Am. Chem. Soc. 1962, 84, 347. (e) Eaborn, C.; Skinner, G. A.; Walton, D. R. M. J. Chem. Soc. (B) 1966, 922. (f) Charton, M. J. Org. Chem. 1972, 37, 3684. (g) Lin, A. C.; Chiang, Y.; Dahlberg, D. B.; Kresge, A. J. J. Am. Chem. Soc. 1983, 105, 5380. (h) Kresge, A. J.; Powell, M. F. J. Org. Chem. 1986, 51, 819. No acidity constants were found in literature for propynylbenzenes. If the transmisstion of electronic effects via a carbon-carbon triple bond is similar to that of a carboxylic group, a large effect can be anticipated upon substituting the phenyl group with the pentafluorophenyl group. For example, substitution of the phenyl group of benzenecarboxylic acid C6H5CO2H (pKa = 4.2042a) by the

63

Chapter 2

43

44

45

46 47

48

49

50

51

pentafluorophenyl group (C6F5CO2H, pKa = 1.5142b) increased the aqueous acidity equilibrium Ka at 25 C almost 500-fold. For pKa values, see: (a) Streitweiser, A., Jr.; Klein, H. S. J. Am. Chem. Soc. 1969, 85, 2759. (b) Mosely, P. G. N. Electrochim. Acta 1980, 25, 1309. An impressive body of experimental evidence has accumulated that demonstrated significant stabilizing effects by fluoro and polyfluoroaryl substituents on carbanions. The evidence includes marked acceleration of the rates of carbanionic rearrangements and enhanced hydrocarbon acidities, relative to analogous all-hydrogen systems. For examples, see: (a) Filler, R.; Wang, C. S. J. Chem. Soc. (D) 1968, 287. (b) Streitweiser, A., Jr.; Mares, F. J. Am. Chem. Soc. 1968, 90, 644. (c) Streitweiser, A., Jr.; Hudson, J. A.; Mares, F. J. Am. Chem. Soc. 1968, 90, 648. (d) Streitweiser, A., Jr.; Scannon, P. J.; Niemeyer, H. M. J. Am. Chem. Soc. 1972, 94, 7936. (e) Bordwell, F. G.; Branca, J. C.; Bares, J. E.; Filler, R. J. Org. Chem. 1988, 53, 780. (f) Korenaga, T.; Kadowaki, K.; Ema, T.; Sakai, T. J. Org. Chem. 2004, 69, 7340. For reviews on carbon acids, see: (a) Cram, D. J. Fundamentals of Carbanion Chemistry, Academic Press: New York, 1965. (b) Jones, J. R. The Ionisation of Carbon Acids, Academic Press: London, 1973. (c) Reutov, O. A.; Beletskaya, I. P.; Butin, K. P. CH-Acids, Pergamon Press: Oxford, 1978. (d) R. Stewart The Proton: Applications to Organic Chemistry, Academic Press: New York, 1985. For examples, see: (e) Closs, G. L.; Closs, L. E. J. Am. Chem. Soc. 1963, 85, 2022. (f) Closs, G. L.; Larrabee, R. B. Tetrahedron Lett. 1965, 287. (g) Streitweiser, A., Jr.; Caldwell, R. A.; Young, W. R. J. Am. Chem. Soc. 1969, 91, 529. (h) Maksi, Z. B.; Eckert-Maksi, M. Tetrahedron 1969, 25, 5113. (i) Maksi, Z. B.; Randi, M. J. Am. Chem. Soc. 1973, 95, 6522. (j) Luh, T.-Y.; Stock, L. M. J. Am. Chem. Soc. 1974, 96, 3712. In an attempt to determine the relative (kinetic) acidity of 1-phenyl-1-propyne (1) and 1pentafluorophenyl-1-propyne (5), n-buyllithium was added to a stirred solution of 1 (1.0 equiv.) and 5 (1.0 equiv.) in THF at -80 C. Instead of the expected formation of allenylic and acetylenic quenchings products after addition of trimethylsilylchloride (TMSCl), a complicated mixture containing a multitude of unidentified compounds was obtained. For example, the reaction of [Cp*2Sm(-H)]2 with propene, 1-butene and allylbenzene form the corresponding 3-allyl complexes in hexane, but give Cp*2SmCH2Ph, exclusively, in toluene.54 Dissociation of the dimer [Cp*2Y(-H)]2 to the monomer Cp*2YH in reactions of [Cp*2Y(-H)]2 with alkenes has been studied in detail. See: Casey, C. P.; Tunge, J. A.; Lee, T.-Y.; Carpenetti III, D. W. Organometallics 2002, 21, 389. The reaction of [Cp*2La(-H)]2 with pyridine (2 equiv.) in benzene-d6 produces a 1:1 mixture of Cp*2La(2-2-C5H4N) and Cp*2La(NC5H8) instantaneously. The formed products were rationalized by ortho-metalation of pyridine resulting in the formation of Cp*2La(2-2-C5H4N) and H2. This reaction is followed by the rapid capture of the formed H2 by Cp*2La(2-2-C5H4N) to afford Cp*2La(NC5H8). In marked contrast to this efficient hydrogen trapping is the independent reaction of Cp*2La(1-2C5H4N)(C5H5N) with excess H2 (4 atm.). In this case, the hydrogenation product Cp*2La(NC5H8)(C5H5N) is formed quantitatively after 1 day at 20 C. See: (a) Ringelberg, S. N. Ph. D. Thesis, University of Groningen, 2001; Chapter 5. For another example of efficient hydrogen trapping, see (b): Hao, L.; Harrod, J. F.; Lebuis, A.-M.; Mu, Y.; Shu, R.; Samuel, E.; Woo, H.-E. Angew. Chem. Int. Ed. 1998, 37, 3126. Sung et al. proposed the steric substituent constant SA, based on isodesmic reactions and ab initio calculations of substituted adamantane systems. According to the SA scale, the phenyl (SA = -5.10) is larger than the methyl group (SA = -2.02), see: (a) Sung, K.; Chen, F.-L. Org. Lett. 2003, 5, 889. For reviews on steric effects, see: (b) Gallo, R. Prog. Phys. Org. Chem. 1983, 14, 115. (c) Frster, H.; Vgtle, F. Angew. Chem. Int. Ed. Engl. 1977, 16, 429. For example, see: (a) Fu, P.-F.; Brard, L.; Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 7157. (b) Molander, G. A.; Romero, J. A. C.; Corrette, C. P. J. Organomet. Chem. 2002, 647, 225. (c) Molander, G. A.; Schmitt, M. H. J. Org. Chem. 2000, 65, 3767. (d) Molander, G. A.; Knight, E. E. J. Org. Chem. 1998, 63, 7009. Reversible insertion of the carbon-carbon triple bond of diphenylacetylene into the metal-hydride of Cp*2LaH should yield Cp*2La(Ph)=C(Ph)D upon standing in benzene-d6, due to the well-known tendency of Cp*2LaH to undergo H/D exchange with benzene-d6.122 However, the intensity of the vinylic proton in Cp*2LaC(Ph)=C(Ph)H relative to that of the Cp* groups did not change after

64

Rare-earth metallocene propargyl/allenyls

52 53

54

55 56 57 58

59

60

61 62 63

64 65 66

standing for 14 days at room temperature, thereby providing evidence against the formation of Cp*2La(Ph)=C(Ph)D via reversible insertion of diphenylacetylene into Cp*2LaH. Catalytic hydrogenation of alkynes and alkenes, via alkenyl and alkyl intermediates, is wellestablished for rare-earth metallocene hydrides.53b,d For examples, see: (a) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Bercaw, J. E. J. Am. Chem. Soc. 1987, 109, 203. (b) Jeske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8111. (c) den Haan, K. H.; Wielstra, Y.; Teuben, J. H. Organometallics 1987, 6, 2053. (d) Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091. (e) Jeske, G.; Schock, L. E.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8103. (f) Watson, P. L.; Parshall, G. W. Acc. Chem. Res. 1985, 18, 51. (g) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1990, 112, 2314. (h) Evans, W. J.; DeCoster, D. M.; Greaves, J. Organometallics 1996, 15, 3210. (i) den Haan, K. H.; Teuben, J. H. J. Chem. Soc., Chem. Commun. 1986, 682. (j) Booij, M.; Deelman, B.-J.; Duchateau, R.; Postma, D. S.; Meetsma, A.; Teuben, J. H. Organometallics 1993, 12, 3531. Interestingly, the samarium allyl complexes Cp*2Sm(3-CH2CHCHR) (R = H, Me, Ph) are also reported to be extremely soluble in alkane solvents and an decreasing tendency to crystallize was found for the methyl and phenyl derivatives. Cp*2Sm(3-CH2CHCHPh) could, in fact, only be obtained as a tractable solid after recrystallization in the presence of a coordinating solvent forming the corresponding adduct. See: Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1990, 112, 2314. Heeres et al. also commented on the extreme solubility of Cp*2CeCH2CCMe and attempts to obtain single crystals for Cp*2LnCH2CCR (Ln = Ce, Y; R = Me, Ph) reported to be unsuccessful.6b Kretschmer, W. P., personal communication den Haan, K. H. Ph. D. Thesis, University of Groningen, 1986; Chapter 6. (a) Kalinowski, H.-O.; Berger, S.; Braun, S. Carbon-13 NMR Spectroscopy; Wiley: Chichester, 1991, Chapter 4. (b) Friebolin, H. Basic One- and Two-Dimensional NMR Spectroscopy; Wiley-VCH: Weinheim, 1998; Chapter 3.3; p. 97. For examples, see: (a) Reich, H. J.; Holladay, J. E. Angew. Chem. Int. Ed. Engl. 1996, 35, 2365. (b) Reich, H. J.; Thompson, J. L. Org. Lett. 2000, 2, 783. (c) Reich, H. J.; Holladay, J. E.; Walker, T. G.; Thompson, J. L. J. Am. Chem. Soc. 1999, 121, 9769. (d) Reich, H. J.; Holladay, J. E.; Mason, J. D.; Sikorsko, W. H. J. Am. Chem. Soc. 1995, 117, 12137. (e) Reich, H. J.; Holladay, J. E. J. Am. Chem. Soc. 1995, 117, 8470. (f) Lambert, C.; von Ragu Schleyer, P.; Wrthwein, E.-U. J. Org. Chem. 1993, 58, 6377. Infrared absorptions at 1850-1900 cm-1 are assigned to allenyl structures and those >2000 cm-1 to propargyl structures in organolithium chemistry, see: (a) Jaffe, F. J. Organomet. Chem. 1970, 23, 5362. (b) Priester, W.; West, R.; Chwang, T. L. J. Am. Chem. Soc. 1976, 98, 8413. (c) West, R.; Jones, P. C. J. Am. Chem. Soc. 1969, 91, 6156. (d) Klein, J.; Becker, J. Y. J. Chem. Soc., Chem. Commun. 1973, 576. For examples of organozinc and -magnesium compounds, see: (e) Ref. 2b. For examples of organotitanium compounds, see: (f) Ishiguro, M.; Ikeda, N.; Yamamoto, H. J. Org. Chem. 1982, 47, 2225. (g) Furuta, K.; Ishiguro, M.; Haruta, R.; Ikeda, N.; Yamamoto, H. Bull. Chem. Soc. Jpn. 1984, 57, 2768. Blosser, P. W.; Gallucci, J. C.; Wojcicki, A. J. Organomet. Chem. 2000, 597, 125. The paramagnetic Cp*2Ti(CH2CCPh) was furthermore characterized by elemental analysis, see: Ogoshi, S.; Stryker, J. M. J. Am. Chem. Soc. 1998, 120, 3514. (a) Jemmis, E. D.; Chandrasekhar, J.; von Ragu Schleyer, P.; J. Am. Chem. Soc. 1979, 101, 2848. (b) Carfagna, C.; Deeth, R. J.; Green, M.; Mahon, M. F.; McInnes, J. M.; Pellegrini, S.; Woolhouse, C. B. J. Chem. Soc., Dalton Trans. 1995, 3975. (c) Graham, J. P.; Wojcicki, A.; Bursten, B. E. Organometallics 1999, 18, 837. (a) den Haan, K. H. Ph. D. Thesis, University of Groningen, 1986, Chapter 6. (b) den Haan, K. H.; Wielstra, Y.; Teuben, J. H. Organometallics 1987, 6, 2053. Evans, W. J.; Meadows, J. H.; Hunter, W. E.; Atwood, J. L. Organometallics 1983, 2, 1252. (a) Ringelberg, S. N. Ph. D. Thesis, University of Groningen, 2001; Chapter 2. (b) Ringelberg, S. N.; Meetsma, A.; Troyanov, S. I.; Hessen, B.; Teuben, J. H. Organometallics 2002, 21, 1759.

65

Chapter 2

67

68 69 70 71 72

73

74

75 76 77 78 79

80

81

82 83

Reported spectral data of phenylallene: (a) 13C NMR (CDCl3): Maercker., A; Fischenich, J. Tetrahedron 1995, 51, 10209. (b) 13C NMR (CDCl3): Schaefer, T.; Kroeker, S.; McKinnon, D. M. Can. J. Chem. 1995, 73, 1478. (c) IR: Fleming, I.; Mwaniki, J. M. J. Chem. Soc., Perkin Trans. 1 1998, 1237. (d) IR: Chen, T. R.; Anderson, M. R.; Grossman, S.; Peters, D. G J. Org. Chem. 1987, 52, 1231. The corresponding protons of pentafluorophenyl-1-propyne and 1-pentafluorophenylpropa-1,2-diene resonate at 1.47 and 4.72 ppm, respectively, see Experimental Section. Miao, W.; Chung, L. W.; Wu, Y.-D.; Chan, T. H. J. Am. Chem. Soc. 2004, 126, 13326. Bushby, R. J.; Patterson, A. S.; Ferber, G. J.; Duke, A. J.; Whitham, G. H. J. Chem. Soc., Perkin Trans. II 1978, 807. Ionic radii for eight-coordinate complexes: La3+ (1.160 ) and Y3+ (1.019 ), see: Shannon, R. D Acta Crystallogr., Sect. A 1976, A32, 751. No inductive/field parameters were found in literature for these metals, but the higher inductive/field effect of Y3+ versus La3+ is apparent from the carbon chemical shift of the metal-bound methine carbon in Cp*2LaCH(SiMe3)2 ( 44.69 ppm) versus Cp*2YCH(SiMe3)2 ( 25.19 ppm) in benzene-d6 solution. In support, the electronegativity of Y3+ (X = 1.22) is higher than that of La3+ (X = 1.10), according to the Pauling scale, see: CRC Handbook of Chemistry and Physics, Lide, D. R. (Ed.); CRC Press: Boca Raton; 1999. (a) Heeres, H. J.; Meetsma, A.; Teuben, J. H. Angew. Chem. Int. Ed. 1990, 29, 420. (b) Scholz, A.; Smola, A.; Scholz, J.; Loebel, J.; Schumann, H.; Thiele, K.-H. Angew. Chem. Int. Ed., 1991, 30, 435. (c) Gagn, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (d) Thiele, K.-H.; Bambirra, S.; Sieler, J. Angew. Chem. Int. Ed. 1998, 37, 2886. (e) Evans, W. J.; Davis, B. L.; Nyce, G. W.; Perotti, J. M.; Ziller, J. W. J. Organomet. Chem. 2003, 677, 89. The average Ln-C(C5Me5) distances parallel the ionic radius of the lanathanide metal, but other distances such as Ln-C(alkyl) do not follow this pattern as closely, see: Evans, W. J.; Foster, S. E. J. Organomet. Chem. 1992, 433, 79. The twist angle is defined as the average of the five smallest dihedral angles formed between the ten planes which consist of a ring carbon and the two centroids. van der Heijden, H.; Schaverien, C. J.; Orpen, A. G. Organometallics 1989, 8, 255. Evans, W. J.; Foster, S. E. J.Organomet.Chem. 1992, 433, 79 (a) den Haan, K. H.; de Boer, J. L.; Teuben, J. H. J. Organomet. Chem. 1988, 327, 31. (b) den Haan, K. H.; de Boer, J. L.; Teuben, J. H. Organometallics 1986, 5, 1726. (a) Evans, W. J.; Seibel, C. A.; Ziller, J. W.; Doedens, R. J. Organometallics 1998, 17, 2103. (b) Evans, W. J.; DeCoster, D. M.; Greaves, J. Organometallics 1996, 15, 3210. (c) Gerald, J.; Schock, L. E.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8103. (d) Ref. 40. (e) Ref. 54. Evidence was reported for two mechanisms for allyl fluxionality in a series of scandocene allyl compounds that occur simultaneously, i.e. an 3-C3H3 to 1-C3H3 coordination change and a 180 rotation of the C3H3 moiety, see: Abrams, M. B.; Yoder, J. C.; Loeber, C.; Day, M. W. Bercaw, J. Organometallics 1999, 18, 1389. (a) Evans, W. J.; Forrestal, K. J.; Ziller, J. W. Angew. Chem., Int. Ed. Engl. 1997, 109, 774. (b) Evans, W. J.; Forrestal, K. J.; Ziller, J. W. J. Am. Chem. Soc. 1995, 117, 12635. (c) Evans, W. J.; Forrestal, K. J.; Ziller, J. W. J. Am. Chem. Soc. 1998, 120, 9273. (a) den Haan, K. H.; Wielstra, Y.; Teuben, J. H. Organometallics 1987, 6, 2053. (b) Heeres, H. J. Ph. D. Thesis, University of Groningen, 1990; Chapter 3. Reported acidity values for phenylacetylene differ by as much as 10 pK units, depending on the solvent and the method used to determine acidity. Benzene is an aprotic, nonpolar solvent. Comparison with acidities in protic, polar solvents, such as reflected by the pKa values which refer to (equilibrium) acidities in water, seems therefore inappropriate. The most extensive acidity scale in a nonprotic solvent is based on dimethylsulfoxide (DMSO).83e According to the pKDMSO values, the (equilibrium) acidities of phenylacetylene (28.7) and methanol (29.0) are similar. According to gasphase equilibrium acidities83f Gacid (kcal/mol, 298 K) in which solvent effects are absent, methanol (372.6) is more acidic than phenylacetylene (362.6). For reviews on carbon acids, see: (a) Cram, D. J. Fundamentals of Carbanion Chemistry, Academic Press: New York, 1965. (b) Jones, J. R. The Ionisation of Carbon Acids, Academic Press: London, 1973. (c) Reutov, O. A.; Beletskaya, I. P.;

66

Rare-earth metallocene propargyl/allenyls

84

85

86

87

88

89

90 91

92

93

94

95 96 97 98

99 100 101

Butin, K. P. CH-Acids, Pergamon Press: Oxford, 1978. (d) R. Stewart The Proton: Applications to Organic Chemistry, Academic Press: New York, 1985. For acidity scales, see: (e) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456. (f) Bartmess, J. E.; Scott, J. A.; McIver, R. T., Jr. J. Am. Chem. Soc. 1979, 101, 6046. (a) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533. (b) Hard and Soft Acids and Bases; Pearson, R. G. (ed.); Dowden, Hutchinson and Ross, Stroudsberg, PA; 1973. (b) Fleming, I. Frontier Orbitals and Organic Chemical Reactions, Wiley-Interscience, London; 1976. (a) Evans, W. J.; Keyer, R. A.; Ziller, J. W. Organometallics 1990, 9, 2628. (b) Heeres, H. J.; Nijhoff, J,; Teuben, J. H.; Rogers, J. D. Organometallics 1993, 12, 2609. (c) Evans, W. J.; Keyer, R. A.; Ziller, J. W. Organometallics 1993, 12, 2618 (d) Forsyth, C. M.; Nolan, S. P.; Stern, C. L.; Marks, T. J.; Rheingold, A. L. Organometallics 1993, 12, 3618. The reactions of Cp*2LuMe5 and Cp*2ScH102 with allene are reported to proceed via C-H activation, forming the allenyl Cp*2Lu(1-CH=C=CH2) and allyl Cp*2Sc(3-CH2CH=CH2), respectively. The lanthanide alkenyl derivatives Cp*2LnC(R)=C(R) are believed to be reactive intermediates in numerous catalytic reactions, such as 1-alkyne oligomerization, alkyne hydroamination and alkyne hydrosilylation, and are presumed to undergo insertion reactions with unsaturated substrates and C-H activation reactions with Brnsted acids.1 However, -to best of our knowledge- no (systematic) reactivity studies of well-defined lanthanidocene alkenyl derivatives are reported in literature. The observation of acetylenic and allenylic phenylacetylene trimers C24H20 (for details, see Chapter 4) with GC-MS provides circumstantial evidence that organic compounds C25H20 are stable under the present experimental conditions. The Lewis-base-free complex Cp*2ZrMe[B(4-C6H4F)4] undergoes selective insertion of 1,2propadiene at -30 C, but a complicated mixture is formed upon warming with the excess amount of 1,2-propadiene present. The reaction with 3-methyl-1,2-butadiene is reported to afford a complex mixture of unidentified products. See: Horton, A. D. Organometallics 1992, 11, 3271. (a) Burns, C. J.; Andersen, R. A. J. Am. Chem. Soc. 1987, 109, 941. (b) Schultz, M.; Burns, C. J.; Schwartz, D. J.; Andersen, R. A. Organometallics 2001, 20, 5690. (c) Perrin, L.; Maron, L.; Eisenstein, O.; Schwartz, D. J.; Burns, C. J.; Andersen, R. A. Organometallics 2003, 22, 5447. Casey, C. P.; Lee, T.-Y.; Tunge, J. A.; Carpenetti II, D. W. J. Am. Chem. Soc. 2001, 123, 10762. Thermolysis of Cp*2CeCH(SiMe3)2 is also reported to proceed via intramolecular C-H activation, forming a tetranuclear fulvene species [Cp*Ce(-5:1:1-C5Me3(CH2)2CeCp*2]2 which was characterized by X-ray crystallography.92a-b IR spectroscopy of the thermolysis product of Cp*2LaCH(SiMe3)2 indicated a similar compound. (a) Booij, M.; Meetsma, A.; Teuben, J. H. Organometallics 1991, 10, 3246. (b) Booij, M. Ph. D. Thesis, University of Groningen, 1989; Chapter 3. (c) Ryu, J.-S.; Marks, T. J.; McDonald, F. E. J. Org. Chem. 2004, 69, 1038 Regiorandom insertion of phenylallene into the La-CH2 of Cp*2La(3-CH2CCPh) affords, in principle, four different products after protonolysis, whereas 1-phenyl-1-propyne yields only two. Because insertion is believed to proceed regiorandomly at 120 C, internal metalation producing 1phenyl-1-propyne is presently favored. No NMR spectral data were found in literature for (monomeric) Cp*2LaCH2R complexes, but analogous Cp*2YCH2R derivatives have -CH proton resonances in the range from 0.66 to -0.08 ppm, see: Casey, C. P.; Tunge, J. A.; Lee, T.-Y.; Fagan, M. A. J. Am. Chem. Soc. 2003, 125, 2641. Watson, P. L. J. Chem. Soc., Chem. Commun. 1983, 276. (a) Deelman, B.-J. Ph. D. Thesis, University of Groningen, 1994; Chapter 5. (b) Deelman, B.-J.; Booij, M.; Meetsma, A.; Teuben, J. H.; Kooijman, H.; Spek, A. L. Organometallics 1995, 14, 2306. Evans, W. J.; Chamberlain, L. R.; Ulibarri, T. A.; Ziller, J. W. Organometallics 1988, 110, 6423. Following the previous line of reasoning93, insertion of 1-phenyl-1-propyne into the La-CH2 bond of Cp*2La(3-CH2CCPh) is likely to occur regioselectively at 50 C, affording only one dimer after protonolysis. Evans, W. J.; Dominguez, R.; Hanusa, T. P. Organometallics 1986, 5, 1291. Ringelberg, S. N. Ph. D. Thesis, University of Groningen, 2001; Chapter 5. Ringelberg, S. N. Ph. D. Thesis, University of Groningen, 2001; Chapter 4.

67

Chapter 2

102

103 104

105 106 107

108

109

110 111

112

113 114 115 116

(a) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D. ; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1987, 109, 203. (b) Thompson, M. E.; Bercaw, H. E. Pure Appl. Chem. 1984, 105, 6491. (a) Deelman, B.-J.; Stevels, W. M.; Teuben, J. H.; Lakin, M. T.; Spek, A. L. Oranometallics 1994, 13, 3881. (b) Deelman, B.-J. Ph. D. Thesis, University of Groningen, 1994; Chapter 4. The deep coloration noted for rare-earth metal pyridyl, furyl and phenyl complexes has tentatively been assigned to products formed from radical coupling or ring-opening of (hetero)aromatic units. For examples, see: (a) Ref. 103. (b) Evans, W. J.; Meadows, J. H.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1984, 106, 1291. (c) Fryzuk, M. D.; Jafarpour, L.; Kerton, F. M.; Love, J. B.; Patrick, B. O.; Rettig, S. J. Organometallics 2001, 20, 1387. (d) Arndt, S.; Elvidge, B. R.; Zeimetz, P. M.; Spaniol, T. P.; Okuda, J. Organometallics 2006, 25, 793. Hazin, P. N.; Bruno, J. W.; Schulte, G. K. Organometallics 1990, 9, 416. Watson, P. L.; Whitney, J. F.; Harlow, R. L. Inorg. Chem. 1981, 20, 3271. (a) Evans, W. J.; Drummond, D. K. J. Am. Chem. Soc. 1986, 108, 7440. (b) [(Cp*2Sm)2(-2:2PhC4Ph)]: Evans, W. J.; Keyer, R. A.; Zhang, H.; Atwood, J. L. J. Chem. Soc., Chem. Commun. 1987, 837. (c) Rausch, M. D.; Moriarty, K. J. Organometallics 1986, 5, 1281. (d) {(Cp*2Sm)2[-4PhCH=C(O)-C(O)=CHPh]}: Evans, W. J.; Drummond, D. K. J. Am. Chem. Soc. 1988, 110, 2772. (a) Gagn, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (b) Giesbrecht, G. R.; Collis, G. E.; Gordon, J. C.; Clark, D. L.; Scott, B. L.; Hardman, N. J. J. Organomet. Chem. 2004, 689, 2177. For examples, see: (a) La(NH-2,6-iPr2C6H3)3(C5H5N)2: Giesbrecht, G. R.; Gordon, J. C.; Clark, D. L.; Hay, P. J.; Scott, B. L.; Tait, C. D. J. Am. Chem. Soc. 2004, 126, 6387. (b) La(O-2,6i Pr2C6H3)3(NH3)4: Butcher, R. J.; Clark, D. L.; Grumbine, S. K.; Vincent-Hollis, R. L.; Scott, B. L.; Watkin, J. G. Inorg. Chem. 1995, 34, 5468. (c) [Cp3La(NCCH3)2]: Spirlet, M. R.; Rebizant, J.; Apostolidis, C.; Kanellakopulos, B. Acta Cryst. 1987, C43, 2322. (d) [Cp3La(NCCH2CH3)2]: XingFu, L.; Eggers, S.; Kopf, J.; Jahn, W.; Fischer, R. D.; Apostolidis, C.; Kanellakopulos, B.; Benetollo, F.; Polo, A.; Bombieri, G. Inorg.Chim.Acta 1985, 100, 183. Koch, W.; Hertwig, R. H. In The Encyclopedia of Computational Chemistry; Schleyer, P. v. R., Editor-in-Chief; Wiley: Chichester, 1998. (a) Maron, L.; Eisenstein, O. J. Phys. Chem. A 2000, 104, 7140. (b) M. Dolg, Stoll, H. Handbook on the Physics and Chemistry of Rare Earths, Vol. 22; Elsevier, Amsterdam, 1996; Chapter 152. (c) M. Dolg, H. Stoll, Theor. Chim. Acta 1989, 75, 369. (d) Dolg, M.; Stoll, H; Preu, H, J. Chem. Phys. 1989, 90, 1730. (e) Dolg, M.; Fulde, P.; Kijchle, W.; Neumann, C. S.; Stoll, H. J. Chem. Phys. 1991, 94, 3011. (f) Dolg, M.; Stoll, H.; Preu, H. J. Mol. Struct. 1991, 235, 67. (g) Dolg, M.; Stoll, H.; Flad, H. T.; Preu, H. J. Chem. Phys. 1992, 97, 1162. (h) Kaupp, M.; Schleyer, P. v. R.; Dolg, M.; Stoll, H. J. Am. Chem. Soc. 1992, 114, 8202. (h) Dolg, M.; Stoll, H.; Preu, H., Theor. Chim. Acta 1993, 85, 441. (i) Brady, E. D.; Clark, D. L.; Gordon, J. C.; Hay, P. J.; Keogh, D. W.; Poli, R.; Scott, B. L.; Watkin, J. G. Inorg, Chem. 2003, 42, 6682. (a) Lauher, J. W.; Hoffmann, R. J. Am. Chem. Soc. 1976, 98, 1729. For examples of rare-earth metallocenes, see: (b) Steigerwald, M. L.; Goddard III, W. A.; J. Am. Chem. Soc. 1984, 106, 308. (c) Ortiz, J. V.; Hoffmann, R. Inorg. Chem. 1985, 24, 2095. (d) Raba, H.; Saillard, J.-Y.; Hoffmann, R. J. Am. Chem. Soc. 1986, 108, 4327. (e) Rapp, A. K. J. Am. Chem. Soc. 1987, 109, 5605. (f) Ouddai, N.; Bencharif, M. J. Mol. Struct. (Theochem) 2004, 709, 109. (g) Ref. 114. (h) Ref. 115. (a) Erker, G.; Rosenfeldt, F. Angew. Chem., Int. Ed. Engl. 1978, 17, 605. (b) Erker, G. Acc. Chem. Res. 1984, 17, 103 and references therein. (a) Deelman, B.-J.; Teuben, J. H.; MacGregor, S. A.; Eisenstein, O. New J. Chem. 1995, 19, 691. (b) Sndig, N.; Koch, W. Organometallics 2002, 21, 1861. Perrin, L.; Maron, L.; Eisenstein, O. New. J. Chem. 2004, 28, 1255 and references therein. (a) Ahlrichs, R.; Br, M.; Baron, H.-P.; Bauernschmitt, R.; Bcker, S.; Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche, F.; Haase, F.; Hser, M.; Httig, C.; Horn, H.; Huber, C.; Huniar, U.; Kattannek, M.; Khn, A.; Klmel, C.; Kollwitz, M.; May, K.; Ochsenfeld, C.; hm, H.; Schfer, A.; Schneider, U.; Treutler, O.; Tsereteli, K.; Unterreiner, B.; Von Arnim, M.; Weigend, F.; Weis, P.; Weiss H. Turbomole Version 5, january 2002. Theoretical Chemistry Group, University of Karlsruhe. (b) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346. (c) Turbomole basisset library, Turbomole Version 5, see 1a. (d) Schfer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571. (e) Andrae,

68

Rare-earth metallocene propargyl/allenyls

117 118

119

120 121

122 123 124 125 126

127

128 129 130 131 132 133 134

D.; Hussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor. Chim. Acta 1990, 77, 123. (f) Schfer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829. Although the fluorine atom is larger than the hydrogen atom, both the pentafluorophenyl group and the phenyl group are planar structures. It can be anticipated that the decrease of electron density at the triple bond will weaken the interaction with the electrophilic lanthanide metal. On the other hand, polarization of the triple bond, as to induce a partial positive charge at the -position and a partial negative charge at the -position of the fourcentered insertion transition state is well-known to facilitate insertion into electrophilic metal-carbon bonds. For examples, see: (a) Burger, B. J.; Santarsiero, B. D.; Trimmer, M. S.; Bercaw, J. E. J. Am. Chem. Soc. 1988, 110, 3134. (b) Doherty, N.; Bercaw, J. E. J. Am. Chem. Soc. 1985, 107, 2670. (c) Halpern, J.; Okamoto, T.; Zakhariev, A. J. Mol. Catal. 1976, 2, 65-68. (d). Lin, Z.; Marks, T. J. J. Am. Chem. Soc. 1990, 112, 5515. (e) Ref. 102a. For a review of insertion reactions in organo rare-earth metal chemistry, see: (a) Zhou, X.; Zhu, M. J. Organomet. Chem. 2002, 647, 28. For examples, see: (b) Evans, W. J.; Wayda, A. L.; Hunter, W. E.; Atwood, J. L. J. Chem. Soc., Chem. Commun. 1981, 706. (c) Ref. 122. Deelman, B.-J,; Stevels, W. M.; Teuben, J. H.; Lakin, M. T.; Spek, A. L. Organometallics 1994, 13, 3881. To the best of our knowledge, only one example of CO insertion into a Ln-C(alkenyl) bond has been reported in literature, see: Evans, W. J.; Hughes, L. A.; Drummond, D. K.; Zhang, H.; Atwood, J. L. J. Am. Chem. Soc. 1986, 108, 1722. Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091. den Haan, K. H.; de Boer, J. L.; Teuben. J. H.; Spek, A. L.; Koji-Prodi, B.; Hays, G. R.; Huis, R. Organometallics 1986, 5, 1726. Vogel, A. I.; Rogers, V.; Hannaford, A. J.; Furniss, B. S. Vogel's Textbook of Practical Organic Chemistry, 4th ed.; Longman: London, 1978; p. 409. Vogel, A. I.; Rogers, V.; Hannaford, A. J.; Furniss, B. S. Vogel's Textbook of Practical Organic Chemistry, 4th ed.; Longman: London, 1981; p. 285. (a) Kaufman, G. B.; Pinnell, R. P. Inorg. Synth. 1960, 6, 3. (b) Hiller, W.; Zybill, C. E. In Synthetic Methods of Organometallic and Inorganic Chemistry, Breitinger, D. K.; Herrmann, W. A. (Eds.); Georg Thieme: Stuttgart, 1999; Vol. 5., Chapter 1, p. 6. For propynyllithium, see: (a) Midland, M. J. Org. Chem. 1975, 40, 2250. (b) Taschner, M. J.; Rosen, T.; Heathcock, C. H. Org. Synth. Coll. Vol. 7, 226. (c) Stang, P. J.; Boeshar, M.; Wingert, H.; Kitamura, T. J. Am .Chem. Soc. 1988, 110, 3272. (d) Marshall, J. A.; Wang, X. J. Org. Chem. 1991, 56, 960. Brandsma, L.; Vasilevsky, S.F.; Verkruijsse, H. D. Application of Transition Metal Catalysts in Organic Synthesis, Springer: Berlin, 1998; Chapter 10, p. 4. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals; 2nd ed.; Pergamon Press: Oxford, 1980. Maercker, A.; Fischenich, J. Tetrahedron 1995, 51, 10209. Chen, T. R.; Anderson, M. R.; Grossmann, S.; Peters, D. G. J. Org. Chem. 1987, 52, 1231. A. Koch Magn. Reson. Chem. 2000, 38, 216. SMART, SAINT, SADABS, XPREP and SHELXK/NT. Smart Apex Software Reference Manuals 2000, Bruker AXS Inc., Madison, Wisconsin, USA. Beurskens, P. T.; Beurskens, G.; De Gelder, R.; Garca-Granda, S.; Gould, R. O.; Isral, R.; Smits, J. M. M. The DIRDIF-99 Program System 1999, University of Nijmegen, The Netherlands.

69

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

3.

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

3.1.

Introduction

Transition-metal catalyzed oligomerization reactions of alkynes constitute an effective synthetic method for forming new carbon-carbon bonds between unsaturated organic compounds.1 Important examples include linear oligomerization, (co-)cyclotrimerization and cyclotetramerization reactions, catalyzed by a variety of transition-metal complexes. Current mechanistic understanding of these reactions entails carbon-carbon bond formation via oxidative coupling or metal alkyne -bonding which activates the carbon-carbon triple bond towards nucleophilic attack. Scheme 3-1. The catalytic cyclodimerization of 1-methylalk-1-ynes.
R catalyst R C6D6, 50 C R = Me, Et (80%, 2.3:1),nPr (89%, 1.3:1) R + R R

catalyst =

Ln CH(SiMe3)2

Ln = La, Ce

Catalytic alkyne cyclodimerization reactions that produce alkylidenecyclobutenes are scarce and only a few examples exist in literature.2 Alkyl-substituted 1-methylalk-2-ynes CH3CCR (R = Me, Et, nPr) have been reported to undergo catalytic cyclodimerization in the presence of Cp*2LnCH(SiMe3)2 (Ln = La, Ce), yielding mixtures of substituted 3-alkylidenecyclobutenes (Scheme 3-1).3 Unfortunately, further development of this catalytic system was inhibited by the small scope of substrates (i.e. 1-methylalk-1-ynes bearing small alkyl groups) and the low catalytic activities (e.g. 2-butyne is converted with a turnover frequency Nt of 2.0 h-1 at 80 C) and (regio)selectivities. The use of Me2Si(C5Me4)2CeCH(SiMe3)2 resulted in lower catalytic rates and failed to broaden the scope of substrates. In contrast to most metal-catalyzed alkyne reactions, a mechanism involving -methyl C-H activation, forming a lanthanide propargyl Cp*2LnCH2CCR, was proposed. Due to the inherent reactivity of functionalized four-membered carbocycles and the possibility of a variety of further synthetic manipulations, the discovery of new methods for their construction is an active area of research.4 Contrary to other four-membered carbocycles, the synthetic access to alkylidenecyclobutenes is not straightforward and has been approached through the thermal coupling of alkynes and allenes5 or coppermediated cyclization of 1,4-enynes.6 Mainly because of their high reactivity and their limited synthetic access, alkylidenecyclobutenes belong to a relatively unexplored class of molecules, their reported reactivity being limited to scattered examples of thermolysis,7a-c singlet oxygenation,7d-f electrophilic addition7b,g,5a and polymerization reactions.8 A study involving the synthesis, structure and reactivity of aryl-substituted 3-allenyl/propargyl metallocene complexes Cp*2Ln(3-CH2CCAr) of the rare-earth metals revealed that 1-phenyl-1-propyne undergoes permethyllanthanocene-catalyzed cyclodimerization, forming mainly (E)-3-benzylidene-2-methyl-1phenylcyclobutene (Chapter 2). The goal of the investigations described in this Chapter was to further explore the scope and selectivity of the permethyllanthanidocene-mediated cyclodimerization reaction of propynylaromatics, searching for highly selective catalytic pathways general to the construction of four-

71

Chapter 3 membered carbocycles. First, the catalytic cyclodimerization of 1-phenyl-1-propyne is described, including reaction kinetics and mechanistic pathways to the observed products. Then, the effect of temperature, metal ion radius, substrate substituent pattern and ancillary ligation on the rate and selectivity is examined. Finally, the possibility of applying the cyclodimerization reaction as a method to prepare novel (cross-)conjugated polymers is investigated.

3.2.
3.2.1.

The cyclodimerization of propynylaromatics

Lanthanocene-catalyzed cyclodimerization

Catalyst precursors When a benzene-d6 solution of Cp*2LaR [R = CH(SiMe3)2 (1a), H (2a), 3-CH2CCPh (3a)] was heated to 80-120 C in the presence of an excess amount of 1-phenyl-1-propyne (10-25 equiv.), slow consumption of 1-phenyl-1-propyne was observed. NMR analysis of the reaction mixture indicated the conversion of 1-phenyl-1-propyne (4) into mainly (E)-3-benzylidene-2-methyl-1-phenylcyclobutene (5) (eq. 3.1). The catalytic nature of the cyclodimerization reaction was confirmed by performing a control experiment without metal complex. A solution of 1-phenyl-1-propyne in perdeuteriobenzene remained unchanged after 12 days at 120 C, as indicated by NMR and GC-MS analysis.
Ph + 4 C6D6 120 C 5 75% Ph Ph 6 10% Ph Ph + Ph 7 9%

Cp*2LaR 5 mol%

(3.1)

Ph

R = CH(SiMe3)2 (1a) , H (2a), CH2CCPh (3a)

The product mixtures obtained with the complexes 1a-3a were identical (NMR, GC/GC-MS) and the only lanthanocene species observed by in situ 1H NMR spectroscopy during substrate conversion was the complex Cp*2La(3-CH2CCPh) (3a). The mixtures containing 1a showed lower rates of substrate conversion than mixtures containing of 2a and 3a, consistent with the relative rate of propargylic metalation of 1-phenyl-1propyne of 1a as compared to 2a (Chapter 2).9 The rates of substrate conversion for mixtures containing 2a and 3a in the presence of a 20-fold molar excess of 1-phenyl-1-propyne are invariant within experimental error. In both cases, complete substrate conversion was observed within 7 days at 120 C. Reaction intermediates When the substrate was completely consumed, 1H NMR resonances attributable to 3a disappeared and new resonances were observed. The identities of these new organometallic species could not be established unequivocally, but NMR and GC-MS analysis of reaction mixtures, treated with trimethylsilylchloride (TMSCl), point to the cyclobutenyl derivatives C and F (Scheme 3-2). For example, the formation of two sets of characteristic 13C NMR resonances at 43.66 and 41.01 ppm and 12.22 and 11.36 ppm was observed after addition of TMSCl. The former set is assigned to the endocyclic CH2 group of the silylated alkylidenecyclobutenes 9 and 10, while the latter set is assigned the exocyclic methyl group, consistent with literature values (vide infra).10 Because the formed methylenecyclobutenes undergo decomposition during GCMS analysis (vide infra), the product mixtures were subjected to catalytic hydrogenation prior to GC analysis. This methodology allowed for the quantitative analysis of stable derivatives by means of FID-GC. GC-MS analysis of the above mixtures after treatment with TMSCl and catalytic hydrogenation indicated the presence of two C21H28Si isomers (in a 71:29 ratio according to FID-GC), corresponding to the expected hydrogenated derivatives of 9 and 10. Attempts to prepare the proposed cyclobutenyl organolanthanide derivatives F and G by reacting 3a with a two-fold molar excess of 1-phenyl-1-propyne gave reaction mixtures in which 3a was the major organometallic species.

72

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

Scheme 3-2. The reactions of the organometallic species present during the catalytic cyclodimerization of 1-phenyl-1-propyne.
Cp*2LaR C6D6 120 C Cp*2La Ph Ph Ph Cp*2La F Me3SiCl -(Cp*2LaCl)n Me3Si Ph Ph D . + 3a CD3OD Ph Ph D Ph Ph -RH Cp*2La Ph Ph Ph C Me3SiCl -(Cp*2LaCl)n Me3Si Ph Ph R = CH(SiMe3)2 (1a), H (2a)

10

Mechanism Catalytic cyclodimerization of alkyl-substituted 1-methyl-alk-1-ynes was previously observed for Cp*2LnCH(SiMe3)2 (Ln = La, Ce) and a catalytic cycle was proposed.3 Analogous reaction sequences (i-v) can be envisioned for the present catalytic system (Scheme 3-3). Additional experimental evidence is provided by the following observations: (i) the reactions of 1a and 2a with 1-phenyl-1-propyne afford 3a (Chapter 2), (ii) the product mixtures obtained with the complexes 1a-3a are identical, (iii) the only organolanthanide species observed during cyclodimerization in the presence of excess substrate is 3a, (iv) the reaction rate of substrate conversion is first-order in 3a (vide infra), (v) the observation of an alk-4-en-1-ynyl permethylyttrocene derivative in the reaction of Cp*2Y(3-CH2CCPh) (3b) with an excess amount of 1-phenyl-1-propyne (vide infra) and (vi) the ability of a permethyllanthanocene alk-1-en-1-yl to undergo propagylic C-H activation with 13a 5 4
Ph +
4

Cp*2LaCH2CCPh (3a) 2 mol% Ph C6D6 100 C

Ph

Ph + Ph 6 Ph 7

* *
6

Ph

5 7 6 7 7 7 6
3a

6.0

5.0

4.0

3.0

2.0

ppm

Figure 3-1. 400 MHz 1H NMR spectra of a benzene-d6 solution of 3a and a 20-fold molar excess of 1phenyl-1-proyne, before (lower spectrum) and after substrate conversion (upper spectrum). Proton resonances denoted by an asterisk (*) are assigned to lanthanocene reaction intermediates.

73

Chapter 3

0.8 4 Concentration (mol/L) 0.6 5 7 6

0.4

0.2

0 0 50 100 time (h) 150 200

Figure 3-2. The conversion of 4 (18 equiv.) and the formation of 5-7 during the cyclodimerization reaction catalyzed by 3a (30.9 mM) in benzene-d6 at 120 C, as monitored by 400 MHz 1H NMR spectroscopy. The lines drawn connecting the experimental points represent fitted first-order exponential functions. phenyl-1-propyne (as demonstrated by the reaction of Cp*2LaC(Ph)=C(Ph)H with 1-phenyl-1-propyne, giving 3a and cis-diphenylethene, Chapter 2). The observation of 3a as the resting state of the active catalyst argues for slow, rate-limiting alkyne insertion (ii, Scheme 3-3), followed by rapid intramolecular cyclization (iv) and protonolysis (v). This picture is consistent with 3a being the major organometallic species after reaction of 3a with a two-fold molar excess of 1-phenyl-1-propyne and the rate of substrate conversion being first-order in substrate (vide infra). Product analysis The major product 5 exhibits three broad singlets at 6.45, 3.02 and 1.75 ppm in a 1:2:3 ratio in the 1 H NMR spectrum of the reaction mixture, assigned to CHPh, endocyclic CH2 and exocyclic CH3, respectively. The proton resonances are broadened, because of long-range couplings, typical of substituted cyclobutenes.11 Characteristic 13C NMR resonances were observed at 15.10 (q, 1JCH = 125.6 Hz), 40.16 (t, 1JCH = 139.2 Hz) and 111.68 (d, 1JCH = 125.6 Hz) ppm, assigned to the CH3, endocyclic CH2 and CHPh groups, respectively. Both the proton and carbon chemical shifts are consistent with literature values of analogous substituted 3methylenecyclobut-1-enes.11 The trans stereochemistry of the exocyclic double bond was indicated by the absence of a NOE cross peak between 6.45 (CHPh) and 3.02 (CH2) ppm, while that between 3.02 and 1.75 ppm (CH3) was observed. The proposed structure of 5 was furthermore corroborated by correlations observed with 2D NMR spectroscopy (i.e. 1H,1H-gCOSY, 1H-13C gHSQC, 1H-13C gHMBC). Attempts to isolate (E)-3-benzylidene-2-methyl-1-phenylcyclobutene on a preparative scale were unsuccessful, as decomposition into unidentified compounds was observed upon contact with air (thereby excluding the facile use of chromatographic techniques) and at high temperatures during in vacuo distillation attempts (>200 C). In addition, suitable crystallization conditions employing a variety of solvents and solvent mixtures (e.g. acetonitrile, toluene, hexanes) were not found. Considerable precedent exists for the highly reactive nature of alkylidenecyclobutenes.3,11,12 Although some exceptions are known which are reported to exhibit limited stability in air, most alkylidenecyclobutenes undergo extensive decomposition or polymerization on exposure to air or attempted purification by chromatographic techniques or distillation and are only stable under an inert atmosphere and in the absence of acids, bases or radical reagents. Quantitative 1H NMR analysis of the cyclodimerization catalyzed by 1a-3a by means of an internal standard (hexamethyldisiloxane) and appropriate long pulse delays indicated that the formation of 5 was accompanied by that of several by-products. GC-MS analysis indicated the presence of phenylallene-d1, 1phenyl-1-propyne-dn and a variety of C18H16 isomers (at least fifteen of which the three major isomers

74

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

Scheme 3-3. The proposed catalytic cycle of the cyclodimerization reaction of 1-phenyl-1-propyne catalyzed by 1a-3a.
Ph 6 Ph Ph Cp*2La Ph Ph F C Ph Ph Ph Cp*2LaR R = CH(SiMe3)2 (1a), H (2a) Ph Ph Ph Ph

RH

ix

Cp*2La Ph

Ph

iv

viii

Cp*2La 3a Cp*2La Ph Ph Cp*2La Ph D Ph Ph

Cp*2La E Ph

vii

iii

vi

ii

Cp*2La Ph A

corresponded to 61, 15 and 13% of the total amount of the C18H16 isomers, as indicated by uncorrected FID-GC values) after quenching the reaction mixture with methanol-d4. The formation of phenylallene-d1 and 1-phenyl-1-propyne-d1 can be ascribed to the deuterolyis reaction of 3a which is known to give a mixture of acetylenic and allenylic deuterolysis products (Scheme 3-2, Chapter 2). The observation of the large amount of C18H16 isomers with GC-MS is inconsistent with the composition of the product mixture as observed with NMR spectroscopy and can be explained by thermal, oxygen- and/or acid-induced decomposition of the formed methylenecyclobutenes during GC analysis (vide infra). In fact, the electrocyclic ring-opening of alkylidenecyclobutenes, forming vinylallene and ultimately benzene derivatives as major products, has been implicated by several studies.13 Because extensive decomposition of the present alkylidenecyclobutenes during GC analysis was only observed to a relatively small degree, GC-MS could still be used to characterize the major reaction products. Because purification of the reaction products and reliable quantitative GC analysis were frustrated by the reactivity of the formed alkylidenecyclobutenes, a methodology was sought to convert the reaction products into stable derivatives. To distinguish the reaction products, furthermore, from the organic products formed from the organometallic species present, the crude reaction mixture was first treated with trimethylsilychloride (TMSCl), producing (Cp*2LaCl)n and silylated organic derivatives (vide supra). Catalytic hydrogenation of crude reaction mixtures, quenched with TMSCl, afforded the desired cyclobutane derivatives, based on the presence of six C18H20 isomers (in a 12:72:2:1:13:1 ratio, according to FID-GC) and two C21H28Si isomers (in a 71:29 ratio) as indicated by GC-MS, but these products could not be purified by column chromatography or characterized unambiguously by standard spectroscopic methods. Gratifyingly, careful analysis of the crude reaction mixture by means of multinuclear 1D and 2D NMR spectroscopy led to the identification of the two major by-products. The presence of 1H NMR resonances at 6.06, 3.27 and 1.73 in a 1:2:3 ratio pointed to the presence of (E)-3-benzylidene-1-methyl-2phenylcyclobutene (6) which was formed with a selectivity of 10-13%, depending on the reaction conditions (vide infra, Table 3-1). This proposal is supported by the corresponding 13C NMR resonances, 2D NMR

75

Chapter 3

Scheme 3-4. The proposed reaction sequences leading to the formation of 7 and 8.
Ph (not observed) 8 Ph Ph Ph Cp*2La L Ph . Cp*2La K vi Cp*2La Ph Ph . Ph J v i Cp*2La Ph G Ph Cp*2La . 3a Ph Cp*2La Ph Cp*2La Ph H Ph viii Ph Ph Ph iv Cp*2La I Ph iii . Ph Ph 7 (observed)

vii

Ph .

ii

spectroscopy and the similarity with reported alkylidenecyclobutenes. The formation of 6 can, furthermore, be explained by 1,2-insertion of 1-phenyl-1-propyne into the propargylic La-CH2 bond (vi and vii, Scheme 3-3), consistent with the catalytic sequences proposed for the formation of 5. Other by-products could, in principle, result from competing protonolysis relative to intramolecular ring-closing, giving rise to linear alk-4-en-1-ynes. The absence of acetylenic carbons ( 80-90 ppm) argues against the (significant) presence of these compounds, however. The absence of multiplets, having a ~7 Hz proton-proton coupling, in the vinylic region of the 1H NMR spectrum exludes, moreover, the presence of linear alk-4-en-1-ynes. Competing intermolecular alkyne insertion into B can also be discarded (Scheme 3-3), because organic products having masses higher than that of 1-phenyl-1-propyne dimers were not observed with GC-MS. It seems, therefore, that intramolecular alkyne insertion of B is rapid under the applied reaction conditions, competing effectively with substrate protonolysis and insertion. The 1H NMR spectrum of the reaction mixture displayed three characteristic proton resonances at 5.07, 4.57 and 4.12 ppm in a 1:1:1 ratio. A 1H-13C gHSQC experiment revealed that the former two resonances corresponded to a triplet (1JCH = 158.0 Hz) at 95.88 ppm in the 13C NMR spectrum, while the latter corresponded to a doublet (1JCH = 158.0 Hz) at 56.50 ppm. The carbon resonance at 95.88 ppm is suggestive of an allenylic methylene group (i.e. =C=CH2), but is too far downfield to correspond to a terminal allenylic methylene group and no resonance attributable to a central allenylic carbon resonance could be observed with 13 C{1H} NMR spectroscopy.14 Methylene groups resonating in the 110-90 ppm region of the 13C NMR spectrum are observed for methylenecyclobutane derivatives, however. Based upon these findings, the presence of 1,3-diphenyl-2-methyl-4-methylidenecyclobutene (7) was inferred which was formed with a selectivity of 911% (Table 3-1). The proposed structure is supported by the observed correlations with 2D NMR spectroscopy and the similarity of the NMR spectral data with reported analogous substituted methylenecyclobutenes (vide supra). In order to account for the (catalytic) formation of 1,3-diphenyl-2-methyl-4-methylidenecyclobutene (7), insertion of 1-phenyl-1-propyne into the La-C bond of the -allenylic tautomer of 3a seems to be the most plausible mechanistic scenario. Indirect evidence for the kinetic accessability of a -allenyl tautomer of 3a has been presented previously, as the reaction of 3a with methanol and phenylacetylene furnished mixtures of acetylenic and allenylic protonolysis products (Chapter 2). Accordingly, the reaction of 3a and 1-phenyl-1-

76

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

8 6 4 2 0 0 20 40 60 80 100 [cat] (mM) y = 0.0522x + 0.0624 R2 = 0.9873

Figure 3-3. A plot of the observed reaction rate kobs versus the catalyst concentration ([3a]) for the cyclodimerization reaction of 1-phenyl-1-propyne in benzene-d6 at 100 C and constant substrate concentration (0.76 M). The line drawn represents the least-squares fit to the data points. propyne may give rise to a lanthanum vinylallene species (H) after 2,1-insertion into the -allenylic tautomer (i and ii, Scheme 3-4). Further reactivity, most likely, involves protonolysis with 1-phenyl-1-propyne, forming a substituted vinylallene, or four-electron electrocyclic ring closure (vide infra), forming 1,3-diphenyl-2-methyl-4methylidenecyclobutene (7) after protonolysis (iii). Electrocyclization of vinylallenes is a well-studied reaction, but its synthetic use has been limited by the high reaction temperatures required (i.e. 200-500 C) and the fact that equilibrium mixtures of starting material and product are obtained.15 Over the last decade, however, appropriate substitution has been found to result in mild and unidirectional electrocyclization of vinylallenes.16 In fact, several titanated vinylallenes undergo facile isomerization to cyclobutenyltitanium compounds at room temperature.16e As a consequence, electrocyclization of H is considered to be more facile than that of the vinylallene formed upon protonolysis of H. This view is supported by the absence of NMR resonances attributable to such a vinylallene or derivatives thereof in the reaction mixture (during and after substrate consumption). These considerations led us to propose that 7 is formed via route (i)-(iv) (Scheme 3-4). It should also be noted that no indirect evidence for the 1,2insertion of 1-phenyl-1-propyne into the La-C bond of the -allenylic tautomer of 3a was found, which is consistent with the observed predominant 2,1-insertion mode of 1-phenyl-1-propyne in the -propargylic tautomer (vide supra). Other minor by-products were formed with a combined selectivity of 1-2% (1H NMR), depending on the reaction conditions. Although their low relative concentration impeded unambiguous characterization, structures similar to (E)-3-benyzlidene-2-methyl-1-phenylcyclobutene (5) and (E)-3-benyzlidene-2-phenyl-1methylcyclobutene (6) can be proposed on the basis of the comparable proton (i.e. 6.05, 6.01, 3.53, 3.46, 1.80 and 1.70 ppm) and carbon resonances (i.e. 15.09, 15.09, 41.45, 42.43, 117.45 and 116.27 ppm). Mechanistic pathways to these products may consist of transmetallation of 3a with 1-phenyl-1-propyne yielding phenylallene (as observed for 3b, vide infra), followed by regiorandom insertion of phenylallene into 3a. It should be noted that no experimental evidence for the formation of 8 was found in this study (Scheme 3-4). Kinetic study The kinetics of the cyclodimerization of 1-phenyl-1-propyne (4) catalyzed by Cp*2La(3-CH2CCPh) (3a) were studied by means of normalized 1H NMR spectroscopy, using long pulse delays to avoid signal saturation under the present anaerobic conditions. Substrate consumption was monitored by normalizing the intensity of the methyl proton resonance ( 1.65 ppm) against that of hexamethyldisiloxane ( 0.11 ppm) as an internal standard. However, the methyl resonance of the substrate overlapped partially with those of the products at relatively high substrate conversion, thereby lowering the accuracy of the kinetic data. Improved kinetic data were obtained by means of line-fitting procedures. Due to the slow rate of the cyclodimerization reaction, the progress of reaction could not be monitored at a constant reaction temperature by NMR spectroscopy. The NMR tubes were heated in an electric oven and transferred to the spectrometer after appropriate time intervals. It

103 k obs (M-1s -1)

77

Chapter 3

Table 3-1. The rate and selectivity of the cyclodimerization reaction of 1-phenyl-1-propyne (4) catalyzed by Cp*2LaCH2CCPh (3a) as a function of reaction temperature, substrate concentration and catalyst concentration.a Entry T 5b Ntc 6b 7b kobs 4 3a (C) (equiv.) (mM) (%) (%) (%) (10-3 M-1s-1) (10-3 h-1) 1 100 47 16.2 71 12 10 0.83(9) 6.9(3) 2 100 13 58.4 72 11 11 11.7(2) 6.6(3) 3 100 10 76.4 71 12 10 17.9(3) 5.7(1) 4 80 20 38.2 75 10 9 1.0(1) 1.6(1) 5 100 20 38.2 71 12 10 2.97(8) 4.7(1) 6 120 20 38.2 69 13 12 12.5(4) 20(1) 7 100 100 38.2 70 12 10 2.9(1) 4.6(2) a Reactions conditions: substrate (382 mol), benzene-d6. b The selectivity towards 5, 6 and 7 was calculated from the amount of product formed divided by the amount of substrate converted, as determined by normalized 1H NMR intensities from line-fitting procedures. Repeated experiments suggested an experimental error of 1%. c The turnover frequency Nt was calculated from the relationship Nt = t1/2[3a] using the half-life t1/2 as determined from nonlinear first-order exponential fitting of a plot of [4] versus time. seems that this methodology limited the accuracy of the kinetic analysis to some extent, especially for the slower reactions.17 Even so, the rate of substrate consumption could be fitted convincingly (R2 > 0.98) to first-order rate dependence on substrate concentration in most cases. For the reaction of Cp*2La(3-CH2CCPh) (3a) and 1-phenyl-1-propyne (4) in benzene-d6 solution, Figure 3-3 presents kinetic data typical of many runs, which show the rate to be first-order in substrate for at least 3 half-lives. The rate of substrate conversion was observed to be first-order in substrate over a 5-fold concentration range (0.76-3.81 M) at constant catalyst concentration (Entries 5 and 7, Table 3-1). When the initial substrate concentration was maintained constant and the concentration of 3a was varied over a 5-fold concentration range (16.2-76.4 mM), a plot of the reaction rate versus catalyst concentration ([3a]) indicates that the reaction is first-order dependent on 3a (Figure 3-3). Thus, the empirical rate law for the cyclodimerization of 4 catalyzed by 3a can be formulated as v = kobs[3a][4]. Influence of the reaction temperature The cyclodimerization reaction of 1-phenyl-1-propyne (20 equiv.) catalyzed by 3a was performed at different temperatures in order to investigate the effect of temperature on both the rate and selectivity. Complete conversion of 1-phenyl-1-propyne was observed after 7 days at 120 C. When the reaction was conducted at 100 C, complete substrate conversion was observed after 28 days. In both cases, the kinetic data obtained agreed well with first-order rate dependence on substrate. The accuracy of the kinetic data for the reaction performed at 80 C was limited by the slow rate of substrate conversion. After following the reaction for 30 days (1.7 halflives) the substrate consumption was fitted only moderately to first-order kinetic behavior in substrate (R2 = 0.9678). Repeated experiments involving fewer measurements (i.e. better temperature control) did not lead to kinetic data of higher accuracy. Thus, the present results clearly indicate that the rate of cyclodimerization is enhanced by increasing the reaction temperature, but the experimental error of the present methodology does not allow for an accurate determination of activation parameters by means of standard Eyring and Arrhenius analyses. The selectivity for the formation of (E)-3-benzylidene-1-methyl-2-phenylcyclobutene (5) decreased with increasing reaction temperature (Table 3-1). The increased formation of (E)-3-benzylidene-1-methyl-2phenyl-cyclobutene (6, from 1,2-insertion into the propargylic La-CH2 bond of 3a, vide supra) and 1,3-diphenyl2-methyl-4-methylidenecyclobutene (7, from 2,1-insertion into the allenylic La-C bond of 3a, vide supra) indicates that other processes compete more effectively with 2,1-insertion of 1-phenyl-1-propyne into the LaCH2 bond of 3a at higher reaction temperatures. 3.2.2. Influence of the metal ion size

The effect of changing the metal in organometallic compounds of rare-earth metals is predominantly steric in nature and many examples exist in literature where the selectivity and rate of reactions can be tuned by

78

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

Scheme 3-5. The observed reaction pathways for 3b in the presence of excess 1-phenyl-1-propyne.
D Ph 14-d1 Ph CD3OD Ph D Ph 13-d1 CD3OD Ph Ph

Cp*2Y 12b Ph -14

Ph ii

Cp*2Y 3b Ph

Ph i

Cp*2Y Ph 11b -13

Cp*Y 15b
1

Cp*Y 15b

varying the ionic metal radius along the lanthanide and group 3 triad. The available metal sizes range from 0.870 for trivalent scandium to 1.160 for trivalent lanthanum in formally eight-coordination.18 Yttrium (1.019 ), representing an intermediate size in this range, was chosen to evaluate the effect on the rate and selectivity of the rare-earth metallocene-catalyzed cyclodimerization reaction upon decreasing the metal ionic radius of the catalyst Cp*2LnCH2CCPh (3a). When a reaction mixture of Cp*2YCH2CCPh (3b) and an excess amount of 1-phenyl-1-propyne (1020 equiv.) was heated to 100 C in benzene-d6 for several days, no significant substrate consumption was observed with 1H NMR spectroscopy. Instead, small amounts of phenylallene (0.31 equiv. per Y) and (E)-3benyzlidene-2-methyl-1-phenylcyclobutene (0.16 equiv. per Y) were formed after several hours, while 3b was slowly consumed. Further heating did not change the amount of phenylallene and (E)-3-benyzlidene-2-methyl-1phenylcyclobutene, but 3b was completely converted within three days at 100 C into a 85:15 mixture of the pent-1-en-4-yn-1-yl yttrocene derivatives 11b and 12b, respectively, according to 1H NMR spectroscopy (Scheme 3-9). The major species 11b was characterized by multinuclear 1D and 2D NMR spectroscopy (see Experimental Section). A particularly interesting feature is the observation that the 13C NMR resonances of the acetylenic carbons in 11b exibit yttrium-carbon coupling. Attempts to isolate 11b on a preparative scale by fractional crystallization have not yet been successful. The proposed structures of 11b and 12b are, furthermore, supported by the identity of their quenching products with methanol-d4, i.e. (E)-1,5-diphenyl-1-deuterio-2methyl-1-penten-4-yne (13-d1) and (Z)-3,6-diphenyl-2-deuterio-2-penten-5-yne (14-d1). When the reaction mixture was allowed to react further at 100 C, proton resonances in the Cp* ( 2.5-1.5 ppm) and aliphatic region ( 1-0 ppm) appeared, while the intensities of resonances due to 11b and 12b diminished slowly. Unfortunately, NMR spectra were not structurally diagnostic due to overlapping signals. The reaction mixture was quenched with methanol-d4 after 24 days at 100 C and analyzed with GC-MS. Besides 13dn and 14-dn, the presence of small amounts of unidentified oligodeuterated, organic compounds C18H16 (several isomers of m/z 232, consistent with the mass of a dimer of 1-phenyl-1-propyne) and C19H24 (three isomers of m/z 252) was indicated. These findings indicate that insertion of 1-phenyl-1-propyne into the Y-C bond of 3b is feasible. In spite of the larger steric constraints of 3b, due to the smaller metal size, the electronically favored 2,1-insertion (i) predominates over the sterically favored 1,2-insertion (ii) of 1-phenyl-1-propyne at 100 C (Scheme 3-9).19 The presence of phenylallene can be explained by transmetallation of the -allenyl tautomer of 3b with 1phenyl-1-propyne (Scheme 3-6), as previously observed for sterically hindered propynylarenes and Cp*2LnCH2CCAr complexes (Chapter 2). Clearly, transmetallation of the -propargyl tautomer of 3b with 1phenyl-1-propyne is nonproductive. In contrast to 3a (Chapter 2), no further reactivity is observed for phenylallene and 3b.

79

Chapter 3

Scheme 3-6. Transmetalation reaction of 3b with 1-phenyl-1-propyne.

Ph Cp*2Y 3b Ph Cp*2Y . + Ph Cp*2Y 3b Ph + Ph + Ph Cp*2Y 3b Ph + Ph

The lack of catalytic substrate consumption can be attributed to the high stability of the pent-1-en-4yn-1-yl yttrocene derivatives 11b and 12b towards both intramolecular cyclization and substrate protonolysis. It seems that the yttrium metal is too small to allow intramolecular alkyne insertion, although an intramolecular metal-to-alkyne interaction is evident from the observed yttrium-carbon couplings of the acetylenic carbon resonances. This intramolecular alkyne coordination is most likely responsible for the absence of substrate protonolysis, as it effectively competes with intermolecular alkyne coordination. Upon prolonged heating, however, evidence for intramolecular ligand metallation was found. The observation of organic compounds C19H24 is consistent with the coupling products, originating from insertion of 1-phenyl-1-propyne into the Y-CH2 bond of the fulvene species (15b). Insertion of unsaturated bonds into the metal-methylene bond of a fulvene species have been reported for analogous Cp*FvM (M = Zr, Ti) complexes.20 Extensive H/D scrambling and the observation of proton resonances in the aliphatic region was also observed, when benzene-d6 solutions of Cp*2Ln(3-CH2CCPh) and Cp*2LnCH(SiMe3)2 were heated to 120 C (Chapter 2).21 The formation of fulvene derivatives [Cp*(C5Me4CH2-5:2)Ln] was put forward to account for these observations. The fact that the functionalized Cp* derivatives are not deuterated after quenching with methanol-d4 implies that the insertion products are not stable under the reaction condition. Regiorandom insertion of 1-phenyl-1-propyne and a 1,3-H shift of the insertion product can be proposed to rationalize the formation of three different isomers. In conclusion, no catalytic cyclodimerization of 1-phenyl-1-propyne was observed with Cp*2Y(3CH2CCPh) (3b). Although the present data indicate that 1-phenyl-1-propyne insertion into the Y-C bond of 3b does take place, both intramolecular alkyne insertion and intermolecular alkyne protonolysis of the formed insertion product are practically absent upon substituting the lanthanum metal center in the catalyst Cp*2LnCH2CCPh with the smaller yttrium. 3.2.3. Influence of the ancillary ligation

Introduction Structurally, a considerable opening of the metal coordination sphere at the -ligand equatorial girdle is obtained by replacing the bis(pentamethylcyclopentadienyl) ligation in Cp*2LnR by the ansa-ligation Me2SiCp2LnR (Cp = C5Me4) and it has been shown that linking the cyclopentadienyl ligands by a Me2Si group greatly influences the rates of insertion and -bond metathesis.22 The present results revealed that intermolecular alkyne insertion is rate limiting in the cyclodimerization reactions and that steric interactions between substrate and catalyst influence the observed selectivity. Reactions of Me2SiCp2LnCH(SiMe3)2 (Ln = Y 23b, Ce 23c) with 1-phenyl-1-propyne (20 equiv.) were conducted in order to investigate the effects on the rate and selectivity of the cyclodimerization reaction upon opening the metal coordination environment of the catalyst. Previous attempts to apply the catalyst precursor 23c in the catalytic cyclodimerization of 2-propyne did not lead to catalytic substrate conversion. Instead, a catalytically inactive product was formed, tentatively formulated as the 2-alkynyl derivative (Me2SiCp2CeCH2CCMe)n.6b

80

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes Me2SiCp2YCH(SiMe3)2 When a benzene-d6 solution of Me2SiCp2YCH(SiMe3)2 (23b) was heated to 120 C in the presence of a 20-fold molar excess of 1-phenyl-1-propyne, slow propargylic metalation took place, as seen from the slow formation of CH2(SiMe3)2. 1H NMR spectroscopy indicated that complete substrate conversion into several products took place after 16 days at 120 C. Upon complete substrate conversion, 12% of 23b was present in the reaction mixture. GC-MS analysis indicated the presence of dimers (m/z 232) and tetramers (m/z 464), but did not provide information on the number of formed products, due to thermal, oxygen- and/or acid-induced decomposition of the products during GC analysis (Section 3.2.1). Assuming a linear relationship between the response factors and the carbon numbers of the oligomers, FID-GC values of the reaction products could be expressed in terms of monomeric units. These normalized FID-GC values revealed that the reaction displayed a 89% selectivity for dimerization and a 11% selectivity for tetramerization. Even though the product mixture was too complex for conclusive NMR analysis, the major products could be identified as (E)-3-benyzlidene-1methyl-2-phenylcyclobutene (5) and 1,3-diphenyl-4-methyl-2-methylene-cyclobutene (7) (eq. 3.2).
Me2SiCp"2YCH(SiMe3)2 (23b) 5 mol% 4 C 6D 6 120 C Ph Ph + 5 13% Ph 7 10% Ph

(3.2)

Ph

The observed catalytic cyclodimerization of 1-phenyl-1-propyne in the presence of Me2SiCp2YCH(SiMe3)2 (23b) is in marked contrast with that of Cp*2YCH(SiMe3)2 (1b) for which no catalytic reactivity was observed. It seems that the increased coordination sphere of the catalyst allows for intramolecular alkyne insertion. The observed major products 5 and 7 are formed from 2,1-insertion into the propargylic Y-CH2 and allenylic Y-CHPh bond of the active catalyst, respectively (Section 3.2.1.). The observed regioselectivity corresponds to the electronically favored, but sterically more hindered insertion at the metal center and reflects the decreased steric demands of alkyne insertion in the more open coordination sphere of the catalyst. The similar preference for 2,1-insertion into both the propargylic Y-CH2 and allenylic Y-CHPh bond is remarkable, because products originating from substrate insertion into the allenylic metal-carbon were only minor byproducts in the reactions of Cp*2La(3-CH2CCPh) with 1-phenyl-1-propyne.
Me2SiCp"2CeCH(SiMe3)2 Ph (23c) 5 mol% 4 C6D6 120 C 5 25% Ph Ph + 6 27% + 7 14% Ph + unidentified tetramer 8 30%

(3.3)

Ph

Ph Ph

Me2SiCp2CeCH(SiMe3)2 Heating a benzene-d6 solution of Me2SiCp2CeCH(SiMe3)2 (23c) and 1-phenyl-1-propyne (20 equiv.) to 120 C resulted within several minutes in a clear color change from purple to brown-red. Monitoring the progress of reaction by 1H NMR spectroscopy revealed that the substrate was completely converted after 16 h. On the basis of NMR and GC-MS analysis, the products were identified as 5 (25%), 6 (27%), 7 (14%) and an unknown tetramer (m/z 464, 30%) (eq. 3.3). Although the complexity of the product mixture hampered attempts to determine the structure of the tetramer unequivocally by means of NMR spectroscopy, the presence of new, non-overlapping 1H NMR resonances did allow for quantitative 1H NMR analysis. After quenching the product mixture with methanol (3 equiv.), FID-GC revealed a 64% selectivity for dimerization and a 36% selectivity for tetramerization. These values agree only moderately with those obtained from quantitative 1H NMR analysis (i.e. 52% selectivity for dimerization and 48% selectivity for tetramerization), even after taking the decomposition product(s) of the catalyst (5%) into account. Apparently, a nonlinear relationship between response factors and carbon numbers exists for the present oligomers. Hence, the normalized FID-GC values represent only an estimate for the degree of oligomerization. When the analogous reaction was performed at 80 C, complete substrate conversion into 5 (9%), 6 (28%) and 8 (58%) was observed after 18 days. In this case, normalized FID-

81

Chapter 3 GC analysis after methanolyis of the reaction mixture revealed a 41% selectivity for dimerization and a 59% selectivity for tetramerization. These results indicate that Me2SiCp2CeCH(SiMe3)2 (23c) is a more active (pre)catalyst for cyclodimerization of 1-phenyl-1-propyne than Cp*2LaCH(SiMe3)2 (1a) and Cp*2La(3-CH2CCPh) (3a). Monitoring the reaction at 80 C with 1H NMR spectroscopy revealed quantitative propargylic metalation within 6 h. The rate of CH2(SiMe3)2 formation was found to be first-order in substrate concentration (R2 = 0.9914, kobs = 0.293(2) M-1min-1). This observation suggests that propargylic metalation of 23c is more rapid than that of 1a (i.e. quantitative within 5.4 h at 120 C). Even so, the observed 6-fold rate increase of substrate conversion for the reaction with 23c relative to that with 1a most likely reflects an increased rate of alkyne insertion in a more open metal coordination sphere of the catalyst. Unfortunately, the increased catalytic activity is accompanied by a lower selectivity. As compared to 3a, 2,1-insertion of substrate into the propargylic metal-carbon bond to form 5 seems to be less favored in this system relative to other modes of substrate insertion into the active catalyst, such as 1,2-insertion into propargylic Ln-CH2 to form 6 and 2,1-insertion into allenylic Ln-CH to form 7. In analogy to the reactions of 3a with 1-phenyl-1-propyne, the formation of products originating from the sterically more hindered insertion at the metal center (i.e. 5 and 8) decreased upon decreasing the reaction temperature. It is interesting to note that the relative formation of tetramer increased upon decreasing the reaction temperature. Apparently, the rate of protonolysis by substrate increases relative to that of multiple alkyne insertions upon increasing the reaction temperature. Concluding remarks The results for Me2SiCp2LnCH(SiMe3)2 (Ln = Y, Ce) reveal that the ancillary ligation influences not only the reaction rate by enhancing both the rate of propargylic metalation and alkyne insertion reaction sequences, but the reaction selectivity to a significant degree as well. Linking the cyclopentadienyl ligands with a Me2Si group led to a decreased relative formation of 5 and an increased tendency to form oligomers higher than dimers. It seems therefore that lowering the steric demands of alkyne insertion decreases the preference for the electronically preferred 2,1-insertion mode (to give 5) over the sterically preferred 1,2-insertion mode (to give 6) and decreases the preference for dimerization over higher oligomerization. The present findings suggest also that alkyne insertions into the allenylic Ln-CH(Ph) bond of the active catalyst are more facile in a Me2SiCp2 ligand environment than in Cp*2 ligand environment. This observation is consistent with the higher steric requirements for the formation of an inital 1-allenyl-like Lewis base adduct relative to that of an initial 1-propargyl-like Lewis base adduct (Chapter 2). 3.2.4. Reactions with 2-propynyltoluene

Introduction In order to assess the effect of ortho-substituents on the rate and selectivity of the lanthanidecatalyzed catalytic cyclodimerization of 1,4-dipropynylphenyl derivatives (Section 3.3), the reaction of 3a with 2-propynyltoluene (16) was investigated as a model substrate. Cp*2LaCH2CCPh When Cp*2LaCH2CCPh (3a) was heated in benzene-d6 to 100 C for several days in the presence of a 20-fold molar excess of 2-propynyltoluene (16), slow substrate consumption was observed with 1H NMR spectroscopy. The progress of reaction could be followed in time by monitoring the intensity of the substrate CH3 1H NMR resonance in time with 1H NMR spectroscopy, thereby revealing first-order rate dependence on substrate concentration (R2 = 0.9974, kobs = 1.64(2) M-1day-1) for at least 3 half-lives. After heating for 37 days, the reaction was stopped by quenching the reaction mixture with methanol-d4. GC/GC-MS analysis indicated the presence of several isomeric dimers (m/z 260, at least fifteen of which the two major isomers corresponded to 34 and 27% of the total amount of the C20H20 isomers, according to normalized FID-GC values), two isomers of C19H18 (m/z 246, corresponding to the mass of a cross-coupled dimer of 1-phenyl-1-propyne and 2propynyltoluene) and three isomeric trimers (m/z 390). The multitude of minor isomers can plausibly be ascribed to oxygen- and/or acid-induced decomposition during GC-analysis (Section 3.2.1). Normalized FID-GC revealed a 96% selectivity for dimerization and a 4% selectivity for trimerization. 1 H NMR analysis of the reaction mixture indicated that the propargyl derivative Cp*2LaCH2CCPh (3a) was completely converted into Cp*2LaCH2CCC6H4Me-2 within 1 day. Further heating led to slow substrate

82

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

0.8 0.7 [Substrate] (mol/L) 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 5 10 15 20 25 time (day) 30 35 40 45 50 1-phenyl-1-propyne 2-propynyltoluene

Figure 3-4. The conversion of 1-phenyl-1-propyne and 2-propynyltoluene catalyzed by 3a (38.2 mM) in benzene-d6 at 100 C as monitored by 400 MHz 1H NMR spectroscopy. The lines drawn connecting the experimental points are fitted first-order exponential functions. consumption and the formation of a complex reaction mixture, exhibiting several vinylic ( 7-5 and 4-3 ppm) and aliphatic ( 2.5-1.5 ppm) proton resonances. The two major products could be identified with NMR spectroscopy by comparison with previous alkylidenecyclobutenes as (E)-3-(2-methylbenzylidene)-2-(2methylphenyl)-1-methylcyclobutene (17) and (E)-3-(2-methylbenzylidene)-2-methyl-1-(2-methylphenyl)cyclobutene (18). After 37 days, 88% of the substrate was consumed and the two major products 17 (4.7 equiv. relative to 3a) and 18 (5.6 equiv.) were formed with a selectivity of 27 and 32%, respectively. The identity of the other dimers and the trimers is presently unknown, but 13C NMR analysis of the reaction mixture indicated several acetylenic signals, suggesting the presence of linear dimers or trimers.

(3.4)

Cp*2LaCH2CCPh (3a) 5 mol% + 16 C6D6 100 C 17 27% 18 32%

These findings reveal that the rate of cyclodimerization is decreased ~3-fold upon substituting the phenyl group of 1-phenyl-1-propyne with an ortho-methyl group. Increasing the steric bulk of the phenyl group also affects the selectivity of the reaction to a significant degree. The selectivity for the cyclodimer, resulting from 2,1-insertion of the propynylarene into the La-CH2 bond, decreased from 75 to 27%, while the major product formed from 1,2-insertion of the propynylarene into the La-CH2 bond of the catalyst. In contrast to analogous reactions with 1-phenyl-1-propyne, ortho-substitution led also to the formation of oligomers higher than dimers. Me2SiCp2CeCH(SiMe3)2 (23c) Upon heating a benzene-d6 solution of Me2SiCp2CeCH(SiMe3)2 (23c) and 2-propynyltoluene (20 equiv.) to 120 C, the purple solution turned brown-red within several minutes. Monitoring the progress of reaction by 1H NMR spectroscopy revealed quantitative propargylic metalation after 4 h and complete substrate conversion after 55 days. Because the formation of the active catalyst is rapid as compared to subsequent catalytic reaction sequences, the rate of reaction could reasonably well be modeled by first-order rate

83

Chapter 3 dependence on substrate concentration (R2 = 0.9807, kobs = 2.09(1) M-1day-1) for at least 3 half-lives. Several products formed and the major products were identified as 17 (33%), 18 (1%), 19 (2%) and 20 (2%) with NMR spectroscopy (eq. 3.5). Normalized FID-GC revealed a 85% selectivity for dimerization (at least 7 isomers, of which the 4 major isomers were present in a 8:8:41:5 ratio) and a 15% selectivity for trimerization (2 major isomers in a 1:1 ratio).
Me2SiCp"2CeCH(SiMe3)2 (23c) 5 mol% 16 C6 D6 120 C

(3.5)

+ 17 33% 18 1%

+ 19 2%

+ 20 2%

These results indicate that the rate of cyclodimerization is decreased dramatically upon substituting the phenyl group of 1-phenyl-1-propyne with an ortho-methyl group. The reaction of 23c with a 20-fold molar excess of 1-phenyl-1-propyne is complete after 16 h at 120 C, whereas the analogous reaction with 2propynyltoluene requires 55 days under identical reaction conditions for complete substrate conversion. Concomitantly, the selectivity for dimerization increased from 64% to 85%, while the nonspecific regioselectivity of cyclodimerization changed into a preference for cyclodimerization via the sterically favored 1,2-insertion mode. Substrate insertion into the allenylic Ln-CH bond is also possible in this catalytic system. Concluding remarks The increased relative formation of the cyclodimer formed from 1,2-insertion rather than 2,1insertion of substrate into the active catalyst can be rationalized, when the insertion of the carbon-carbon triple bond into the lanthanide-carbon bond of the propargyl is taken into account (Scheme 3-1). Insertion most likely takes place via a concerted, four-centered transition state (Scheme 3-7).23 When 1-aryl-1-alkynes or 1-aryl-1alkenes add to the metal-carbon bond of electrophilic metal complexes, the electronically preferred 2,1-insertion mode commonly outweighs the sterically preferred 1,2-insertion mode.24 The higher observed preference for 1,2insertion suggests that the steric hindrance resulting from ortho-methyl substitution decreases the rate of 2,1insertion relative to that of 1,2-insertion. Another effect of ortho-methyl substitution is the occurrence of trimerization in the reactions of 3a. The reaction of 3a with 2-propynyltoluene displayed a 8% selectivity for trimerization and a 92% selectivity for dimerization, whereas exclusive dimerization was observed for the analogous reaction with 1-phenyl-1-propyne. Apparently, ortho-methyl substitution leads also to a decrease of the relative rate of intramolecular alkyne insertion (ii) and/or alkyne protonolysis (ii, Scheme 3-1), thereby favoring intermolecular alkyne insertion. A decreased relative rate of intramolecular insertion is also consistent with the observation of acetylenic carbons in the reaction mixture with 13C NMR spectroscopy, indicative of linear dimers and/or trimers. Scheme 3-7. The insertion of propynylarene into the Ln-CH2 bond of a rare-earth metallocene propargyl.
R 1,2 L2Ln Me D R R L2Ln R A Me R 2,1 R

Me

Me

84

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

Scheme 3-8. The formation of cross-conjugated polymer M and conjugated polymer N via selective headto-head (via 2,1-metal insertion) and head-to-tail (via 1,2-metal insertion) propargylic cyclodimerization of propynylarenes, respectively, and the formation of cross-conjugated polymer O and conjugated polymer P via selective head-to-head (via 2,1-metal insertion) and head-to-tail (via 1,2-metal insertion) allenylic cyclodimerization of propynylarenes, respectively.

n M 2,1 2,1 O

1,2

1,2

n N
n

Interestingly, ortho-methyl substitution has a different effect on the degree of oligomerization in the reactions of 23c. The reaction of 23c with 1-phenyl-1-propyne displayed a 70% selectivity for dimerization and a 30% selectivity for tetramerization, whereas the analogous reaction with 2-propynyltoluene exhibited a 85% selectivity for dimerization and a 15% selectivity for trimerization. In this case, the increased steric bulk of the substrate lowers the rate of multiple alkyne insertions into the active catalyst as compared to that of propargylic metalation and/or intramolecular alkyne insertion.

3.3.
3.3.1.

The cyclodimerization of dipropynylaromatics

Introduction

A catalytic system capable of catalyzing the cyclodimerization of propynylarenes selectively to headto-head dimers (via 2,1-metal insertion into the propargylic metal-carbon bond) should, in principle, yield a cross-conjugated polymer M, when allowed to react with 1,4-dipropynylbenzene (Scheme 3-8).25 Similarly, a catalytic system capable of selective head-to-tail cyclodimerization (via 1,2-metal insertion into the propargylic metal-carbon bond) should give rise to a conjugated polymer N. Both type of polymers are unprecedented and represent as such an interesting new class of (cross)-conjugated polymer in this very active field of research.26,27 These polymerization reactions involving bifunctional substrates may be considered as taking place via a step growth mechanism.28 Polymerization reactions taking place by a step growth mechanism give rise to high molecular weight polymers only at high degrees of monomer conversion and the average length of the polymer chain is mostly limited by the presence of side reactions, the solubility of the growing polymer and the viscosity of the reaction medium. As a consequence, step growth polymerization reactions place generally stringent requirements on any reaction to be used for polymerization, such as very high conversions and selectivities. The factors that govern the rate and selectivity of the lanthanidocene-catalyzed cyclodimerization reaction of propynylarenes as model substrates has been investigated. Unfortunately, the observed selectivities were not high, the rates were low and the major products were not stable under ambient conditions. In some cases, minor products originating from substrate insertion into the allenylic metal-carbon bond of the catalyst

85

Chapter 3 were observed. Similar to propargylic coupling (insertion into the propargylic metal-carbon bond), allenylic coupling may take place in a 1,2- (head-to-tail) and 2,1-manner (head-to-head), thereby introducing structural defects into the polymer backbone (Scheme 3-8). Despite the above limitations, explorative reactions of several catalyst precursors with dipropynylarenes were performed in order to assess the feasibility of the present cyclodimerization reaction as a route towards novel (cross)-conjugated polymers. 3.3.2. 1,4-Dipropynylbenzene

When Cp*2LaCH(SiMe3)2 (1a) was allowed to react in benzene-d6 at 100 C with an 8-fold molar excess of 1,4-dipropynylbenzene (21) and monitored with 1H NMR spectroscopy, slow metalation was observed, as evidenced by the formation CH2(SiMe3)2. After 1 day 66% of 1a was converted and 1H NMR analysis indicated that one major 3-propargyl/allenyl derivative (one propargylic CH2 proton resonance at 2.81 ppm and one new Cp* resonance at 1.89 ppm in a 1:15 ratio) and one cyclodimer (broad singlets at 6.28, 2.88, 1.70, 1.70 and 1.71 in a ratio of 1:2:3:3:3, respectively) were present in the reaction mixture. Further heating led to substrate consumption (~90% after 4 days as determined by line-shape analysis) and the formation of broad signals around 6.5, 3.0 and 1.7 ppm, accompanied by a gradual change of the initially light-yellow solution into a viscous deep-red liquid and ultimately the precipitation of a dark-red solid. Because the formed polymer was found to be insoluble in both aromatic (i.e. benzene-d6, toluene-d8) and halogenated deuteriosolvents (i.e dichloromethane-d2, tetrachloroethylene-d2, bromobenzene-d5), even after prolonged heating under an inert atmosphere, characterization of the formed polymer by means of 13C NMR spectroscopy could not be performed. The soluble portion of the reaction mixture displayed a 1H NMR spectrum too complex for unambiguous assignments, especially after exposure to air. Also, GC/GC-MS analysis was not diagnostic in this respect. Similar results were obtained after increasing the amount of substrate or the use of Cp*2LaCH2CCPh (3a). For instance, when 3a was allowed to react with an 50-fold molar excess of 1,4-dipropynylbenzene in benzene-d6 at 100 C, a dark red mixture was obtained after 14 days, containing a red precipitate. No solvents were found that allowed 13C NMR analysis of the formed polymer. 3.3.3. 1,4-Dipropynyl-2,5-di-n-hexylbenzene.

Introduction In order to facilitate the characterization of the polymer, it was decided to increase the solubility of the polymer by substituting 1,4-diethynylbenzene with flexible aliphatic side chains.29 It is likely that the incorporation of substituents onto the monomer will influence both the reactivity of the monomer and the regioselectivity of the reaction, as observed for 2-propynyltoluene (16) relative to propynylbenzene in the analogous cyclodimerization reaction (Section 3.2.4). The choice of the substituent is not clear-cut, however. On the one hand, it can be expected that both solubility and the degree of polymerization will increase with the length of the side-chain. On the other hand, however, the steric size of the substituent will most likely hinder metal coordination, thereby diminishing the observed reactivity as well.30 The n-hexyl substituent was chosen based on reports that the degree of polymerization and the solubility of poly(p-phenyleneethynylene)s substituted with 2,5-di-n-hexyl groups did not increase significantly upon chain elongation of the substituents.31 Reactions with Me2SiCp2CeCH(SiMe3)2 (23c) Encouraged by the high activity of the catalyst precursor Me2SiCp2CeCH(SiMe3)2 (23c) in the catalytic cyclodimerization of 1-phenyl-1-propyne, reactions of 23c and a 20-fold molar excess of 1,4dipropynyl-2,5-di-n-hexylbenzene (22) were performed in benzene-d6 at 120 C. 1H NMR spectroscopy indicated the quantitative formation of CH2(SiMe3)2 within 2 h, while the intensity of the aromatic CH and propargylic CH3 1H NMR resonances of 22 decreased slowly upon further heating. Substrate conversion was accompanied by the formation of new resonances in the vinylic ( 7.0-3.0 ppm) and aromatic region of the 1H NMR spectrum. The newly formed 1H NMR resonances became increasingly broad and their intensity slowly decreased upon further heating (Figure 3-5). Complete substrate conversion was observed after 15 days and continued heating led to the precipitation of a red solid after 22 days.

86

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

6.0

5.0

4.0

G E C Si
23c

H H C6D 6 120 C D E-G

Ce CH(SiMe3) 2

+ 20

22

B
#

*
6.0 5.0 4.0 3.0 2.0 1.0 0.0 ppm

9.0

8.0

7.0

Figure 3-5. 400 MHz 1H NMR spectra in benzene-d6 of 23c and 22 (20 equiv.) before (lower spectrum) and after heating for 10 days at 120 C (upper spectrum). Resonances in the lower spectrum due to the solvent (#) and CH2(SiMe3)2 (*) are denoted, while resonances due to the catalyst precursor are unassigned. The relatively large number of 1H NMR resonances in the vinylic region observed during the 23ccatalyzed polymerization reaction of 22 point to nonselective cyclodimerization (Figure 3-7). The 1H NMR resonances of the products formed from cyclodimerization of 1-phenyl-1-propyne can be used to identify the nature of the formed C-C linkages in the polymer of 22. The resonances at 6.52 and 3.85 ppm are assigned to the CH and CH2 groups formed from head-to-head propargylic C-C coupling, while the resonances at 6.24 and 3.29 ppm are assigned to CH and CH2 groups formed from head-to-tail propargylic coupling. These assignments are supported by the analogous CH and CH2 1H NMR resonances of (E)-3-benyzlidene-1-methyl-2phenylcyclobutene (5) (i.e. 6.45 and 3.03 ppm, respectively) and (E)-3-benyzlidene-2-methyl-1phenylcyclobutene (6) (i.e. 6.10 and 3.27 ppm, respectively). Because only allenylic coupling of 1-phenyl-1propyne led to products resonating in the region of 5.1-4.1 ppm, the presence of 1H NMR resonances in the region of 6.0-4.0 ppm demonstrates that insertion of 22 into the allenylic metal-carbon bond of the catalyst is also feasible in this system. These considerations imply that head-to-head propargylic coupling is the dominant mode of C-C coupling in the reaction of 23c and 22. Reactions with Cp*2La(3-CH2CCPh) (3a) To determine the effect of catalyst structure on the mode of C-C coupling in the present polymerization reactions, the reaction of Cp*2La(3-CH2CCPh) (3a) and a 20-fold molar excess of 1,4dipropynyl-2,5-di-n-hexylbenzene (22) was conducted in benzene-d6 at 120 C. 1H NMR spectroscopy indicated very slow substrate consumption. For example, only 82% substrate conversion was observed after 50 days. After 3 months, the reaction mixture was found to be completely solidified into a dark-red solid. 1H NMR resonances observed during polymerization indicated that allenylic coupling was practically absent in this system and that propargylic head-to-head coupling was the dominant mode of C-C coupling in the reaction of 3a and 22. Product analysis In analogy to the methylenecyclobutenes formed from cyclodimerization of 1-phenyl-1-propyne, the formed polymers were found to be both acid- and air-sensitive. When the polymer solution was opened to air or quenched with methanol, an instantaneous color change form dark-red to light-yellow was observed. Attempts to isolate and identify products from these yellow suspensions under aerobic conditions afforded complicated mixtures that defied characterization. Attempts to remove the catalyst from the reaction mixture by quenching the solidified reaction mixture with a small amount of dry methanol (3 equiv. relative to the precatalyst) also led

87

Chapter 3

Scheme 3-10. Proposed polymer structure after quenching with trimethylsilylchloride.

Me3SiCl A -(L2LnCl)n Me3Si C6H13 H13C6 C6H13 A = C6H13 H13C6 y . SiMe3 A

L2Ln

H13C6

to complicated mixtures of ill-defined compounds, accompanied by a color change from red to yellow. Catalyst removal in the absence of air and acids was accomplished successfully by quenching the reaction mixture with trimethylsilylchloride. This methodology did not lead to a color change of the product mixture. Quenching with trimethylsilylchloride introduces trimethylsilyl groups onto the formed polymer. Because the resting state of the catalyst was found to be the 3-propargyl/allenyl derivative Cp*2La(3CH2CCPh) in the permethyllanthanocene-catalyzed cyclodimerization of 1-phenyl-1-propyne (Section 3.2.1), a similar species may reasonably be assumed to represent the resting state of the catalyst in the present polymerization reactions. In analogy to other electrophilic reagents (e.g. methanol, phenylacetylene, Chapter 2), the reaction of such an 3-propargyl/allenyl derivative with trimethylsilylchloride is likely to produce both allenylic and propargylic quenching products. Hence, plausible end-group structures of the polymer formed after quenching with trimethylsilylchloride are believed to be: Me3SiCH2CC-, Me2SiCH=C=CH- and CH3CC(Scheme 3-10). Poly(3a+22) When the solidified reaction mixture of 3a and 22 was treated with trimethylsilylchloride after polymerization, a dark-red solution was obtained containing a dark-red solid. The suspension was filtered and washed with chloroform, affording a chloroform-soluble and chloroform-insoluble polymer fraction. The chloroform-soluble polymer fraction poly(3a+22) was isolated as a red solid in 27% yield, while the insoluble fraction was isolated as a darker red solid in 35% yield. Both fractions were analyzed by infrared spectroscopy and the close resemblance of the infrared spectra of both polymer fractions suggests a similar structure. The insolubility of the chloroform-insoluble polymer fraction may be due to its higher molecular weight and/or the occurrence of cross-linking reactions. It can be envisioned that C-C bond forming reaction sequences that give rise to the formation of higher oligomers in the reaction of the monofunctional substrates introduce oligofunctional structural defects in the polymer (Section 3.2.4). These defects are likely to produce cross-linked structures upon further reaction. The infrared spectra of both polymer fractions did not display symmetrical C C stretching vibrations (observed at 2225 cm-1 for the monomer), thereby suggesting a relatively high degree of polymerization and the absence of linear oligomeric structures (Figure 3-6). Both polymer fractions displayed absorptions at ~1670, ~1600 and 1015 cm-1 that are not present in the infrared spectrum of the monomer. The absorptions near 1670 and 1600 cm-1 are presently assigned to symmetric and asymmetric coupled C=C-C=C stretching vibrations of the four-membered carbocycles, respectively, while the absorption at 1015 cm-1 is assigned to the out-of-plane =C-H bending vibration of the exo-methylene group.32,33 The vibrations at 900 and 722 cm-1 point to the presence of vinylidene (R2C=CH2) and cis-alkene (cis-RCH=RCH) groups, respectively (Figure 3-6). In addition, no evidence for the proposed end-group structures was obtained with infrared spectroscopy, however.34 The chloroform-soluble polymer fraction could also be analyzed with NMR spectroscopy (Figure 3-7). Previous 1H NMR assignments observed during the polymerization reaction are supported by the observed 13 C NMR resonances of the isolated, chloroform-soluble polymer fraction. The 13C NMR resonances at 109.40, 40.72 and 14.12 ppm are assigned to the CH, CH2 and CH3 groups, respectively, of the cyclodimer formed from head-to-head propargylic coupling, while the 13C NMR resonances at 116.71, 39.62 and 10.98 ppm are

88

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

100 80 60

22
Transmittance (%)
100

90

insoluble fraction poly(3a+22)

100

poly(23c+22)

90

3200

2800

2400

2000

1600
-1

1200

800

Wavenumber (cm )

Figure 3-6. IR spectra of the 1,4-dipropynyl-2,5-di-n-hexylbenzene (22) (KBr, upper spectrum), the chloroform- insoluble fraction of poly(3a+22) (KBr, middle spectrum) and poly(23c+22) (KBr, lower spectrum). assigned to the CH, CH2 and CH3 groups, respectively, of the cyclodimer formed from head-to-tail propargylic coupling. These assignment are consistent with the analogous 13C NMR resonances of (E)-3-benyzlidene-1methyl-2-phenylcyclobutene (5) (i.e. 111.66, 40.16 and 14.16 ppm, respectively) and (E)-3-benyzlidene-2methyl-1-phenylcyclobutene (6) (i.e. 112.42, 36.26 and 10.81 ppm, respectively). Low-intensity 13C NMR resonances in the 100-90 ppm region are reminescent of the resonance previously observed for the CH2=C group of 1,3-diphenyl-2-methyl-4-methylidenecyclobutene (7) (i.e. 95.88 ppm). However, the presence of such groups, originating from 2,1-insertion into the allenylic metal-carbon bond of the catalyst, is ruled out by the absence of corresponding 1H NMR resonances (i.e. 5.05, 4.55 and 4.12 ppm of the CHH=C, CHH=C and CHPh groups in 7, respectively). The significance of the 13C NMR resonances in the 100-90 ppm region is therefore not understood at present. Attempts to determine the number-average degree of polymerization Pn by means of end-group analysis were hampered by the fact that the existence of propynyl end-groups could neither be confirmed nor discarded spectroscopically. If propynyl end-groups are neglected, quantitative 1H NMR analysis of the trimethylsilyl groups relative to the aromatic or methylenecyclobutenic CH and CH2 groups (of the soluble polymer fraction in dichloromethylene-d2 at 80 C) reveals a Pn of 36(2).35 The solubility of the polymer was too low in aromatic and halogenated deuteriosolvents to allow quantitative 13C NMR analysis and its air-sensitive nature precluded MALDI-TOF and GPC analysis under aerobic conditions. Poly(23c + 22) When the solidified reaction mixture of 23c and 22 was treated with trimethylsilylchloride after polymerization, a dark-red, chloroform-insoluble solid was obtained in virtually quantitative yield. The infrared spectrum of the polymer (23a+22) resembles that of poly(3a+22), but the intensity of the vibrations at 900 and 722 cm-1 point to an increased presence of vinylidene (R2C=CH2) and cis-alkene (cis-RCH=RCH) groups (Figure 3-6).32 Hence, infrared spectroscopy supports previous notion, based on 1H NMR resonances observed during polymerization (vide supra), that the reaction of 23c and 22 takes place via nonselective C-C bond forming reaction sequences.

89

Chapter 3

F D

C6H 13 X X

G Me 3SiCH 2C H = Me3 SiCH C

C CH C

H 13C6 C 6H13 X H13C6 A

E H13 C6

I CH3 C

C 6H13 B C C F I y

B D
6.0 5.0 4.0

G H

9.0

8.0

7.0

3.0

2.0

1.0

0.0

pp

3 6 H 13 C6 C 6 H13 X H 13C 6 1 C 6H 13 2 3 1 4 5 y 2 6 4 C 6H 13 X

x
100

5 H 13 C6

140
1

120

80

60
13 1

40

20

0 ppm

Figure 3-7. 500 MHz H (upper spectrum) and 125.7 MHz benzene-d6 at 80 C of the soluble fraction of poly(3a+22).

C{ H} (lower spectrum) NMR spectrum in

3.4.

Conclusions

The present mechanistic study has demonstrated that the cyclodimerization reaction of 1-methylalk-2ynes catalyzed by rare-earth permethylmetallocenes proceeds via well-established elementary reactions, such as protonolysis of a metal-carbon -bond by the acidic propargylic hydrogen of 1-methylalk-2-ynes and insertion of carbon-carbon triple bonds into a metal-carbon -bond. Intermolecular insertion of substrate into the metalcarbon bond of an 3-propargyl/allenyl derivative was found to be rate-limiting. In accord with previous investigations of rare-earth metallocene 3-propargyl/allenyl derivatives, the involvement of both the 1-allenyl and 1-propargyl tautomer was implicated by the identities of the reaction products. The practical utility of this catalytic reaction was found to be limited by the small scope of substrates, the modest selectivities and the low catalytic rate. The observed catalytic activity and selectivity were found to be governed by a delicate balance between the steric properties of both the catalyst and substrate. The combination of Cp*2LaCH2CCPh and 1-phenyl-1-propyne performed best in catalysis. Changes in the metal ion

90

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes size (Cp*2La Cp*2Y), the ancillary ligation (Cp*2Ln Me2Cp2Ln) or substrate structure (propynylbenzene 2-propynyltoluene) did not lead to improved catalytic behavior. Explorative reactions of Cp*2LaCH2CCPh with 1,4-dipropynylbenzenes indicated that the slow rate, the low selectivity and the reactive nature of the formed polymer hampered the use of the present cyclodimerization reaction considerably as a practical route towards novel soluble (cross)-conjugated polymers.

3.5.

Experimental section

General considerations. For general remarks and physical and analytica measurements, see Section 2.7. The compounds Me2SiCp2LnCH(SiMe3)2 (Ln = Y36, Ce37) and 2,5-di-n-hexyl-1,4-propynylbenzene38 were prepared according to literature procedures. 1-Phenyl-1-propyne (distilled over CaH2) and chlorotrimethylsilane (distilled over KOH) were purchased from Aldrich and dried before use as recommended.39 General purification procedure for 1-methylalk-2-ynes. The liquids obtained after synthesis or received after purchase were brought in a flask containing freshly ground CaH2 and stirred at room temperature under nitrogen for at least 24 h after several freeze-thaw-pump degassing cycles. Vacuum transfer afforded colorless oils for 1-phenyl-1-propyne and 2-propynyltoluene. 1,4-Di(prop-1-ynyl)benzene was dried as a pentane solution with CaH2, filtered and crystallized at low temperature to afford an off-white crystalline material. 2,5Di-n-hexyl-1,4-propynylbenzene was dried as a pentane solution with CaH2 and filtered through a plug of alumina (neutral, activated) to afford a colorless oil after evaporation to dryness. These substrate samples were subsequently stored at -30 C under a nitrogen atmosphere. 1,4-Di(prop-1-ynyl)benzene (21). A similar procedure as described for 2-propynyltoluene (Chapter 2) was applied, using 1,4-diiodobenzene (4.29 g, 13.0 mmol), Pd(PPh3)4 (0.75 g, 0.65 mmol), propynylmagnesium bromide (80 mL, 40 mmol, 0.5 M in THF) and anhydrous ZnBr2 (8.10 g, 36.0 mmol). After 6 h stirring at room temperature, the reaction mixture was quenched with 50 mL of a saturated aqueous NaCl solution. Filtration, separation of the organic layer, drying over MgSO4 and flash column chromatography (silica, petroleum ether) afforded yellow crystalline material after rotatory evaporation. Sublimation (80 C/1 mmHg) and repeated recrystallization from toluene produced colorless crystals. Yield: 1.84 g (92%). 1 H NMR (300 MHz, CDCl3, 25 C): 1.99 (s, CH3, 3 H), 7.24 (s, CH, 2 H). 13C NMR (75 MHz, CDCl3, 25 C): 4.35 (CH3), 79.49 (C C), 87.28 (C C), 123.14 (i-C), 131.28 (CH). IR (neat, [cm-1]): 2950 (m), 2925 (s), 2845 (m), 2230 (w), 1490 (s), 1465 (m), 1450 (m), 1425 (m), 1260 (w), 1020 (w), 990 (w), 905 (s), 870 (s), 725 (s). GC-MS, m/z (relative intensity): 154 (M+, 100), 153 (M+ - H, 43), 152 (65), 151 (19), 139 (9), 126 (6), 115 (22), 77 (5), 76 (17), 75 (7), 63 (11), 51 (7). Anal. Calcd. for C12H10 (154.21): C, 93.46%; H, 6.54%. Found: C, 93.75%; H, 6.62%. Reaction of Cp*2La(3-CH2CCPh) (3a) with 1-phenyl-1-propyne. NMR scale. Cp*2La(3CH2CCPh) (10.0 mg, 19.1 mol) was dissolved in 0.50 mL of a benzene-d6 solution of hexamethyldisiloxane (1.4 mM). After addition of 1-phenyl-1-propyne (47.7 L, 381 mol) with a microsyringe, the mixture was heated to 120 C and monitored with 1H NMR spectroscopy. After 7 days at 120 C, the substrate was consumed completely and the mixture was quenched with methanol-d4. GC/GC-MS analysis indicated the presence of 1phenylprop-1-ene, Cp*D, phenylpropadiene-d1, 1-phenyl-1-propyne-dn and at least 15 isomers of C18H16. The three major C18H16 isomers were assigned to the three major products, as observed with NMR spectroscopy. The major product of the reaction mixture could be characterized with HR-MS. (E)-3-benzylidene-1-methyl-2Scheme 3-11. Numbering scheme of 5. phenylcyclobutene (5): 1H NMR (C6D6, 25 C, 500 MHz): 1.75 (dt, 4JHH = 1.6 Hz, 6JHH = 0.6 Hz, CH3, 3 H, A), 3.03 (dt, 4JHH = 1.6 Hz, 4JHH = 0.6 Hz, CH2, 2 H, B), 6.45 (sept., 4JHH = 0.6 Hz, D =CH, 1 H, C) ppm. The aromatic signals could 7 not be assigned due to considerable overlap with 5C B other signals. 13C NMR (C6D6, 25 C, 125.7 4 3 MHz): 15.09 (q, 1JCH = 126.7 Hz, CH3, 6), 40.16 1 2 1 (t, JCH = 138.5 Hz, CH2, 4) , 111.66 (d, 1JCH = 8 A 6 152.2 Hz, =CH, 5), 127.62 (s, C, 7), 138.89 (s, C, E 8), 140.31 (s, C, 3), 142.64 (s, C, 2), 146.72 (s, C, 1) ppm. The aromatic signals could not be

91

Chapter 3 assigned due to considerable overlap with other signals. 1H-1H gCOSY (500-500 MHz, C6D6, 25 C): ABC, BAC, CAB. 1H-1H gNOESY (500-500 MHz, C6D6, 25 C): ABE, BAD, CDE. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A6, B4, C5. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A1-4, B1-3, C2-4,7. GC-MS, m/z (relative intensity): 232 (M+, 96), 231 (30), 217 (M+ - CH3, 100), 216 (46), 215 (59), 203 (14), 202 (M+ - 2 CH3, 49), 191 (13), 190 (6), 189 (17), 153 (8), 152 (8), 128 (8), 116 (8), 115 (31), 107 (6), 101 (6), 89 (7), 77 (6), 63 (6). GC-MS, m/z (calc., found): 235 (0.1, 0.1), 234 (1.8, 2.0), 233 (19.6, 20.7), 232 (100.0, 100.0). HR-MS: C18H16, calc.: 232.12520, found: 232.12585. (E)-3-benzylidene-1-phenyl-2-methylcyclobutene (6): 1H NMR (C6D6, 25 C, 500 MHz): 1.73 (dt, 5 JHH = 2.0 Hz, 5JHH = 0.6 Hz, CH3, 3 H, A), 3.27 (dq, 4JHH = 2.0 Hz, 4JHH = 1.5 Hz, CH2, 2 H, B), 6.10 (qt, 4JHH = 1.5 Hz, 5JHH = 2.0 Hz, =CH, 1 H, C) ppm. The aromatic signals could not be assigned due to considerable overlap with other signals. 13C NMR (C6D6, 25 C, 125.7 MHz): 10.81 (q, 1JCH = 126.8 Hz, Scheme 3-12. Numbering scheme of 6. CH3, 8), 36.26 (t, 1JCH = 137.8 Hz, CH2, 4), 112.42 (d, 1JCH = 151.2 Hz, =CH, 5) ppm. Due to overlapping signals, the aromatic and D cyclobutenic carbon signals could not be 7 assigned. 1H-1H gCOSY (500-500 MHz, C6D6, 25 5C C): ABC, BAC, CAB. 1H-1H gNOESY (500-500 B 4 3 MHz, C6D6, 25 C): ACE, BDE, CAD, DBC, EAB. 1H1 2 13 E C gHSQC (500-125.7 MHz, C6D6, 25 C): A8, 8 A 6 B4, C5. GC-MS, m/z (relative intensity): 232 (M+, 97), 231 (13), 218 (21), 217 (M+ - CH3, 100), 216 (34), 215 (52), 203 (12), 202 (M+ - 2 CH3, 47), 155 (10), 153 (8), 141 (10), 128 (12), 116 (16), 115 (36). GC-MS, m/z (calc., found): 235 (0.1, 0.3), 234 (1.8, 3.3), 233 (19.6, 26.3), 232 (100.0, 100.0). 1,3-Diphenyl-2-methyl-4-methylidenecyclobutene (7): 1H NMR (C6D6, 25 C, 500 MHz): 1.75 (m, CH3, 3 H, A), 4.11 (m, CHPh, 1 H, B), 4.57 (q, 6JHH = 0.8 Hz, =CHH, 1 H, C), 5.07 (dq, 4JHH = 1.2 Hz, 6JHH = 0.8 Hz, =CHH, 1 H, D) ppm. Coupling patterns could not be resolved completely. The aromatic signals could not be assigned due to considerable overlap with Scheme 3-13. Numbering scheme of 7. other signals. 13C NMR (C6D6, 25 C, 125.7 MHz): 13.22 (q, 1JCH = 125.2 Hz, CH3, 6), 56.50 1 JCH = 158.0 Hz, CHPh, 4) , 95.88 (t, 1JCH = (d, F 158.0 Hz, =CH2, 5) ppm. Due to overlapping C 5 D signals, the aromatic and cyclobutenic carbon B 4 3 signals could not be assigned. 1H-1H gCOSY 1 2 E (500-500 MHz, C6D6, 25 C): ABCD, BACD, CABD, DABC 1H-1H gNOESY (500-500 MHz, C6D6, 25 A 6 C): ABE, BACF, CBDF, DCE,. 1H-13C gHSQC (500125.7 MHz, C6D6, 25 C): A6, B4, C5, D5. GC-MS, m/z (relative intensity): 232 (M+, 42), 231 (22), 218 (18), 217 (M+ - CH3, 100), 216 (44), 215 (58), 203 (11), 202 (M+ - 2 CH3, 42), 115 (27), 108 (8). GC-MS, m/z (calc., found): 235 (0.1, 0.1), 234 (1.8, 1.8), 233 (19.6, 18.7), 232 (100.0, 100.0). Reaction of Cp*2Y(3-CH2CCPh) (3b) with 1-phenyl-1-propyne. NMR scale. Cp*2Y(3CH2CCPh) (18.1 mg, 38.1 mol) was dissolved in benzene-d6 (500.0 L) in a Teflon-capped NMR tube and 1phenyl-1-propyne (48.0 L, 384 mol, 10.1 equiv.) was added with a microsyringe. The reaction mixture was heated to 100 C and monitored with 1H NMR spectroscopy. After 3 days at 100 C, when Cp*2Y(3-CH2CCPh) was found to be completely converted into a 85:15 mixture of 11b and 12b, respectively, the mixture was evaporated to dryness, redissolved in benzene-d6 and analyzed with multinuclear 1D and 2D NMR spectroscopy. Subsequent addition of methanol-d4 (10 L) was followed by NMR and GC/GC-MS analysis. GC/GC-MS analysis indicated the presence of Cp*D, C18H15D (two isomers 13-d1 and 14-d1 in a 86:14 ratio, respectively) and C18H16 (four isomers). 11b: 1H NMR (C6D6, 25 C, 500 MHz): 1.95 (s, C5Me5, D), 2.08 (s, CH3, A), 3.46 (s, CH2, B) ppm. The aromatic signals could not be assigned due to overlap with other signals. 13C NMR (C6D6, 25 C, 125.7 MHz): 11.60 (dq, 1JCH = 125.4 Hz, 3JYC = 1.4 Hz, CH3, 9), 19.74 (q, 1JCH = 128.1 Hz, CH3, 7) , 27.44 (t, 1JCH = 125.4 Hz, CH2, 3), 89.00 (d, 1JYC = 2.8 Hz, C, 5), 91.84 (d, 1JYC = 1.6 Hz, C, 4), 110.40 (d, 1JYC = 6.1 Hz, i-C, 6), 118.20 (m, 10) 216.17 (d, 1JYC = 43.2 Hz, YC, 1) ppm. Due to overlapping signals, the aromatic carbon signals

92

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes could not be assigned. 1H-1H gNOESY (500-500 MHz, C6D6, 25 C): AC, CD. 1H-13C gHSQC (125.7-500 MHz, C6D6, 25 C): A7, B3. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A1,4,6, 9 B2,4-6,8. 8 5 10 D 4 12b: 1H NMR (C6D6, 25 C, 500 Y 1 2 3B MHz): 1.98 (s, C5Me5, 30H), 4.22 (s, CH2, 2 H) ppm. Unambiguous assignment of the methyl and 6 7 A aromatic signals was not possible due to overlap C with other signals. (E)-1,5-diphenyl-1-deuterio-2-methyl1-penten-4-yne (13-d1): 1H NMR (C6D6, 25 C, 300 MHz): 1.99 (t, 4JHH = 0.6 Hz, CH3), 3.54 (m, CH2) ppm. The aromatic signals could not be assigned due to overlap with other signals. 13C NMR (C6D6, 25 C, 75 MHz): 19.50 (CH3), 25.95 (CH2), 82.71 (CC), 87.03 (CC) ppm. The aromatic carbon signals could not be assigned due to overlap with other signals. The proton and carbon NMR resonances are consistent with literature values of analogous alk-4-en-1-ynes.10 GC-MS, m/z (relative intensity): 233 (M+, 100), 232 (31), 218 (51), 217 (51), 216 (63), 215 (35), 203 (55), 202 (50), 189 (10), 156 (15), 155 (20), 154 (37), 153 (35), 128 (30), 118 (35), 116 (53), 115 (48), 91 (30), 28 (34). GC-MS, m/z (calc., found): 236 (0.1, 0.2), 235 (1.8, 1.8), 234 (19.6, 19.9), 232 (100.0, 100.0). (Z)-3,6-diphenyl-2-deuterio-2-hexen-4-yne (14-d1): 1H NMR (C6D6, 25 C, 300 MHz): 1.60 (m, CH3), 3.54 (q, 5JHH = 0.6 Hz, CH2) ppm. The aromatic signals could not be assigned due to overlap with other signals. 13C NMR (C6D6, 25 C, 75 MHz): 21.26 (CH3), 30.18 (CH2) ppm. The acetylenic and aromatic carbon signals could not be assigned. The proton and carbon NMR resonances are consistent with literature values of analogous alk-4-en-1-ynes.10 GC-MS, m/z (relative intensity): 233 (M+, 62), 232 (34), 218 (100), 217 (76), 216 (92), 215 (39), 203 (65), 202 (40), 153 (25), 116 (25), 115 (32), 103 (31). GC-MS, m/z (calc., found): 236 (0.1, 0.1), 235 (1.8, 1.8), 234 (19.6, 18.1), 232 (100.0, 100.0). Reaction of Cp*2Y(3-CH2CCPh) (3b) with 1-phenyl-1-propyne after prolonged heating. NMR scale. Cp*2Y(3-CH2CCPh) (9.1 mg, 19.2 mol) was dissolved in 0.50 mL of a benzene-d6 solution of hexamethyldisiloxane (1.4 mM). After addition of 1-phenyl-1-propyne (48.0 L, 381 mol) with a microsyringe the mixture was heated to 100 C and monitored with 1H NMR spectroscopy. After 24 days at 100 C, no significant changes were observed with 1H NMR spectroscopy and the reaction mixture was quenched with methanol-d4. GC/GC-MS analysis indicated the presence of Cp*D, 1-phenyl-1-propyne, 13-dn, 14-dn, C19H24 (three isomers of m/z 252) and C18H16 (several isomers of m/z 232). C19H24: GC-MS, m/z (calc., found): 255 (0.1, -), 254 (2.0, 1.9), 253 (20.8, 21.6), 252 (100.0, 100.0); 255 (0.1, -), 254 (2.0, 2.9), 253 (20.8, 21.4), 252 (100.0, 100.0); 255 (0.1, -), 254 (2.0, 2.1), 253 (20.8, 21.0), 252 (100.0, 100.0). Reaction of Cp*2La(3-CH2CCPh) (3a) with 2-propynyltoluene. NMR scale. Cp*2La(3CH2CCPh) (10.1 mg, 19.3 mol) was dissolved in 0.50 mL of a benzene-d6 solution of hexamethyldisiloxane (1.4 mM). After addition of 2-propynyltoluene (50.0 mg, 384 mol) the mixture was heated to 100 C and monitored with 1H NMR spectroscopy. After 37 days at 100 C, no significant changes were observed with 1H NMR spectroscopy and the reaction mixture was quenched with methanol-d4. GC/GC-MS analysis indicated the presence of Cp*D, 2-propynyltoluene-dn, 2-methylphenylallene-d1, C20H20 (at least fifteen isomers of m/z 130), C19H18 (two isomers of m/z 246) and C30H30 (three isomers of m/z 390). The two major C20H20 isomers were assigned to the two major products as observed with NMR spectroscopy. 2-Propynyltoluene-dn: GC-MS, m/z (relative intensity): 130 (M+, 100), 129 (M+ - H, 62), 128 (65), 127 (29), 116 (9), 115 (M+ - CH3, 83), 102 (8), 89 (7), 77 (10), 64 (11), 63 (14), 51 (14), 50 (7), 39 (7). GC-MS, m/z (calc., found): 132 (0.5, 0.7), 131 (11.0, 11.1), 130 (100.0, 100.0). The observed values differ somewhat from those obtained after synthesis (Chapter 2) due to the presence of 2-propynyltoluene-d1. 2-Methylphenylallene-d1: GC-MS, m/z (relative intensity): 131 (M+, 10), 130 (M+ - H, 89), 129 (100), 128 (88), 127 (38), 116 (10), 115 (93), 103 (4), 102 (11), 91 (6), 89 (9), 77 (15), 74 (7), 65 (9), 64 (14), 63 (20), 51 (11), 39 (14). GC-MS, m/z (calc., found): 133 (0.5, 0.9), 131 (11.0, 10.7), 130 (100.0, 100.0). (E)-3-(2-methylbenzylidene)-2-methyl-1-(2-methylphenyl)cyclobutene (17). 1H NMR (C6D6, 25 C, 500 MHz): 1.66 (br. s, CH3, 3 H), 2.16 (s, CH3, 3 H), 2.33 (s, CH3, 3 H), 3.53 (br. s, CH2, 2 H), 6.28 (br. s, =CH, 1 H) ppm. The aromatic signals could not be assigned due to considerable overlap. 13C NMR (C6D6, 25 C, 125.7 MHz): 10.86 (q, 1JCH = 124.9 Hz, CH3), 20.18 (q, 1JCH = 125.8 Hz, CH3), 20.41 (q, 1JCH = 125.8 Hz, CH3), 39.36 (t, 1JCH = 137.6 Hz, CH2), 109.16 (d, 1JCH = 151.5 Hz, =CH) ppm. The aromatic signals and Scheme 3-14. Numbering scheme of 11b.

93

Chapter 3 cyclobutenic carbon signals could not be assigned due to considerable overlap. GC-MS, m/z (relative intensity): 260 (M+, 100), 246 (13), 245 (M+ - CH3, 61), 231 (18), 230 (M+ - 2 CH3, 58), 229 (33), 228 (13), 219 (8), 217 (19), 216 (24), 215 (64), 204 (12), 203 (18), 202 (22), 165 (8), 155 (16), 153 (16), 152 (13), 130 (10), 129 (27), 128 (38), 127 (14), 116 (8), 115 (40), 105 (17), 77 (9). GC-MS, m/z (calc., found): 263 (0.1, 0.2), 262 (2.2, 2.5), 261 (22.0, 23.0), 260 (100.0, 100.0). (E)-3-(2-methylbenzylidene)-2-(2-methylphenyl)-1-methylcyclobutene (18). 1H NMR (C6D6, 25 C, 500 MHz): 1.77 (br. s, CH3, 3 H), 2.19 (s, CH3, 3 H), 2.28 (s, CH3, 3 H), 3.07 (br. s, CH2, 2 H), 6.35 (br. s, =CH, 1 H) ppm. The aromatic signals could not be assigned due to considerable overlap. 13C NMR (C6D6, 25 C, 125.7 MHz): 15.47 (q, 1JCH = 126.7 Hz, CH3), 20.28 (q, 1JCH = 126.5 Hz, CH3), 21.28 (q, 1JCH = 126.2 Hz, CH3), 40.22 (t, 1JCH = 139.3 Hz, CH2), 109.24 (d, 1JCH = 151.3 Hz, =CH) ppm. The aromatic and cyclobutenic carbon signals could not be assigned due to considerable overlap. GC-MS, m/z (relative intensity): 260 (M+, 100), 246 (16), 245 (M+ - CH3, 65), 231 (13), 230 (M+ - 2 CH3, 49), 229 (88), 228 (11), 217 (8), 216 (16), 215 (46), 213 (13), 202 (8), 155 (16), 153 (14), 152 (10), 143 (8), 142 (8), 141 (10), 130 (14), 129 (35), 128 (42), 127 (15), 116 (10), 115 (50), 114 (12), 105 (26), 91 (12), 77 (10). GC-MS, m/z (calc., found): 263 (0.1, 0.2), 262 (2.2, 3.0), 261 (22.0, 25.5), 260 (100.0, 100.0). C19H18: GC-MS, m/z (calc., found): 249 (0.1, 0.3), 248 (2.0, 2.3), 247 (20.8, 20.6); 249 (0.1, 0.2), 248 (2.0, 2.0), 247 (20.8, 21.0), 246 (100.0, 100.0). C30H30: GC-MS, m/z (calc., found): 393 (0.5, -), 392 (5.2, 7.0), 391 (32.8, 35.5), 390 (100.0, 100.0); 393 (0.5, 0.8), 392 (5.2, 5.3), 391 (32.8, 34.2), 390 (100.0, 100.0); 393 (0.5, -), 392 (5.2, 5.8), 391 (32.8, 34.3), 390 (100.0, 100.0). Reaction of Cp*2LaCH(SiMe3)2 (1a) with 1-phenyl-1-propyne, followed by treatment with trimethylsilylchloride and catalytic hydrogenation. NMR scale. Cp*2LaCH(SiMe3)2 (10.2 mg, 17.7 mol) was dissolved in benzene-d6 (0.50 mL) and transferred to a Teflon-valve NMR tube. After addition of 1-phenyl1-propyne (44.4 L, 355 mol) the mixture was heated to 120 C and monitored with 1H NMR spectroscopy. After 7 days at 120 C, the substrate was completely consumed and trimethylsilylchloride (3.0 L, 24 mol) was added. Upon heating 6 h to 50 C, the deep red solution was found to be converted into a light-yellow solution containing a white solid. The 1H NMR resonances attributed to Cp* groups had disappeared, while new 1H NMR resonances had formed (assigned to 9 and 10). The white solid was isolated by filtration, dissolved in THF-d8 and identified with 1H and 13C NMR spectroscopy as Cp*2LaCl(THF-d8) by comparison with reported data.40 The filtrate was subjected to catalytic hydrogenation (room temperature, 12 h, 4 atm. of H2) using Pd/C (~100 mg). The reaction mixture was filtered through a plug of neutral alumina, analyzed with NMR spectroscopy and quenched with methanol-d4. GC/GC-MS analysis indicated the presence of CH2(SiMe3)2, 1,2,3,4,5pentamethylcyclopentane, six C18H20 isomers (in a 12:72:2:1:13:1 ratio) and two C21H28Si isomers (in a 71:29 ratio). C18H20: GC-MS, m/z (calc., found): 238 (1.8, 6.8), 237 (19.8, 20.0), 236 (100.0, 100.0); major isomer: 239 (0.1, 0.1), 238 (1.8, 2.5), 237 (19.8, 20.3), 236 (100.0, 100.0); 238 (1.8, 1.3), 237 (19.8, 22.9), 236 (100.0, 100.0); 238 (1.8, 2.0), 237 (19.8, 21.5), 236 (100.0, 100.0); 238 (1.8, 1.5), 237 (19.8, 20.3), 236 (100.0, 100.0); 238 (1.8, 2.4), 237 (19.8, 20.2), 236 (100.0, 100.0). C21H28Si (major isomer): GC-MS, m/z (relative intensity): 308 (M+, 2), 293 (M+ - CH3, 8), 234 (14), 143 (27), 142 (13), 135 (12), 117 (11), 91 (19), 73 (100). GC-MS, m/z (calc., found): 311 (1.1, 0.8), 310 (7.0, 7.7), 309 (28.1, 31.5), 308 (100.0, 100.0). C21H28Si (minor isomer): GC-MS, m/z (relative intensity): 308 (M+, 1), 293 (M+ - CH3, 6), 234 (11), 143 (23), 142 (13), 135 (10), 117 (11), 91 (19), 73 (100). GC-MS, m/z (calc., found): 311 (1.1, 1.4), 310 (7.0, 6.9), 309 (28.1, 30.0), 308 (100.0, 100.0). Trimethyl[(E)-(2-methyl-3-phenyl-2-cyclobuten-1-ylidene)(phenyl)methyl]silane (9): 1H NMR (C6D6, 25 C, 500 MHz): -0.02 (s, SiCH3) ppm. The methyl, methylene and aromatic signals were obscured by overlapping signals. 13C NMR (C6D6, 25 C, 125.7 MHz): -2.16 (SiCH3), 12.22 (CH3), 43.66 (CH2), 111.6 (CSiMe3) ppm. The cyclobutenic and aromatic carbon signals could not be assigned unambiguously. The carbon NMR resonances are consistent with reported values of alkylidenecyclobutenes (vide supra).11 Trimethyl[(E)-(3-methyl-1-phenyl-2-cyclobuten-1-ylidene)(phenyl)methyl]silane (10): 1H NMR (C6D6, 25 C, 500 MHz): -0.02 (s, SiCH3), 3.24 (CH2) ppm. The methyl and aromatic signals were obscured by overlapping signals. 13C NMR (C6D6, 25 C, 125.7 MHz): -2.16 (SiCH3), 11.36 (CH3), 41.01 (CH2), 111.6 (CSiMe3) ppm. The cyclobutenic and aromatic carbon signals could not be assigned unambiguously. The proton and carbon NMR resonances are consistent with reported values of alkylidenecyclobutenes (vide supra).11 Reaction of Me2SiCp2YCH(SiMe3)2 (19b) with excess 1-phenyl-1-propyne. NMR scale. Me2SiCp2YCH(SiMe3)2 (12.2 mg, 22.3 mol) was dissolved in 0.50 mL of a benzene-d6 solution. After

94

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes addition of 1-phenyl-1-propyne (49.7 L, 397 mol, 17.8 equiv.) with a microsyringe, the mixture was heated to 120 C and monitored with 1H NMR spectroscopy. After 16 days at 120 C, the substrate was consumed completely and the mixture was quenched with methanol. GC-MS analysis indicated the presence of dimers (m/z 232) and tetramers (m/z 464). Normalized FID-GC values revealed a 89% selectivity for dimerization and a 11% selectivity for tetramerization. Even though the product mixture was too complex for conclusive NMR analysis, the major products could be identified as (E)-3-benyzlidene-1-methyl-2-phenylcyclobutene (5) and 1,3-diphenyl4-methyl-2-methylene-cyclobutene (7) in yields of 13 and 10%, by integration against the SiMe3 1H NMR resonances of CH2(SiMe3)2 and intact Me2SiCp2YCH(SiMe3)2. Reaction of Me2SiCp2CeCH(SiMe3)2 (19c) with excess 1-phenyl-1-propyne. NMR scale. Me2SiCp2CeCH(SiMe3)2 (11.8 mg, 19.7 mol) was dissolved in 0.50 mL of a benzene-d6 solution. After addition of 1-phenyl-1-propyne (48.4 L, 387 mol, 20 equiv.) with a microsyringe, the mixture was heated to 120 C and monitored with 1H NMR spectroscopy. Within 1 day at 120 C, the substrate was consumed completely and the mixture was quenched with methanol. On the basis of NMR and GC-MS analysis, the products were identified as 5 (25%), 6 (27%), 7 (14%) and a tetramer of unknown structure (m/z 464, 30%), while the yields were determined by integration against CH2(SiMe3)2. Normalized FID-GC revealed a 64% selectivity for dimerization and a 36% selectivity for tetramerization. These values agree only moderately with those obtained from quantitative 1H NMR analysis (i.e. 52% selectivity for dimerization and 48% selectivity for tetramerization). Tetramer (C36H32): major isomer, GC-MS, m/z (relative intensity): 464 (M+, 44), 450 (39), 449 (100), 343 (8), 265 (11), 253 (8), 252 (10), 241 (11), 239 (9), 215 (11), 202 (10), 191 (10), 178 (14), 165 (16), 139 (8), 128 (8), 115 (19), 105 (22), 91 (51), 77 (8), 28 (11). GC-MS, m/z (calc., found): 467 (0.9, 1.0), 466 (7.5, 7.2), 465 (39.3, 37.2), 464 (100.0, 100.0). Reaction of Me2SiCp2CeCH(SiMe3)2 (19c) with excess 2-propynyltoluene. NMR scale. Me2SiCp2CeCH(SiMe3)2 (11.1 mg, 18.6 mol) was dissolved in 0.50 mL of a benzene-d6 solution. After addition of 2-propynyltoluene (54.5 L, 395 mol, 20 equiv.) with a microsyringe, the mixture was heated to 120 C and monitored with 1H NMR spectroscopy. Complete substrate conversion was observed after 55 days and the mixture was quenched with methanol. On the basis of NMR and GC-MS analysis, the products were identified as 17 (33%), 18 (1%), 19 (2%) and 20 (2%), while the yields were determined by integration against CH2(SiMe3)2. Normalized FID-GC revealed a 85% selectivity for dimerization (at least 7 isomers, of which the 4 major isomers were present in a 8:8:41:5 ratio) and a 15% selectivity for trimerization (2 major isomers in a 1:1 ratio). 1,3-Di(2-methylphenyl)-2-methyl-4-methylenecyclobut-1-ene (19) and 1,2-di(2-methylphenyl)-3methyl-4-methylenecyclobut-1-ene (20). The complexity of the 1H NMR spectrum hampered unambiguous assignments, but characteristic 1H NMR resonances at 4.68, 4.61, 4.48 and 4.22 ppm in a 1:1:1:1:1 ratio are reminescent of those of the exo-cyclic methylene group in 6 (i.e. 4.57 and 5.07 ppm). Under the assumption that these 1H NMR resonances correspond to 19 and 20, the yields of these dimers, based on 1H NMR, agree well with those of two new major isomeric cyclodimers (GC-MS) based on FID-GC. It was not possible to distinguish between these two isomers. Isomer 1. GC-MS, m/z (relative intensity): 260 (M+, 100), 246 (14), 245 (M+ - CH3, 63), 230 (M+ - 2 CH3, 65), 229 (37), 228 (15), 217 (18), 216 (26), 215 (75), 204 (14), 203 (21), 202 (25), 165 (11), 155 (22), 153 (23), 152 (18), 145 (26), 142 (14), 141 (13), 130 (14), 129 (41), 128 (58), 127 (22), 117 (12), 116 (11), 115 (58), 114 (23), 108 (10), 105 (29), 101 (14), 91 (17), 77 (15). GC-MS, m/z (calc., found): 263 (0.1, 0.2), 262 (2.2, 2.4), 261 (22.0, 22.7), 260 (100.0, 100.0). Isomer 2. GC-MS, m/z (relative intensity): 260 (M+, 29), 246 (21), 245 (M+ - CH3, 100), 231 (20), 230 (M+ - 2 CH3, 87), 229 (26), 228 (15), 217 (18), 216 (20), 215 (73), 202 (10), 153 (10), 152 (12), 129 (20), 128 (36), 127 (14), 117 (11), 115 (48), 114 (28), 101 (12), 91 (13), 77 (11). GC-MS, m/z (calc., found): 263 (0.1, 0.3), 262 (2.2, 2.8), 261 (22.0, 22.5), 260 (100.0, 100.0). Trimers (C30H30): GC-MS, m/z (calc., found): 393 (0.5, 0.7), 392 (5.2, 5.9), 391 (32.8, 36.7), 390 (100.0, 100.0); 393 (0.5, 0.6), 392 (5.2, 5.9), 391 (32.8, 36.0), 390 (100.0, 100.0). Kinetic studies of the cyclodimerization reaction. A catalyst stock solution was prepared by weighing the amount of precatalyst and dissolving the solid in a specified volume of benzene-d6, as determined by volumetric glassware. After preparation, the catalyst solution (typically 40 mM) was transferred into a preweighted vial with screw-cap and weighted. A specified amount of internal standard (hexamethyldisiloxane) was added with a microsyringe. After use, the catalyst stock solution was stored at -40 C in the glovebox. A 1H NMR experiment of the sample containing the catalyst solution (prior to substrate addition) ensured the presence of the prerequisite amount of catalyst after long-term storage or handling. In a typical experiment, an NMR tube was charged with 500.0 L of a catalyst stock solution (40 mM) using a 500.0-L microsyringe. The volume of substrate needed for the kinetic experiment was calculated

95

Chapter 3 from the density which was determined experimentally. The tube was heated in an electric oven at a specified temperature and taken out for NMR analysis after appropriate intervals. NMR data were acquired using appropriate long pulse delays (at least 60 s) in order to avoid signal saturation under anaerobic conditions. In most cases, the reaction kinetics were monitored from the normalized intensity changes in the substrate resonance over 3 or more half-lives. The substrate concentration was determined from the normalized intensity of the methyl substrate protons relative to the methyl protons of hexamethyldisiloxane. Typical polymerization reaction. Cp*2La(3-CH2CCPh) (10.2 mg, 19.4 mol) was dissolved in 0.50 mL of a benzene-d6 solution of hexamethyldisiloxane (1.4 mM). After addition of 2,5-di-n-hexyl-1,4dipropynylbenzene (123.6 mg, 383.2 mol, 19.8 equiv.), the mixture was heated to 120 C and the progress of reaction was followed in time by means of 1H NMR spectroscopy. After 3 months at 120 C, the reaction mixture was completely solidified. The contents of the tube was transferred into a Teflon-capped reaction vial and suspended in toluene (3.0 mL). Addition of trimethylsilylchloride (24.6 L, 194 mol, 10 equiv.) was followed by heating to 120 C and stirring for 7 days in a closed Teflon-capped flask. The formed red suspension was filtered over a small glass filter and washed with (dry) chloroform. The chloroform-soluble fraction was evaporated to dryness and washed with copious amount of pentane over a filter, affording a bright red rubbery solid after drying in vacuo. Yield: 33.4 mg (27%).

3.6.
1

References and notes

(a) Henkelmann, J. In Applied Homogeneous Catalysis with Organometallic Compounds; Cornils, B., Herrmann, Wolfgang A., Eds.; VCH: Weinheim, 1996; Chapter 2.3.2. (b) Parshall, G. W. In Homogeneous Catalysis; Wiley-Interscience: New York, 1980; Chapter 8. (c) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry, University Science Books: Mill Valley, 1987. (d) Keim, W., Behr, A., Rper, M. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, 1982; Vol. 8; Chapter 52. (e) Grotjahn, D. B. In Comprehensive Organometallic Chemistry II; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, 1994; Vol. 7.3; Chapter 52. (f) Winter, M. J. In The Chemistry of the Metal-Carbon Bond; Hartley, F. R., Patai, Saul, Eds.; Jonh Wiley & Sons: New York, 1985; Vol. 3; Chapter 5. (a) Mixtures containing alkylidenecyclobutenes are formed by oligomerization of 2-butyne and 3heptyne catalyzed by a cationic nickel hydride complex, [HNi(dppe)2]OCOCF3, see: Inoue, Y.; Itoh, H.; Hashimoto, H. Chem. Lett. 1978, 911. (b) Propagylic alcohols are cyclodimerized catalytically by Cp*RuCl(COD) affording alkylidenecyclobutenes, see: Le Paih, J.; Drien, S.; Bruneau, C.; Demerseman, B.; Toupet, L.; Dixneuf, P. H. Angew. Chem. Int. Ed. 2001, 40, 2912. (a) Heeres, H. J.; Heeres, A.; Teuben, J. H. Organometallics 1990, 9, 1508. (b) Heeres, H. J. Ph. D. Thesis, University of Groningen, 1990; Chapter 5. (a) Warrener, R. R.; Abbenante, G.; Kennard, C. H. L. J.Am.Chem.Soc., 1994, 116, 3645 and references therein. For reviews, see: (b) Namyslo, J. C.; Kaufmann, D. E. Chem. Rev. 2003, 103, 1485. (c) Houben-Weyl Methods of Organic Chemistry; de Meijere, A., Ed.; Thieme: Stuttgart, 1997; Vol. E17e. For examples, see: (a) Applequist, D. E.; Roberts, J. D. J. Am. Chem. Soc. 1956, 78, 4012. (b) Kirk, B. E.; Taylor, D. R. J. Chem. Soc., Perkin Trans. 1 1974, 1844. (c) Trabert, L.; Hopf, H. Liebigs Ann. Chem. 1980, 40, 1786. (d) Klop, W.; Klusener, P. A. A.; Brandsma, L. Recl. Trav. Chim. Pays-Bas 1984, 103, 85. (e) Pasto, D. J.; Kong, W. J. Org. Chem. 1988, 53, 4807. (f) Kimura, M.; Horino, Y.; Wakamiya, Y.; Okajima, T.; Tamaru, Y. J. Am. Chem. Soc. 1997, 119, 10869. (g) Shen, Q.; Hammond, G. B. J. Am. Chem. Soc. 2002, 124, 6534. (h) Wei, L.-L.; Xiong, H.; Hsung, R. P. Acc. Chem. Res. 2003, 36, 773. Piers, E.; Boehringer, E. M.; Yee, J. G. K. J. Org. Chem. 1998, 63, 8642. (a) Fhlisch, B. Tetrahedron Lett. 1969, 35, 3008. (b) Ezcurra, J. E.; Pham, C.; Moore, H. W. J. Org. Chem. 1993, 57, 4787. (c) Charrouf-Chaf Chaouni, Z.; Maurin, R.; Guillaume, D. Tetrahedron Lett. 1996, 37, 5099. (d) Frimer, A. A.; Pizem, H. Tetrahedron 1999, 55, 12175. (e) Pizem, H.; Sharon, O.; Frimer, A. A. Tetrahedron 2002, 58, 3199. (f) Sharon, O.; Frimer, A. A. Tetrahedron 2003, 59, 8153. (g) Bushmelev, V. A.; Genaev, A. M.; Osadchii, S. A.; Shakirov, M. M.; Shuin, V. G. Russ. J. Org. Chem. 2003, 39, 1301.

3 4

6 7

96

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

8 9 10

11

12

13

14 15

16

17

18 19

20 21

22

For a review, see: Hall, H. K. Jr.; Ykman, P. J. Polymer. Sci.: Macromol. Rev. 1976, 11, 1 In accord, slow formation of CH2(SiMe3)2 was observed with 1H NMR spectroscopy. Linear alk-4-ene-1-yne derivatives can be ruled out by NMR analysis, as 13C NMR resonances attributable to acetylenic carbons ( 95-80 ppm) were not observed and the corresponding methylene ( 28-23 ppm) and methyl groups ( 20-24 ppm) are known to resonate at higher and lower field, respectively. For NMR data of 4-penten-1-yne derivatives, see: D.-M. Cui, J. Org. Chem. 1995, 60, 5752. (a) Lpez, S.; Rodrguez, J.; Garca Rey, J.; de Lera, A. R. J. Am. Chem. Soc., 1996, 118, 1881. (b) Hther, H.; Brune, H. A. Org. Magn. Res., 1971, 3, 737. (c) Maurin, R.; Leandri, G.; Bertrand, M. Bull. Soc. Chim. Fr. 1971, 2, 541. (d) Hopf, H.; Kretschmer, O.; Ernst, L.; Witte, L. Chem. Ber. 1991, 124, 875. (a) Blomquist, A. T.; Meinwald, Y. C. J. Am. Chem. Soc. 1959, 81, 667. (b) Pasto, D. J.; Kong, W. J. Org. Chem. 1988, 53, 4807. (c) Pasto, D. J.; Kong, W. J. Org. Chem., 1989, 54, 4028. (d) Applequist, D. E.; Roberts, J. D. J. Am. Chem. Soc. 1956, 78, 4012. (e) Griesbaum, K.; El-Abed, M. Chem. Ber. 1973, 106, 2001. (f) Criegee, R.; Dekker, J.; Engel, W.; Ludwig, P.; Noll, K. Chem. Ber.1963, 96, 2362. (g) Howton, D. R.; Buchman, E. R. J. Am. Chem. Soc. 1956, 78, 4011. (h) Inoue, Y.; Itoh, H.; Hashimoto, H. Chem. Lett. 1978, 911. (i) Klop, W.; Klusener, P. A. A.; Brandsma, L. Recl. Trav. Chim. Pays-Bas 1984, 103, 85. (j) Taylor, D. R.; Warburton, M. R.; Wright, D. B. J. Chem. Soc., Perkin Trans.1, 1972, 1365. (k) Delas, C.; Urabe, H.; Sato, F. J. Am. Chem. Soc. 2001, 123, 7937. (a) Ezcurra, J. E.; Pham, C.; Moore, H. W. J. Org. Chem., 1993, 57, 4787. (b) Charrouf-Chaf Chaouni, Z.; Maurin, R.; Guillaume, D. Tetrahedron Lett. 1996, 37, 5099, (c) Maurin, R.; Piscot, C.; Charrouf-ChafChaouni, Z. Tetrahedron Lett. 1980, 21, 2425. For example, phenylallene (75 MHz, benzene-d6, 27 C): 78.61 (t, 1JCH = 168.2 Hz, CH2), see: Schaefer, T.; Kroeker, S.; McKinnon, D. M. Can. J. Chem. 1995, 73, 1478. For reviews, see: (a) Marvell, E. N. Thermal Electrocyclic Reactions; Academic Press: New York, 1980; pp 145. (b) Schuster, H. F.; Coppola, G. M. Allenes in Organic Synthesis; Wiley: New York, 1984; pp 89. (c) Okamura, W. H. Acc. Chem. Res. 1984, 40, 2805. (d) Durst, T.; Breau, L. In Comprehensive Organic Synthesis; Trost, B. M.; Fleming, I.; Paquette, L. A., Eds.; Pergamon Press: Oxford, 1991; Vol. 5, pp. 675. For examples, see: (e) Pasto, D. J.; Kong, W. J. Org. Chem. 1989, 54, 4028. (f) Gil-Av, E.; Herling, J. Tetrahedron Lett. 1967, 1. (g) Schneider, R.; Siegel, H.; Hopf, H. Liebigs Ann. Chem. 1981, 1812. (a) Rey, J. G.; Rodrguez, J.; de Lera, A. R. Tetrahedron Lett. 1993, 34, 6293. (b) Murakami, M.; Amii, H.; Itami, K.; Ito, Y. Angew. Chem., Int. Ed. Engl. 1995, 34, 1476. (c) Lpez, S.; Rodrguez, J.; Garca Rey, J.; de Lera, A. R. J. Am. Chem. Soc. 1996, 118, 1881. (d) Delas, C.; Urabe, H.; Sato, F. J. Am. Chem. Soc. 2001, 123, 7937. (e) Murakami, M.; Ashida, S.; Matsuda, T. J. Am. Chem. Soc. 2004, 126, 10838. Because the rate constant k varies exponentially with temperature T and activation energy Ea according to the Arrhenius equation (i.e. k = Ae-Ea/RT), the error in k associated with a particular uncertainity in T is proportional with Ea/RT (i.e. k/k = Ea/RT T/T). The error due to poor temperature control will therefore increase with decreasing reaction temperature. Moreover, slow reactions have in general high activation energies Ea and require more precise temperature control for decent precision in rate constants than fast reactions. Ionic radii for eight-coordinate complexes: La3+ (1.160 ), Ce3+ (1.143 ), Y3+ (1.019 ) and Sc3+ (0.870 ), see: Shannon, R. D Acta Crystallogr., Sect. A 1976, A32, 751. The regiosomeric ratio of products arising from 1,2- and 2,1-insertion of phenylacetylene into the LnC bond of Cp*2LnCCPh at 25 C was found to be greatly influenced by the metal size (Chapter 4). For example, the use of a 50-fold excess of phenylacetylene gives a 95:5 ratio of products arising from 1,2- and 2,1- insertion, respectively, for Ln = Y, whereas a ratio of 0:100 was found for Ln = La. (a) Ti: Pattiasina, J. W. Ph. D. Thesis, University of Groningen, 1988; Chapter 4. (b) Zr: Horton, A. D. Organometallics 1992, 11, 3271. (a) Booij, M.; Meetsma, A.; Teuben, J. H. Organometallics 1991, 10, 3246. (b) Booij, M. Ph. D. Thesis, University of Groningen, 1989; Chapter 3. (c) Ryu, J.-S.; Marks, T. J.; McDonald, F. E. J. Org. Chem. 2004, 69, 1038 (a) Hajela, S.; Bercaw, J. E. Organometallics 1994, 13, 1147. (b) Jeske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8111-8118. (c) Gagn, M. R.; Marks, T.

97

Chapter 3

23

24

25

26

27

28 29 30 31 32

J. J. Am. Chem. Soc. 1989, 111, 4108-4109. (d) Giardello, M. A.; Conticello, V. P.; Brard, L.; Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10241-10254. (e) Shin, J. H.; Parkin, G. Chem. Commun. 1999, 887-888. (f) Fendrick, C. M.; Schertz, L. D.; Day, V. W.; Marks, T. J. Organometallics 1988, 7, 1828-1838. (g) Dash, A. K.; Gourevich, I.; Wang, J. Q.; Wang, J.; Kapon, M.; Eisen, M. S. Organometallics 2001, 20, 5084-5104. (h) Fontaine, F.G.; Tilley, D. T. Organometallics 2005, 24, 4340. Indirect evidence for this general accepted view comes from the electronic effects of -H elimination from lanthanide alkyls (the microscopic reverse of alkene insertion into a Ln-H bond), alkene/alkyne insertion into Ln-N bonds and alkene insertion into transition metal hydrogen bonds and actinide hydrogen bonds. For examples, see: (a) Burger, B. J.; Santarsiero, B. D.; Trimmer, M. S.; Bercaw, J. E. J. Am. Chem. Soc. 1988, 110, 3134. (b) Doherty, N.; Bercaw, J. E. J. Am. Chem. Soc. 1985, 107, 2670. (c) Halpern, J.; Okamoto, T.; Zakhariev, A. J. Mol. Catal. 1976, 2, 65-68 (d). Lin, Z.; Marks, T. J. J. Am. Chem. Soc. 1990, 112, 5515. 2,1-Insertion of aromatic substituted alk-1-enes and alk-1-ynes has been observed in a variety of organolanthanide-catalyzed reactions, such as hydrosilylation, hydroamination and oligomerization, but also in analogous reactions catalyzed by organoactinides and other relatively electrophilic metals. For examples, see: (a) Fu, P.-F.; Brard, L.; Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 7157. (b) Molander, G. A.; Romero, J. A. C.; Corrette, C. P. J. Organomet. Chem. 2002, 647, 225. (c) Molander, G. A.; Schmitt, M. H. J. Org. Chem. 2000, 65, 3767. (d) Molander, G. A.; Knight, E. E. J. Org. Chem. 1998, 63, 7009. (d) Gagn, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (e) Li, Y.; Marks, T. J. Organometallics 1996, 15, 3770. (f) Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1996, 118, 9295. (g) Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 1757. (h) Ryu, J.S.; Li, G. Y.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 12584. A cross-conjugated compound may be defined as a compound possessing three unsaturated groups, two of which although conjugated to a third unsaturated center are not conjugated to each other, see: Phelan, N. F.; Orchin, M. J. Chem. Educ. 1968, 45, 633. For reviews on conjugated polymers, see: (a) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew. Chem. Int. Ed. 1998, 37, 403. (b) Martin, R. E.; Diederich, F. Angew. Chem., Int. Ed. 1999, 38, 1350. (c) Bunz, U. H. F. Chem. Rev. 2000, 100, 1605. (d) Skotheim, T. A.; Elsenbaumer, R. L.; Reynolds, J. R. Handbook of Conducting Polymers, Dekker: New York, 1997. (e) Salaneck, W. R.; Lundstrm, I.; Rnby, B. Conjugated Polymers and Related Materials; Oxford University Press: Oxford, 1993. For reviews on cross-conjugated compounds, see: (e) Hopf, H. Angew. Chem., Int. Ed. Engl. 1984, 23, 948. (f) Hopf, H. Classics in Hydrocarbon Chemistry; Wiley-VCH: Weinheim, 2000; Chapter 11. (g) Tykwinski, R. R.; Zhao, Y. Synlett 2002, 1937. For examples of cross-conjugated oligomers and polymers, see: (h) Zhao, U.; Tykwinski, R. R. J. Am. Chem. Soc. 1999, 121, 458. (i) Brnsted Nielsen, M.; Schreiber, M.; Baek, Y. G.; Seiler, P.; Lecomte, S.; Bouden, C.; Tykwinski, R. R.; Gisselbrecht, J.-P.; Gramlich, V.; Skinner, P. J.; Bosshard, C.; Gnter, P.; Gross, M.; Diederich, F. Chem. Eur. J. 2001, 7, 3263. (j) Zhao, Y.; Cambell, K. ; Tykwinski, R. R. J. Org. Chem. 2002, 67, 336 and references therein. Conjugated organic molecules are a primary focus of a new generation of optical and electronic materials.26a-g The most intensely studied conjugated molecules with all-carbon frameworks have extended, linearly conjugated systems. The electronic properties that result from alternative modes of -electron communication, in particular cross conjugation, have been less frequently studied and the utility of materials with a fully cross-conjugated backbone is currently an active field of research. (a) Odian, G. Principles of Polymerization, Wiley: New York, 2004; 4rd ed. (b) Synthetic Methods in Step-Growth Polymers, Rogers, M. E.; Long, T. E. (Eds.), Wiley-Interscience: New Jersey, 2003. Typical examples of this approach include poly(thiophene)s, poly(arylenevinylene)s and poly(aryleneethynylene)s. Examples in which the degree of polymerization and the solubility of the polymer increased with side-chain elongation include poly(p-phenyleneethynylene)s and poly(thiophene)s.26 For reviews on poly(p-ethynylenephenylene)s, see: (a) Giesa, R. J. M. S.-Rev. Macromol. Chem. Phys. 1996, 36, 631. (b) Ref. 26c. (a) Lin-Vien, D.; Colthup, N. B.; Fately, W. G.; Grasselli, J. G. The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules; Academic Press: London, 1991. (b) Silverstein, R.

98

The cyclodimerization of 1-methylalk-2-ynes catalyzed by rare-earth metallocenes

33

34

35 36 37 38 39 40

M.; Bassler, G. C. ; Morrill, T. C. Spectrometric Identification of Organic Compounds, Wiley: New York, 1981. Infrared spectral data of cyclobutenes containing substituted exo-methylene groups (=CHR) are lacking in literature, but cyclobutenes containing unsubstituted exo-methylene groups are reported to exhibit vibrations at 1670-1680 and 860-880 cm-1, due to the exocyclic =CH2 groups.5a,12a Characteristic vibrations reported for 1-(trimethylsilyl)-3-phenyl-1-propyne (i.e. 2180 (C C), 1250, 840 (C-Si) cm-1) and 1-(trimethylsilyl)-1-phenylpropa-1,2-diene (i.e 1912 (C=C=C) cm-1) were not observed. For examples of infrared spectral data for these compounds, see: (a) Najafi, M. R.; Wang, M.-L.; Zweifel, G. J. Org. Chem. 1991, 56, 2468. (b) Ma, S.; Zhang, A. J. Org. Chem. 2002, 67, 2287. The experimental error was obtained by repeated integration. Coughlin, E. B.; Henling, L. M.; Bercaw, J. E. Inorg. Chim. Acta 1996, 242, 205. Heeres, H. J.; Renkema, J.; Meetsma, A.; Teuben, J. H. Organometallics 1988, 7, 2495 (a) Kloppenburg, L.; Song, D.; Bunz, U. H. F. J. Am. Chem. Soc. 1998, 120, 7973. (b) Kloppenburg, L.; Jones, D.; Bunz, U. H. F. Macromolecules 1999, 32, 4194. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals; 2nd ed.; Pergamon Press: Oxford, 1980. Booij, M. Ph. D. Thesis, University of Groningen, 1989; Chapter 7.

99

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

4.

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

4.1.

Introduction

The high atom efficiency renders the direct coupling of two 1-alkynes to produce but-1-en-3-ynes (I) or butatrienes (II) a highly attractive process (Scheme 4-1).1 The dimerization of 1-alkynes is catalyzed by various well-defined organometallic complexes of early2, first-row3 and late transition metals4 as well as by complexes of lanthanides5 and actinides.6 However, applications based on the linear dimerization of substituted 1-alkynes are still in their infancy, due to the lack of control of chemo-, stereo- and regioselectivity.7 Enynes are organic molecules of both practical and fundamental interest, as they represent valuable intermediates in organic synthesis8 and are applied in the preparation of both natural products9 and optoelectronic materials.10 Scheme 4-1. The metal-catalyzed linear dimerization of 1-alkynes.
R R R R R Ia . R . IIa R + R IIb + Ib . . R R + Ic R

The factors that affect the selectivity of catalytic dimerization have been found to depend highly on both the steric and electronic properties of the alkyne substituent and the nature of the catalyst. Exclusive formation of a single isomer has been achieved only with a few catalysts.11,12 The catalytic dimerization of aromatic 1-alkynes has attracted considerable interest recently and a survey of the reported catalytic behavior towards phenylacetylene places the permethyllanthanocene alkyl derivative Cp*2LaCH(SiMe3)2 among the most active and selective (pre)catalysts for the formation of the trans-head-to-head dimer (Ia).13 From the desire to apply the rare-earth metallocene-catalyzed, linear dimerization of aromatic 1alkynes to substituted (hetero)aromatic and bifunctional substrates in order to prepare novel (cross-)conjugated oligomers (Chapter 5) and polymers (Chapter 6), it was considered of interest to investigate the scope and selectivity of these reactions. In this chapter, the rare-earth metallocene-catalyzed oligomerization of phenylacetylene is described, including the influence of metal ion radius, ancillary ligation and substrate concentration on the rate and selectivity. The reaction kinetics, the identification of reaction intermediates and the reactivity of these reaction intermediates are also discussed. A plausible mechanistic scenario is proposed to account for the observed behavior.

4.2.

Catalytic oligomerization of phenylacetylene

4.2.1.

Introduction

Many well-defined rare-earth metal complexes are known to catalyze the dimerization of phenylacetylene (Table 4-1 and Scheme 4-2), but efforts to extend this reaction to other (hetero)aromatic 1alkynes have not yet been reported.14 Besides the scope, the regioselectivity is also of concern in the present

101

Chapter 4

Scheme 4-2. Well-defined rare-earth metal complexes that are reported to be active in the catalytic oligomerization of phenylacetylene.
Ph O Ln CH(SiMe3)2 Lu SiMe3 SiMe3 N

Me3Si N Y Me3Si N H N Ph SiMe3 7

Si N R 2

Ln

n SiMe3

5A : Ln = Y 5B : Ln = Pr 5C : Ln = Ce 5D : Ln = La N N N 9 La

8Aa : Ln = Y; R = Ph; n = 2 8Ea : Ln = Yb; R = Ph; n = 2 8Fa : Ln = Lu; R = Ph; n = 2 8Fb : Ln = Lu; R = C6H3Me2-2,6; n =1 8Fc : Ln = Lu; R = C6H2Me3-2,4,6: n = 1 8Fd : Ln = Lu; R = tBu; n = 1

/ [PhNMe2H][B(C6F5)4] SiMe3 SiMe3

Y[N(SiMe3)2]3 / 4-ClC6H4NH2 10

context, as the selective formation of the head-to-head dimer will result in the formation of conjugated oligomers and/or polymers, whereas catalysts favoring the formation of the head-to-tail dimer are expected to produce cross-conjugated oligomers and/or polymers. It should also be noted that other well-defined rare-earth metal complexes, such as Cp*2ScR5a (R = H, Me), Cp*Ce[CH(SiMe3)2]215 and Cp*(OAr)YCH(SiMe3)25c (Ar = C6H3tBu2-2,6), have been reported to be active catalysts for the dimerization of alkyl- and silyl-substituted 1alkynes, but their reactivity towards aryl-substituted 1-alkynes is not documented. It can be seen that the alkyl derivative Cp*2LaCH(SiMe3)2 (5D) represents the most active and selective (pre)catalyst for the trans-head-tohead dimeriation of phenylacetylene (12a) among the reported well-defined rare-earth metal complexes that are active for the catalytic dimerization of phenylacetylene (Table 4-1). Several studies have been dedicated to aspects relevant to the mechanism of the lanthanidocenecatalyzed 1-alkyne oligomerization (Section 4.3).16 The catalytically active species is presumed to be a monomeric alkynyl derivative, which undergoes alkyne insertion and subsequent protonolysis by the 1-alkyne to yield but-1-en-3-ynes. The relative formation of the head-to-tail (11) and trans-head-to-head dimer (12) is believed to be determined by the regioselectivity of alkyne insertion into the carbon-metal bond of the alkynyl derivative, while the factors determining the formation and nature of higher oligomers is presently not understood. The monomeric alkynyl derivatives are unstable and form dimers with bridging alkynyl ligands. In some cases, these dimeric alkynyl derivatives undergo intramolecular C-C coupling to afford 1,4-disubstituted butatriene-1,4-diyl derivatives (Cp2Ln)2(-2:2-RC=C=C=CR).16

4.2.2.

Influence of the precatalysts

Earlier it was observed that phenylacetylene (2a) was converted catalytically by rare-earth metallocene alkyl derivatives Cp*2LnCH(SiMe3)2 into mixtures of 2,4-diphenylbut-1-en-yne (11a), (E)-1,4diphenylbut-1-en-yne (12a) and trimers (Table 4-1). The product mixtures reported for the Cp*2Ln-type (Ln = Lu, Y, Pr, Ce, La) precatalysts show that the relative formation of the desired trans-head-to-head dimer (12a) increases with the ionic radius of the metal (Figure 4-1).17 Although the details of the experimental conditions are not known for all reactions considered (viz. 5B and 6), the present study reveals that the substrate-to-catalyst molar ratio affects the relative formation of dimers versus trimers, but not the regioselectivity of dimerization (Section 4.2.4). Based on the reported data, no clear-cut correlation is observed for the ionic metal radius and the degree of oligomerization and more experimental data are needed to investigate this issue. Even so, it is believed that the reported data for the Cp*2Ln-type precatalysts provide reasonable evidence that the lanthanum-based precatalyst Cp*2LaCH(SiMe3)2 (5D) represents the most selective and active catalyst for the formation of the trans-head-to-head dimer (12a) within the available size range of rare-earth metals.

102

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

Table 4-1. The product mixtures obtained for the oligomerization of phenylacetylene catalyzed by rareearth metal complexes.
Ph Ph Ph 11a + Ph 12a 13a Ph + Ph Ph + oligomers

Entry 1 2 3 4 5 6 7 8 9 10 11 12 13 14

(Pre)catalyst 5A 5B 5C 5D 6 7 8Aa 8Ea 8Fa 8Fb 8Fc 8Fd 9 10

11a 89

96 100

12a 11 73 82 86 4

13a

Oligomers 27 18 14

89 92 100 100 100 72 99 91

Ref. 5f,g 5i 5f,g 5f,g 5i 5c 5h 5h 5h 5h 5h 5h 5j 5e

Apart from the report that a 100-fold molar excess of phenylacetylene was converted within 10 min at room temperature by 5C-D, the effect of metal size on the rate of substrate conversion has not been addressed by previous investigations. Thus, experiments with Cp*2YCH(SiMe3)2 (5A) and Cp*2LaCH(SiMe3)2 (5D) were conducted in benzene-d6 with a 50-fold molar excess of phenylacetylene to shed more light on this issue. The La3+ and Y3+ metal ions correspond to the largest and middle size, respectively, of the available size range of the rare-earth metals and are both diamagnetic, thereby allowing the convenient use of NMR spectroscopy.17 In accord with previous observation, the complete conversion of phenylacetylene took place within 10 min at room temperature for the reaction catalyzed by 5D. Three Cp* 1H NMR resonances were observed after substrate conversion, assigned to [(Cp*2La)2(-PhC4Ph)] ( 2.04 ppm), Cp*2LaC(Ph)=C(H)CCPh ( 1.97 ppm) and Cp*2LaC(H)=C(Ph)CCPh ( 1.92 ppm), present in a 1.0:1.4:1.3 ratio, respectively. Their identification is discussed in Sections 4.3.2-4.3.3. The analogous reaction catalyzed by 5A was significantly slower, as complete substrate conversion was achieved only after 2.5 h at room temperature. In this case, only one Cp* 1H NMR resonance was observed during and after substrate conversion. The relatively slow substrate conversion

100 Product mixture (%) 80 60 40 20 0 0.95 1 1.05 1.1 1.15 1.2 11a 12a trimers

Ionic radius ()

Figure 4-1. The reported relative amount of oligomers, formed from the reaction of the rare-earth metallocene precatalysts (5A-D, 6) with phenylacetylene (Table 4-1) as a function of the ionic radius of the metal.17

103

Chapter 4 Table 4-2. The oligomerization of phenylacetylene catalyzed by different precatalysts.a

Ph catalyst Ph C6D6 Ph 11a + Ph 12a + Ph 15a Ph Ph . + Ph 16a Ph Ph Ph

Entry Precatalyst 11a 12a 15a 16a 1 Cp*2YCH(SiMe3)2 (5A) 93.7 5.2 0.4 0.7 2 Cp*2LaCH(SiMe3)2 (5D) 0.1 97.8 0.5 1.6 3 Me2SiCp2YCH(SiMe3)2 (14A) 0.0 71.9 0.1 26.1 4 Me2SiCp2CeCH(SiMe3)2 (14C)b a Reaction conditions: [precatalyst] = 8.8-9.2 mM, benzene-d6 (0.50 mL), substrate (55-57 equiv.) and 25 C. Yields are determined by normalized, in situ 1H NMR spectroscopy and represent average values of two or more runs. The experimental error was found to be 0.2. b No catalytic activity was observed. allowed for the use of in situ 1H NMR spectroscopy to follow substrate conversion in time (Figure 4-2). Normalization of the acetylenic proton resonance of the substrate against CH2(SiMe3)2, formed from quantitative and rapid protonolysis of 5A (Section 4.3.3), indicated that the rate of reaction was first-order in substrate for at least 3.5 half-lives (R2 = 0.9971, kobs = 4.49(7) M-1min-1). The details of the kinetic analysis and the mechanistic consequences are postponed to a later section (Section 4.4.4). The relative amount of products was determined by in situ 1H NMR spectroscopy and their structures are depicted in Table 4-2 (for more details see Section 4.2.3). It can be seen that the present data support the above conclusion that the preference for trans-head-to-head dimerization is favored by a relatively large metal ion radius. The pronounced effect on the regioselectivity of dimerization is remarkable upon substituting the lanthanum metal center in 5D with the smaller yttrium metal center. The smaller relative amount of trimers formed relative to previous reported data (Table 4-1) can be ascribed to the smaller substrate-to-catalyst ratios employed in the present study. The effect of substrate-to-catalyst molar ratios on the selectivity of the present oligomerization reaction is discussed in Section 4.2.4. Structurally, a considerable opening of the metal coordination sphere at the -ligand equatorial girdle is obtained by replacing the bis(pentamethylcyclopentadienyl) ligation in Cp*2LnR by the ansa-ligation Me2SiCp2LnR (Cp = C5Me4) and many examples exist in literature where this ligand transposition has led to substantial effects on the rate and selectivity of reactions catalyzed by rare-earth metallocenes.18 Because the formation of the desired trans-head-to head dimer 12 was found to be favored by more open metal coordination spheres in the (pre)catalysts of the type Cp*2LnR, it was considered of interest to study the effect of opening the metal coordination sphere of Cp*2LaCH(SiMe3)2 (5D) by the use of the (pre)catalyst Me2SiCp2CeCH(SiMe3)2 (14C). No catalytic activity was observed for the reactions of 14C and excess phenylacetylene (50-100 equiv.). Instead, precipitation of the organometallic species took place. The solid was isolated and allowed to react with D2O in benzene-d6. Subsequent NMR and GC/GC-MS analysis indicated that phenylacetylene-d1 was the major product, suggesting the formation of insoluble alkynyl derivatives in the reaction of 14C with phenylacetylene. On the basis of infrared spectral analysis, a similar explanation was previously put forward to account for the absence of catalytic activity in the reaction of 14C with an excess amount of tertbutylacetylene.16a Interestingly, substitution of the bis(pentamethylcyclopentadienyl) ligation in Cp*2YCH(SiMe3)2 (1A) by the silylene-bridged ligation Me2SiCp2 did result in the anticipated increased preference for trans-headto-head dimerization over head-to-tail dimerization, albeit at a considerably slower rate and accompanied by significant trimerization. When Me2SiCp2YCH(SiMe3)2 (14A) was allowed to react with a 50-fold molar excess of phenylacetylene, complete substrate conversion was achieved only after 40 h at room temperature. Substrate conversion was followed in time by monitoring the intensity of the acetylenic proton resonance of the substrate by means of normalized, in situ 1H NMR spectroscopy. The kinetic data obtained after 12 h (corresponding to 50% substrate conversion) indicated that the rate was zero-order in substrate (R2 = 0.9929, kobs = 54.4(4) 10-3

104

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

0.6 concentration (mol/L) 5A 0.4 14A

0.2

0.0 0 100 200 300 400 500 600 700 time (min)

Figure 4-2. Plot of the substrate concentration versus time for the oligomerization of phenylacetylene catalyzed by 5A and 14A. Line and curve connecting the data points represent fitted linear plot and fitted first-order exponential, respectively. min-1). It is evident that the reaction kinetics changed upon opening up the coordination sphere of the metal center (Cp*2Ln Me2SiCp2Ln). In marked contrast to many organo-f-element catalyzed alkene transformations, a dramatic rate deceleration is observed for the present catalytic process, when more open Me2SiCp2Ln-coordinate spheres are used in the precatalysts. These results likely reflect the stability of the presumed active catalyst, the monomeric alkynyl species, and this possibility is discussed in more detail in Section 4.4. The present data on the rare-earth metallocene-catalyzed oligomerization of phenylacetylene indicate also that trimerization and trans-head-to-head dimerization of phenylacetylene are favored by more open coordination metal environments in the (pre)catalysts.

Ph + 5D Ph 11a Ph Ph Ph

Ph Ph Ph

(4.1)

H2 12a Pd/C Ph

+ 17a Ph Ph 18a Ph

+ Ph 15a Ph Ph 16a Ph Ph Ph 19a

4.2.3.

Product analysis

The composition of the product mixtures was determined by in situ 1H NMR spectroscopy, using appropriate long pulse delays to avoid signal saturation under the present anaerobic conditions. Other methods to determine the relative amount of products yielded values that corresponded only moderately to those determined by in situ 1H NMR analysis. For example, normalized FID-GC values for the reaction products were obtained under the assumption that the response factors are linearly correlated to the carbon numbers of the oligomers. The modest agreement between 1H NMR and normalized FID-GC values was attributed to a nonlinear relationship between response factors and carbon numbers and/or the well-recognized (thermal) instability of the head-to-tail dimer19 11a and trimer20 15a. Differences in product volatility were believed to be responsible for

105

Chapter 4

Table 4-3. The effect of substrate concentration on the 5D-catalyzed oligomerization of phenylacetylene (2a) at a constant precatalyst concentration.a Entry [2a] [Cp*H] Cat. deact. 2a 11a 12a 15a 16a (equiv.) (mM) (%)b 1 64 0.29 0.06 0.2 98.2 1.2 0.4 0.6 2 128 0.60 0.22 0.2 95.9 2.9 1.0 2.3 3 258 1.14 0.29 0.1 92.6 5.5 1.8 3.3 4 359 1.58 0.45 0.1 89.1 8.4 2.4 5.1 5 804 3.62 0.70 0.2 85.4 11.3 3.2 7.7 6 1118 5.03 1.44 0.1 82.7 13.5 3.8 16.1 7 258 + 258 0.36 0.1 92.3 5.6 2.1 4.5 8 359 + 359 0.53 0.1 88.8 8.7 2.4 7.1 a Reaction conditions: [5D] = 4.4-4.7 mM, C6D6 and 25 C. Relative amounts of products were determined by normalized in situ 1H NMR spectroscopy and represent average values of two or more runs with an experimental error of 0.2. The experimental error in [Cp*H] was found to be 0.05. b Catalyst deactivation (experimental error of 0.6), calculated from [Cp*H] and [5D]. the poor agreement between the 1H NMR values obtained after vacuum transfer of volatiles or procedures involving in vacuo solvent removal and those obtained by in situ 1H NMR spectroscopy. The identity of the trimers 15a and 16a was established by multinuclear 1D and 2D NMR spectroscopy (see Experimental Section). This identification differs from the structures proposed previously for these products. On the basis of 1H and 13C NMR spectroscopy, Heeres et al. proposed an isomer of 16a having a carbon skeleton different from that of 15a.5f,g Additional evidence for the present structures is provided by the catalytic hydrogenation of the crude product mixture, yielding 1,3-diphenylbutane (17a), 1,4-diphenylbutane (18a) for the dimers and 1,3,6-triphenylhexane (19a), exclusively, for the trimers (eq. 4.1). This result reveals unequivocally that trimers 15a and 16a possess a common carbon skeleton.

4.2.4.

Effect of substrate concentration

Based on the high selectivity and activity for the formation of the desired trans-head-to-head dimer (12a), further investigations were conducted with the precatalyst Cp*2LaCH(SiMe3)2 (5D). In order to determine the effect of substrate structure on the rate of the 5D-catalyzed 1-alkyne conversion (Chapter 5) and monitor the reaction progress by in situ 1H NMR spectroscopy, it was necessary to prolong the reaction time. This was achieved by increasing the substrate concentration relative to precatalyst concentration (Table 4-3). However, both the rate of trimerization and catalyst deactivation increased relative to that of dimerization upon doing so. The increased rate of catalyst deactivation upon increasing the substrate concentration at constant precatalyst concetration is seen from the increased formation of Cp*H (Table 4-3). The adventitious presence of moisture in phenylacetylene (after drying with CaH2) is ruled out by comparing the amounts of Cp*H formed after addition of phenylacetylene either in one or in two portions. If the substrate is contaminated with an unknown Brnsted acid, doubling the amount of substrate is expected to produce twice the amount of Cp*H. Two experiments reveal that doubling the substrate amount produced less than twice the amount of Cp*H (Entry 7 versus 5 and entry 8 versus 6, Table 4-3). It is obvious that catalyst deactivation is not directly proportional to the amount of substrate. The absence of air and moisture in phenylacetylene after drying with CaH2 was also demonstrated by independent reactions of phenylacetylene with Cp2ZrMe2. Cp2ZrMe2 is known to react with trace amounts of O2 and H2O to give soluble and NMR-observable products, while it reacts only slowly with 1-alkynes.21 As a consequence, the presence of unreacted Cp2ZrMe2 in solutions containing large excess amounts of phenylacetylene provided strong assurance that the concentration of Brnsted acids (other than phenylacetylene), undetected by GC and NMR spectroscopy, are extremely low after drying with CaH2.22 Because no catalytic activity was observed for mixtures of 5D with excess phenylacetylene in the presence of Cp2ZrMe2, the present oligomerization reactions could not be performed in the presence of this desiccant. Considerable precedent for the abstraction of cyclopentadienyl ligands in the metallocene chemistry of the rare earth metals exists in literature and the mechanistic and kinetic consequences are discussed in more detail in later sections (Sections 4.4.3-4.4.4).

106

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

1.6 2a Concentration (mol/L) 1.2 0.8 0.4 0 0 50 100 time (min) 150 200 12a 15a 16a

Figure 4-3. Plot of the concentration of 2a, 12a, 15a and 16a versus time for the catalytic oligomerization of 2a (1110 equiv.) in the presence of 5D (1.4 mM), as monitored by normalized, in situ 1H NMR spectroscopy in benzene-d6 at 25C.

4.2.5.

Reaction kinetics

A kinetic study of the catalytic oligomerization of phenylacetylene (1200 equiv.) mediated by Cp*2LaCH(SiMe3)2 (5D) was carried out by means of single-pulse, in situ 1H NMR spectroscopy (Figure 4-3). The 1H NMR resonances of the acetylenic proton C CH ( 2.74 ppm) of the substrate, CH=CH ( 6.29 ppm) of the trans-head-to-head dimer (12a), CHCH2 ( 4.00 ppm) of trimer 15a and CH2CC ( 3.50 ppm) of trimer 16a were normalized to an internal standard (cyclooctane). Although the 1H NMR resonance of the acetylenic proton C CH ( 2.71) of 2a overlapped with that of CHCH2 ( 2.81 ppm) of 16a, the intensity of the acetylenic proton could be calculated by taking the intensity of CHCH2 ( 4.00) of 16a into account. The kinetic concentration profiles of the reaction products show that the concentrations of 12a, 15a and 16a increase with substrate conversion. The concentration of the head-to-tail dimer (11a) and Cp*H were too low to allow reliable quantitative analysis by means of single-pulse, in situ 1H NMR spectroscopy, even at higher precatalyst concentrations.23 A detailed kinetic study of the present catalytic oligomerization reaction by means of single-pulse, in situ 1H NMR spectroscopy under ambient conditions was frustrated by the following circumstances: (i) the acetylenic proton of the substrate requires ~500 s for complete relaxation under the present anaerobic conditions,24 (ii) the high catalytic activity, (iii) the occurrence of non-negligable catalyst deactivation and (iv) competing catalytic trimerization at relatively high substrate-to-catalyst molar ratios. For example, a 400-fold molar excess of substrate was completely converted within 22 min in the presence of 5D (4.5 mM). In order to obtain a reliable kinetic data set (e.g. 10 data points), the reaction time (e.g. 83 min for single-pulse 1H NMR experiments with delays of 500 s) was prolonged by increasing the substrate concentration. When a 1118-fold molar excess was employed, complete substrate conversion was observed after 69 min, thereby allowing the acquisition of 13 data points (Figure 4-3). However, catalytic dimerization was accompanied by significant catalyst deactivation (13%) and catalytic trimerization (17%) at a substrate-to-catalyst molar ratio of 1200 and the kinetic data obtained could not be modeled convincingly to either zero- or first-order rate dependence over the entire substrate conversion range. Varying the substrate concentration over a 17-fold range at constant precatalyst concentration (4.5 mM) revealed that competing catalytic trimerization increased from 1.6% to 17.3% and the degree of catalyst deactivation increased from 0.6% to 13.1% (Table 4-3). Clearly, catalytic trimerization and catalyst deactivation compete more effectively with catalytic dimerization at higher substrate concentrations.

107

Chapter 4

Table 4-4. The effect of precatalyst concentration on the 5D-catalyzed oligomerization of phenylacetylene (2a) at a constant substrate concentration.a Entry [5D] [Cp*H] Cat. deact. 2a 11a 12a 15a 16a (equiv.) (mM) (%)b 1 359 4.23 0.43 0.1 85.6 11.1 3.2 5(1) 2 430 3.53 0.48 0.2 85.5 11.8 2.5 6(1) 3 791 1.92 0.36 0.2 85.4 11.1 3.3 9(1) 4 820 1.85 0.38 0.1 85.3 11.5 3.2 10(1) 5 1047 1.45 0.39 0.2 84.9 11.7 3.5 13(2) 6 1186 1.28 0.40 0.1 84.7 11.8 3.3 16(2) 7 2232 0.68 0.47 0.1 84.8 12.1 3.1 35(4) a Reaction conditions: [2a] = 1.518 M, C6D6 and 25 C. Relative amounts of products were determined by normalized in situ 1H NMR spectroscopy and represent average values of two or more runs with an experimental error of 0.2. The experimental error in [Cp*H] was found to be 0.05. b Catalyst deactivation, calculated from [Cp*H] and [5D]. When the catalyst concentration was decreased over a 6-fold range at constant substrate concentration (1.518 M), competing catalytic trimerization and catalyst deactivation remained invariant within experimental error (Table 4-4). Decreasing the precatalyst concentration below 0.5 mM afforded poorly reproducible kinetic data.25 This suggested that catalyst poisoning by trace impurities (e.g. H2O, O2) dominates the kinetic behavior at very low precatalyst concentration. Hence, precatalyst concentrations above 0.5 mM were used in the present study and the reproducible kinetic data obtained (experimental error within 5%), using different batches of precatalyst, substrate and solvent, provided strong assurance that catalyst poisoning by traces of protonolytic agents was negligible under these conditions. When precatalyst concentrations above 0.5 mM were used in combination with substrate concentrations that lead to predominant catalytic dimerization, substrate was converted in less than 60 min, thereby hampering the acquisition of a reliable kinetic data set. Plots of the substrate concentration versus time during catalytic conversion of phenylacetylene at constant precatalyst concentration (2.25 mM) showed a gradual shift from a mostly substrate independent regime (linear consumption plot) at relatively low initial substrate concentration to a substrate dependent regime (approximately exponential consumption plot) at a relatively high initial substrate concentration (Figure 4-4). At relatively low substrate concentrations, the substrate conversion displays predominantly a zero-order rate dependence on substrate concentration, as demonstrated by the oligomerization of phenylacetylene in the

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 50 100 150 200 250 time (min) 300 350 400 450 500 1.05 M 1.80 M 2.46 M 3.19 M

Figure 4-4. Plot of the substrate concentration versus time for the catalytic oligomerization of phenylacetylene mediated by 5D at a constant precatalyst concentration (2.25 mM) and different initial substrate concentrations.

108

substrate concentration (mol/L)

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

2.4 substrate concentration (M) 2 1.05 M 1.6 1.2 0.8 0.4 0 0 40 80 time (min) 120 160 200 1.80 M 2.46 M

Figure 4-5. Plot of the substrate concentration versus time for the catalytic conversion of phenylacetylene in the presence of 5D (2.25 mM) at different initial substrate concentrations.

presence of a 404-fold molar excess of substrate and 5D (Figure 4-5). Catalytic substrate conversion was achieved within 27 min under these conditions and the reaction products revealed the occurrence of catalytic dimerization (94%), catalytic trimerization (7%) and catalyst deactivation (5%). In spite of the occurrence of these three processes, the reaction rate corresponds well to zero-order rate dependence on substrate concentration (R2 = 0.9963, kobs = 18(1) min-1) over the entire substrate conversion range. When the substrate concentration was increased at constant precatalyst concentration, the kinetic behavior changed and became increasingly more first-order dependent on substrate concentration (Figure 4-4). Concomitantly, the product distributions revealed that the rate of catalytic trimerization and catalyst deactivation increased relative to catalytic dimerization. The catalytic oligomerization of phenylacetylene in the presence of a 1416-fold molar excess of substrate and 5D represents a limiting case of such behavior in the present study (Figure 4-4). Complete substrate conversion was achieved after 500 min and the reaction products revealed

0 0 -0.5 -1 ln([S]t /[S]0) -1.5 -2 -2.5 -3 -3.5 -4

50

100

150

200

250

300

350

400

450

500

3.19 M

2.46 M

1.80 M

time (min)

Figure 4-6. Integrated rate plot of the substrate concentration versus time for the catalytic conversion of phenylacetylene in the presence of 5D (2.25 mM) at different initial substrate concentrations.

109

Chapter 4

1.2 substrate concentration (M) 1.0 run 1 0.8 0.6 0.4 0.2 0.0 0 25 50 75 100 time (min) 125 150 175 200 run 2 run 3

Figure 4-7. Plot of the substrate concentration versus time for the catalytic conversion of phenylacetylene in the presence of 5D (2.15 mM) after addition of consecutive portions of phenylacetylene (488 equiv.).

conversion), but deviated from first-order kinetic behavior in substrate at higher substrate conversion. At higher substrate conversion, the kinetic behavior of the reaction rate conforms to zero-order rate dependence on substrate concentration, but the observed rate is considerably lower as observed at relatively low intial substrate concentration (Figure 4-4). Catalyst deactivation (24%) cannot account for this significant decrease in reaction rate and product inhibition may be invoked to explain this behavior. To investigate the possibility of product inhibition, several equimolar portions of substrate (488 equiv.) were added successively to a solution of 5D in benzene-d6 (2.15 mM). The first portion of substrate (run 1) was completely consumed within 30 min (Figure 4-7), accompanied by minor catalyst deactivation (3%). The rate of substrate conversion could convincingly be modeled by zero-order dependence on substrate concentration (R2 = 0.9946, kobs = 17(1) min-1). Addition of a second portion of substrate (run 2) led to complete substrate

0 0

25

50

75

100

125

150

175

200

run 3 -1 ln([S]t /[S]0)

run 2

run 1

-2

-3

-4 time (min)

Figure 4-8. Integrated rate plot of the substrate concentration versus time for the catalytic conversion of phenylacetylene in the presence of 5D (2.15 mM) after addition of consecutive portions of phenylacetylene (488 equiv.).

110

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

2.30

2.20

2.10

2.00

1.90

1.80

1.70 ppm

Figure 4-9. 400 MHz 1H NMR spectra of the reaction of the precatalyst 5D (1.92 mM) and a 430-fold molar excess of phenylacetylene in benzene-d6 at 25 C: (top) white-wash stack plots displaying the timedependent chemical shifts of characteristic Cp* 1H NMR resonances during and after substrate conversion and (bottom) contour plots of the same resonances (see text for details). The first spectrum is taken 8 min after substrate addition, while ensuing spectra are taken with intervals of 5 min. conversion after 52 min, accompanied by additional catalyst deactivation (3%). Deviation from zero-order rate dependence on substrate concentration (R2 = 0.9976, kobs = 8.4(2) min-1) was observed after 70% substrate conversion. A third successive portion of substrate (run 3) was consumed only after 170 min, accompanied by yet another 3% of catalyst deactivation. In this case, deviation from zero-order rate dependence on substrate concentration (R2 = 0.9946, kobs = 3.5(1) min-1) was observed after 62% substrate conversion. Clearly, catalyst deactivation cannot account for the observed decrease in reaction rate and the proposed inhibitory effect of the reaction products on the catalytic activity seems to be justified. Alternatively, the effects of product inhibition may also be interpreted as altering the kinetic order in substrate concentration from zero- to first-order (Figure 4-8).26 The reaction rate after the third portion of substrate may in this scenario convincingly be modeled by first-order rate dependence on substrate concentration (R2 = 0.9893, kobs = 6.5(2) M-1min-1) during the first 96 min (69% substrate conversion).

4.2.6.

Reaction intermediates

Monitoring the reaction of 5D (1.92 mM) with phenylacetylene (791 equiv.) in benzene-d6 by in situ H NMR spectroscopy revealed that one predominant organometallic species (~90% of the total amount of precatalyst added) was present during substrate conversion, as evidenced by one major Cp* 1H NMR resonance at 2.12 ppm (Figure 4-9). This resonance was, furthermore, found to shift gradually to 2.14 ppm with increasing substrate conversion. As substrate depleted (~89% substrate conversion), this major lanthanocene derivative was converted into three new derivatives, based on the appearance of three new Cp* 1H NMR resonances at 2.04, 1.97 and 1.92 ppm. Upon complete substrate conversion, these derivatives were the major lanthanocene species present and the composition of the reaction mixture did not change upon standing at room temperature. When another portion of substrate was added to the reaction mixture, analogous observations were made with 1H NMR spectroscopy: one major Cp* signal was observed at 2.12 ppm during substrate conversion which shifted gradually to 2.14 ppm and gave rise to the three aforementioned signals at a more advanced stage of substrate conversion. To identify the lanthanocene derivatives involved in the catalytic oligomerization of phenyacetylene, quenching experiments with THF and D2O were performed in conjunction with stoichiometric reactions of several catalyst precursors with phenylacetylene (Section 4.3). Addition of excess THF to a reaction mixture of 5D (17.6 mM) and phenylacetylene (50 equiv.) during substrate conversion gave Cp*2LaCCPh(THF) as the only observable organometallic species with NMR spectroscopy. The identity of Cp*2LaCCPh(THF) was established by its independent preparation and its reaction with D2O affording phenylacetylene-d1 (Chapter 5). Based on this
1

6 0 tim e(m in )

1 2 0

2.30

2.20

2.10

2.00

1.90

1.80

1.70 ppm

111

Chapter 4

2.5 2.0 Concentration (mM) 1.5 1.0 0.5 0.0 0 20 40 60 time (min) 80 100

100 80 60 40 20 0 120 Substrate conversion (%)

20a2a

22a

24a

23a

conv

Figure 4-10. Concentration profiles of Cp*2LaCCPh(PhCCH) (20a2a), [(Cp*2La)2(-PhC4Ph)] (22a), Cp*2LaC(Ph)=C(H)CCPh (24a) and Cp*2LaC(H)=C(Ph)CCPh (23a) during the reaction of the precatalyst 5D (1.92 mM) and a 430-fold molar excess of phenylacetylene in benzene-d6 at 25 C, as monitored by normalized in situ 400 MHz 1H NMR spectroscopy (Figure 4-9). The lines connecting the data points are drawn as guides for the eye. observation, the broad signal at 2.12 ppm was assigned to the base adduct Cp*2LaCCPh(PhCCH) (20aPhCCH). Additional evidence for this assignment is provided by the gradual downfield shift of its Cp* 1H NMR resonance during substrate conversion, thereby indicating that the corresponding derivative is in equilibrium with free phenylacetylene (Eq. 4.2).27 The signals observed after substrate conversion at 2.04, 1.97 and 1.92 ppm were assigned to a butatrienediyl and two but-1-en-3-yn-1-yl derivatives, i.e. [(Cp*2La)2(PhC4Ph)] (22a), Cp*2LaC(Ph)=C(H)CCPh (24a) and Cp*2LaC(H)=C(Ph)CCPh (23a), respectively, and their identification is discussed in Sections 4.3.2-4.3.3.

Cp*2La

Ph

(4.2)

Cp*2La 20a

Ph

+ 2a

Ph Ph 20a . 2a

The failure to prepare Lewis base adducts of dimeric alkynyl derivatives [Cp*2Ln(-CCR)]2 in which the dinuclear character is retained argues against the involvement of dinuclear adducts [Cp*2La(CCPh)(PhCCH)]2.28 The possibility that the broad signal at 2.12 ppm corresponds to a base free, alkynyl derivative is ruled out by an in situ 1H NMR spectroscopic studies (Section 4.3.3). The dimeric alkynyl derivative [Cp*2La(-CCPh)]2 was only observed at low temperatures (as a bright-yellow and thermo-labile solid that was only sparingly soluble in toluene at -50 C), displaying a Cp* 1H NMR resonance at 2.22 ppm in toluene-d8 at -50 C, whereas no spectroscopic evidence for the monomeric, base-free alkynyl derivative Cp*2LaCCPh was obtained (down to -60 C in toluene-d8). When the reaction of 5D (2.25 mM) and phenylacetylene (1416 equiv.) was monitored in benzene-d6 by in situ 1H NMR spectroscopy, two major Cp* 1H NMR resonance were initially observed at 2.09 and 2.23 ppm during substrate conversion. Upon substrate conversion, the broad resonance at 2.09 ppm shifted gradually to 2.14 ppm, while the resonance at 2.23 ppm decreased gradually in intensity. Both Cp* resonances disappeared completely upon substrate depletion, forming the three aforementioned Cp* 1H NMR resonances at 2.04, 1.97 and 1.92 ppm. It seems reasonable to assign the Cp* 1H NMR resonance at 2.23 ppm to an intermediate species on the trimerization pathway, since the degree of trimerization is considerable (20%)

112

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes Scheme 4-3. The formation of lanthanidocene alkynyl derivatives and their rearrangement products.a
R Cp*2LnR' 2 - R'H Cp*2Ln 20 R R Cp*2Ln R R LnCp*2 LnCp*2 21 Cp*2Ln R

22

Entry 1 2 3 4 5
c c

Ln a h i Ci Ch Ea Ej
1

R (ppm) Ph
t b

21 T ( C)
b

22 (ppm) 2.08 2.07 2.06 T ( C) 20 C 20 C 21 C

Ref. 16d 16d 16a 16a 16a 16c 16b,c

La La La Ce Ce Sm Sm

Bu

2.21 2.19

20 C -10 C

Me Me
t

Bu

6c 7c
a

Ph CH2CH2Ph
b c

Reported Cp* H NMR resonances of 21 in C7D8 and of 22 in C6D6. Not observed. Paramagnetic 1H NMR resonances have been reported, but are not relevant in the present context.

under these reaction conditions (Section 4.2.5). The observation that the Cp* 1H NMR resonance at 2.09 ppm shifts gradually to 2.14 ppm under these reaction conditions supports the above notion that the active catalyst of dimerization is Cp*2LaCCPh which is in equilibrium with free phenylacetylene (Eq. 4.2).

4.3.

Stoichiometric reactions with phenylacetylene

4.3.1.

Introduction

As mentioned before, the catalytically active species of the catalytic 1-alkyne dimerization is thought to be a monomeric, alkynyl derivative Cp*2LnC CR (20). Although these species have been trapped with hard Lewis bases, direct evidence for the existence of monomeric, unsolvated, terminal alkynyl derivatives is -to the best of our knowledge- lacking in literature.29 The hypothetical, base-free monomeric Cp2LnCCR derivatives are unstable and rearrange to form dimeric bridged alkynyl complexes [Cp2Ln(-C CR)]2 (21) and/or butatrienediyl complexes (Cp2Ln)2(-2:2-RC=C=C=CR) (22). Dinuclear lanthanidocene alkynyl derivatives 21 have been synthesized via reactions of alkyl and hydride precursors with the corresponding 1-alkynes, but syntheses involving the thermolysis of Cp*2LnCCR(THF) complexes (20THF) have also been reported.16b,c Several dimeric, bridging alkynyl 21 and butatrienediyl derivatives 22 have been characterized both spectroscopically and structurally (Scheme 4-3). The rate of lanthanidocene alkynyl C-C coupling (21 22) is believed to be governed by a complex interplay between the alkyne substituent, the ligand Cp and the metal. In some cases (e.g. 21i), this process was found to be reversible, producing equilibrium mixtures at ambient temperatures, and, in other cases (e.g. 21h), irreversible, exhibiting clean first-order kinetic behavior. In an effort to identify the catalytic reaction sequences of the 5D-catalyzed oligomerization reaction of phenylacetylene, stoichiometric reactions of catalyst precursors with phenylacetylene were performed. Stoichiometric reactions of the hydride [Cp*2La(-H)]2 (25D) and alkyl derivative Cp*2LaCH(SiMe3)2 (5D) with phenylacetylene afforded butatrienediyl and butenynyl lanthanocene derivatives, previously observed during the 5D-catalyzed oligomerization of phenylacetylene (Section 4.2.6). Reactivity studies of the butatrienediyl

113

Chapter 4 derivative towards Lewis bases and protonolytic reagents provided evidence for its relative importance in the conversion of phenylacetylene into its observed oligomers.

4.3.2.

The hydride derivative

The reactions of the hydride derivative [Cp*2La(-H)]2 (25D) with a stoichiometric amount of phenylacetylene in benzene-d6 or toluene-d8 were clean at room temperature, affording [(Cp*2La)2(-PhC4Ph)] (22a) as the single observable product (Eq. 4.3). The identity of 22a was established by multinuclear 1D and 2D NMR spectroscopy and X-ray structure determination. The preparation of [(Cp*2La)2(-PhC4Ph)] (22a) had previously been reported via the reaction of phenylacetylene with excess Cp*2LaCH(SiMe3)2 (5D) at room temperature.16d In the present study, the reaction of the hydride [Cp*2La(-H)]2 (25D) and phenylacetylene (1 equiv.) in hexanes afforded 22a as dark red crystals in 82% isolated yield after crystallization at low temperature. Another method to prepare 22a involves vapor phase transfer of phenylacetylene into a stirred solution of an equimolar amount of 5D in hexanes at room temperature (77% isolated yield). Recrystallization in toluene allowed for the formation of 22a2C7H8 as single-crystals suitable for X-ray analysis. The details of the solidstate molecular structure of 22a and comparison with (hetero)aromatic analogues are discussed in Chapter 5. In the present context, it suffices to note that the two Cp*2La moieties bind each in an 3 fashion to a bridging, formally dianionic butatrienediyl ligand [PhC=C=C=CPh]2- in 22a. The butatrienediyl derivative displays a characteristic Cp* 1H NMR resonance at 2.04 ppm in benzene-d6 solution which has also been observed after full conversion of the substrate during the catalytic oligomerization of phenylacetylene mediated by 5D (Section 4.2.6).

Ph

(4.3)

[(Cp*2La(H)]2 25D

25 C + 2a Ph -H2 Cp*2La 22a

LaCp*2

Ph

When the above reaction was performed at low temperatures by condensing phenylacetylene onto a solution of 5D at -100 C in toluene, the formation of a yellow suspension was observed within several minutes. Upon warming to -50 C, the yellow suspension changed within several hours into a dark red solution. In situ 1H NMR experiments in toluene-d8 revealed the transient formation of a new lanthanum species, exhibiting a Cp* 1 H NMR resonance at 2.22 ppm. This species was identified as the dimeric, alkynyl derivative [(Cp*2La(CCPh)]2 (21a), based on the formation of phenylacetylene-d1 upon reaction with D2O.30 Within several hours at -50 C, 21a was completely converted into [(Cp*2La)2(-PhC4Ph)] (22a). Attempts to observe the proposed monomeric, alkynyl derivative 20a in the reaction mixture at -78 C were unsuccessful. Interestingly, the yellow color of 21a was also noted for other dimeric, alkynyl lanthanidocene derivatives and seems to be independent of both metal and alkyne substituent (Chapter 5). Unfortunately, the observation of clean first-order kinetic behavior for the conversion of 21a into 22a by means of in situ 1H NMR spectroscopy was hampered by the low solubility of 21a in toluene-d8 at -50 C. The solubility of 21a in methylcyclohexane-d14 at -50 C was even lower, while dichloromethane-d4 and mixtures of dichloromethane-d4 and toluene were found to react with the lanthanocene derivatives at these temperatures, affording numerous species that defied characterization. In spite of the low solubility of 21a, reasonable kinetic data (R2 = 0.9922, t1/2 = 3.2(2) h, -50 C) was obtained by (i) lowering the concentration of 5D to such an extent that quantitative NMR analysis was well possible within the time scale of the reaction and (ii) discarding the first data points, due to the low solubility of 21a (Figure 5-11). In this manner, the rate of rearrangement into 22a was found to be first-order in 21a for at least 2.5 half-lives (R2 = 0.9922, t1/2 = 3.2(2) h, 50 C). These observations provide evidence that 22a is formed from 21a and that this transformation is both rapid and quantitative (Eq. 4.4). It is interesting to note that the methyl analogue [Cp*2La(-C CMe)]2 (21i)

114

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes rearranged at room temperature (t1/2 = 3.2 h, 25 C) to produce an equilibrium mixture of [(Cp2Ln)2(MeC4Me)] (22i), while the tert-butyl analogue [Cp*2La(-C CtBu)]2 (21h) rearranged much more slowly (t1/2 = 2.2 h, 50 C), but irreversibly, into 22h and was isolated at 0 C.16a These results indicate that the alkyne substituent plays an important role in determining the rate of C-C coupling in dimeric lanthanidocene alkynyl derivatives 21. These alkyne substituent effects are discussed in more detail in Chapter 5. When the stoichiometric reaction of 25D and phenylacetylene was performed at low temperature using a small excess of phenylacetylene (1-2 equiv.), the formation of a small amount (1-2% relative to 22a) of an unidentified, organometallic by-products was observed, displaying one Cp* 1H NMR resonance at 2.01 ppm. Attempts to isolate and identify this species failed. Similar observations have been reported previously for reactions of [Cp*2Ce(-CCtBu)]2.16a On the basis of infrared analysis, the authors suggested that the unknown species is analogous to the dimeric alkynyl derivative [Cp*2Ce(-CCtBu)]2, but differs only in the degree of association.

Ph

(4.4)

[Cp*2La(-H)]2 + Ph

Ph LaCp*2

-50 C - H2

Cp*2La 20a

Ph

Cp*2La Ph

LaCp*2 21a

Cp*2La 22a

Ph

4.3.3.

The alkyl derivative

The reaction of Cp*2LaCH(SiMe3)2 (5D) with phenylacetylene (1 equiv.) in non-coordinating solvents, such as benzene, toluene or hexanes, at room temperature was rapid, as indicated by the immediate color change of the solution from pale yellow to dark red. 1H NMR experiments in benzene-d6 indicated the formation of a mixture, containing unreacted 5D, the 1,4-diphenylbutatrienediyl derivative [(Cp*2La)2(-3:3PhC4Ph)] (22a), two new lanthanum species, assigned to Cp*2LaC(H)=C(Ph)CCPh (23a) and Cp*2LaC(Ph)=C(H)CCPh (24a), and the organic compounds, (E)-1,4-diphenyl-but-1-en-3-yne (12a) and 2,4diphenylbut-1-en-3-yne (11a) (Eq. 4.5). The butenynyl derivatives 23a and 24a display characteristic Cp* 1H NMR resonances at 1.93 and 1.97 ppm, respectively, previously observed after full conversion of substrate in the catalytic oligomerization of phenylacetylene mediated by 5D (Section 4.2.6).

Cp*2LaCH(SiMe3)2 + PhCCH 5D Ph LaCp*2 Cp*2La 22a Ph 1.21 : + Cp*2La 24a 0.21 2a

-CH2(SiMe3)2 Ph + Cp*2La Ph : Ph

Cp*2LaCH(SiMe3)2 + 0.72 Ph + 23a Ph : 11a 0.12 Ph + 12a Ph : 0.07 : Ph

(4.5)

0.17

Unfortunately, 23a and 24a could not be isolated on a preparative scale,31 but NMR analysis and quenching experiments with D2O and H2O provided reasonable evidence for their identities. The 1H NMR spectrum of a mixture containing 23a and 24a displayed two singlets at 6.69 (correlating with 120.85 ppm in the 13C NMR spectrum, as indicated by a 1H-13C gHSQC experiment) and 1.93 ppm in a 1:30 ratio, assigned to =CH and Cp* groups of 23a and two singlets at 7.74 (correlating with 132.55 ppm in the 13C NMR spectrum) and 1.97 ppm in a 1:30 ratio, assigned to =CH and Cp* groups of 24a.

115

Chapter 4

2.30

2.20

2.10

2.00

1.90 ppm

Figure 4-11. 500 MHz 1H NMR spectra of the Cp* region of a mixture containing the precatalyst 5D and a stoichiometric amount of phenylacetylene in toluene-d8 at -50 C. The bottom spectrum is taken 14 min after warming to -50 C (see text for details) and the consecutive upper spectra are taken with intervals of 10 min. The far-downfield location of the vinyl group resonance is consistent with that of the alkenyl derivative Cp*2LaC(Ph)=C(Ph)H, exhibiting a proton resonance at 6.99 ppm and a carbon resonance at 134.28 ppm due to the =CH group (Chapter 2). Based on the LaC 13C NMR resonances at 217 ppm for both 22a and Cp*2LaC(Ph)=C(Ph)H, two characteristic, low-field singlets at 190 and 183 ppm in the 13C NMR spectrum were assigned to the LaC groups of 24a and 23a, respectively. When a mixture containing 22a, 23a and 24a in a 1.8:0.6:1.0 ratio, respectively, and in the absence of 11a and 12a was treated with D2O, (E)-1,4-diphenyl-1,4-dideuteriobutatriene (27a-d2), 1-deuterio-2,4diphenylbut-1-en-3-yne (11a-1-d) and (E)-1-deuterio-1,4-diphenyl-but-1-en-3-yne (12a-1-d) formed in a 1.8:0.6:1.0 ratio, as indicated by NMR and GC-MS analysis. The formation of 27a-d2 is attributed to the reaction of 22a with D2O (Section 4.3.4), while that of 11a-1-d1 and 12a-1-d1 is attributed to the reaction with 23a and 24a with D2O, respectively. The analogous reaction with H2O provided 27a and 12a in the corresponding ratios. Addition of excess THF (1-5 equiv.) to a mixture of 23a and 24a resulted in their rapid conversion into several new, unidentified lanthanocene derivatives. When the reaction of 5D was performed at low temperatures by condensing phenylacetylene onto a solution of 5D at -196 C in toluene-d8, 1H NMR spectroscopy revealed the transient formation of the dimeric, alkynyl derivative [(Cp*2La(-CCPh)]2 (21a) ( 2.22 ppm) alongside that of the previously observed insertion products ( 2.05 ppm, Section 4.3.2), Cp*2LaC(H)=C(Ph)CCPh (23a) ( 2.00 ppm) and Cp*2LaC(Ph)=C(H)CCPh (24a) ( 1.97 ppm) (Figure 4-11). Within several hours at -50 C, 21a was completely converted into [(Cp*2La)2(-PhC4Ph)] (22a) ( 2.09 ppm, overlapping with the residual protio signal of toluened8). The above observations may plausibly be interpreted as proceeding via a monomeric alkynyl species 20a. At room temperature the rate of alkyne insertion into 20a is more rapid than the rate of protonolysis of 5D by alkyne. The rate of alkyne insertion into 20a relative to protonolysis of 5D by alkyne can be decreased to some extent by lowering the reaction temperature. In the absence of phenylacetylene, 20a undergoes dimerization into 21a. Subsequent C-C coupling of 21a to yield 22a is very rapid at room temperature. Even though [Cp*2La(-H)]2 (25D) can reasonably be assumed to exist as an equilibrium between dimer and monomer Cp*2LaH in solution,32 alkyne insertion into 20a does not compete effectively with dissociation of 25D and subsequent protonolysis of Cp*2LaH by alkyne.33 As proposed before, 20a is believed to yield 1-alkyne dimerization products via sequential alkyne insertion and protonolysis, affording the head-to-tail (11a) and the trans-head-to-head dimer (12a).5b,f

4.3.4.

The butatrienediyl derivative

Introduction The observation of reaction intermediates during 5D-catalyzed oligomerization of phenylacetylene provided evidence that [(Cp*2La)2(-PhC4Ph)] (22a) was formed only at a more advanced stage of substrate conversion and plays therefore a minor role in the observed catalytic activity (Section 4.2.6). However, the addition of excess phenylacetylene to mixtures containing 22a, Cp*2LaC(H)=C(Ph)CCPh (23a) and Cp*2LaC(Ph)=C(H)CCPh (24a) revealed that the active catalyst Cp*2LaCCPh (20a) can be regenerated from 22a. To investigate this issue in more detail, reactions of 22a with Brnsted acids were performed.

116

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes Protonolysis of lanthanide butatrienediyl derivatives [(Cp*2Ln)2(-RC4R)] (Ln = La, Ce; R = Me, Bu) with alcohols is known to produce mixtures of 1,4-disubstituted but-1-en-3-ynes and butatrienes.16a In contrast, no reactivity with phenylacetylene was observed for [(Cp*2Sm)2(-PhC4Ph)] (22Ea).16d It seems not unlikely that the details of the protonolysis reaction of [(Cp*2Ln)2(-RC4R)] depend on the nature of the Brnsted acid, the metal and the alkyne substituent. Because dimeric lanthanidocene alkynyl derivatives [(Cp*2Ln(-CCR)]2 react with hard Lewis bases to afford the corresponding Lewis base adducts of Cp*2LnCCR, the reaction of butatrienediyl derivatives [(Cp*2La)2(-RC4R)] (22) with hard Lewis bases can be used to probe the rate of the transformation of 22 into [(Cp*2La(-CCR)]2 (21).16 As noted before, the methyl analogue was found to produce an equilibrium mixture between [Cp*2La(-C CMe)]2 (21i) and [(Cp2Ln)2(-MeC4Me)] (22i), whereas no evidence for the formation of [Cp*2La(-C CtBu)]2 (21h) from [(Cp2La)2(-tBuC4tBu)] (22i) was obtained for the tert-butyl analogue.16a
t

Reactivity towards 1-alkynes 1 H NMR experiments of [(Cp*2La)2(-PhC4Ph)] (22a) with phenylacetylene (2 equiv.) in benzene-d6 led to product mixtures, containing residual starting material 22a, the but-1-en-3-ynyl derivatives 23a and 24a and the phenylacetylene dimers 11a and 12a (Eq. 4.6), as determined by NMR spectroscopy. The addition of smaller amounts of phenylacetylene (1 equiv.) produced mixtures containing more starting material 22a, while the addition of larger amounts (5 equiv.) led to product mixtures containing more oligmerization products 11a and 12a. These reaction are in accord with previous observations, indicating that 22a can be converted into the active catalyst Cp*2LaCCPh (Section 4.2.6). However, the fate of the butatrienediyl ligand of 22a remained ambiguous, because both 11a and 12a may originate from [PhC=C=C=CPh]2- upon protonolysis of 22a by phenylacetylene, as observed for lithiated butatrienes.34

Ph LaCp*2 Cp*2La + Ph Ph LaCp*2 Cp*2La 22a Ph 0.17 : + Cp*2La 24a 0.13 Ph : + Cp*2La Ph + 23a Ph 0.16 : 12a 0.03 + 11a Ph : 0.02 Ph Ph Ph Ph 2 PhCCH

(4.6)

Ph LaCp*2 + 10 Cp*2La S

(4.7)

22a

Ph

2d

S Ph

Cp*2La
S

+ 24d S 12d

+ Ph 12a

To solve this issue, reactions of [(Cp*2La)2(-PhC4Ph)] (22a) with another 1-alkyne, 2ethynylthiophene (2d), were performed. When a 10-fold molar excess of 2d was employed to ensure complete conversion of 22a, a product mixture was obtained that consisted of the but-1-en-3-yn-1-yl derivative

117

Chapter 4 Cp*2LaC(2-C4H3S)=C(H)CC(2-C4H3S) (24d), (E)-1,4-di(2-thienyl)but-1-en-3-yne (12d) and 12a (Eq. 4.7). Thus, the butatrienediyl ligand of 22a is released exclusively as 12a upon protonolysis of 22a by 2ethynylthiophene. Evidence for the identity of 24d and 12d will be discussed in Chapter 5. When a larger molar excess of 2d (20 equiv.) was added, more 12d formed. It is interesting to note that the reactivity of [(Cp*2La)2(-PhC4Ph)] (22a) contrasts with that observed for [(Cp*2Sm)2(-PhC4Ph)] (22Ea). In the case of samarium, no reactivity with phenylacetylene was observed in toluene up to temperatures of 100 C with up to a 100-fold excess of alkyne. In marked contrast, both 22a and 5D catalyze the oligomerization of phenylacetylene and both the rate of substrate conversion and the product mixtures obtained are indistinguishable. The low reactivity of 22Ea relative to 22a towards phenylacetylene can plausibly be ascribed to the smaller metal size of samarium relative to lanthanum, since the difference in ionic radius for eight-coordinate complexes is 0.047 .17

Ph

(4.8)
Cp*2La

LaCp*2

H2O Ph

Ph

Ph

27a

Ph LaCp*2 + Ph Ph Ph + Ph : 12a 2.2 : + 13a 1.0 Ph Ph OH

(4.9)

Cp*2La

Cp*2La O

+ Ph

. 27a 5.1

Reactivity towards alcohols The above results provide evidence that the butatrienediyl ligand is converted selectively into trans1,4-diphenylbut-1-en-3-yne in the reaction of [(Cp*2La)2(-PhC4Ph)] (22a) with 1-alkynes. This behavior differs from that observed for the reactions of the butatrienediyl derivatives [(Cp*2Ln)2(-RC4R)] (Ln = La, Ce; R = Me, tBu) with alcohols, affording mixtures of 1,4-disubstituted but-1-en-3-ynes and butatrienes.16a To investigate whether this discrepancy can be attributed to a difference in the electronic properties of the Brnsted acid,35 reactions of 22a with H2O and a sterically hindered alcohol were performed. The reaction of [(Cp*2La)2(-PhC4Ph)] (22a) with excess H2O furnished (E)-1,4-diphenylbutatriene (27a) and Cp*H, as the only observed organic products with NMR and GC-MS (Eq. 4.8). Upon standing at room temperature for several hours the concentration of 27a slowly decreased, but no 1-alkyne oligomerization products 11a and 12a were formed. In fact, 27a is well-known to form poly(2-butyne-1,4-diyl) thermally.36 These findings indicate that the butatrienediyl ligand of 22a is released as (E)-1,4-diphenylbutatriene (27a) upon reaction with H2O and as (E)-1,4-diphenylbut-1-en-3-yne (12a) upon reaction with 2ethynylthiophene. Because this difference in reactivity may be attributable to both the steric and electronic properties of the Brnsted acid, the reaction of 22a with 4-methyl-2,5-di-tert-butylphenol (ArOH) was performed. The choice for this substituted phenol was based on the reasonable assumption that its electronic properties resemble those of H2O, while its steric requirements for reactivity are considerably larger than that of H2O. In contrast to the reaction with H2O, the reaction of 22a with a stoichiometric amount of 4-methyl2,5-di-tert-butylphenol was not instantaneous. It took several seconds before the color changed from dark red to

118

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

Scheme 4-4. The proposed reaction sequences leading to the observed protonolytic quenching products of 22a.
Cp*2La Ph H Ph Ph Cp*2La Ph Cp*2La 22a Ph Ph Cp*2La Ph LaCp*2 Ph Cp*2La Ph LaCp*2 HX Ph + Ph H Ph 12a . . LaCp*2 HX Cp*2La Ph H Ph LaCp*2 HX Ph . . Ph Ph H HX Ph . . Ph 12a Ph HX

27a

light yellow. 1H NMR spectroscopy indicated that the reaction mixture consisted of Cp*2LaOAr,37 (E)-1,4diphenylbutatriene (27a), (Z)- (13a) and (E)-1,4-diphenylbut-1-en-3-yne (12a) (Eq. 4.9). Discussion of protonolytic reactions of the butatrienediyl derivative The above reactions of [(Cp*2La)2(-PhC4Ph)] (22a) with different Brnsted acids can arguably be explained by proposing that the delocalized 3-bonding of the Cp*2La moiety with the butatrienediyl ligand in 22a can be represented by several tautomers (Scheme 4-4). Analogous to the Cp*2La(3-CH2CCPh) complexes, no evidence for fluxional behavior was obtained for 22a with 1H and 13C NMR spectroscopy in the temperature range from -80 C to 100 C in both aliphatic (i.e. cyclohexane-d12, methylcyclohexane-d14) and aromatic (i.e. benzene-d6, toluene-d8) solvents. Although spectroscopic and structural evidence argued for a static 3propargyl/allenyl bonding in the Cp*2LnCH2CCAr complexes, protonolysis reactions afforded acetylenic (i.e. CH3CCAr) and allenylic products (i.e. CH2CCHAr) originating from protonolysis of the 1-propargylic and 1allenylic ligand, respectively (Chapter 2). The relative amount of the protonolysis products for the reactions of Cp*2La(3-CH2CCPh) with phenylacetylene was found to be under electronic control, producing mainly phenylallene.

H Cp*2La Ph . . Ph H Cp*2La H Ph Ph Ph 13a HX Ph . . Ph

(4.10)

Cp*2La Ph

119

Chapter 4 The present results of the protonolytic reactions of [(Cp*2La)2(-PhC4Ph)] (22a) suggest that the major quenching product is determined mainly by the electronic properties of the Brnsted acid. 1-Alkynes favor the formation of the butenyne quenching products (i.e. 12a and 13a), while alcohols favor the formation of butatriene quenching products (i.e. 27a). The preference for the butatriene quenching product can be diminished to some degree by increasing the steric bulk of the Brnsted acid. The proposed reaction sequences, involving 1,3-metal shifts analogous to those proposed for the 3-propargyl/allenyl Cp*2Ln(3-CH2CCAr) complexes, account for the formation of (E)-1,4-diphenylbutatriene (12a) and (E)-1,4-diphenylbut-1-en-3-yne (27a) (Scheme 4-4). The formation of (Z)-1,4-diphenylbut-1-en-3-yne (13a) may be explained by proposing cis-trans isomerization of an intermediate butatrienyl derivative (Eq. 4.10).38 Reactivity towards hard Lewis bases No reactivity of [(Cp*2La)2(-PhC4Ph)] (22a) towards THF (1-10 equiv.) was observed at room temperature after standing for several days. This behavior agrees well with previous observations, involving analogous samarium, neodymium and cerium butatrienediyl derivatives [(Cp*2Ln)2(-PhC4Ph)].16b,c When [(Cp*2La)2(-PhC4Ph)] (22a) was heated in benzene-d6 to 80 C in the presence of an excess amount of THF, 1H NMR spectroscopy indicated the formation of Cp*2LaCCPh(THF) (20aTHF), accompanied by the formation of other unidentified organometallic products (Eq. 4.11). Further heating led to the consumption of 20aTHF under formation of more by-products. Similar 1H NMR resonances were observed after heating benzene-d6 solutions of 20aTHF to 80 C, thereby revealing that the by-products are most likely formed from thermolysis of 20aTHF. Because [(Cp*2Sm)2(-PhC4Ph)] did not form Cp*2SmCCPh(THF) at 120 C in THF as a solvent,16d the reactivity of [(Cp*2Ln)2(-PhC4Ph)] seems to be enhanced by a larger metal size.

Ph

(4.11)
Cp*2La 22a

LaCp*2

xs THF toluene

Cp*2La 20aTHF

C CPh O

unidentified products

Ph

4.4.

Mechanism of catalytic oligomerization

4.4.1.

Catalytic dimerization

Stoichiometric reactions of the catalyst precursors Cp*2LaCH(SiMe3)2 (5D), [Cp*2La(-H)]2 (25D) and [(Cp*2La)2(-PhC4Ph)] (22a) provided experimental evidence for the formation of (E)-1,4-diphenyl-but-1en-3-yne (12a) and 2,4-diphenylbut-1-en-3-yne (11a) from the insertion of phenylacetylene into the monomeric, alkynyl derivative Cp*2LaCCPh (20a), followed by protonolysis by phenylacetylene. One major lanthanocene derivative was observed during catalytic oligomerization of phenylacetylene and this resting state of the catalyst has been identified as a monomeric, alkynyl derivative which is in equilibrium with its substrate adduct. Upon substrate depletion, the alkynyl derivative Cp*2LaCCPh (20a) undergoes dimerization and subsequent C-C coupling to afford the buatrienediyl derivative [(Cp*2La)2(-PhC4Ph)] (22a), while the butenynyl derivatives Cp*2LaC(Ph)=C(H)CCPh (24a) and Cp*2LaC(H)=C(Ph)CCPh (23a) remain unaltered in the reaction mixture. The butatrienediyl derivative [(Cp*2La)2(-PhC4Ph)] (22a) can be converted back to the catalytically active species by its protonolysis reaction with 1-alkyne. These results are consistent with and add substantial support to the catalytic cycle as proposed previously (Scheme 4-5).5b,f The observation that the preference for trans-head-to-head dimerization and the relative amount of trimerization products increase with increasing metal ionic radius of the Cp*2LnR (Ln = Lu, Y, Pr, Ce, La) precatalysts revealed the importance of steric effects in the regio- and chemoselectivity of the organolanthanidecatalyzed 1-alkyne oligomerization (Figure 4-1). This behavior can be rationalized, when the insertion of alkyne into the Ln-C bond of the monomeric, alkynyl derivative Cp*2LnCCPh (20) is taken into account. Substrate insertion into the catalyst may take place in a 2,1- (route ii, Scheme 4-5) and a 1,2-manner (route iv), providing

120

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

Scheme 4-5. The proposed catalytic cycle for the lanthanocene-catalyzed dimerization of phenylacetylene.
11a Ph Ph Ph Cp*2La Ph Ph Cp*2LaC CPh Ph viii Cp*2LaC CPh HC CPh x Ph v iv Ph iii ii 20a Ph 12a Ph Ph xii vi 2,1 23a ix Ph Cp*2LaR i RH 12a Ph vii Ph Ph Cp*2La Ph 24a

Ph Ph

Cp*2La H 1,2

Cp*2La Ph

Ph H

Ph

Cp*2LaC CPh PhC CH

Ph Cp*2La 21a Ph LaCp*2 xi

Ph LaCp*2 Cp*2La Ph 22a

12a and 11a, respectively, after protonolysis of the formed butenynyl derivative 24a and 23a, respectively, by 1alkyne. Insertion most likely takes place via a concerted, four-centered transition state in which a partial positive charge develops at the -carbon of the incipient metal alkenyl bond (Scheme 4-6). Indirect evidence for this hypothesis comes from the electronic effects of -H elimination from lanthanide alkyls32c (the microscopic reverse of alkene insertion into a Ln-H bond), alkene/alkyne insertion into Ln-N bonds39 and alkene insertion into transition metal hydrogen bonds40a-c and actinide hydrogen bonds.40d The observed regiochemistry in which insertion takes place to orient the metal center at the sterically most hindered carbon has been observed in many reactions of electrophilic metal complexes with 1-arylalk-ynes and 1-arylalk-1-enes.41,42 Marks et al. invoked an aryl-directed process in which the aromatic moiety serves as a Lewis base that interacts with the Lewis acidic metal center in the transition state to account for the observed regioselectivity.42a Although there is substantial precedent for n-benzylic structures in organo-f-element chemistry,43 the observed kinetic preference can also be explained by the electronic demands of the alkyne insertion in which an electron-withdrawing group at the position and an electron-donating group at the -position of the four-centered transition state are expected to accelerate the insertion reaction (Scheme 4-6).44 Scheme 4-6. The polarization of the transition state of 1-alkyne insertion into the Ln-C bond of an alkynyl derivative. R''' + Ln
+

R'

R''

The observation that the tendency for head-to-tail (11) formation increases in the rare-earth metalcatalyzed 1-alkyne oligomerization by decreasing the metal ion radius in the (pre)catalysts Cp*2Ln can thus be interpreted as the result of an increase in the steric interaction between the ancillary ligand system and the

121

Chapter 4

Scheme 4-7. Proposed pathway for the formation of the observed trimers.
R Cp*2Ln 24 R R R Cp2*Ln R R R H Cp*2Ln R H 1,3-H R R H H Cp*2Ln R R R + - 20 R 15 R R . R 16 R R R - 20 R R R

coordinated substrate. It seems that the electronically preferred 2,1-insertion for 1-arylalk-1-enes and 1-arylalk1-ynes, as observed for the larger metals, is outweighted by the sterically preferred 1,2-insertion for the smaller metals in the (pre)catalysts Cp*2Ln.45 This picture is supported by the observation that the preference for 1,2insertion (11a:12a = 93.7:5.2) changes into a preference for 2,1-insertion (11a:12a = 0.0:71.9), when the coordination space around the yttrium metal center in Cp*2YCH(SiMe3)2 is increased as in Me2SiCp2YCH(SiMe3)2 (Table 4-2).

4.4.2.

Catalytic trimerization

The factors determining the formation and nature of higher oligomers was previously not understood (Section 4.2.3). Insertion of the formed enynes into the alkynyl species 20 was originally proposed to rationalize the formation of the observed trimers.5g The present study does not support this proposal, however. A plot of the concentration of the trimers versus time during 1-alkyne oligomerization shows that the rate of trimer formation does not increase with increasing enyne concentration.46 Moreover, the kinetic concentration profile of the products formed during alkyne oligomerization and the observed increase of trimerization with increasing substrate concentration suggest that the trimers are primary products orginating from insertion of substrate into an intermediate of the dimerization reaction. This hypothesis is supported by the observed increase in trimerization relative to dimerization upon opening the metal coordination sphere of the catalyst (i.e. larger metal ions in the precatalyst Cp*2LnR or ligand substitution Cp*2Ln Me2SiCp2Ln for a given metal Ln, Section 4.2.2). A more open coordination sphere of the catalyst increases the rate of trimerization relative to that of dimerization, because the steric requirements for protonolysis by 1-alkyne are less demanding than that of insertion of 1-alkyne. Linear trimerization of 1-alkynes has been reported predominantly as a side-reaction of the metalcatalyzed 1-alkyne dimerization,47 while selective catalytic 1-alkyne trimerization has been achieved only in a few cases.48 In all of these cases, the linear trimers were identified as substituted dienynes. Interestingly, dienynes were formed in the trimerization of propyne by permethyllanthanidocenes, whereas trimethylsilylacetylene yielded allenylic and acetylenic trimers similar to those observed for phenylacetylene.5g,f It seems therefore that the trimerization of 1-alkynes into allenic and diynic trimers represents a unique feature of the rare-earth metallocene catalysts in combination with aryl- or silyl-substituted 1-alkynes. Both the experimental conditions under which the reported dienynes have been isolated and their general reported

122

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

Scheme 4-8. The proposed reaction sequences leading to catalyst deactivation via Cp* abstraction by substrate.
Cp*2LnR' R Cp*2Ln 23 R v R iv ii Cp*Ln(C iii R R 11 R Cp*2LnC 20 CR + Ln(C CR)3 12 R i Cp*2LnC 20 R + - HR' R vi Cp*2Ln R vii + Cp*H R R

CR

Cp*2LnC 20

CR

CR)2

Cp*2LnC 20

CR

+
R

reactivity suggest that substituted dienynes should be stable under the reaction conditions of the present 1-alkyne oligomerization reactions.49,50 Also, in situ 1H NMR experiments do not provide evidence for intermediates in the formation of the trimers, but suggest that the trimers formed are primary products. These considerations led to the proposal of an interaction between the initially formed hexa-1,3-dien-5-yn-1-yl ligand with the lanthanide metal center, affording the observed mixtures of allenic and diynic trimers (Scheme 4-7). Insertion of substrate into the alkenyl species 24 most likely proceeds in a 2,1-manner for aromatic and silyl-substituted 1-alkynes (Section 4.4.1). Instead of protonolysis by 1-alkyne (forming a dienyne) or intramolecular alkyne insertion (yielding a fulvene species51), a 1,3-hydrogen shift is proposed to take place via an unknown mechanism, affording a (thermodynamically favorable) 3-propargyl/allenyl derivative that undergoes protonolysis by phenylacetylene to yield the observed trimers. Protonolysis of 3-propargyl/allenyl lanthanocene derivatives by phenylacetylene is known to produce mixtures of alkynes and allenes (Chapter 2). Although 1,3-hydrogen shifts are symmetry-forbidden,52 they are commonly observed for allenylic, propargylic and acetylenic lithium compounds.53 Further study is needed to allow a more detailed discussion of this rearrangement which is presently only tentative. However, the proposed reaction sequences do reconcile the plausible alkyne insertion into the alkenyl species 24 with the formation of the observed allenic and diynic trimers.

4.4.3.

Catalyst deactivation

Catalyst deactivation repesents another aspect of the permethyllanthanidocene-catalyzed oligomerization of phenylacetylene worth of consideration. Although the cyclopentadienyl-metal -bond is generally considered to be very stable, metallocene complexes of rare-earth metals undergo protonolytic reactions forming cyclopentadiene in the presence of compounds with acidic protons (e.g. HCl, H2O, ROH, HCN, etc.).54 -Bond cleavage has been observed in the reactions of bis(cyclopentadienyl)lanthanide alkynyl derivatives Cp2LnCCR with a five-fold molar excess of water, methanol and tert-butanol.54f The amount of CpH abstraction was found to increase with the acidity of the reagent. -Bond cleavage has also been reported for permethylmetallocenes of rare-earth metals in the presence of tert-butanol55, acetonitrile56 and amines.57,58 The observed partial Cp*H abstraction in the present oligomerization reactions is therefore not entirely unexpected, considering the acidity of phenylacetylene and the relatively large substrate-to-catalyst molar ratios.59,60 The absence of similar observations in transition-metal and actinide61 chemistry is most likely related to the more polar nature of the rare-earth metal carbon bonds.62,63 Indeed, cyclopentadienyl metal compounds in which electrostatic interactions dominate the bonding between the Cp fragments and the electropositive metal, are well-known Cp transfer agents in synthetic chemistry.64 Additional support is provided by the chemistry of

123

Chapter 4 bis(alkoxysilylamido) yttrium complexes, as it was believed that their highly polar character was responsible for the observed rapid ligand protonolysis and the concomitant formation of [Ln(C CR)n] in the reactions of [Me2Si(NCMe3)(OCMe3)]2YR (R = alkyl) with 1-alkynes.65 On the basis of the above considerations, catalyst deactivation in the present alkyne oligomerization reactions is believed to involve the formation of a dialkynyl species Cp*La(C CR)2 from the reaction of Cp*2LaCCPh and phenylacetylene (Scheme 4-8). In view of the known reactive nature of Cp*La(OC6H3-iPr22,6)266 and Cp*La[CH(SiMe3)2]2,67 due to electronic and coordinative unsaturation, the independent synthesis of Cp*La(C CPh)2 was deemed problematic. Indeed, Heeres et al. reported that the isolation of the bis(alkynyl) cerium complex, [Cp*Ce(C CtBu)2]n was precluded by its instability.15 Interestingly, [Cp*Ce(C CtBu)2]n was found to effect the catalytic dimerization of tert-butylacetylene, albeit at a much lower rate than the Cp*2Ln systems. It was believed that ligand redistribution took place, generating the mono(alkynyl) species Cp*2CeCCtBu with concomitant formation of [Ce(C CtBu)3]n. In the present context, it is very likely that the anticipated instability of [La(C CPh)3]n renders the ligand rearrangement reaction of the bis(alkynyl) species Cp*La(C CPh)2 into the mono(alkynyl) species Cp*2LaCCPh irreversible.67,68

4.4.4.

The mechanistic interpretation of the observed kinetic behavior

The time-dependence of substrate conversion during the lanthanocene-catalyzed oligomerization of phenylacetylene was found to be complex under the studied reaction conditions (Section 4.2.5). At relatively low substrate concentration, substrate conversion displayed zero-order rate dependence on substrate concentration. The reaction products revealed the occurrence of catalytic dimerization under these conditions, accompanied by both minor catalytic trimerization and catalyst deactivation. Upon increasing the substrate concentration, the kinetic behavior changed and became increasingly more first-order dependent on substrate concentration (Figure 4-4). Concomitantly, the rate of catalytic trimerization and catalyst deactivation increased relative to catalytic dimerization. Even so, catalytic dimerization remained the dominant reactive pathway during substrate conversion at relatively high substrate concentration under the studied reaction conditions. In addition, experiments demonstrated that the reaction products alter the kinetic behavior of the lanthanocene-catalyzed oligomerization of phenylacetylene as well. The presence of reaction products resulted in a decrease of the reaction rate and qualitatively changed the reaction order in substrate concentration from zero-order towards first-order (Figure 4-7). These results led to the proposal that the rate dependence on substrate concentration of the present catalytic oligomerization reaction can be described by concurrent zero-order catalytic dimerization and firstorder catalytic trimerization, accompanied by catalyst deactivation and product inhibition. At relatively low substrate concentration, catalytic dimerization dominates the kinetic behavior of the oligomerization reaction and zero-order rate dependence on substrate concentration is observed. At relatively high substrate concentration, catalytic trimerization and/or product inhibition dominate the kinetic behavior of the oligomerization reaction and first-order rate dependence on substrate concentration is observed. The origin of first-order rate dependence on substrate concentration may be ascribed to the increasing importance of catalytic trimerization at relatively high substrate concentration, the increasing importance of product inhibition at relatively high substrate concentration or the increasing importance of both catalytic trimerization and product inhibition at relatively high substrate concentration. Unfortunately, the present data do not allow discrimination between these three kinetic models. As the rate dependence on substrate concentration was seemingly unaffected by catalyst deactivation under reaction conditions where catalytic dimerization dominated the kinetic behavior, the rate of catalyst deactivation may plausibly be zero-order dependent on substrate concentration as well. These considerations imply that both catalytic dimerization and catalyst deactivation take place via a rate-limiting intramolecular reaction step, whereas catalytic trimerization proceeds via a rate-limiting intermolecular reaction step. Based on the present observations and implications, the mechanism of oligomerization is proposed to take place according to the reaction sequences depicted in Scheme 4-9. The first step, protonolysis of the metal alkyl (5), hydride (25) or butatrienediyl (22) derivative to generate the catalytically active, monomeric alkynyl derivative is rapid and irreversible, as indicated by 1H NMR experiments. The following step takes into account the propensity of the monomeric, alkynyl derivative Cp*2LnCCPh to bind Lewis bases, as observed for Cp*2LaCCPh(THF) (Chapter 5), and involves a rapid equilibrium between the phenylacetylene adduct Cp*2LnCCPhPhCCH and its base-free form.69 The monomeric,

124

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes Scheme 4-9. Proposed mechanistic scenario for the rare-earth metallocene-catalyzed oligomerization of phenylacetylene. Catalyst initiation
Cp*2LnR + PhCCH k1 Cp*2LnCCPh + RH

Cp*2LnCCPh + PhCCH

k2 k-2

Cp*2LnCCPhPhCCH

Catalyst deactivation
Cp*2LnCCPhPhCCH k3 Cp*Ln(CCPh)2 + Cp*H

Cp*Ln(CCPh)2

k4

C p2*LnCCPh +

Ln(CCPh)3

Catalyst inhibition
Cp*2LnCCPh + PhCCR' k5 k-5 Cp*2LnCCPhPhCCR'

Catalytic trans-head-to-head 1-alkyne dimerization

Ph Cp*2LnCCPhPhCCH k6 Cp*2Ln Ph
Ph Cp*2Ln Ph + PhCCH k7 Cp*2LnCCPh + Ph Ph

Catalytic head-to-tail 1-alkyne dimerization

Ph Cp*2LnCCPhPhCCH k8 Cp*2Ln Ph
Ph Cp*2Ln Ph + PhCCH k9 Cp*2LnCCPh + Ph Ph

Catalytic 1-alkyne trimerization

Ph Cp*2Ln Ph + PhCCH k10 k-10 Ph Cp*2Ln Ph Ph

Ph Cp*2Ln Ph Ph k11 Cp*2Ln Ph

Ph

Ph

125

Chapter 4
Ph Cp*2Ln Ph Ph
Ph Cp*2Ln Ph Ph Ph Cp*2Ln Ph Ph + PhCCH Ph k14 Ph Ph + Cp*2LnCCPh + PhCCH Ph k13 . Ph Ph + Cp*2LnCCPh

Ph k12 Cp*2Ln Ph Ph

alkynyl derivative Cp*2LnCCPh may also bind to the reaction products, however. The resulting competition between substrate and product for the metal center leads to catalyst inhibition and becomes increasingly important at higher substrate concentration. The adduct Cp*2LnCCPhPhCCH may either dissociate the substrate or undergo irreversible migratory insertion to yield the single-insertion products Cp*2LaC(H)=C(Ph)CCPh (23a) and Cp*2LaC(Ph)=C(H)CCPh (24a). When the but-1-en-3-yn-1-yl derivatives undergo protonolysis by phenylacetylene, the base-free, monomeric alkynyl derivative is regenerated under formation of the head-to-tail (23a) and the trans-head-to head dimer (12a). The but-1-en-3-yn-1-yl derivatives may, in principle, also be engaged in a rapid equilibrium with phenylacetylene, forming the corresponding base adduct prior to protonolysis or insertion. Contrary to insertion, however, initial base adduct formation is not required for protonolysis, whereas the reaction products reveal that only Cp*2LaC(Ph)=C(H)CCPh (24a) undergoes base adduct formation with phenylacetylene, followed by substrate insertion (vide infra). When Cp*2LaC(Ph)=C(H)CCPh undergoes insertion of phenylacetylene, a hexa-1,3-dien-5-yn-1-yl derivative is formed which is presently believed to undergo an intramolecular 1,3-H shift to yield an 3-propargyl/allenyl derivative (Section 4.4.2). Irreversible protonolysis of this 3-propargyl/allenyl derivative by phenylacetylene reforms the base-free, monomeric alkynyl species and gives rise to the observed acetylenic (15a) and allenylic (16a) trimers.

(4.12)

k1 E + S k-1 ES

kp

(4.13)

d[S] dt =

kp [ET][S] KM + [S]

Vmax [S] KM + [S]

The present findings led to the proposal that the resting state of the catalyst corresponds to the adduct Cp*2LaCCPhPhCCH (20a-2a) and that intramolecular, alkyne insertion (k6 and k8) is rate-limiting in the catalytic dimerization reaction. When the effects of catalytic trimerization (i.e. k10 and k11 is small relative to k7 and k9) and catalyst deactivation on the kinetic behavior are neglected (i.e. k3 and k4 are small relative to k6 and k8) relative to those of catalytic dimerization, the proposed mechanism of 1-alkyne oligomerization reduces to a simple scheme involving a pre-equilibrium, followed by reaction to product (eq. 4.12). This general mechanism is frequently observed in chemistry70 and enzyme kinetics.71 In a Henri-Michaelis-Menten description (eqs 4.12 and 4.13; E = enzyme, S = substrate, P = product), the Michaelis constant KM = (k-1 + kp)/k1 refers to the overall

126

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes effectiveness of substrate capture summed over all catalyst species ([ET] = [E] + [ES]) and Vmax = kp[ET] refers to the maximum reaction rate. In a simple situation, such as equation 4.12, the effects of product inhibition, the (reverse) reaction of product to substrate and catalyst deactivation are assumed to be negligable. When the substrate-binding step is rapid relative to the rate of breakdown of the substrate adduct ES (i.e. k1 > kp k-1), KM KS, where KS = k-1/k1 refers to the dissociation constant for the substrate adduct. The rate at high substrate concentrations is zero-order with respect to [S] (KM < [S]), falling until it shows a first-order dependence in the limit of low [S] (KM > [S]). In the present situation, substrate capture is slow relative to the substrate-binding step (i.e. k1 > kp) and the reaction rate is zero-order with respect to [S] (KM < [S]). When the effects of reversible, nonproductive product coordination to the catalyst are included, standard rapid equilibrium analysis yields a modified Henri-Michelis-Menten equation, which reproduces the observed change from zero-order to first-order rate dependence on substrate concentration.72 In general, reaction sequences involving pre-equilibria are conveniently analyzed in terms of initial rates. Such an analysis of the present catalytic system is seriously hampered by the present methodology (i.e. long relaxation times of the acetylenic proton in combination with rapid reaction times at practically meaningful catalyst concentrations) and the catalytic behavior of the present oligomerization reaction, however (i.e. concurrent catalyst deactivation, catalytic trimerization and product inhibition). Alternatively, it is possible to derive values for the catalytic reaction parameters kp and KM from linear ln([S]0/[S]t) versus time plots, if catalyst deactivation, product inhibition and the (reverse) reaction of product to substrate are practically absent. Obviously, the first two criteria are not met in the present catalytic system. Even so, an estimate of kp can be obtained from the intercept of the observed linear ln([S]0/[S]t) versus time plots (e.g. kp = 1.09(4) 103 s-1 from Figure 4-5). The value of kp represents the maximum turnover frequency of the present catalyst. Under the assumption that the proposed reaction sequences leading to catalyst deactivation can be modeled by the irreversible formation of a catalytically inactive complex E* from ES (i.e. k3 large relative to k4), a simplified kinetic scheme for catalyst deactivation can be written as shown in equation 4.14. A standard steady-state approximation (i.e. kd smaller than kp, k-1 and k1, hence d[ES]/dt 0) yields equation 4.15. This scenario implies that rate dependence of catalyst deactivation (d[Cp*H]/dt = kd[ES]) on substrate concentration parallels that of catalytic dimerization (d[P]/dt = kp[ES]), because the rate of both processes is determined only by the corresponding reaction constants and the concentration of the substrate adduct ES. From this, it follows that the ratio of the corresponding reaction products remains constant throughout the course of reaction and that the ratio of the reaction constants kp and kd can, in principle, be determined from the ratio of the corresponding reaction products (eq. 4.16). The observations that the ratio ([11a]+[12a])/[Cp*H] is constant within experimental error (vide infra) and that the oligomerization reaction displays zero-order rate dependence under reaction conditions where catalytic dimerization dominates the kinetic behavior (vide supra) support the proposed model for catalyst deactivation.73

(4.14)

k1 E + S k-1 ES kp E + P

kd

E* + Cp*H

(4.15)

d[Cp*H] dt = kd [ES] =

kd KM+[S] [ET][S]

(4.16)

kp kd =

d[P]dt d[Cp*H]dt =

[P] [Cp*H]

The occurrence of concurrent catalytic trimerization and the experimental error in the final concentration of Cp*H do not allow an accurate determination of kp/kd, however (e.g. kp/kd = 3(1) 103, based on selected data presented in Table 4-3). Even so, the present data reveal that the ratio ([11a]+[12a])/[Cp*H] is

127

Chapter 4

Scheme 4-10. The proposed mechanism of the permethyllanthanocene-catalyzed oligomerization of phenylacetylene.

Cp'2LnC RC CR' Cp'2Ln R R' R R' R LnCp'2 Cp'2Ln R R R Cp'2LnC R CR R R R R R . R R Cp'2Ln R Cp'H R Cp'Ln(C CR)2 RC Cp'2Ln R R R CH R Cp'2Ln R R Cp'2LnC RC CH Cp'2Ln R R R Cp'2Ln CR R + R R R R R R CR R R LnCp'2

R R

constant within experimental error and suggest that intramolecular substrate insertion in 20a2a is ~3000 more rapid than intramolecular protonolysis of 20a2b leading to catalyst deactivation. The reaction products of the catalytic oligomerization reaction revealed that catalytic trimerization becomes more competitive with catalytic dimerization upon increasing the substrate concentration at constant catalyst concentration. This behavior is in accord with the observed change in kinetic behavior and the observed rate dependence on substrate concentration in the proposed mechanism where catalytic dimerization and trimerization are zero- and first-order rate dependent on substrate concentration, respectively. The proposed firstorder rate dependence of catalytic trimerization implies a rate-limiting intermolecular reaction with substrate. Both substrate insertion into Cp*2LaC(Ph)=C(H)CCPh (24a) (k11) and protonolysis of the 3-propargyl/allenyl intermediate by substrate (k13 and k14) rationalize this behavior. No Cp* 1H NMR resonances corresponding to 24 were observed by in situ 1H NMR spectroscopy during catalytic substrate conversion of phenylacetylene under reaction conditions where significant catalytic trimerization takes place (Section 4.2.6). As a result, protonolysis of the 3-propargyl/allenyl intermediate is presently favored as the rate-limiting reaction step in the catalytic trimerization reaction. Even though a detailed quantitative analysis of the lanthanocene-catalyzed oligomerization reaction of phenylacetylene is precluded by the present methodology and the occurrence of several concurrent processes, the proposed mechanism of the permethyllanthanocene-catalyzed oligomerization reaction of phenylacetylene is supported by stoichiometric reactions of catalyst precursors, the identification of reaction intermediates and a plausible mechanistic interpretation of the observed kinetic behavior (Figure 4-10).

4.5.

Conclusions

The present study demonstrated that the permethyllanthanocene-catalyzed oligomerization reaction of phenylacetylene proceeds via well-established elementary reactions, such as insertion of alkyne into a metalcarbon -bond and protonolysis of a metal-carbon -bond by the acetylenic proton of the 1-alkyne. The

128

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes mechanism and the kinetics of the oligomerization reaction were found to depend strongly on catalyst structure, i.e. the metal ion size and the ancillary ligation (Cp*2Ln versus Me2Cp2Ln). The precatalyst Cp*2LaCH(SiMe3)2 performed best in catalysis, representing one of the most active and selective (pre)catalysts for the trans-head-tohead dimerization of phenylacetylene reported in literature. On the basis of a kinetic study and the identification of the reaction intermediates, a plausible mechanistic scenario was proposed for the lanthanocene-catalyzed oligomerization reaction. This allows for a more comprehensive understanding of substrate effects to be discussed in the following chapter.

4.6.

Experimental section

General considerations. For general considerations and physical and analytic measurements, see Sections 2.7 and 3.5. Phenylacetylene (Merck) was distilled from CaH2, passed through a plug of activated neutral alumina and stored at -40 C under nitrogen. Typical NMR-Scale Catalytic 1-Alkyne Oligomerization Reactions Mediated by Cp*2LnCH(SiMe3)2 (Ln = La, Y) and Me2SiCp2LnCH(SiMe3)2 (Ln = Ce, Y). (a) General procedure. A catalyst stock solution was prepared in benzene-d6. The amount of (pre)catalyst was weighted, while the volume of the solvent was determined using of volumetric glassware. In cases of relatively high substrate-to-catalyst ratios, a specified amount of cyclooctane was added with a microsyringe to the catalyst solution to serve as an internal standard. The catalyst stock solution was transferred into a vial with screw-cap and stored after use at -40C in the glovebox. The substrate was added with a microsyringe to the catalyst solution in the NMR tube. After substrate addition, the sample was inserted within 5 min (at room temperature) into the probe of the spectrometer and the reaction was followed by single-pulse, in situ 1H NMR spectroscopy, using appropriate long pulse delays (at least 300 s for the acetylenic proton of the substrate) to avoid signal saturation under anaerobic conditions. As soon as the substrate was consumed, the reaction mixture was analyzed with quantitative 1H NMR spectroscopy (appropriate long pulse delays and long experiment times so as to obtain reliable proton intensities and signal-to-noise ratios, respectively). Finally, the reaction mixture was quenched with methanol-d4, methanol, H2O or D2O and the final organic products were identified by 1H, 13C, 13C{1H} and 2D NMR, GC, GC-MS and high-resolution mass spectroscopy. In most cases, the products were characterized in situ, but in some cases the major product could be isolated by performing several purification steps (i.e. filtration over a plug of neutral alumina using hexanes as eluent to remove inorganic solids, evaporation of volatiles, sublimation of dimers, fractional crystallization, vide infra). (b) Phenylacetylene (2a). As described above, 100.0 L (910.5 mol, 1071 equiv.) of phenylacetylene was converted by Cp*2LaCH(SiMe3)2 (0.85 mol in 500.0 L of benzene-d6) into a mixture of 2,4-diphenylbut-1-en-3-yne (11a), trans-1,4-diphenylbut-1-en-3-yne (12a), 1,3,6-triphenyl-1,5-hexadiyne (15a) and 1,3,6-triphenylhexa-1,2-5-yne (16a). Scheme 4-11. Numbering scheme of 2,4-diphenylbut-1-en-3-yne (11a).
7 6 5 4 3 10 11
1

A
1 2 9

B F G
12

2,4-Diphenyl-1-en-3-yne (11a): H NMR (500 MHz, C6D6, 25 C) 5.69 (d, 2JHH = 1.0 Hz, B), 5.75 (d, 2JHH = 1.0 Hz, A), 6.97 (m, H + D), 7.09 (t, 3JHH = 7.6 Hz, E), 7.15 (t, 3JHH = 7.7 Hz, G), 7.43 (d, 3JHH = 7.3 Hz, F), 7.72 (d, 3JHH = 7.2 Hz, C). 13C NMR (125.7 MHz, C6D6, 25 C): 89.33 (dd, 3JCH = 8.4 Hz, 3JCH = 14.8 Hz, 3), 91.51 (t, 4JCH = 5.2 Hz, 4), 120.80 (dd, 1JCH = 159.4 Hz, 1JCH = 163.0 Hz, 1), 123.63 (s, 9), 126.57 (dt, 1 JCH = 158.5 Hz, 3JCH = 6.9 Hz, 6), 128.53 (dd, 1JCH = 154.3 Hz, 3JCH = 7.1 Hz, 11), 128.59 (8 + 12), 128.70 (dd, 1 JCH = 160.2 Hz, 3JCH = 7.4 Hz, 7), 131.29 (s, 2), 131.98 (dt, 1JCH = 150.2 Hz, 3JCH = 8.6 Hz , 10), 137.78 (s, 9). 1 H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A1, B1, C6, D7, E8, F10, G11, H12. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A2,3,9, B2,3,9, C2,6,8, D5,6,8, E6, F10,12, G9,11,12, H10. GC-MS, m/z (relative intensity): 204 (M+; 100), 203 (M+ - H; 97), 202 (M+ - 2H; 97), 201 (14), 200 (12), 102 (10), 101 (23), 89 (9).

129

Chapter 4 (E)-1,4-Diphenylbut-1-en-3-yne (12a): 1H NMR (500 MHz, C6D6, 25 C): 6.29 (d, 2JHH = 16.3 Hz, B), 6.97 (d, 2JHH = 16.3 Hz, A), 6.99-7.00 (m, D + G + H), 7.08 (d, 3JHH = 7.1 Hz, F), 7.14 (E), 7.51 (d, 3JHH = 7.5 Hz, C). 13C{1H} NMR (125.7 MHz, C6D6, 25 C): 89.73 (d, 2JCH = 8.5 Hz, 3), 92.47 (q, 3JCH = 5.1 Hz, 4), 108.55 (dd, 1JCH = 160.9 Hz, 2JCH = 2.2 Hz, 1), 124.15 (t, 2JCH = 7.9 Hz, 5), 126.57 (ddd, 1JCH = 158.2 Hz, 3JCH = 11.7 Hz, 2JCH = 6.0 Hz, 6), 128.31 (dt, 1JCH = 161.1 Hz, 3JCH = 7.7 Hz, 8), 128.67 (dt, 1JCH = 160.5 Hz, 3JCH = 7.6 Hz, 12), 128.71 (11/7), 128.89 (11/7), 131.80 (dd, 1JCH = 161.7 Hz, 3JCH = 6.5 Hz, 2), 136.58 (q, 2JCH = 6.4 Hz, 3 JCH = 6.4 Hz, 9), 141.63 (dq, 1JCH = 156.6 Hz,, 2JCH = 4.3 Hz, 3JCH = 4.3 Hz, 1). 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A1, B2, C6, D7, E8, F10, G11, H12. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A1,4,9, B3,6,10, C4,6,8, D5,7, E6, F10,12, G10,11, H10. GC-MS, m/z (relative intensity): 204 (M+; 100), 203 (M+ - H; 88), 202 (M+ - 2 H; 85), 201 (M+ - 3 H; 13), 200 (M+ - 4 H; 11), 176 (6), 126 (6), 102 (8), 101 (M+ - 2 H - C6H5C C; 18), 89 (8), 88 (6), 76 (8). HR-MS: C16H12, calc.: 204.09390, found: 204.09436. Scheme 4-12. Numbering scheme of trans-1,4-diphenylbut-1-en-3-yne (12a).
11 10 9

H 12 G F

A
3 4 5 2

D
8E

1,3,6-Triphenyl-1,5-hexadiyne (15a): 1H NMR (500 MHz, C6D6, 25 C): 2.78 (dd, 3JHH = 6.8 Hz, JHH = 16.6 Hz, B, 1 H), 2.84 (dd, 3JHH = 6.8 Hz, 2JHH = 16.6 Hz, C, 1 H), 4.01 (t, 3J HH = 6.8 Hz, A, 1 H). Aromatic proton signals obscured. 13C{1H} NMR (125.7 MHz, C6D6, 25 C): 29.65 (dt, 1JCH = 128.9 Hz, 3JCH = 6.1 Hz, 4), 38.58 (dquint., 1JCH = 134.4 Hz, nJCH = 4.6 Hz, 3), 83.58 (s, 6), 84.59 (s, 1), 87.78 (s, 5), 90.84 (s, 2), 124.06 (s, 7), 131.93 (d, 1JCH = 161.5 Hz, 10), 140.85 (s, 8). Other aromatic carbon signals obscured. 1H-1H gCOSY (500-500 MHz, C6D6, 25 C): ABC, BAC, CAB. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A3, B4, C4. 1 H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A1,2,4,5,7,8, B2,3,5-8,10, C2,3,5-8,10. GC-MS, m/z (relative intensity): 306 (M+; 6), 192 (M+; 19), 191 (M+ - C6H5C CCH2; 100), 189 (M+ - C6H5C CCH2 - 2H; 16), 165 (5), 115 (M+ - C6H5CHC6H5; 7).
2

Scheme 4-13. Numbering scheme of 1,3,6-triphenyl-1,5-hexadiyne (15a).

C B 4
9 1 2 8 3A

7 10

1,3,6-Triphenylhexa-1,2-diene-5-yne (16a): 1H NMR (500 MHz, C6D6, 25 C): 3.51 (d, 5J HH = 2.8 Hz, B, 2 H), 6.48 (t, 5J HH = 2.8 Hz, A, 1 H). Aromatic proton signals obscured. 13C{1H} NMR (75 MHz, CDCl3, 25 C): 22.34 (dt, 1J CH = 129.7 Hz, 4JCH = 3.4 Hz, 4), 99.43 (dt., 1J HH = 163.0 Hz, nJCH = 5.1 Hz, 3), 83.56 (s, 3), 87.26 (s, 5), 106.69 (s, 6), 207.27 (s, 2). Aromatic carbon signals obscured. 1H-1H gCOSY (500-500 MHz, C6D6, 25 C): AB. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A1, B4. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A2,3,5,6, B2. GC-MS, m/z (relative intensity): 306 (M+; 100), 305 (68), 304 (15), 303 (22), 302 (22), Scheme 4-14. Numbering scheme of 1,3,6-triphenylhexa-1,2-diene-5-yne (16a).

B
8

4 3

6 7

. A
1 2 9

130

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes 291 (28), 290 (23), 289 (41), 229 (85), 228 (84), 227 (28), 226 (43), 215 (50), 203 (26), 202 (38), 191 (26), 189 (M+ - C6H5C CCH2 - 2H; 25), 151 (20), 145 (22), 139 (19), 115 (M+ - C6H5CHC6H5; 13), 28 (35). Kinetic Studies of the catalytic oligomerization reactions. A catalyst stock solution was prepared by weighing the amount of precatalyst and dissolving the solid in a specified volume of benzene-d6, as determined by volumetric glassware. After preparation, the catalyst solution was transferred into a pre-weighted vial with screw-cap and weighted. For experiments with relatively high molar substrate-to-catalyst ratios, a specified amount of internal standard (cyclooctane) was also added with a microsyringe. After use, the catalyst stock solution was stored at -40 C in the glovebox. A 1H NMR experiment of the sample containing the catalyst solution (prior to substrate addition) ensured the presence of the prerequisite amount of catalyst after long-term storage or handling. In a typical experiment, an NMR tube was charged with 500.0 L of a catalyst stock solution using a 500.0-L microsyringe. The volume of substrate needed for the kinetic experiment was calculated from the density (determined experimentally by weighing a specified volume of substrate by the use of a microsyringe) and this amount was added with a microsyringe to the catalyst solution. The time period between substrate addition and the start of an arrayed NMR experiment was measured for each experiment and found to be ususally in the range of 6-10 min. The time needed to transfer the sample tube from the glove box into the probe of the Inova-500 or Unity-400 spectrometer was ~5 min after substrate addition. Prior to sample insertion, the probe had been set to the appropriate temperature (T 0.2 C; checked with a methanol temperature standard). Data were acquired using one scan per time interval. Long time intervals (at least 300 s) were used to avoid signal saturation under anaerobic conditions. In most cases, the reaction kinetics were monitored from the intensity changes in the substrate resonance (the acetylenic proton) and in the product resonances over 3 or more half-lives on the basis of acetylene consumption. For experiments involving relatively low substrate-to-catalyst ratios, the substrate concentration was measured from the normalized integral of the acetylenic proton relative to that of CH2(SiMe3)2. The CH2(SiMe3)2 is present as a result of rapid and quantitative protonolytic ligand cleavage during catalyst generation. For experiments, involving (i) relatively high substrate-to-catalyst ratios, (ii) hydride and (iii) butatrienediyl catalyst precursors, the substrate concentration was measured from normalization of the acetylenic substrate signal against that of internal cyclooctane. The reproducibility of kinetic data, using different batches of substrate and catalyst stock solutions, was within 5%. The reaction of MeSi(C5Me4)2CeCH(SiMe3)2 with phenylacetylene. Phenylacetylene (39.6 L, 418 mol, 25 equiv.) was added with a microsyringe to an NMR tube containing MeSi(C5Me4)2CeCH(SiMe3)2 (10.0 mg, 16.7 mol) in benzene-d6. Immediately, precipitation took place and 1H NMR spectroscopy indicated the sole presence of phenylacetylene in solution. The suspension was evaporated to dryness and suspended in benzene-d6. D2O (2.0 L, 0.1 mmol) was added with a microsyringe. The resulting solution was characterized with GC/GC-MS and NMR spectroscopy and shown to consist of phenylacetylene-d1 and MeSi(C5Me4H)2. D2O and THF quenching experiment of the Cp*2LaCH(SiMe3)2-catalyzed phenylacetylene oligomerization reaction. Phenylacetylene (167.0 L, 1758 mol, 50 equiv.) was added with a microsyringe to a stirred benzene solution (2.0 mL) of Cp*2LaCH(SiMe3)2 (20.0 mg, 35.2 mol) in a Schlenk vessel. Within several seconds, THF (143.0 L, 1760 mol) was added with a microsyringe. The reaction mixture was evaporated to dryness and the remaining solid was dissolved in benzene-d6. The major organometallic species present in solution was shown to be Cp*2LaCCPh(THF) by comparison of NMR data with an authentic sample (vide infra). D2O (2.0 L, 0.1 mmol) was added with a microsyringe. The resulting solution was characterized with GC/GC-MS and NMR spectroscopy and shown to consist of 1-alkyne oligomers (11a, 12, 15a and 16a), THF, phenylacetylene-d1 and Cp*H. When THF (3.6 L, 44 mol, 5 equiv.) was added within several seconds after addition of phenylacetylene (48.3 L, 440 mol, 50 equiv.) to a benzene-d6 solution of Cp*2LaCH(SiMe3)2 (5.0 mg, 8.8 mol) in a Teflon-capped NMR tube, 1H NMR spectroscopy indicated that Cp*2LaCCPh(THF) was the only observable organometallic species in solution. The reaction mixture was evaporated to dryness to remove unreacted substrate and the formed solid was dissolved in benzene-d6. GC/GC-MS and NMR analysis after addition of D2O gave identical results as above. Typical NMR tube reaction of Cp*2LaCH(SiMe3)2 (5D) with phenylacetylene. Phenylacetylene (3.86 L, 35.2 mol) was added with a microsyringe to an NMR tube containing Cp*2LaCH(SiMe3)2 (20.0 mg, 35.2 mol) in benzene-d6. The resulting solution was characterized by NMR spectroscopy and shown to consist of [(Cp*2La)2(-PhC4Ph)] (22a), Cp*2LaC(Ph)=C(H)CCPh (24a), Cp*2LaC(H)=C(Ph)CCPh (23a), 2,4diphenylbut-1-en-3-yne (11a) and trans-1,4-diphenylbut-1-en-3-yne (12a). After addition of D2O, GC/GC-MS and NMR analysis indicated the presence of 2,4-diphenylbut-1-en-3-yne-1-d1 (11a-1-d1) and trans-1,4-

131

Chapter 4 diphenylbut-1-en-3-yne-1-d1 (12a-1-d1), 2,4-diphenylbut-1-en-3-yne (11a) and trans-1,4-diphenylbut-1-en-3-yne (12a), Cp*H and trans-1,4-diphenylbutatriene (27a), by comparison with reported spectral data.36a-c Cp*2LaC(H)=C(Ph)CCPh (23a): 1H NMR (1H NMR (500 MHz, C6D6, 25 C): 1.93 (s, C5Me5, 30 H), 6.69 (s, CH=, 1 H) ppm. Aromatic resonances obscured due to overlap. 13C NMR (125.7 MHz, C6D6, 25 C): 11.27 (Cp*), 120.85 (CH=), 183 (LaC) ppm. The remaining 13C resonances were obscured by overlapping resonances. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): 1.93 11.27, 6.69 120.85. Reaction with D2O led to the formation of 2,4-diphenylbut-1-en-3-yne-1-d1 (11a-1-d1), as indicated by NMR and GC-MS analysis by comparison with spectral data of 11a. Cp*2LaC(Ph)=C(H)CCPh (24a): 1H NMR (1H NMR (500 MHz, C6D6, 25 C): 1.97 (s, C5Me5, 30 H), 7.74 (s, CH=, 1 H) ppm. Aromatic resonances obscured due to overlap. 13C NMR (125.7 MHz, C6D6, 25 C): 11.27 (Cp*), 132.55 (CH=), 190 (LaC) ppm. The remaining 13C resonances were obscured by overlapping resonances. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): 1.93 11.27, 7.74 132.55. Reaction with D2O led to the formation of trans-1,4-diphenylbut-1-en-3-yne-1-d1 (12a-1-d1), as indicated by NMR and GC-MS analysis by comparison with spectral data of 12a. Typical NMR tube reaction of [Cp*2La(-H)]2 (25D) with phenylacetylene. Phenylacetylene (1.34 L, 12.2 mol) was added with a microsyringe to an NMR tube containing [Cp*2La(-H)]2 (5.0 mg, 6.1 mol) in benzene-d6. The resulting solution was characterized by NMR spectroscopy and shown to consist only of [(Cp*2La)2(-PhC4Ph)] (22a) (vide infra). After addition of D2O GC/GC-MS and NMR analysis indicated the presence of Cp*H and trans-1,4-diphenylbutatriene-1,4-d2 (by comparison with trans-1,4-diphenylbutatriene). Preparative-scale phenylacetylene oligomerization reaction catalyzed by Cp*2LaCH(SiMe3)2. Phenylacetylene (132.0 L, 1.20 mmol, 68 equiv.) was added to a stirred hexane solution of Cp*2LaCH(SiMe3)2 (10.0 mg, 17.6 mol) in hexane (5.0 mL) in a Schlenk vessel. After stirring for 2 h at room temperature, the reaction was quenched by exposing the reaction mixture to air. The crude product mixture was filtered through a plug of neutral alumina (hexanes as eluent) to remove inorganic residues. After solvent removal by rotatory evaporation, trans-1,4-diphenylbut-1-en-3-yne (12a) was isolated as a white solid by vacuum sublimation (80 C, 1 mmHg). Yield: 83.0 mg (68%). IR (KBr, [cm-1]): 3027 (m), 2921 (m), 2781 (w), 2712 (w), 2358 (w), 1952 (1), 1880 (w), 1615 (m), 1591 (m), 1485 (s), 1438 (m), 1280 (m), 1068 (m), 1023 (m), 949 (s), 748 (s), 689 (s), 534 (m). Anal. Calcd. for C16H12 (204.27): C, 94.08%; H, 5.92%. Found: C, 93.85%; H, 5.95%. Catalytic oligomerization of phenylacetylene by Cp*2LaCH(SiMe3)2, followed by catalytic hydrogenation. Cp*2LaCH(SiMe3)2 (3.6 mg, 6.3 mol) was dissolved in benzene-d6 (0.7 mL) and transferred to a Schlenk vessel. The reaction mixture was stirred for 1 h at room temperature after addition of phenylacetylene (500 L, 4.55 mmol, 722 equiv.). Trimethylsilylchloride (3.0 L, 24 mol) was added and the reaction was stirred for 4 h at 50 C. Then, the reaction mixture was subjected to catalytic hydrogenation (room temperature, 12 h, 4 atm. of H2) using Pd/C (~10 mg). After hydrogenation, the mixture was filtered through a plug of neutral alumina and analyzed with NMR spectroscopy and GC/GC-MS. The presence of 1,3-diphenylbutane, 1,4diphenylbutane and 1,3,6-triphenylhexane was indicated by comparison of the spectral data (1H/13C NMR, MS) with reported data.74 NMR tube reaction of [(Cp*2La)2(-3:3-PhC4Ph)] (22a) with excess H2O. H2O (2.0 L, 0.1 mmol) was added with a microsyringe to an NMR tube containing [(Cp*2La)2(-3:3-PhC4Ph)] (5.0 mg, 4.9 mol) in benzene-d6. An immediate color change from red to colorless was observed. NMR spectroscopy and GC/GC-MS analyis indicated the presence of trans-1,4-diphenylbutatriene (27a) and Cp*H. The presence of trans-1,4-diphenylbutatriene was established by comparison of the spectral data (1H/13C NMR, MS) with reported data.36 NMR tube reaction of [(Cp*2La)2(-3:3-PhC4Ph)] (22a) with 2,6-di-tert-butyl-4-methylphenol. 2,6-Di-tert-butyl-4-methylphenol (1.3 mg, 5.9 mol) was added to an NMR tube containing [(Cp*2La)2(-3:3PhC4Ph)] (6.0 mg, 5.9 mol) in benzene-d6. An immediate color change from red to colorless was observed. NMR spectroscopy indicated the clean formation of Cp*2LaOC6H2tBu2-2,6-Me-4 (by comparison with reported NMR data37), accompanied by the formation of trans-1,4-diphenylbutatriene (27a), 2,4-diphenylbut-1-en-3-yne (11a), trans-1,4-diphenylbut-1-en-3-yne-1 (12a) and cis-1,4-diphenylbut-1-en-3-yne (13a). After addition of H2O (100 L, 0.91 mmol), NMR and GC/GC-MS analysis indicated the presence of 11a, 12a, 13a, 27a, 2,6-ditert-butyl-4-methylphenol and Cp*H. NMR tube reaction of [(Cp*2La)2(-3:3-PhC4Ph)] (22a) with THF. THF (0.50 L, 6.2 mol) was added with a microsyringe to an NMR tube containing [(Cp*2La)2(-3:3-PhC4Ph)] (6.2 mg, 6.1 mol) in benzene-d6. The reaction was followed for 16 h at room temperature, but no change was observed. Similarly, the addition of excess THF (4.5 L, 56 mol) did not result in a change in the reaction mixture as observed with 1H NMR spectropy after 1 day at room temperature. When the reaction mixture was increased to 80 C, a color

132

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes change from dark-red to light-yellow was observed after 1 h. After 11 h, 1H NMR spectroscopy indicated the presence of Cp*2LaCCPh(THF) (by comparison with an authentic sample, vide infra) and unidentified Cp* 1H NMR resonances. D2O was added to reaction mixture and GC/GC-MS analyis indicated the presence of phenylacetylene-d1, Cp*D and small amounts of unidentified compounds (e.g. m/z = 107 and 205). Low-temperature reactions of Cp*2LaCH(SiMe3)2 and [Cp*2La(-H)]2 with phenylacetylene. A stock solution was prepared by dissolving Cp*2LaCH(SiMe3)2 ( mg, mol) in toluene-d8 ( L). The solution was transferred into a pre-weighted vial with screw-cap and weighted. In a typical experiment, an NMR tube was charged with 500.0 L of the stock solution using a 500.0-L microsyringe. The NMR tube was connected to the high-vacuum line and a specified volume of substrate was condended onto the solution at -196 C. After substrate addition, the tube was sealed off in vacuo and kept at -100 C in a cooling bath of ethanol and liquid nitrogen. Prior to sample insertion, the tube was shaken and warmed quickly. When a homogeneous liquid formed, the sample was inserted into the probe of the Inova-500 spectrometer which had previously been set to 50 C (T 0.2 C; checked with a methanol temperature standard). Data were acquired using one scan per time interval. Long time intervals (at least 300 s) were used to avoid signal saturation under anaerobic conditions. The reaction kinetics were monitored from the normalized intensity changes in the Cp* resonance of the dimeric, alkynyl derivative [Cp*2La(-CCR)]2. In all cases, instantaneous formation of CH2(SiMe3)2 was observed and the intensity of its SiMe3 1H NMR resonance was used an internal reference. The reactions were performed twice and the reproducibility of the kinetic data was within 5%. Synthesis of [(Cp*2La)2(-3:3-PhC4Ph)] (22a). An aliquot of phenylacetylene (53.5 L, 487 mol) was added with a microsyringe to a stirred suspension of [Cp*2La(-H)]2 (200 mg, 244 mol) in hexane (20 mL). The pale yellow suspension turned dark red immediately. After being stirred for 1 h at room temperature, the resulting dark red solution was concentrated in vacuo and low-temperature crystallization afforded dark-red crystals. Yield: 203 mg (82%). 1 H NMR (500 MHz, C6D6, 25 C): 2.06 (s, Cp*, 30 H), 6.79 (d, 3JHH = 7.3 Hz, o-CH, 2 H), 7.02 (t, 3 JHH = 7.5 Hz, p-CH, 1 H), 7.25 (t, 3JHH = 7.5 Hz, m-CH, 2 H). 13C NMR (125.7 MHz, C6D6, 25 C): 11.49 (q, 1 JCH = 125.3 Hz, C5Me5), 120.56 (s, C5Me5), 126.84 (dt, 1JCH = 158.8 Hz, 3JCH = 7.6 Hz, p-CH), 129.30 (ddd, 1JCH = 157.8 Hz, 3JCH = 7.9 Hz, m-CH), 131.58 (dt, 1JCH = 158.8 Hz, 3JCH = 7.6 Hz, o-CH), 141.39 (s, i-C), 159.07 (s, LaCC), 217.31 (s, LaCC). 1H-1H gCOSY (500-500 MHz, C6D6, 25 C): 6.79 7.25, 7.02 7.25, 7.25 6.79-7.03. 1H NMR (500 MHz, C7D14, 25 C): 2.03 (s, C5Me5, 30 H), 6.67 (dd, 3JCH = 8.1 Hz, 4JCH = 1.3 Hz, oCH, 2 H), 7.17 (tt, 3JCH = 7.4 Hz, 4JCH = 1.2 Hz, p-CH, 1 H), 7.35 (t, 3JCH = 7.6 Hz, m-CH, 2 H). 13C NMR (125.7 MHz, C7D14, 25 C): 11.25 (q, 1JCH = 125.1 Hz, C5Me5), 120.36 (s, C5Me5), 126.39 (d, 1JCH = 126.3 Hz , p-CH), 128.8 (dd, 1JCH = 158.2 Hz, 3JCH = 7.8 Hz, m-CH), 131.45 (dt, 1JCH = 153.3 Hz, 3JCH = 7.1 Hz, o-CH), 141.40 (t, 3 JCH = 7.7 Hz, i-C), 158.86 (s, LaCC), 217.15 (s, LaCC). 1H-13C gHSQC (500-125.7 MHz, C7D14, 25 C): 2.03 11.25, 6.67 131.45, 7.17 126.39, 7.35 128.95. 1H-13C gHMBC (500-125.7 MHz, C7D14, 25 C): 2.03 120.36, 6.67 217.15-131.45-126.39, 7.17 131.45-128.95, 7.35 141.41-128.95. Anal. Calcd. for C36H70La2 (780.76): C, 55.38%; H, 9.84%. Found: C, 55.25%; H, 9.72%.

4.7.
1

References and notes

(a) Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis, 2nd ed.; John Wiley & Sons: New York, 1992. (b) Keim, W.; Behr, A.; Rper, M. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: New York, 1982; Vol. 8. (c) Winter, M. J. In The Chemistry of the Metal-Carbon Bond; Hartley, F. R., Patai, S., Eds.; John Wiley & Sons: Chichester, 1985; Vol. 3. (d) Grotjahn, D. B. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: New York, 1995; Vol. 12. (e) Trost, B. M. In Homogeneous Transition Metal Catalyzed Reactions; Moser, W. R., Slocum, D. W., Eds.; American Chemical Society: Washington, D. C., 1992. (a) Ti: Akita, H.; Yasuda, H.; Nakamura, A. Bull. Chem. Soc. Jpn 1984, 57, 480. (b) Zr: Horton, A. D. J. Chem. Soc., Chem. Commun. 1992, 185. (c) Hf: Yoshida, M.; Jordan, R. F. Organometallics 1997, 16, 4508. (a) Cr: Hagihara, N.; Tamura, M.; Yamazaki, H.; Fujiwara, M. Bull. Chem. Soc. Jpn. 1961, 34, 892. (b) Cu: Strauss, F Justus Liebigs Ann. Chem. 1905, 342, 190. (a) Ru: Katayama, H.; Ozawa, F. Coord. Chem. Rev. 2004, 102, 1731. (b) Rh: Schfer, M., Mahr, N.; Wolf, J.; Werner, H. Angew. Chem. Int. Ed. Engl. 1993, 32, 1315. (c) Ir: Jun, C.-H.; Lu, Z.; Crabtree,

3 4

133

Chapter 4

6 7

8 9

10

11

12

13

R. H. Tetrahedron Lett. 1992, 33, 7119. (d) Pd: Trost, B. M.; Chan, C.; Ruhter, G. J. Am. Chem. Soc. 1987, 109, 3486. (e) Os: Esteruelas, M. A.; Oro, L. A.; Ruiz, N. Organometallics 1994, 13, 1507. (f) Ru: Katayama, H.; Yari, H.; Tanaka, M.; Ozawa, Chem. Commun. 2005, 4336. (g) Ru: Che, X.; Xue, P.; Sung, H. H. Y.; Williams, I. D.; Peruzzini, M.; Bianchini, C.; Jia, G. Organometallics 2005, 24, 4330. (h) Ni: Ogoshi, S.; Ueta, M.; Oka, M.; Kurosawa, H. Chem. Commun. 2004, 2732. (a) Sc: Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D. ; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1987, 109, 203. (b) Y: den Haan, K. H.; Wielstra, Y.; Teuben, J. H. Organometallics 1987, 6, 2053. (c) Y: Schaverien, C. J. Organometallics 1994, 13, 69. (d) Y: Duchateau, R.; van Wee, C. T.; Meetsma, A.; Teuben, J. H. J. Am Chem. Soc. 1993, 115, 4931. (e) Y: Komeyama, K.; Takehira, K.; Takaki, K. Synlett 2004, 1062. (f) Y, Ce, La: Heeres, H. J.; Teuben, J. H. Organometallics 1991, 10, 1980. (g) Y, Ce, La: Heeres, H. J. Ph.D. Thesis, University of Groningen, 1990; Chapter 6. (h) Yb: Nishiura, M.; Hou, Z.; Wakatsuki, Y.; Yamaki, T.; Miyamoto, T. J. Am. Chem. Soc. 2003, 125, 1184. (i) Yb, Pr, Y: Nishiura, M.; Hou, Z. J. Mol. Catal. A: Chem. 2004, 213, 101. (j) La: Tazelaar, C. G. J.; Bambirra, S.; van Leusen, D.; Meetsma, A.; Hessen, B.; Teuben, J. H. Organometallics 2004, 23, 936 (a) U: Wang, J. Q.; Dash, A. K.; Berthet, J. C.; Ephritikhine, M.; Eisen, M. Organometallics 1999, 18, 2407. (b) Th, U: Straub, T.; Haskel, A.; Eisen, M. J. Am. Chem. Soc. 1995, 117, 6364. The copper-catalyzed linear dimerization of 1-alkynes is termed the Strauss coupling and has found industrial application in the production of vinylacetylene and divinylacetylene. For examples, see: (a) Nieuwland, J. A.; Calcott, W. S.; Downing, F. B.; Carter, S. A. J. Am. Chem. Soc. 1931, 53, 4197. (b) Akhtar, M.; Weedon, B. C. L. Proc. Chem. Soc. 1958, 303. (c) Akhtar, M.; Richards, T. A.; Weedon, B. C. L. J. Chem. Soc. 1959, 933. (d) Balcioglu, N.; Uraz, I.; Bozkurt, C.; Sevin, F. Polyhedron 1997, 16, 327. For a review, see: (e) Neunhoeffer, H.; Franke, W. K. In Methoden der Organische Chemie (Houben-Weyl), Georg Thieme Verlag: Stuttgart; 1972; Vol. 39, Bd. V/1d. For reviews, see: (a) Stang, P. J.; Diederich, F. Modern Acetylene Chemistry, VCH, New York; 1996. (b) Kanis, D. R.; Ratner, M. A.; Marks, T. J. Chem. Rev. 1994, 94, 195. For examples of biologically active compounds containing enyne motifs, see: (a) Trost, B. M. Science 1991, 254, 1471. (b) Nicolaou, K. C.; Dai, W. M.; Tsay, S. C.; Estevez, V. A.; Wrasidlo, W. Science 1992, 256, 1172. (c) Grissom, J. W.; Gunawardena, G. U.; Klingberg, D.; Huang, D. Tetrahedron 1996, 52, 6453. For examples of conjugated polymers containing enyne units, see: (a) Takayama, Y.; Delas, C.; Muraoka, K.; Sato, F. Org. Lett. 2003, 5, 365. (b) Takayama, Y.; Delas, C.; Muraoka, K.; Uemura, M.; Sato, F. J. Am. Chem. Soc., 2003, 126, 14163 and references therein. (c) Materials for Nonlinear Optics: Chemical Perspectives, Stucky, G. D.; Marder, S. R.; Sohn, J., Eds.; ACS Symposium Series 455; American Chemical Society: Washington, D. C.; 1991. (d) Ref. 1a. For examples of crossconjugated materials containing enyne units, see: (e) Ciulei, S. C.; Tykwinski, R. R. Org. Lett. 2000, 2, 3607. (f) Brnsted Nielsen, M.; Schreiber, M.; Baek, Y. G.; Seiler, P.; Lecomte, S.; Bouden, C.; Tykwinski, R. R.; Gisselbrecht, J.-P.; Gramlich, V.; Skinner, P. J.; Bosshard, C.; Gnter, P.; Gross, M.; Diederich, F. Chem. Eur. J. 2001, 7, 3263. (g) Tykwinski, R. R.; Zhao, Y. Synlett 2002, 1937. (h) Zhao, Y.; Cambell, K. ; Tykwinski, R. R. J. Org. Chem. 2002, 67, 336. The selective formation of 1,2,3-butatrienes from the dimerization of terminal alkynes has been rarely observed, in part due to their instability relative to the enynes. For examples, see: (a) Yi, C.S.; Liu, N. Organometallics 1996, 15, 3968. (b) Esteruelas, M. A.; Herrero, J.; Lpez, A. M.; Olivn, M. Organometallics 2001, 20, 3202. (c) Ohmura, T.; Yorozuya, S.; Yamamoto, Y.; Miyaura, N. Organometallics 2000, 19, 365. (d) Wakatsuki, Y.; Yamazaki, H.; Kumegawa, N.; Satoh, T.; Satoh, J. Y. J. Am. Chem. Soc. 1991, 113, 9604. (e) Wakatsuki, Y.; Yamazaki, H.; Kumegawa, N.; Johar, P. S. Bull. Chem. Soc. Jpn. 1993, 66, 987. Examples of catalysts that combine high activity with high selectivity for cis-1,4-diphenylbut-1-en-3yne, see (a) Ti: Ref. 2a. (b) Ti: Varga, V.; Petrusov, L.; Cejka, J.; Mach, K. J. Organomet. Chem. 1996, 509, 235. (c) Zr: Ref. 2b. (d) Al: Dash, A. K.; Eisen, M. Org. Lett. 2000, 2, 737. Other examples of catalysts that combine high activity with high selectivity for trans-1,4diphenylbut-1-en-3-yne, see (a) Ru: Yi, C.S.; Liu, N. Synlett 1999, 281. (b) Ru: Baratta, W.; Herrmann, W. A.; Rigo, P.; Schwartz, J. J. Organomet. Chem. 2000, 593-594, 489. (c) Rh: Werner, H.; Schwab, P.; Heinemann, A. Steinert, P. J. Organomet. Chem. 1995, 496, 207. (d) Pd: Yang, C.; Nolan, S. P. J. Org. Chem. 2003, 67, 591.

134

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

14

15 16

17 18

19

20

21 22

23

24

25

26

Phenylethynes substituted with iodo, bromide, methyl, and CF3 groups on para- and ortho-positions have been applied recently in the organolanthanide-catalyzed 1-alkyne dimerization reactions, see: (a) Ref. 5e. (b) Ref. 5i. (a) Heeres, H. J.; Meetsma, A.; Teuben, J. H.; Rogers, R. D. Organometallics 1989, 8, 2637. (b) Heeres, H. J. Ph. D. Thesis, University of Groningen, 1990; Chapter 8. (a) Heeres, H. J.; Nijhoff, J,; Teuben, J. H.; Rogers, J. D. Organometallics 1993, 12, 2609. (b) Evans, W. J.,; Keyer, R. A.; Ziller, J. W. Organometallics 1990, 9, 2628. (c) Evans, W. J.; Keyer, R. A.; Ziller, J. W. Organometallics 1993, 12, 2618 (d) Forsyth, C. M.; Nolan, S. P.; Stern, C. L.; Marks, T. J.; Rheingold, A. L. Organometallics 1993, 12, 3618. Ionic radii for eight-coordinate complexes: La3+ (1.160 ), Ce3+ (1.143 ), Pr (1.126 ), Sm3+ (1.079 ), Y3+ (1.019 ) and Sc3+ (0.870 ), see: Shannon, R. D Acta Crystallogr., Sect. A 1976, A32, 751. For examples, see: (a) Hydrosilylation: Fu, P.-F.; Brard, L.; Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 7157. (b) Hydrogenation and 1-alkene polymerization: Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091. (c) Hydrogenation: Jeske, G.; Schock, L. E.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8103. (d) Hydrogenation and 1-alkene polymerization: Mauermann, H.; Swepston, P. N.; Marks, T. J. J. Am. Chem. Soc. 1985, 4, 200. (e) Hydroamination: Ryu, J.-S.; Li, G. Y.; Marks, T. J. J. Am. Chem. Soc. 2004, 125, 12584. (f) Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1989, 111, 4108. (g) Hydrophosphination: Douglass, M. R.; Marks, T. J. J. Am. Chem. Soc. 2000, 122, 1824. The instability of 2,4-diphenylbut-1-en-3-yne has been reported in literature. For examples, see: (a) Strauss, F. Annalen 1905, 342, 190, (b) Carlton, L.; Read, G. J. Chem. Soc., Perkin Trans. I 1978, 1631. (c) Ohshita, J.; Furumori, K.; Matsuguchi, A.; Ishikawa, M. J. Org. Chem. 1990, 55, 3277. (d) Bianchini, C.; Innocenti, P.; Peruzzini, M.; Romerosa, A.; Zanobini, F. Organometallics 1996, 15, 272. (e) Kirss, R. U.; Ernst, R. D.; Arif, A. M. J. Organomet. Chem. 2004, 689, 419. Polysubstituted allenes and aryl-substituted allenes, in particular, are known to be thermally unstable. For examples, see (a) Jacobs, T. L.; Singer, S. J. Org. Chem. 1951, 17, 475. (b) Jacobs, T. L.; Danker, D. J. Org. Chem. 1957, 22, 1424. (c) Jacobs, T. L.; Prempree, P. J. Am. Chem. Soc. 1967, 89, 6177. (d) Pasto, D. J.; Shults, R. H.; McGrath, J. A.; Waterhouse, A. J. Org. Chem. 1978, 43, 1382 (e) Pasto, D. J.; Chou, S.-K.; Waterhouse, A.; Shults, R. H.; Hennion, G. F. J. Org. Chem. 1978, 43, 1385. Wailes, P. C.; Weigold, H.; Bell, A. P. J. Organomet. Chem. 1972, 34, 155 For examples of the use of Cp2ZrMe2 as a powerfull desiccant, see: (a) Proulx, G.; Bergman, R. G. J. Am. Chem. Soc. 1995, 117, 6382. (b) Proulx, G.; Bergman, R. G. J. Am. Chem. Soc. 1996, 15, 684. (c) Johnson, J. S.; Bergman, R. G. J. Am. Chem. Soc. 2001, 123, 2923. The concentration limit for reliable quantitative determination of the concentration of Cp*H by means of single-pulse, in situ 1H NMR spectroscopy was found to be ~0.5 mM. This implies, for example, that a precatalyst concentration of 10 mM is necessary to monitor 2.5% catalyst deactivation accurately. Under these set of reaction conditions, the addition of a maximum amount of substrate (i.e. running the reaction in pure substrate corresponds to a substrate concentration of 9.1 M) results in a substrate-to-catalyst molar ratio of 910. The degree of catalyst deactivation is ~15% under these reaction conditions (from data presented in Tables 4-3 and 4-4), thereby excluding the possibility to monitor catalyst deactivation accurately over at least two half-lives or 75 percent reaction by means of in situ 1H NMR spectroscopy. The intensity of the acetylenic proton of phenylacetylene, normalized against cyclooctane, was measured under the present reaction conditions (i.e. in benzene-d6 solution (1.5 M) under a nitrogen atmosphere) using different delay times. The data could be fitted convincingly to a first-order exponential function, revealing that 99.90(16)% of the intensity (relative to the maximum observed intensity) was observed for an experiment with a delay time of 500 s, 99.61(16)% for 400 s, 98.43(16) for 300 s and 93.72(15)% for 200 s. A similar threshold value for the concentration of precatalyst Cp2LnCH(SiMe3)2 was observed before, see: Giardello, M. A.; Conticello, V. P.; Brard, L.; Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10241. Similar behavior has also been observed for the organolanthanide-catalyzed hydroamination of amino olefins, see: Gagn, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275.

135

Chapter 4

27

28

29

30 31

32

33

34

35

The absence of d-* back-bonding renders the coordination of alkynes to d0 metals weak. As a consequence, alkyne complexes are quite rare for d0 metals. For examples of spectroscopically observed alkyne d0 metal complexes, see: (a) Ta: Curtis, M. A.; Finn, M. G.; Grimes, R. N. J. Organomet. Chem. 1998, 550, 469. (b) Zr: Stoebenau III, E. J.; Jordan, R. F. J. Am. Chem. Soc. 2003, 125, 3222. (c) Zr: Stoebenau III, E. J.; Jordan, R. F. J. Am. Chem. Soc. 2004, 126, 11170. For examples of chelated alkyne d0 metal complexes, see: (c) Temme, B.; Erker, G.; Frhlich, R.; Grehl, M. Angew. Chem., Int. Ed. Engl. 1994, 33, 1480. (d) Venne-Dunker, S.; Ahlers, W.; Erker, G.; Frhlich, R. Eur. J. Inorg. Chem. 2000, 1671. (e) Ahlers, W.; Erker, G.; Frhich, R.; Peuchert, U. J. Organomet. Chem. 1999, 578, 115. (f) Ahlers, W.; Temme, B.; Erker, G.; Frhlich, R.; Fox, T. J. Organomet. Chem. 1997, 527, 191. (g) Erker, G.; Venne-Dunker, S.; Kehr, G.; Kleigrewe, N.; Frhlich, R.; Mck-Lichtenfeld, C.; Grimme, S. Organometallics 2004, 23, 4391. (h) Burlakov, V. V.; Arndt, P.; Baumann, W.; Spannenberg, A.; Rosenthal, U. Organometallics 2004, 23, 5188. (i) Ref. 2b. (j) Ref. 2c. The reaction of [Cp*2Ce(-CCtBu)]2 with THF (1 equiv.) forms Cp*2CeCCtBuTHF over the course of 24 hours in benzene-d6 at room temperature.16a The analogous [Cp*2La(-CCPh)]2 rearranged within several hours at -50 C into [(Cp*2La)2(-PhC4Ph)] and its thermal instability precluded its isolation (Section 4.3.2). No reaction of [(Cp*2La)2(-PhC4Ph)] with THF (1 equiv.) was observed after 7 days at ambient conditions. Heating several hours to 80 C led to the formation of Cp*2LaCCPhTHF and several unidentified organometallic compounds (Section 4.3.4). The crystal structure of the unsolvated, dimeric alkynyl derivative (Cp*2SmCCtBu)2 has been reported.16c The dimeric structure in the solid state is achieved through weak Sm-Me(Cp*) interactions rather than a bridging alkynyl likange. Treatment of [Cp*2La(-CCtBu)2] with D2O has been reported to produce tBuCCD by GC-MS.16d Attempts to prepare the but-1-en-3-yn-1-yl derivatives 24a and 25a selectively by reaction of [Cp*2La(-H)]2 (26D) or Cp*2LaCH(SiMe3)2 (5D) with 2 equiv of phenylacetylene in noncoordinating solvents at both low and ambient temperatures were unsuccesfull, as mixtures containing 22a, 24a, 25a and 1-alkyne dimerization products were obtained. Analogous reactions in THF produced the corresponding base adduct Cp*2LaCCPh(THF), exclusively, and only trace amounts of 1-alkyne dimerization products were formed upon increasing the reaction temperature. Mixtures rich in 24a and 25a were, nonetheless, obtained by subjecting the above mixtures to fractional crystallization of 22a and extraction of 1-alkyne oligomers with cold hexane. It is generally accepted that the dissociation of [Cp*2Ln(-H)]2 into the reactive monomer Cp*2LnH is the first step in the reaction of [Cp*2Ln(-H)]2 with alkenes. For examples, see: (a) Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091. (b) Jeske, G.; Schock, L. E.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8103. (c) Burger, B. J.; Thompson, M. E.; Cotter, W. D.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112, 1566. (d) Evans, W. J.; DeCoster, D. M.; Greaves, J. Organometallics 1996, 15, 3210. For a detailed study of the dissociation of [Cp*2Y(-H)]2 during its reaction with 1-alkenes, see: Ref. 33. For relatively sterically less hindered terminal alkenes, such as 1-hexene, [Cp*2Y(-H)]2 is known to undergo alkene insertion at a rate faster than dimer dissociation. Kinetic studies established a twocomponent rate law involving a second-order term for direct attack on the hydride dimer and a firstorder term involving rate-determining dimer dissociation, followed by rapid alkene reaction with monomeric Cp*2YH, see: Casey, C. P.; Tunge, J. A.; Lee, T.-Y.; Carpenetti II, D. W. Organometallics 2002, 21, 389. The transformation of a metalated butatriene into a metalated enyne via a 1,3-metal shift has previously been suggested for the protonolysis reaction of [(Cp*2Ce)2(-2:2-tBuC4tBu)]16a and has been demonstrated for lithiated buatrienes, see: Neugebauer, W.; Geiger, G. A. P; Kos, A. J.; Stezowski, J. J.; Schleyer, P. R. Chem. Ber. 1985, 118, 1504. Phenylacetylene and 2-ethynylthiophene are soft Lewis bases, when interacting either with carboncarbon triple bond, the sulfur atom or the aromatic group, according to Pearsons hard-soft-acid-base (HSAB) principle. Alcohols (ROH), on the other hand, are classified as a hard Lewis. For examples, see: (a) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533. (b) Hard and Soft Acids and Bases; Pearson, R. G. (ed.); Dowden, Hutchinson and Ross, Stroudsberg, PA; 1973. (b) Fleming, I. Frontier Orbitals and Organic Chemical Reactions, Wiley-Interscience, London; 1976.

136

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

36

37 38

39

40

41

42

43

44

45

46

47

48

(a) Pollack, S. K.; Fishea, A. Macromolecules 1998, 31, 2002. (b) Ref. 11b. (c) Chow, H.-F.; Cao, X.P.; Leung, M.-K. J. Chem. Soc., Perkin Trans. 1 1995, 193. For reviews on butatrienes, see: (d) The Chemistry of Ketenes, Allenes and Related Compounds, Parts 1 and 2; Patai, S. (Ed.), Wiley, New York; 1980. (e) Schuster, H. F.; Coppola, G. M. Allenes in Organic Synthesis, Wiley, New York; 1984. (a) Heeres, H. J. Ph. D. Thesis 1990; Chapter 3. (b) Heeres, H. J.; Teuben, J. H. Recl. Trav. Chim. Pays-Bas 1990, 109, 226. Although theoretical studies indicate that the barriers to internal rotation of butatrienes are lower than those of analogous alkenes, the experimental investigation of cis-trans isomerization of substituted butatrienes is just beginning. For examples, see: (a) Kuwatani, Y.; Yamamoto, G.; Iyoda, M. Org. Lett. 2003, 5, 3371. (b) Jarowski, P. D.; Diederich, F.; Houk, K. N. J. Phys. Chem. A 2006, 110, 7237. (a) Gagn, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (b) Li, Y.; Marks, T. J. Organometallics 1996, 15, 3770. (c) Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1996, 118, 9295. (e) Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 1757. (d) Ryu, J.-S.; Li, G. Y.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 12584. (a) Burger, B. J.; Santarsiero, B. D.; Trimmer, M. S.; Bercaw, J. E. J. Am. Chem. Soc. 1988, 110, 3134. (b) Doherty, N.; Bercaw, J. E. J. Am. Chem. Soc. 1985, 107, 2670. (c) Halpern, J.; Okamoto, T.; Zakhariev, A. J. Mol. Catal. 1976, 2, 65-68 (d). Lin, Z.; Marks, T. J. J. Am. Chem. Soc. 1990, 112, 5515. (e) Guram, A. S.; Jordan, R. F. Organometallics 1990, 9, 2190. (a) Haskel, A.; Straub, T.; Dash, A. K.; Eisen, M. S. J. Am. Chem. Soc. 1999, 121, 3014. (b) Haskel, A.; Wang, J. Q.; Straub, T.; Neyroud, T. G.; Eisen, M. S. J. Am. Chem. Soc. 1999, 121, 3025. (b) Dash, A. K.; Gourevich, I.; Wang, J. Q.; Wang, J.; Kapon, M.; Eisen, M. S. Organometallics 2001, 20, 5084. (a) Fu, P.-F.; Brard, L.; Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 7157. (b) Molander, G. A.; Romero, J. A. C.; Corrette, C. P. J. Organomet. Chem. 2002, 647, 225. (c) Molander, G. A.; Schmitt, M. H. J. Org. Chem. 2000, 65, 3767. (d) Molander, G. A.; Knight, E. E. J. Org. Chem. 1998, 63, 7009. (e) Zambelli, A.; Longo, P.; Pellecchia, C.; Grassi, A. Macromolecules 1987, 20, 2035. (f) Nelson, J. E.; Bercaw, J. E.; Labinger, J. A. Organometallics 1989, 8, 2484. (g) Guram, A. S.; Jordan, R. F. Organometallics 1990, 9, 2190. (h) Zambelli, A.;Pellecchia, C.; Proto, A. Makromol. Chem., Macromol. Symp. 1995, 89, 373. (i) LaPointe, A. M.; Rix, F. C.; Brookhart, M. 1997, 119, 906. (a) Mintz, E. A.; Moloy, K. G.; Marks, T. J. J. Am. Chem. Soc. 1982, 104, 4692. (b) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1990, 112, 219. (c) Giardello, M. A.; Conticello, V. P.; Brard, L.; Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10241. For a review, see: (d) Bockarev, M. N. Chem. Rev. 2002, 102, 2089. (a) Ref. 40a. (b) Ref. 40b. (c) Guram, A. S.; Jordan, R. F. Organometallics 1990, 9, 2190. (d) Halpern, J.; Okamoto, T. Inorg. Chim. Acta 1984, 89, L53. (e) Burger, B. J.; Thompson, M. E.; Cotter, W. D.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112, 1566. Tuning the selectivity and activity of reactions by decreasing or increasing the metal ion radius is a common strategy in organometallic chemistry of rare-earth metals. For examples, see: (a) Molander, G. A.; Dowdy, E. D.; Noll, B. C. Organometallics 1998, 17, 3754. (b) Molander, G. A.; Dowdy, E. D. In Topics in Organometallic Chemistry. Kobayashi, S., Ed.; Springer: New York, 1999; Vol. 2, pp 120-154. (c) Enyne insertion into a metal-acetylide bond has been reported as a deactivation route in the catalytic 1-alkyne dimerization reaction by Cp*Ru(PPh3)C CPh, see: Yi, C. S.; Liu, N. Organometallics 1997, 16, 3910. (a) Meriwether, L. S.; Colthup, E. C.; Kennerly, G. W.; Reusch, R. N. J. Org. Chem. 1961, 26, 5155. (b) Pittman, C. U., Jr.; Smith, L. R. J. Organomet. Chem. 1975, 90, 203. (c) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F. Organometallics 1990, 9, 1146. (d) Matsuzaka, H.; Takagi, Y.; Ishii, Y.; Nishio, M.; Hidai, M. Organometallics 1995, 14, 2153. (e) Straub, T.; Haskel, A.; Eisen, M. S. J. Am. Chem. Soc. 1995, 117, 6364. (f) Haskel, A.; Wang, J. Q.; Straub, T.; Neyroud, T. G.; Eisen, M. S. J. Am. Chem. Soc. 1999, 121, 3025. (a) Klein, H.-F.; Beck-Hemetsberger, H.; Reitzel, L.; Rodenhuser, B.; Cordier, G. Chem. Ber. 1989, 122, 43. (b) Klein, H.-F.; Mager, M.; Isringhausen-Bley, S.; Flrke, U.; Haupt, H.-J. Organometallics 1992, 11, 3174. (c) Ref. 2b. (d) Klein, H.-F.; Zettel, B. D. Chem. Ber. 1995, 128, 343. (e) Werner, H.; Schfer, M.; Wolf, J.; Peters, K.; von Schnering, H. G. Angew. Chem. Int. Ed. Engl. 1995, 34, 191. (f)

137

Chapter 4

49

50

51

52

53

54

55 56 57

58

59

Haskel, A.; Straub, T.; Dash, A. K.; Eisen, M. S. J. Am. Chem. Soc. 1999, 121, 3014. (g) Yoshimitsu, S.; Hikichi, S.; Akita, M. Organometallics 2002, 21, 3762. Interconversion of allenes and alkynes, as well as isomerization of allenes to conjugated and nonconjugated dienes, has been accomplished with strong bases and acids or thermally at relatively high temperatures. These conditions are unlikely to occur during substrate conversion in the present rareearth metallocene-catalyzed oligomerization reactions. Among the acyclic isomers with formula CnH2n-2, (substituted) conjugated 1,3-dienes are by far more stable than (substituted) 1,2-dienes and 1-alkynes, see: (a) Jacobs, T. L.; Singer, S. J. Org. Chem. 1951, 17, 475. (b) Jacobs, T. L.; Meyers, R. A. J. Am. Chem. Soc. 1964, 86, 5244. (c) Ref. 20c. (d) Okuyama, T.; Izawa, K.; Fueno, T. J. Am. Chem. Soc. 1973, 95, 6749. For reviews on the general chemistry of conjugated dienes, allenes and acetylenes see: (e) Taylor, D. R. Chem. Rev. 1967, 67, 317. (f) Wotiz, J. H. In Chemistry of Acetylenes; Viehe, H. G., Ed.; Marcel Dekker: New York, 1969. (g) Thron, F., Verny, M., Vessire, R. In The Chemistry of the Carbon-Carbon Triple Bond; Patai, S., Ed.; Wiley: Chichester, 1978. (h) Huntsman, W. D. In The chemistry of ketenes, allenes and related compounds; Patai, S., Ed.; Wiley: New York, 1980. (i) Landor, S. R., Ed. Chemistry of Allenes, Academic Press: New York, 1982. (j) Schuster, H. F.; Coppola, G. M. Allenes in Organic Synthesis, Wiley-Interscience: New York, 1984. (k) Bruneau, C., Dixneuf, Pierre H. In Comprehensive Organic Functional Group Transformations; Katritzky, A. R., Meth-Cohn, O., Rees, C. W., Eds.; Pergamon: Oxford, 1995. The thermal reaction of a hexasilyl-substituted hexa-3,5-dien-1-yne to hexa-4,5-dien-1-yne has been reported to take place at 110 C after 10 hours, see: Sekiguchi, A.; Ebata, K.; Terui, Y.; Sakurai, H. Chem. Lett. 1991, 1417. The stoichiometric and catalytic trimerization of 1-alkynes to form fulvenes is quite rare and the few examples reported involve mostly metal cyclopentadienyl systems, see: (a) Moreto, J.; Maruya, K.; Bailey, P. M.; Maitlis, P. M. J. Chem. Soc., Dalton Trans. 1982, 1341. (b) Moran, G.; Green, M.; Orpen, A. G. J. Organomet. Chem. 1983, 250, C15. (c) O'Connor, J. M.; Hibner, K.; Merwin, R.; Gantzel, P. K.; Fong, B. S.; Adams, M.; Rheingold, A. L. J. Am. Chem. Soc. 1997, 119, 3631. (d) Johnson, E. S.; Balaich, G. J.; Fanwick, P. E.; Rothwell, I. P. J. Am. Chem. Soc. 1997, 119, 11086. (a) Woodward, R. B.; Hoffman, R. The Conservation of Orbital Symmetry, Verlag Chemie, Weinheim, 1970. (b) Woodward, R. B.; Hoffmann, R. J. Am. Chem. Soc. 1965, 87, 2511. (c) Hudson, C. E.; McAdoo, D. J. J. Org. Chem. 2003, 68, 2735 and references therein. (a) Maercker, A.; Wunderlich, H.; Girreser, U. Tetrahedron 1996, 52, 6149. (b) Maercker, A.; Fischenisch, J. Tetrahedron 1995, 51, 10209. (c) Klein, J.; Becker, J. Y. Tetrahedron 1972, 28, 5385. (d) Maercker, A.; Tatai, A.; Grebe, B.; Girreser, U. J. Organomet. Chem. 2002, 642, 1. (a) Birmingham, J. M.; Wilkinson, G. J. Am. Chem. Soc. 1956, 78, 42. (b) Fischer, E. O.; Fischer, H. J. Organomet. Chem. 1965, 3, 181. (c) Coutts, R. S. P.; Wailes, P. C. J. Organomet. Chem. 1970, 25, 117. (d) Kanellakopoulos, B.; Dornberger, E.; Billich., H. J. Organomet. Chem. 1974, 76, C42. (e) Qian, C.; Ye, C.; Lu, H.; Li, Y.; Huang, Y. J. Organomet. Chem. 1984, 263, 333. (f) Qian, C.; Ge, Y. J. Organomet. Chem. 1986, 299, 97. (g) Ref. a. (h) Tsutsui, M.; Ely, N. J. Am. Chem. Soc. 1974, 96, 4042. (i) Ely, N. M.; Tsutsui, M. Inorganic Chemistry 1975, 14, 2680. Heeres, H. J. Ph. D. Thesis, University of Groningen, 1990, Chapter 3. Bercaw, J. E.; Davies, D. L.; Wolczanski, P. T. Organometallics 1986, 5, 443. (a) Giardello, M. A.; Conticello, V. P.; Brard, L.; Sabat, M.; Rheingold, A. L.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10212. (b) Giardello, M. A.; Conticello, V. P.; Brard, L.; Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10241. (c) Douglass, M. R.; Ogasawara, M.; Hong, S.; Metz, M. V.; Marks, T. J. Organometallics 2002, 21, 283. (d) Ryu, J.-S.; Marks, T. J.; McDonald, F. E. J. Org. Chem. 2004, 69, 1038. The observed epimerization of chiral C1-symmetric Me2SiCp(CpR*)2Ln and Me2Si(OHF)(CpR*)Ln amine-amido complexes in the presence of excess amine has been explained in terms of transient CpR* protonolysis and the rate of the epimerization was reported to depend on the structure of the amine substrate, reaction temperature, catalyst and amine concentration.57 Large metals, relatively unhindered, linear amines and high amine concentrations were found to favor epimerization. Reported acidity values for phenylacetylene differ by as much as 10 pK units, depending on the solvent and the method to measure acidity. For example, pK values for phenylacetylene were reported to be 23.2 in cyclohexylamine with lithium cation60c, 28.760j and 29.160f in DMSO with potassium counterion, 31.1 with [2.1.1]cryptated lithium in THF60l, and 21.2 in water based on the hydrogen-

138

The oligomerization of phenylacetylene catalyzed by rare-earth metallocenes

60

61 62 63

64

65

66

67

68

69

exchange rate.60k Benzene is an aprotic, nonpolar solvent. In nonpolar solvents the medium has little influence on the acidity and the measured equilibrium constants do not reflect the free ion acidity, but rather refer to ion pairs or larger associates. 1-Alkynes behave, moreover, both as normal and pseudo-acids and solvent effects are expected to be large in the proton transfer reactions (see Appendix of Chapter 5 for more details).60b,e,g,i Comparison with acidities in aprotic, polar solvent such as water seems therefore inappropriate in the present context. The most extensive, absolute acidity scale in which ion pairs are virtually absent is based on DMSO (D = 46.7).60f The following order of increasing pKDMSO has been found: CpH (18.0) < Cp*H (26.1) < PhC CH (28.7) < MeOH (29.0) < CH3C N (31.3) < H2O (31.2). The gas-phase equilibrium acidities60c Gacid (kcal/mol, 298 K) in which solvent effects are absent decrease in the following order: MeOH (372.6) > tBuOH (366.7) > CH3C N (364.4) > PhC CH (362.6) > CpH (349.9). For reviews on carbon acids, see: (a) Cram, D. J. Fundamentals of Carbanion Chemistry, Academic Press: New York, 1965. (b) Jones, J. R. The Ionisation of Carbon Acids, Academic Press: London, 1973. (c) Reutov, O. A.; Beletskaya, I. P.; Butin, K. P. CH-Acids, Pergamon Press: Oxford, 1978. (d) R. Stewart The Proton: Applications to Organic Chemistry, Academic Press: New York, 1985. (a) Dessy, R. E.; Okuzumi, Y.; Chen, A. J. Am. Chem. Soc. 1962, 84, 2899. (b) Eigen, M. Angew. Chem. Int. Ed. Engl. 1964, 3, 1 (c) Streitweiser, A., Jr.; Reuben, D. M. J. Am. Chem. Soc. 1971, 93, 1794. (d) Bartmess, J. E.; Scott, J. A.; McIver, R. T., Jr. J. Am. Chem. Soc. 1979, 101, 6046. (e) Lin, A. C.; Chiang, Y.; Dahlberg, D. B.; Kresge, A. J. J. Am. Chem. Soc. 1983, 105, 5380. (f) Terekhova, M. L.; Petrov, E. S.; Vasilevskii, S. F.; Ivanov, V. F.; Shvartsberg, M. S. Bull. Acad. Sci. USSR, Div. Chem. Sci. 1984, 33, 850. (g) Kresge, A. J.; Powell, M. F. J. Org. Chem. 1986, 51, 822. (h) Antipin, I. S.; Vedernikov, A. N.; Konovalov, A. I. Zh. Org. Khim. 1986, 22, 446. (i) Aroella, T.; Arrowsmith, C. H.; Hojatti, M.; Kresge, A. J.; Powell, M. F.; Tang, Y. S.; Wang, W.-H. J. Am. Chem. Soc. 1987, 109, 7198. (j) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456. (k) Kresge, A. J.; Pruszynski, P.; Stang, P. J.; Williamson, B. L. J. Org. Chem. 1991, 56, 4808. (l) Antipin, I. S.; Gareyev, R. F.; Vedernikov, A. N.; Konovalov, A. I. J. Phys. Org. Chem. 1994, 7, 181. The reaction of Me2Si(C5Me4)2ThnBu2 with a quite large excess of 1-alkyne (alkyne-to-catalyst molar ratio up to 2000) has been reported.41b (a) Raymond, K. N.; Eigenbrot, Jr., C. W. Acc. Chem. Res. 1980, 13, 276. (b) Evans, W. J. Adv. Organomet. Chem. 1985, 24, 131. (c) Bursten, B. E.; Burns, C. J. Comm. Inorg. Chem. 1989, 2, 61. Cp3Ln complexes behave as Cp-transfer agents in contrast to the cyclopentadienyl compounds of actinides and transition metals, see: Elsenbroich, Ch.; Salzer, A. Organometallics, Wiley: Weinheim, 1992, 2nd ed.; Chapter 17.15. (a) Jutzi, P.; Reumann, G. J. Chem. Soc., Dalton Trans. 2000, 2237. (b) Jutiz, P.; Burford, N. In Metallocenes: Synthesis, Reactivity, and Applications, Togni, A.; Halterman, R. L. (Eds.); WileyVCH, Weinheim, 1998, Vol. 1, p. 354. (a) Duchateau, R., Ph. D Thesis, University of Groningen, 1996; Chapters 4 and 7. (b) Duchateau, R.; Tuinstra, T.; Brussee, E. A. C.; Meetsma, A.; van Duijnen, P. Th.; Teuben, J. H. Organometallics 1997, 16, 3511. (c) Duchateau, R.; Brussee, E. A. C.; Meetsma, A.; Teuben, J. H. Organometallics 1997, 16, 5506. Cp*Ln(OC6H3-iPr2-2,6)2 (Ln = La, Ce) complexes form stable Lewis adducts with THF, tBuCN and PMe3 and disproportionate in solution to form equilibrium mixtures of Cp*2Ln(OC6H3-iPr2-2,6) and Ln(OC6H3-iPr2-2,6)3, see: Heeres, H. J. Ph. D. Thesis, University of Groningen, 1990; Chapter 7. The lack of steric bulk of the alkynyl ligands is expected to allow high reactivity of the electrondeficient metal center, as observed for the homoleptic alkyl complexes Ln[CH(SiMe3)2]3 (Ln = La, Ce), see: Booij, M.; Kiers, N. H.; Heeres, H. J.; Teuben, J. H. J. Organomet. Chem. 1989, 364, 79. Although the compounds Ln(CCPh)3 (Ln = Pr, Sm, Eu, Gd, Tb, Er, Yb) have been described for smaller rare-earth metals, very little structural information is available and it is generally believed that these species form polymers in the absence of donor ligands. For examples, see: (a) Bochkarev, L. N.; Shustov, S. B.; Guseva, T. V.; Shiltsov, S. F. Zh. Obshch. Khim. 1988, 58, 923. (b) Shustov, S. B.; Bochkarev, L. N.; Zhiltsov, S. F. Metalloorg. Khim. 1990, 3, 624. Low-temperature 1H NMR studies of Cp*2YCH2CH2CH(CH3)2 and propene revealed a rapid equilibrium between the yttrium alkyl and its propene adduct prior to insertion. It was alo found that propene dissociation at -100 C is much faster than migratory insertion, see: Casey, C. P.; Lee, T.-Y.; Tunge, J. A.; Carpenetti II, D. W. J. Am. Chem. Soc. 2001, 123, 10762.

139

Chapter 4

70

71 72

(a) Wilkins, R. G. Kinetics and Mechanism of Reactions of Transition Metal Complexes, VCH, Weinheim: 1991; 2nd ed.; Chapter 1. (b) Espenson, J. H. Chemical Kinetics and Reaction Mechanism, McGraw-Hill, Inc., New York: 1995; 2nd ed.; Chapter 4. (a) Segel, I. H. Enzyme Kinetics, Wiley-Interscience, New York; 1993. (b) Marangoni, A. G. Enzyme Kinetics, Wiley-Interscience, New York; 2003. When the effects of reversible, nonproductive substrate and product coordination to the catalyst are included (Eq. 4.18), standard rapid equilibrium analysis yields a modified Henri-Michelis-Menten equation (Eq. 4.19).70,71 This relationship reproduces the observed kinetic behavior, as the reaction rate changes from zero-order (KM + KM[P]/KP < [S]) to first-order dependence on substrate concentration at higher substrate conversion (KM + KM[P]/KP > [S]). (4.18)

KP E
d[S] dt

+
=

EP
kp[ET][S] [P] KM 1 + + [S] KP

(4.19)

73

74

Similar analyses assuming steady-state concentrations for E and ES (i.e. d[E]/dt 0 and d[E]/dt 0) entail that the [P]-to-[Cp*H] ratio is also constant, when the rate of catalyst deactivation is first-order dependent on substrate concentration, see for similar kinetic analyses of enzyme deactivation reactions: Do, D. D.; Weiland, R. H. Biotech Bioeng. 1980, 22, 1087. Izzo, L.; Napoli, M.; Oliva, L. Macromolecules 2003, 36, 9340. (b) Engel, P. S.; Duan, S.; Arhancet, G. J. Org. Chem. 1997, 62, 3537. (c) Akiyama, R.; Kobayashi, S. J. Am. Chem. Soc. 2003, 125, 3412. (d)

140

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

5.

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

5.1.

Introduction

The rare-earth metallocene-catalyzed oligomerization of phenylacetylene has been studied in detail (Scheme 5-1) and the factors determining the regio- and chemoselectivity have been identified (Chapter 4). High selectivities and activities for the formation of (E)-1,4-diphenylbut-1-en-3-yne (12a) were found for reactions with the precatalyst Cp*2LaCH(SiMe3)2, while 2,4-diphenylbut-1-en-3-yne (11a) was formed selectively in reactions with Cp*2YCH(SiMe3)2. However, significant catalyst deactivation via Cp* abstraction and the formation of oligomers higher than dimers was observed at relatively large substrate-to-catalyst molar ratios. A survey of the reported catalytic behavior towards phenylacetylene places the permethyllanthanocene alkyl derivative Cp*2LaCH(SiMe3)2 among the most active and selective precatalysts for the formation of the transhead-to-head dimer.1 Scheme 5-1. The rare-earth metallocene-catalyzed oligomerization of phenylacetylene.
Ph catalyst Ph C 6D 6 Ph 11a + Ph 12a + Ph 15a Ph Ph . + Ph 16a Ph Ph Ph

catalyst =

Ln CH(SiMe3)2 ,

Si

Ln CH(SiMe3)2

Ln = La, Ce, Y

From the desire to apply the catalytic linear dimerization of 1-alkynes to bifunctional substrates in order to prepare conjugated polymers (Chapter 6), it was considered important to assess the scope of variation in the aromatic moiety of this reaction. In this context, the tolerance of the catalyst towards sterically hindered substrates (viz. ortho-substitution of phenyl group) and substrates with heteroatoms was of particular interest. Because large excesses of Lewis basic heteroatom-containing substrates are well-known to inhibit olefin and alkyne interactions with strongly Lewis acidic metal centers and lead commonly to a significant decrease in catalyst activity or even rapid catalyst deactivation, the compatibility of Lewis acidic metal catalysts with Lewis basic heteroatom-containing functionalities is an ongoing challenge in the development of catalytic carboncarbon bond-forming transformations based on early transition and f-block metals. The scope and mechanism of the catalytic oligomerization reaction of substituted (hetero)aromatic 1alkynes, mediated by rare-earth metallocenes, is described in this chapter. The factors governing the activity and selectivity of the catalytic oligomerization reactions with selected (hetero)aromatic 1-alkynes are determined, based on a kinetic and mechanistic study of both catalytic and stoichiometric reactions.

141

Chapter 5

5.2.

Synthesis of (hetero)aromatic 1-alkynes

Introduction Six representative (hetero)aromatic 1-alkynes were synthesized to probe the scope of the permethyllanthanocene-catalyzed 1-alkyne dimerization (Scheme 5-2). Several general routes are available for the preparation of hetero(aromatic) 1-alkynes,2 including (i) the palladium-catalyzed coupling3 (the Negishi reaction) of ethynylzinc, prepared in situ from the ethynyl Grignard reagent, and the corresponding halo(hetero)arene, (ii) the palladium-copper-catalyzed coupling4 (the Castro-Stephens/Hagihara-Sonogashira reaction) of trimethylsilylacetylene (TMSA) and halo(hetero)arene, followed by desilylation, and (iii) the palladium-catalyzed coupling of 2-methyl-3-butyn-2-ol and halo(hetero)arene, followed by base-induced cleavage of -acetylenic alcohol (the reverse Favorskii reaction) into the corresponding carbonyl and ethynyl compounds.5 Although some 1-alkynes were commercially available, they were found to contain impurities that were reactive towards the present catalyst (e.g. aryl halide, 2-methyl-4-arylbut-3-yn-2-ol) and difficult to remove completely from the substrate. The Negishi reaction The Negishi reaction (route i, Scheme 5-2) proved to be unsuitable in the present study, because the commercially available ethynylmagnesium bromide contained trace amounts of the magnesium salt of 2,6-ditert-butyl-4-methylphenolate (BHT). Experiments indicated that the small amounts of this compound in the substrate reacted with the catalyst to yield catalytically inactive, phenolate derivatives, as identified by 1H NMR spectroscopy.6 Taking into account that subsequent catalytic oligomerization experiments are performed in the presence of 2-0.08 mol% (pre)catalyst, contamination of BHT is undesirable, as it affects substrate reactivity. The complete separation of BHT from 2-ethynylanisole was found to be detrimental to the yield. The solvent most commonly used in the Negishi reaction constitutes a second drawback of this procedure, especially for the preparation of 2-ethynyltoluene. Traces of THF in 2-ethynyltoluene were difficult to remove completely by several techniques. Contamination of THF in the substrate has adverse effects on the catalytic performance, as it leads to catalyst deactivation (Chapter 5). Other solvents did not give satisfactorily conversions, thereby contaminating the substrate with halide (hetero)arenes (1) which are known to react with rare-earth metallocene derivatives.7

Scheme 5-2. The synthesis of (hetero)arylalk-1-ynes.

O S S N N

2b 88%

2c 91% i

2d 78% Ar 2

2e 75%

2f 20%

2g 75%

total yield via route iii and iiia:

iia SiMe3

Ar

Ar

ii

Ar 3

iib OH iiia

Ar Ar

iii

Ar 4

Reagents: (i) HC CMgBr, ZnBr2, Pd(PPh3)2Cl2, THF, RT. (ii) HC CSiMe3, Pd(PPh3)2Cl2, CuI, HNiPr2, THF/toluene, 80 C; (iia) K2CO3, THF/MeOH; (iib) nBu4NF, THF. (iii) HC CMe2OH, Pd(PPh3)2Cl2, CuI, HNiPr2, THF/toluene, 80 C; (iiia) , toluene, KOH.

142

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes The Sonogashira reaction The Sonogashira reaction (route ii, Scheme 5-2), on the other hand, produced 1-alkynes which were contaminated by varying amounts of 1,4-bis(trimethylsilyl)butadiyne. Diynes are a common by-product of palladium-catalyzed 1-alkyne cross-coupling methodologies.8 The contaminating small amounts of butadiyne reacted with the present catalyst forming organic products that were not studied further.9 Complete removal of the butadiyne proved to be difficult and the use of substoichiometric amounts of TMSA led to substrate which was contaminated with halide precursors. The synthesis of desired (hetero)aromatic 1-alkynes The substrates 2b-g were prepared satisfactorily by palladium-catalyzed coupling of (hetero)aryl iodide with 2-methyl-3-butyn-2-ol (route iii, Scheme 5-2), according to a published general procedure.10 Additional purification steps were found necessary to obtain the requisite purity for the present catalytic reactions. In most cases, the acetylenic carbinol (3) was purified by vacuum distillation, column chromatography and repeated crystallizations, while deprotection involved base-catalyzed cleavage followed by vacuum distillation and flash column chromatography. The properties of the substrates It must be noted that the prepared 1-alkynes should be handled with care. Especially the heteroaromatic 1-alkynes, such as 2-ethynylthiophene and 1-methyl-2-ethynylpyrrole, are thermally unstable. Thermal decomposition was accompanied by a coloration which was apparent within several hours after storage as pure oils or as solutions at room temperature and within minutes at higher temperatures. The (thermal) instability of terminal acetylenes is well-documented in literature, especially for 1-alkynes containing electronwithdrawing substituents.11,12 For example, it has been reported that 2-ethynylthiophene polymerizes at room temperature within 5 days to insoluble dark brown resins.11f IR and solid 13C NMR of the formed solid indicated that most acetylenic carbons had been converted to olefinic carbons. Some 1-alkynes are, in addition, fairly volatile and must be handled accordingly. The colored liquids, obtained after synthesis, were dried on CaH2 and produced colorless liquids after vacuum transfer. If stored under nitrogen at -30 C, these liquids remained indefinitely stable and colorless.

5.3.

Catalytic oligomerization of (hetero)aromatic 1-alkynes

5.3.1.

Introduction

To study the effects of 1-alkyne substituents on the rate and selectivity of the 1-alkyne oligomerization reaction catalyzed by Cp*2LaCH(SiMe3)2 (5D), a selected group of (hetero)aromatic 1-alkynes was applied under standardized reaction conditions. However, the relative amount of trimerization was found to increase upon increasing the substrate-to-catalyst ratio. Significant substrate and product inhibition was observed as well at relatively high substrate-to-catalyst ratios. As a consequence, two standardized reaction conditions (i.e. substrate-to-catalyst molar ratios of 50 and 400) at constant catalyst concentration were chosen to evaluate the substrate effects on rate and selectivity of the 5D-catalyzed 1-alkyne oligomerization reaction. The organic products were identified by multinuclear 1D and 2D NMR spectroscopy and MS, while their relative amounts were determined by normalized, in situ 1H NMR spectroscopy, using appropriate long pulse delays to avoid signal saturation under the present anaerobic conditions. Other methods to determine the relative amount of products yielded values that corresponded only moderately with those determined by in situ 1 H NMR analysis (Section 4.2.3).

5.3.2.

Substrate effects at relatively low initial substrate concentration

As discussed previously, the reaction of Cp*2LaCH(SiMe3)2 (5D) with a 50-fold molar excess of phenylacetylene (2a) takes place both rapidly and selectively to form trans-1,4-diphenylbut-1-en-3-yne (12a) (Table 5-1). Complete substrate conversion was observed within 10 min, accompanied by the formation of three

143

Chapter 5 Cp* 1H NMR resonances, assigned to [(Cp*2La)2(-PhC4Ph)] ( 2.04 ppm), Cp*2LaC(Ph)=C(H)CCPh ( 1.97 ppm) and Cp*2LaC(H)=C(Ph)CCPh ( 1.92 ppm), present in a 1.0:1.4:1.3 ratio, respectively (Chapter 4). The application of ortho-substituted analogues, i.e. 2-ethynyltoluene (2b) and 2-ethynylanisole (2c), led to a decreased catalytic activity, accompanied by an increased selectivity for the formation of the head-to-tail dimer (9) and a decreased selectivity for trimerization. This catalytic behavior is most likely due to their larger steric requirements of the substrate. The steric effects of 1-alkyne substituents are discussed in more detail in Section 5.5.3. Unfortunately, the ortho-methyl 1H NMR resonances of 2b and its oligomers 11b and 12b overlapped with the Cp* 1H NMR resonances of the organometallic reaction intermediates, thereby impeding attempts to determine the nature and number of reaction intermediates during and after substrate conversion. In the case of 2c, two major Cp* 1H NMR resonances were observed at 1.96 and 1.86 ppm during and after substrate conversion. In analogy to the reactions with phenylacetylene (vide supra), these resonances are presently assigned to [(Cp*2La)2(-RC4R)], Cp*2LaC(R)=C(H)CCR and Cp*2LaC(H)=C(R)CCR, respectively (R = C6H4OMe-2). After substrate conversion, these resonances were present in a 10.0:1.0:0.8 ratio, respectively. Monitoring the substrate concentration by means of normalized, single-pulse in situ 1H NMR spectroscopy revealed that the rate of dimerization is first-order in substrate for at least three half-lives (R2 = 0.9974, kobs = 4.25(5) M-1min-1).13 When five-membered heteroaromatic 1-alkynes, i.e. 2-thienylacetylene (2d), 3-thienylacetylene (2e) and 1-methyl-2-ethynylpyrrole (2g), were employed, the observed catalytic rates and selectivities for the formation of the trans-head-to-head dimer (12) resembled those previously observed for phenylacetylene (Table 5-1). After complete conversion of 2d, the precatalyst 5D was quantitatively converted into a single new lanthanocene derivative, as indicated by one Cp* 1H NMR resonance at 1.97 ppm. Stoichiometric reactions provided evidence that the observed Cp* 1H NMR resonance corresponds to the but-1-en-3-yn-1-yl derivative Cp*2LaC(R)=C(H)CCR (R = C4H3S-2) (Section 5.4.1). In marked contrast, the Cp* 1H NMR region of the reaction mixture after conversion of 2e was considerably more complex. Four major Cp* 1H NMR resonances at 2.17 (broad), 2.03, 1.95 and 1.91 ppm in a 1.0:3.7:1.0:1.5 ratio were observed, representing only ~80% of the total amount of organometallic species present. In analogy to the reactions with phenylacetylene, the latter three resonances may be assigned to [(Cp*2La)2(-RC4R)], Cp*2LaC(R)=C(H)CCR and Cp*2LaC(H)=C(R)CCR (R = C4H3S-3), respectively. After consumption of 2g, the reaction mixture resembled that of the analogous reaction with phenylacetylene in the sense that 5D was quantitatively converted into three species, as evidenced by three Cp* 1H NMR resonances at 2.03, 2.01 and 1.99 ppm in a 1.0:3.3:0.4 ratio, respectively. Based on this analogy, the resonances were tentatively assigned to [(Cp*2La)2(-RC4R)], Cp*2LaC(R)=C(H)CCR and Cp*2LaC(H)=C(R)CCR (R = 2C3H3NMe-1), respectively. Encouraged by the displayed tolerance of the catalyst towards 1-alkynes possessing relatively weakly coordinating heteroatoms (Section 5.4.3), it was decided to study the reactivity of 5D towards 2-ethynylpyridine (2f) as well. In the light of the well-recognized kinetic lability of permethyllanthanidocene derivatives and their preference to coordinate hard heteroatoms,14 2f represents an interesting substrate to probe the competition between strong Lewis base coordination to the catalyst and catalytic substrate conversion. Gratifyingly, 2f (50 equiv.) was converted for 95% within 30 min under the present reaction conditions, affording the corresponding trans-head-to-head dimer (12f) as the only observed organic product (Table 5-1). However, substrate conversion was very slow beyond this point and full conversion was only achieved after 16 h. A more detailed discussion of the reactions of 5D with 2f is deferred to Section 5.3.5.

5.3.3.

Substrate effects at relatively high initial substrate concentration

Introduction To determine the influence of the 1-alkyne substituent on the rate of 1-alkyne oligomerization catalyzed by Cp*2LaCH(SiMe3)2 (5D) and monitor substrate conversion conveniently by means of in situ 1H NMR spectroscopy, the reaction time was increased by applying relatively large substrate-to-catalyst ratios. Unfortunately, the kinetic behavior of the 5D-catalyzed 1-alkyne oligomerization reaction under these reaction conditions was complicated by significant catalyst deactivation and product/substrate inhibition to varying degrees, depending on the substrate. In spite of these complications, the use of a substrate-to-catalyst molar ratio of 400 was found to allow for the convenient evaluation of reaction rates for all of the studied substrates. The

144

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Table 5-1. The product distribution of the 5D-catalyzed 1-alkyne oligomerization reactions conducted at relatively low and high substrate concentrations.a
R R 5D R 2 C6 D6 25 C 11 R R + 12 R R + R 15 R + 16 R R .

Entry 1 2 3 4 5 6 7 8 9 10 11 12 13 14
S

2 (equiv.) 50

11 0.1 0.2 4.9 4.4 37.0 23.2 0.0 0.1 1.8 0.0 0.0 0.0 0.1 0.2

12 97.8 89.1 92.9 89.3 63.0 76.6 99.4 98.3 97.8 87.5 100.0 100.0 99.2 97.3

15 0.5 8.4 2.2 6.3 0.0 0.1 0.6 1.2 0.3 7.7 0.0 0.0 0.5 2.0

16 1.6 2.4 0.0 0.0 0.0 0.0 0.0 0.5 0.1 4.8 0.0 0.0 0.1 0.5

time (conv.) < 10 min (100%) 22 min (100%) 1h (100%) 6.3 h (100%) 2.5 h (100%) 10.1 h (94%) <10 min (100%) 37 min (100%) <10 min (100%) 6.6 h (93%) 1h (95%) 8.2 h (83%) <10 min (100%) 26 min (100%)

Cat. deact. b (%) 0.7 5.0 0.0 0.0 0.6 7.3 0.8 9.6 0.7 12.1 -c -c -c -c

400 50

b
O

400 50 400 50 400 50

c
S

e
N

400 50 400 50 400

f
N

a Reaction conditions: [5D] = 4.1-4.4 mM, C6D6 and 25 C. Yields are determined by normalized in situ 1H NMR spectroscopy and represent average values of two or more runs. The experimental error was found to be 0.2. b Percentage of catalyst deactivated based on the amount of Cp*H formed relative to CH2(SiMe3)2, as determined by in situ 1H NMR spectroscopy c Not determined, due to overlapping 1H NMR resonances.

standardized reaction conditions were performed at constant catalyst concentration. Substrate conversion was monitored by single-pulse, in situ 1H NMR spectroscopy, using appropriate delay times to avoid signal saturation under the present anaerobic conditions.13 In most cases, normalization of substrate and product 1H NMR resonances against cyclooctane as an internal standard provided consistent kinetic data and product distributions (Table 5-1).

145

Chapter 5

1.6 Concentration (mol/L) 7a 1.2 7c 7d 7e 7f 7g

0.8

0.4

0 0 30 60 90 time (min) 120 150 180

Figure 5-1. Plot of the substrate concentration versus time for the catalytic oligomerization reactions of 2ag mediated by 5D. Lines and curves connecting the data points represent fitted linear plots and first-order exponentials, respectively (see text for details). Phenylacetylene The reaction of Cp*2LaCH(SiMe3)2 (5D) with a 400-fold molar excess of phenylacetylene is rapid. 1H NMR spectroscopy revealed that phenylacetylene was completely converted within 22 min, thereby allowing the acquisition of four data points. Even so, the rate of substrate conversion is more likely to be zero-order in substrate (R2 = 0.9944, kobs = 20(1) min-1) than first-order in substrate, as seen from plots of substrate concentration [S] versus time (Figure 5-1) and ln([S]t/[S]0) versus time (Figure 5-2). In addition, catalyst deactivation via Cp* ligand abstraction took place to a relatively small extent (5%). A detailed kinetic and mechanistic study of the reaction of 5D with phenylacetylene revealed that the rate of substrate conversion may plausibly be described by saturation kinetics involving a rate-limiting preequilibrium of a monomeric, alkynyl derivative Cp*2LaCCPh (20a) with its Lewis base adduct Cp*2LaCCPhPhCCH (20a2a) (Chapter 4). In accord with this model, the rate is zero-order in substrate at high substrate concentration and becomes first-order in substrate at lower substrate concentrations. 2-Ethynyltoluene Substrate conversion in the catalytic oligomerization of 2-ethynyltoluene (2b) was followed in time by monitoring the intensity decrease of the substrate 1H NMR resonance C CH ( 2.92 ppm) relative to that of cyclooctane. Full conversion of 2b was observed after 6.2 h and the rate of reaction was found to be first-order in substrate for at least 4 half-lives (R2 = 0.9982, kobs = 9.46(8) M-1min-1). Unfortunately, the product 1H NMR resonance at 2.01 ppm obscured Cp* 1H NMR resonances and in situ 1H NMR spectroscopy did not provide insight into the organometallic reaction intermediates during and after substrate conversion. The formed products reveal that more head-to-tail dimerization and less trimerization take place in the reaction of 2b as compared to the analogous reaction with phenylacetylene. The selectivity for head-to-tail dimerization versus trans-head-to-head dimerization in the present oligomerization reaction was rationalized in terms of a competition between 1,2- and 2,1-insertion of substrate into a monomeric alkynyl derivative (Chapter 4). The present results show that the presence of an ortho-methyl group diminishes the preference for 2,1insertion. The selectivity for dimerization versus trimerization was rationalized in terms of an alkenyl derivative (formed from substrate insertion into a monomeric alkynyl derivative) undergoing either protonolysis by substrate or insertion of substrate, respectively. It seems that the presence of an ortho-methyl group decreases the rate of substrate insertion into the alkenyl derivative relative to that of protonolysis by substrate. The presence of an ortho-methyl group affects the regioselectivity of catlytic trimerization as well, as trimer 15b was formed exclusively.

146

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

0 0

10

20

30

40

50

60

7a
-1 ln([S]0/[S]t )

7d

7g

-2

-3

-4 t (min)

Figure 5-2. Integrated rate plot of the substrate concentration versus time for the catalytic oligomerization reactions of 2a, 2d and 2g mediated by 5D. Lines represent fitted linear plots (see text for details). 2-Ethynyanisole The catalytic oligomerization of 2-ethynylanisole (2c) is considerably slower than that of 2ethynyltoluene (2b) and phenylacetylene. The substrate was found to be converted for 94% after 10.1 h. In spite of significant catalyst deactivation (10%), the kinetic data indicate that the rate of substrate conversion is firstorder in substrate during the first 1.5 half-lives (R2 = 0.9981, kobs = 3.51(2) M-1min-1). Significant deviation from first-order kinetic behavior in substrate was observed after 204 min at which point 72% of the substrate had been converted (1.9 half-lives). In contrast to the four Cp* 1H NMR resonances observed during the analogous reaction of 5D with excess phenylacetylene, only two major Cp* 1H NMR resonances (at 1.98 and 1.87 ppm in a 1.0:1.5 ratio, respectively, after substrate conversion) were observed during and after conversion of 2ethynylanisole (2c). These results confirm the above notion that the presence of an ortho-substituent disfavors trimerization over dimerization, trans-head-to-head dimerization over head-to-tail dimerization and catalytic trimerization to produce 16 over catalytic trimerization to produce 15 relative to analogous reactions with phenylacetylene. Remarkably, the selectivity for head-to-tail dimerization of 2c seems to decrease with increasing substrate-to-catalyst molar ratios (Table 5-1, Entries 5 and 6). This behavior was studied in more detail and is discussed in Section 5.3.6. The slower rate of reaction as compared to analogous reactions with 2b and phenylacetylene and the observed deviation from first-order rate dependence on substrate concentration are presently attributed to substrate and/or product inhibition via La-O interactions (Section 5.5.3). More catalyst deactivation via Cp*H ligand abstraction was observed for 2c than for phenylacetylene and 2b. 2-Ethynylthiophene Complete conversion of 2-ethynylthiophene (2d) was observed after 40 min. Only one major 1H NMR resonance at 1.97 ppm was observed during substrate conversion in the Cp* region ( 2.5-1.5 ppm). Despite significant catalyst deactivation (10%), the first 7 data points (during which 99% of the substrate is converted, covering 6.6 half-lives of the reaction) indicate that the rate of reaction is first-order in substrate (R2 = 0.9981, kobs = 36(1) M-1min-1). The formed products reveal that less trimerization and more catalyst deactivation via Cp* abstraction take place in the reaction of 2d as compared to the analogous reaction with phenylacetylene. The decreased tendency for catalytic trimerization as displayed by 2-ethynyltoluene (2b) and 2ethynylanisole (2c) relative to phenylacetylene was previously rationalized in terms of an increased steric bulk of the 1-alkyne substituent. This view is consistent with the larger steric requirements for 1-alkyne insertion relative to those for protonolysis by 1-alkyne, but contradicts the present result.15 In spite of its smaller size, 2ethynylthiophene (2d) displays a significant lower tendency for catalytic trimerization than phenylacetylene.

147

Chapter 5

0 0

100

200

300

400

500

2c -1 ln([S]0/[S]t) -2 -3 -4 -5 time (min)

2e

2f

2b

Figure 5-3. Integrated rate plot of the substrate concentration versus time for the 5D-catalyzed oligomerization reaction of 2b, 2c, 2e and 2f. Lines represent fitted linear plots (see text for details). Clearly, electronic effects play an additional role in determining the reactivity of the alkenyl derivative, undergoing either substrate insertion (to yield trimerization products) or protonolysis by substrate (to yield dimerization products). The increased rate of catalyst deactivation as observed for 2d relative to phenylacetylene can be explained by the higher (kinetic) acidity, thereby promoting Cp*H abstraction (Section 5.5.3). Another explanation involves the more electron-withdrawing character and smaller size of the 2-thienyl group relative to the phenyl group, thereby rendering the metal center both electronically and sterically less saturated and therefore more reactive. The reactivity of lanthanide complexes is well-known to be correlated with steric and electronic saturation at the metal center, while the stability of alkynyl complexes decreases in general with the electron-withdrawing ability of the 1-alkyne substituent.16 The steric and electronic 1-alkyne substituent effects and their importance in the present catalytic reactions are discussed in more detail in later sections (Sections 5.5.1 and 5.5.3). One major Cp* 1H NMR resonance at 1.97 ppm was observed during catalytic conversion of 2ethynylthiophene (2d). Interestingly, this Cp* 1H NMR resonance was also observed in the product mixture resulting from the stoichiometric reaction of Cp*2LaCH(SiMe3)2 (5D) with 2d (Section 5.4.1). In the latter mixture, NMR spectroscopy showed the presence of two Cp* 1H NMR resonances at 2.00 and 1.97 ppm, assigned plausibly to the butatrienediyl derivative {(Cp*2La)2[-(C4H3S-2)C4(C4H3S-2)]} (22d) and the but-1en-3-yn-1-yl derivative Cp*2LaC(2-C4H3S)=C(H)CC(2-C4H3S) (24d), respectively. The but-1-en-3-yn-1-yl derivative 24d is formed from insertion of 2d into the metal-carbon bond of a monomeric alkynyl species Cp*2LaCC(2-C4H3S) (20d) and seems to represent the resting state of the catalyst during the catalytic conversion of 2d. This finding implies that the reaction of 24d with 1-alkyne is rate-limiting in the present reaction, which is consistent with the observed first-order rate dependence on substrate concentration. 3-Ethynylthiophene The reaction of 5D with 3-ethynylthiophene (2e) exhibited both a lower reaction rate and a lower selectivity towards the trans-head-to-head dimer 12e than the analogous reaction with 2-ethynylthiophene (2d) (Table 5-1). Remarkably, no head-to-tail dimer 11e was observed and the relatively low selectivity for 12e stems from an increased preference for trimerization. The first 2.13 h (5.6 half-lives) correspond to a region of firstorder rate dependence on substrate concentration (R2 = 0.9967, kobs = 6.1(1) M-1min-1) during which 97% of the substrate was consumed. After 6.6 h, only a conversion of 98% was reached and longer reaction times did not lead to complete substrate conversion. One dominant Cp* 1H NMR resonance at 2.10 ppm was observed

148

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Scheme 5-3. Model for competing Lewis base coordination via heteroatom-metal interactions during transhead-to-head dimerization (S = heteroatom of substrate, P = heteroatom of product).
P Cp*2Ln R R R R P Cp*2Ln R R P -P R Cp*2Ln R S -S Cp*2Ln R S R R P -P Cp*2Ln R S -S R Cp*2Ln S

during the majority of the reaction which transformed into several new, unidentified Cp* 1H NMR resonances after ~2 h. In view of the formation of multiple Cp* 1H NMR resonances at a relatively high substrate conversion, the deviation from first-order kinetic behavior in substrate at higher substrate conversion and the absence of other products after quenching with methanol-d1 (GC/GC-MS), reversible catalyst inhibition via metal sulfur interactions is believed to be responsible for the observed catalytic behavior. The preparation of Lewis base adduct of monomeric alkynyl derivatives Cp*2LaCCR (Section 5.4.3) and the observed reactivity of but-1-en-3-ynyl lanthanocene derivatives with THF (Chapter 4) provide indirect evidence that the proposed reaction intermediates are capable of coordinating to substrate and product(s). Competing Lewis base coordination of substrate and product via metal heteroatom interactions during trans-head-to-head dimerization is shown schematically in Scheme 5-3. Deviation from first-order kinetic behavior in substrate at higher substrate conversion was also observed for analogous reactions with 2-ethynylanisole (2c) and 2-ethynylpyridine (2f), but was absent in reactions with phenylacetylene (2a), 2-ethynyltoluene (2b) and 1-methyl-2-ethynylpyrrole (2g). This finding suggests that competing Lewis base coordination via heteroatom metal interactions take place in the oligomerization reactions of 2c, 2e and 2f, but not in the oligomerization reactions of 2a and 2g (Section 5.5.3). The 6-fold higher reaction rate and the absence of a significant deviation from first-order kinetic behavior in substrate at relatively high substrate conversion in the reaction of 2-ethynylthiophene (2d) as compared to 3-ethynylthiophene (2e) reveals that competing Lewis base coordination via heteroatom metal interactions depends on the substitution pattern of the thienyl ring. Indeed, the stronger -electron withdrawing character of the 2-thienyl group versus the 3-thienyl group is well-established (Appendix) and electronic substituent effects are presently believed to influence both catalytic rate and selectivity (Section 5.5.3). The lower degree of reversible catalyst inhibition via metal sulfur interactions in the reactions with 2d relative to reactions with 2e is also attributed to the close proximity of the ethynyl group and the sulfur atom in 2d. The proximity of the thienyl group and the sulfur atom in 2d is presently believed to assist catalytic reaction sequences (Section 5.5.3). The present results for the reactions of 5D and 3-ethynylthiophene (2e) at relatively high initial substrate concentration indicate that the catalytic reacton sequences can compete effectively with Lewis base coordination at low degrees of substrate conversion. At a higher degree of substrate conversion, the substrate concentration decreases relative to the concentration of products, rendering nonproductive sulfur coordination more competitive with the catalytic reaction sequences. Because complete substrate conversion is achieved at a relatively low initial substrate concentration (Section 5.3.2), the extent of competing Lewis base coordination via metal heteroatom interactions seems to be determined by the absolute substrate concentration. Catalytic rate depression due to metal heteroatom interactions was demonstrated by catalytic reactions in the presence of exogenous Lewis bases (Sections 5.3.4 and 5.3.5) and is commonly observed in rare-earth metal-catalyzed reactions.17

149

Chapter 5 2-Ethynylpyridine The catalytic dimerization of 2-ethynylpyridine (2f) slowed down in a remarkable fashion after 1.4 h (57% conversion). In the first hour 44% of the substrate was converted, but only a conversion of 63% and 73% was reached after the second and third hour, respectively. Finally, after 3.9 days 89% of the substrate was found to be consumed. Exclusive trans-head-to-head dimerization was indicated by NMR spectroscopy and no evidence for the formation of the head-to-tail dimer and trimers was obtained with GC/GC-MS. Several major Cp* 1H NMR resonances were observed at low substrate conversion, giving rise to a multitude of Cp* 1H NMR resonances at higher substrate conversion. The kinetic data of the first 1.4 h (1.2 half-lives) correspond to a region, exhibiting first-order rate dependence on substrate concentration (R2 = 0.9976, kobs = 2.08(3) M-1min-1). Exclusive trans-head-to-head dimerization is presently attributed to the high -electron withdrawing character of the 2-pyridyl group and precomplexation to the nitrogen atom, rendering the reaction a heteroatom-directed C-C coupling reaction (Section 5.5.3). The relatively low reaction rate and the failure to achieve full substrate conversion are ascribed to product inhibition and catalyst deactivation (Section 5.3.5). When the analogous reaction was performed at 50 C, the high selectivity for trans-head-to-head dimerization was preserved, but full conversion was still not achieved (86% after 9.5 h). The reaction of 5D with 2f was studied in more detail and these results are discussed in Section 5.3.5. 1-Methyl-2-ethynylpyrrole In marked contrast to the other five-membered heteroaromatic 1-alkynes, 1-methyl-2-ethynylpyrrole (2g) was competely converted within 26 min (allowing the acquisition of only four data points), while the selectivity was comparable to that of 2-ethynylthiophene (2d).18 In spite of the limited number of kinetic data, the rate of reaction can be modeled more accurately by zero-order rate dependence on substrate concentration (R2 = 0.9971, kobs = 14.2(4) min-1) than by first-order rate dependence on substrate concentration, as seen from plots of substrate concentration [S] versus time (Figure 5-1) and ln([S]t/[S]0) versus time (Figure 5-2). One major, broad Cp* 1H NMR resonance at 2.11 ppm was observed during substrate conversion which transformed into three Cp* 1H NMR resonances at 2.03, 2.01 and 1.99 ppm in a 0.9:3.4:0.5 ratio, respectively, after complete substrate conversion. The Cp* 1H NMR resonances of the reaction intermediates are analogous to those observed during the corresponding reaction of Cp*2LaCH(SiMe3)2 (5D) and phenylacetylene (Chapter 4). Also, the kinetics of the reaction of 5D with 2g resembles that of the corresponding reaction of 5D with phenylacetylene. These considerations imply that the nitrogen atom of 2g does not interact significantly with the metal center of the catalyst. This notion is, in fact, in accord with the common belief that the nitrogen lone pair in 1-methylpyrrole is part of the aromatic system and therefore not available for interaction with electrophiles.19

5.3.4.

Experiments with 2-ethynylthiophene

Effect of substrate and catalyst concentration In view of the favorable catalytic behavior, combining a high selectivity for trans-head-to-head dimerization with a high activity, it was decided to study the reaction of Cp*2La(SiMe3)2 (5D) with excess 2ethynylthiophene (2d) in more detail. Monitoring the substrate consumption during the 5D-catalyzed oligomerization of 2d by normalized, single-pulse in situ 1H NMR spectroscopy revealed that the rate of reaction is first-order dependent on substrate concentration over a 3-fold substrate concentration range (0.6-1.6 M) and a 3-fold catalyst concentration range (1.4-4.4 mM) for at least 3 half-lives. When the substrate concentration at constant catalyst concentration (1.4 mM) was increased from 0.601 M to 1.52 M, deviation from first-order rate dependence on substrate concentration was observed at a lower degree of substrate conversion (i.e. after 5.4 and 3.1 half-lives, respectively). This observation suggests that nonproductive product coordination to the catalyst competes effectively with productive substrate coordination to the catalyst at relatively high product concentrations and low substrate concentrations. Similar, but more pronounced behavior was also observed in the analogous reactions with 2-ethynylanisole, 3ethynylthiophene and 2-ethynylpyridine (Section 5.3.3). The product distribution of the oligomerization reaction of 2d at different reaction conditions revealed that the rates of catalytic trimerization and catalyst deactivation are promoted relative to catalytic

150

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Table 5-3. The 5D-catalyzed oligomerization of 2-ethynylthiophene at different substrate and catalyst concentrations and in the presence of thiophene.a Entry [5D] [2d] [Cp*H] time Cat. deact.b 11d 12d 15d 16d (mM) (M) (mM) (min) (%) 1 2 3 4 5d 8.46 4.41 1.42 1.35 1.35 0.47 1.63 0.601 1.52 1.52 0.24 0.79 0.31 0.77 0.95 0.0 0.1 0.0 0.0 0.0 99.5 97.0 98.9 96.8 96.8 0.6 2.5 1.0 2.7 2.6 0.0 0.4 0.1 0.6 0.6 <10 32 83 105 282 1.4(3) 9(1) 11(2) 27(2) 34(2)

a Reaction conditions: benzene-d6 and 25 C. Yields are determined by normalized in situ 1H NMR spectroscopy and represent average values of two or more runs. The experimental error in the relative amounts and [Cp*H] were found to be 0.2 and 0.05 mM, respectively. b Catalyst deactivation, calculated from [Cp*H] and [5D]. c Not determined, due to rapid substrate conversion. d In the presence of 1130 equiv. of thiophene.

dimerization by the use of relatively high substrate concentrations (Table 5-3). It can also be seen that the rates of catalytic trimerization and catalyst deactivation relative to that of catalytic dimerization are independent of precatalyst concentration and remain invariant within experimental error. Similar substrate and catalyst concentration effects on the relative rates of concurrent catalytic dimerization, catalytic trimerization and catalyst deactivation have also been observed in analogous oligomerization reactions of phenylacetylene (Chapter 4). Substrate and product inhibition To obtain additional evidence for the proposal that nonproductive metal heteroatom interactions play a role in the catalytic oligomerization reactions of 2-ethynylthiophene (2d), the reaction of 5D and 2d (1200 equiv.) was performed in the presence of thiophene (Entry 5, Table 5-3). The rate of 2d conversion was found to be first-order in 2d for only 2.7 half-lives and a ~1.6-fold rate decrease was observed (Figure 5-4). This result clearly demonstrates that nonproductive metal heteroatom interactions lead to catalyst rate depression by competing for and blocking coordination sites at the metal center of the catalyst in the present catalytic system. As thiophene is capable of competing effectively with productive Lewis base coordination of the substrate to the catalyst, it seems likely that both substrate and product are engaged in competitive inhibition of the catalyst (Scheme 5-3). Additionally, the observation that the presence of thiophene leads to more catalyst deactivation (34% versus 26%, Table 5-3) suggests that the nonproductive coordination of thiophene to the catalyst retards the rate of catalytic dimerization relative to that of catalyst deactivation. Mechanism In analogy to the oligomerization reactions of Cp*2LaCH(SiMe3)2 (5D) with phenylacetylene, the proposed mechanism for the analogous oligomerization of 2-ethynylthiophene (2d) is believed to involve the reaction of Cp*2LaCC(2-C4H3S) (20d) with substrate to form the substrate adduct 20d2d which may undergo intramolecular protonolysis leading to catalyst deactivation or intramolecular substrate insertion leading to Cp*2LaC(2-C4H3S)=CHCC(2-C4H3S) (24d) (Scheme 5-4). Protonolysis of the formed but-1-en-3-ynyl derivative 24d by 2d leads to the formation of 12d and 20d, while insertion of 2d into 24d leads to catalytic trimerization. It can also be envisaged that protonolysis of 24d by substrate proceeds via the formation of an alkyne coordinated substrate adduct or via the formation of a sulfur coordinated substrate adduct which may give rise to an alkyne coordinated substrate adduct (Scheme 5-15). Although these possibilities are neither supported nor discarded by the present data (vide infra), sulfur and alkyne coordination to the metal center in substrate adducts of the catalyst are treated separately in the following analysis. Recognizing that the reaction of 2d and 24d is rate-limiting in the present oligomerization reaction leads to the following empirical rate law: v = kobs[2d][5D]. When the effects of reversible, nonproductive coordination of the substrate (substrate self-inhibition) and product (competitive product inhibition) to the catalyst are included, a standard rapid equilibrium analysis reproduces the observed kinetic behavior, as increasing substrate and product concentrations lead to decreasing reaction rates and deviations from first-order rate dependence on substrate concentration.20,21 It is difficult to determine the significance of non-productive substrate coordination to the catalyst during catalytic conversion of 2d based on the present data, however.22

151

Chapter 5

0 0

40

80

120

160

-1 ln([S]t/[S]0)

-2 2d -3

2d + C 4H4S

-4 time (min)

Figure 5-4. Integrated rate plot of the substrate concentration versus time for the 5D-catalyzed 1-alkyne oligomerization of 2d in the presence and absence of thiophene (Entries 5 and 4, respectively, Table 5-3). Lines connecting the data points represent fitted linear plots (see text for details). In analogy to the oligomerization reactions of Cp*2LaCH(SiMe3)2 (5D) with phenylacetylene, catalyst deactivation may be modeled by the irreversible formation of a catalytically inactive complex from the catalyst precursor (Chapter 4). A standard steady-state approximation reveals that the rate of catalyst deactivation may be zero-, first- or mixed-order dependent on substrate concentration. The observation that the ratio between the final concentration of Cp*H and that of 12d is invariant (within experimental error) suggests that catalyst deactivation is first-order rate dependent on substrate concentration under the studied reaction conditions. For this scenario implies that the ratio of reaction products of catalyst deactivation (d[Cp*H]/dt = kd[ET][S]) and catalytic dimerization (d[P]/dt = kp[ET][S]) is determined solely by the ratio of the corresponding reaction constants. The experimental error in the final concentration of Cp*H does not allow an accurate determination of kp/kd, however (e.g. [12d]/[Cp*H] = 2.0(4) 103, based on selected data presented in Table 5-3). It should be noted that a scenario in which the rate of catalyst deactivation is zero- or mixed-order dependent on substrate concentration entails a [12d]-to-[Cp*H] ratio that is not constant, but varies with substrate concentration. Influence of the reaction temperature Interestingly, the preference for trans-head-to-head dimerization decreased only to a small extent, when the reaction of Cp*2LaCH(SiMe3)2 (5D) with a 400-fold molar excess of 2-thienylacetylene (2d) was performed at 80 C (i.e. 11d:12d:15d:16d = 1.25:97.10:1.10:0.56). Because 2d was found to be somewhat thermally unstable (Section 5.2), control experiments without catalyst were performed in order to determine the rate of thermal oligomerization of 2d. Experiments conducted at 80 C revealed that thermal oligomerization is not competitive with 5D-catalyzed oligomerization. For example, under identical reaction conditions, but in the absence of metal (pre)catalyst, a substrate conversion of 28% was indicated by 1H NMR spectroscopy after 17 h at 80 C. Longer reaction times did not lead to a significantly higher degree of substrate conversion. After 26 days (92% conversion) the sample was analyzed with GC- and GC-MS. Unidentified compounds were found having the following mass-to-charge ratios and relative flame ionization detector responses (FID-GC yields): a dimer (m/z 216, 31%), an unknown compound (m/z 298, 50%) and three trimers (m/z 324, 7/8/5%). The radical nature of the thermal polymerization of arylacetylenes was established several decades ago, but the exact mechanism is not known presently.23 Even so, there seems to be a general consensus that the low molecular weight fraction consists mainly of dimers and trimers, whose structures have been identified as substituted naphtalenes, and that the higher molecular weight fraction consists of oligo(ene)s.

152

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Scheme 5-4. The proposed mechanism for the catalytic oligomerization of 2-ethynylthiophene (2d) mediated by Cp*2LaCH(SiMe3)2 (5D) (S = heteroatom of substrate and products).
Cp*2LaR -RH 2d 2d S Cp*2La S S 20d.2d Cp*2La Cp*La S S S -S 24d Cp*2La S S + Cp*H S -15d -16d -12d Cp*2La 20d -2d S S -S 2d S Cp*2La S

S Cp*2La S S

S 2d Cp*2La S

5.3.5.

Experiments with 2-ethynylpyridine

Introduction Monitoring the substrate consumption during the reaction of Cp*2LaCH(SiMe3)2 with 2ethynylpyridine (55 equiv.) by normalized, single-pulse in situ 1H NMR spectroscopy revealed three regions of kinetic behavior (Figure 5-6). The first region is defined by the first 18 min of substrate conversion in which 87% of the substrate is converted. Because the corresponding four data points can be modeled both to first- and zero-order rate dependence, the determination of the rate order in substrate is rather tentative, however. In analogy to the observed kinetic behavior of the reactions of Cp*2LaCH(SiMe3)2 (5D) with other coordinating heteroaromatic 1-alkynes (e.g. 2- and 3-ethynylthiophene), the first kinetic region is presently assumed to exhibit first-order kinetic behavior in substrate (R2 = 0.9935, kobs = 14.4(8) M-1min-1). A transition is observed between 22 and 52 min followed by a third kinetic region after 95% substrate conversion where the rate appears to be approximately zero-order in substrate (R2 = 0.9543, kobs = 8.2(4) 10-3 min-1). Additional mechanistic information for this reaction was obtained by monitoring the 1H NMR resonances in the Cp* region during substrate conversion. During the first 30 min of the reaction, corresponding to a substrate conversion of 91%, one major Cp* 1H NMR resonance at 2.16 ppm was observed. Further substrate conversion led to its disappearance, accompanied by the formation of at least eight new Cp* 1H NMR resonances. The complexity of the 1H NMR spectrum after substrate conversion thwarted attempts to identify the organometallic reaction products. Upon addition of D2O or MeI to the reaction mixture after substrate conversion, most of the 1H NMR resonances in the Cp* region ( 2.5-1.5 ppm) disappeared and the amount of Cp*H formed could be determined (by 1H NMR integration versus CH2(SiMe3)2 formed in situ from 5D and substrate). Catalyst deactivation resulting in cyclopentadienyl ligand abstraction occurred only to a small extent (0.5%), thereby advocating that the observed catalytic rate depression at a relatively high substrate conversion is not due to catalyst deactivation via Cp* H abstraction. The above results argue that catalytic dimerization of 2-ethynylpyridine (2f) represents the only productive pathway in the reaction of 5D with a 55-fold molar excess of 2-ethynylpyridine. Upon reaching a substrate conversion of 87%, however, nonproductive, competing reaction sequences become increasingly important, thereby altering the observed kinetic behavior of the reaction of 5D with 2f and giving rise to several organometallic reaction products. In view of the formation of several Cp* 1H NMR resonances at a relatively high substrate conversion, the deviation from first-order kinetic behavior in substrate at higher substrate

153

Chapter 5

0.6 0.5 Concentration (mol/L) 0.4 0.3 0.2 0.1 0 0 25 50 75 time (min) 100 125 150 run 1 run 2 run 3

Figure 5-5. Plot of the substrate concentration versus time for the 5D-catalyzed dimerization of a single portion of 2f (50 equiv., run 1), a second portion of 2f (50 equiv., run 2) and 2f (50 equiv.) in the presence of pyridine (50 equiv., run 3). Curves connecting the data points are fitted second-order exponentials. conversion and the absence of other products after quenching with methanol-d1 (GC/GC-MS), it seems natural to ascribe the nature of these nonproductive, competing reaction sequences to reversible heteroatom coordination of the reaction intermediates to substrate and/or product, as proposed previously for the reactions of 5D with 3ethynylthiophene (Section 5.3.3) and 2-ethynylthiophene (Section 5.3.4). Product and substrate inhibition In order to provide additional evidence for the proposed competition between nitrogen coordination of substrate and/or product to the catalyst and catalytic reaction sequences, analogous reactions of Cp*2LaCH(SiMe3)2 (5D) with 2-ethynylpyridine (2f) were performed in the presence of product and pyridine. When a second portion of substrate (55 equiv.) was added to the above reaction mixture, in situ 1H NMR spectroscopy revealed the slow conversion of substrate. The kinetic profiles of the reactions in the presence (run 2) and absence of product (run 1) were similar (Figure 5-5), but the rates of substrate conversion during the first and third region were lower and the third region started at a lower degree of substrate conversion (71% versus 95%). Instead of a single, major Cp* 1H NMR resonance at 2.16 ppm, two major Cp* 1H NMR resonances were observed at 1.75 and 1.71 ppm during substrate conversion. It seems therefore that other organometallic reaction intermediates are involved in the catalytic conversion of 2f, when a second portion of substrate is added to the reaction mixture. Catalyst deactivation via Cp* abstraction took place only to a small extent (~1%) and no products other than those expected from exclusive trans-head-to-head dimerization of 2f were observed with GC/GC-MS after quenching the reaction mixture with D2O. Under the assumption that the degree of irreversible catalyst deactivation can be neglected (vide infra), the catalytic conversion of 2f in the presence of product (~25 equiv. relative to catalyst) indicates that the presence of product lowers both the catalytic rate and the attainable degree of substrate conversion. In order to investigate the possibility of irreversible catalyst deactivation during the catalytic conversion of 2f, product inhibition was modeled by pyridine by performing the reaction of 5D with 2-ethynylpyridine (55 equiv.) in the presence of pyridine (55 equiv.). In situ 1H NMR spectroscopy revealed slow substrate conversion and a kinetic profile (run 3) similar to those observed before (Figure 5-5). The observed Cp* 1H NMR resonances were reminiscent of those observed during the reaction of 5D and 2f (55 equiv.) in the sense that one major species ( 2.15 ppm, ~50% of the total amount of organometallic species present) was present during the first stage of substrate conversion which transformed into a multitude of unidentified species at higher substrate conversion. The rates of substrate conversion for the reaction of 5D in the presence of product (run 2) and pyridine (run 3) are similar during the first kinetic region and this finding reveals that the decreased reaction rate at higher degrees of substrate conversion is due to reversible catalyst inhibition and not irreversible catalyst deactivation. The observation that a higher degree of substrate conversion is achieved in the presence of pyridine (93%) than in the presence of product (71%) implies also that the dimerization product inhibits catalytic activity more effectively than pyridine. The above reactions of 5D and 2f in the presence of product and pyridine clearly

154

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

0 0

25

50

75 run 1

100 run 2

125 run 3

150

-1 ln([S]t/[S]0) -2 -3 -4 t (min)

Figure 5-6. Integrated rate plot of the substrate concentration versus time for the 5D-catalyzed dimerization of a single portion of 2f (50 equiv., run 1), a second portion of 2f (50 equiv., run 2) and 2f (50 equiv.) in the presence of pyridine (50 equiv., run 3). Lines drawn represent fitted linear plots (see text for details). establish that competing reversible nitrogen coordination to the catalyst results in both a lower rate and a lower degree of substrate conversion. Similar catalytic consequences from competing metal heteroatom interactions during catalytic oligomerization of heteroaromatic 1-alkynes were also observed in the reactions of 3ethynylthiophene (Section 5.3.3) and 2-ethynylthiophene (Section 5.3.4), albeit less pronounced. The strong preference for nitrogen coordination over sulfur coordination is well-recognized in rare-earth metal chemistry (Sections 5.4.3 and 5.4.4). An inverse relationship between the degree of substrate conversion and the substrate-to-catalyst molar ratio was found. For example, the use of a 30-fold molar excess led to 100% substrate conversion (after 12 h at room temperature), whereas the use of a 200-, 400 and 1000-fold molar excess led to 95% (after 28 h), 89% (after 94 h) and 71% (after 102 h) substrate conversion, respectively. It seems therefore that the degree of catalyst inhibition increases upon inceasing the substrate concentration at constant precatalyst concentration. This finding provides additional support to the above view that competing reversible nitrogen coordination of both product and substrate compete with catalytic reaction sequences (Scheme 5-3), thereby lowering the degree of substrate conversion. Similar catalytic behavior was also observed for the reactions of 5D with 3ethynylthiophene (2d). Even though a 55-fold molar excess of 2d was converted completely, the use of a 400fold molar excess led to incomplete substrate conversion (Section 5.3.3).

5.3.6.

Experiments with 2-ethynylanisole

Substrate and product inhibition Reactions with a 40-, 200- and 600-fold molar excess of 2c were conducted at constant precatalyst concentration and substrate conversion was followed in time by normalized, in situ 1H NMR spectroscopy (Figure 5-7). The reaction of 5D with a 40-fold molar excess of 2c exhibited first-order rate dependence on substrate concentration over at least 7 half-lives (R2 = 0.9974, kobs = 20.6(2) M-1min). Complete substrate conversion was observed after 150 min, accompanied by 0.6% catalyst deactivation via Cp*H abstraction. The analogous reaction with a 200-fold molar excess of 2c displayed first-order dependence on substrate concentration over at least 3 half-lives (R2 = 0.9983, kobs = 4.70(3) M-1min). Only 99.8% substrate conversion was observed after 10 h, accompanied by 5% catalyst deactivation via Cp*H abstraction, and longer reaction times did not lead to complete substrate conversion. When the reaction was performed in the presence of a 600fold molar excess of 2c, the reaction rate was first-order dependent on substrate concentration for 1.5 half-lives (R2 = 0.9981, kobs = 3.83(2) M-1min) and only 94% substrate conversion was achieved after 13 h, accompanied by 10% catalyst deactivation via Cp*H abstraction.

155

Chapter 5

0 0

100

200

300

400

500

600

-1 ln([S]t/[S]0)

0.99 M -2

0.42 M -3 0.072 M -4 time (min)

Figure 5-7. Integrated rate plot of the substrate concentration versus time for the 5D-catalyzed oligomerization of 2c at different initial substrate concentrations. The lines connecting the data points represent fitted linear plots (see text for details). These results reveal a lower reaction rate and a lower degree of substrate conversion upon increasing the molar substrate-to-catalyst ratio. This catalytic behavior points to nonproductive metal oxygen interactions that compete with catalytic reation sequences, as discussed previously for the reactions of 3-ethynylthiophene (Section 5.3.3), 2-ethynylthiophene (Section 5.3.4) and 2-ethynylpyridine (Section 5.3.5). No evidence for other products than those expected from catalytic oligomerization was obtained with GC/GC-MS after quenching the above product mixtures with D2O. This observation argues against the occurrence of irreversible catalyst deactivation via other routes than Cp*H abstraction and supports the notion that the above observed catalyst inhibition is due to reversible metal oxygen coordination of the catalyst with substrate and product (Scheme 5-3.).

Table 5-4: The 5D-catalyzed oligomerization of 2-ethynylanisole at different substrate-to-catalyst ratios.a Cat. deact. Entry [2c] kobs 2c 11 12 15 16 (%) (M-1min-1) (equiv.) (M) 1 40 0.072 36.9 63.1 0.0 0.0 20.6(2) 0.6 2 230 0.42 18.7 81.3 0.0 0.0 4.70(3) 5.4 3 593 0.99 10.1 89.9 0.1 0.0 3.83(2) 9.5 c 15.2 4 1108 2.00 4.8 95.2 0.3 0.0 a Reaction conditions: [5D] = 1.6-1.8 mM, benzene-d6 and 25 C. Yields are determined by normalized in situ 1H NMR spectroscopy and represent average values of two or more runs. The experimental error was found to be 0.2. The second-order reaction constants kobs were determined from linear regression analysis. b Catalyst deactivation, calculated from the amount of Cp*H formed (1H NMR). c Not determined.

Effects of substrate-to-catalyst ratios on the regioselectivity As noted before in Section 5.3.3, the selectivity for head-to-head dimerization in the reactions of 5D with 2f increases with increasing substrate concentration (Table 5-4). This trend was also supported by the product distribution of an analogous reaction with a 1200-fold molar excess of 2c, but is not understood at present. Based on the reasonable assumption that substrate and/or product coordination to the catalyst will be favored at relatively high substrate-to-catalyst molar ratios, the relative increase of the sterically hindered 2,1metal insertion mode with increasing initial substrate concentration seems counter intuitive.

156

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes Table 5-5. The substrate effects of the 5A-catalyzed 1-alkyne oligomerization.a Entry Conv. b 2 11 12 15 16 (%) 1 a 2 3 4
S

kobsc (M-1 min-1) 4.5(1) 0.391(4) 0.46(2) 9.1(1)

93.7 0.0 d
N

5.2 51.2 100.0 69.9

0.4 0.0 0.0 6.4

0.7 0.0 0.0 3.2

100 (2 h) 100 (44 h) 74 (72 h) 100 (0.83 h)

0.0 f 20.6

g Reaction conditions: [5A] = 8.8-8.9 mM, substrate (54-55 equiv.), benzene-d6 and 25 C. Yields are determined by normalized in situ 1H NMR spectroscopy and represent average values of two or more runs. The experimental error was found to be 0.2. b Conversion as determined determined by in situ 1H NMR. c The observed second-order reaction constant kobs obtained from linear regression analysis (see text for details).
a

5.3.7.

Experiments with yttrium precatalyst

Introduction The precatalyst Cp*2YCH(SiMe3)3 (5A) was found to be an active and selective catalyst for the formation of the head-to-tail dimer 11a of phenylacetylene (Chapter 4). The formation of the dimer, arising from the sterically preferred 1,2-metal insertion, was rationalized by the small metal size of yttrium relative to lanthanum, thereby increasing the steric interaction between substrate and catalyst during alkyne insertion into the monomeric, alkynyl species. To investigate the substrate effects on the rate and selectivity of a precatalyst Cp*2LnCH(SiMe3)2 that favors head-to-tail dimerization due to steric control, reactions of 5A with representative substrates were performed. Catalytic reactions Even though 1H NMR spectroscopy indicated rapid and quantitative formation of CH2(SiMe3)2, no catalytic reactivity was observed with excess amounts of 2-ethynyltoluene and 2-ethynylanisole (50-200 equiv.). The reaction of 5A with 2-ethynylthiophene (54 equiv.) was significantly slower than the analogous reaction with phenylacetylene (by a factor of 11) and full substrate conversion was observed after 2 days at room temperature. The rate of reaction was first-order in substrate for at least 3 half-lives (kobs = 0.391(4) M-1min-1, t1/2 = 201(2) min), suggesting either rate-limiting alkyne insertion or protonolysis. The product mixture after substrate conversion consisted of the trans-head-to-head dimer 12d and two higher oligomers of unknown structure. The reaction of 5A with excess 2-ethynylpyridine (55 equiv.) was reminescent of the analogous reaction catalyzed by 5D (Section 5.3.5). Three kinetic regions were observed during substrate conversion, but the rate of reaction was lower and significant catalyst deactivation took place at a lower degree of substrate conversion. The first 53 min (27% substrate conversion) correspond to a region of approximately first-order rate dependence in substrate (R2 = 0.9811, t1/2 = 170(7) min, kobs = 0.46(2) M-1min-1) which is followed successively by a region of intermediate kinetic behavior and a region of approximately zero-order rate-dependence in substrate (R2 = 0.9649, kobs = 1.34(5) 10-4 min-1) after 198 min (56% conversion). After monitoring the reaction for three days, a substrate conversion of 74% was found. Quite surprisingly, only the trans-head-to-head dimer (12d) was formed. No evidence for other reaction products than those expected from trans-head-to-head dimerization was obtained with GC/GC-MS analysis after quenching the reaction mixture with D2O. The reaction of 5A with 1-methyl-2-ethynylpyrrole (55 equiv.) was more rapid than the analogous reaction with phenylacetylene. Complete substrate conversion was observed after 50 min at room temperature.

157

Chapter 5

0.5 2f Substrate concentration (mol/L) 0.4 2d 2a 2g

0.3

0.2

0.1

0 0 100 200 300 t (min) 400 500

Figure 5-9. Plot of the substrate concentration versus time for the 5A-catalyzed oligomerization of 2a, 2d and 2f. The curves connecting the data points of 2a, 2d and 2g represent fitted first-order exponentials, while the curve connecting the data points of the reaction with 2f is drawn as a guide for the eye (see text for details). The rate of reaction was first-order in substrate for at least 5 half-lives (kobs = 9.1(1) M-1min-1, t1/2 = 8.5(1) min). The reaction products were identified as 8f:9f:11f:12f in a 20.6:69.9:6.4:3.2 molar ratio, respectively. Substrate effects It seems that the tolerance of Cp*2YCH(SiMe3)2 (5A) towards sterically hindered substrates is low. Indeed, no catalytic activity for the reaction of 5A with -ethynyltoluene (2b) and 2-ethynylanisole (2c) was observed, despite rapid and quantitative protonolysis of 5A by 1-alkyne forming CH2(SiMe3)2. This observation implies that ortho-substituted phenylacetylenes are sterically too crowded for alkyne insertion into the metalcarbon bond of the monomeric alkynyl species Cp*2YCCR (R = C6H4-2-CH3, C6H4-2-OCH3). Another observation is the relative importance of steric and electronic effects in the observed regioselectivity of the 1-alkyne reactions catalyzed by Cp*2YCH(SiMe3)2 (5A). For the reactions of 5A with 2ethynylthiophene (2d), 2-ethynylpyridine (2f) and 1-methyl-2-ethynylpyrrole (2g), major products originating from 2,1-insertion are observed. These results may plausibly be rationalized by a combination of steric and electronic effects in the case of 2d and 2g (Section 5.5.3), while exclusive trans-head-to-head dimerization of 2f seems to be the result of both electronic effects and an interaction between heteroatom and metal center, resulting in a heteroatom-directed reaction (Section 5.5.3). These results provide also additional evidence that the extent of oligomerization is determined by the steric properties of the substrate, as larger relative amounts of oligomers higher than dimers are observed for 2d and 2g as compared to phenylacetylene. The influence of the steric size of the 1-alkyne substituent on the relative rates of 1-alkyne dimerization and trimerization was proposed before for the reactions of Cp*2LaCH(SiMe3)2 (5D) with 2-ethynyltoluene and 2-ethynylanisole relative to that with phenylacetylene (Section 5.3.3). The higer reactivity of 2g relative to phenylacetylene may be explained by its higher (kinetic) acidity, thereby suggesting that protonolysis of a butenyl yttrocene derivative is rate-limiting in the present catalytic dimerization reactions mediated by 5A. The lower reactivities of 2d and 2f are most likely the result of catalyst inhibition, due to reversible, nonproductive metal heteroatom interactions.

158

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

5.4.

Stoichiometric reactions of (hetero)aromatic 1-alkynes

5.4.1.

The reactivity of alkyl and hydride derivatives

Introduction In an effort to understand the substrate effects in the present catalytic 1-alkyne oligomerization reactions, mediated by Cp*2LaCH(SiMe3)2 (5D), stoichiometric reactions with representative substrates were performed. Stoichiometric reactions of 5D with phenylacetylene revealed that the reaction between 5D and the substrate, producing the monomeric alkynyl derivative Cp*2LaCCPh and CH2(SiMe3)2, is slower than that of Cp*2LaCCPh with substrate, giving rise to two regioisomeric insertion products. The monomeric alkynyl species was not observed, as it readily rearranged into a butatrienediyl derivative via a dimeric alkynyl species. This dimeric alkynyl species was observed at low temperature by means of in situ 1H NMR spectroscopy. The butatrienediyl derivative was formed selectively from the reaction of [Cp*2La(-H)]2 with phenylacetylene (2 equiv.) at room temperature.

S Cp*2LaCH(SiMe3)2 5D + 2d S LaCp*2 Cp*2La 22d S + Cp*2La S 24d + S

25 C -CH2(SiMe3)2 S S

(5.1)

+ 11d S S 12d

S Cp*2LaCH(SiMe3)2 5D + 2d

-60 C - CH2(SiMe3)2

S Cp*2La 20d

(5.2)

S LaCp*2 Cp*2La 22d S Cp*2La S

LaCp*2 21d

Reactions with 2-ethynylthiophene The reaction of 5D with 2-ethynylthiophene (2d) was rapid in non-coordinating solvents, such as toluene and hexane, at room temperature, as evidenced by an instantaneous color change from colorless to dark red. 1H NMR experiments in benzene-d6 indicated the formation of a product mixture consisting of the butatrienediyl derivative {(Cp*2La)2[-(C4H3S-2)C4(C4H3S-2)]} (22d), the but-1-en-3-yn-1-yl derivative Cp*2LaC(2-C4H3S)=CHCC(2-C4H3S) (24d) and the organic compounds, (E)-1,4-di(2-thienyl)but-1-en-3-yne (12d) and 2,4-di(2-thienyl)but-1-en-3-yne (11d) (eq. 5.1). The reaction of the hydride derivative [Cp*2La(-H)]2 (25D) with a stoichiometric amount of 2d in benzene-d6 was very clean at room temperature, affording 22d as the only observable product. The identity of 22d was determined unequivocally by single-crystal X-ray diffraction (details are discussed in Section 5.4.6), while NMR analysis in combination with quenching experiments with H2O and D2O

159

Chapter 5

a e
2.20 2.10 2.00 1.90 ppm

Figure 5-10. Variable-temperature 500 MHz 1H NMR spectra of the stoichiometric reaction of 5D and 2d at low temperature in C7D8 (see text for details): (a, upper spectrum) 5D at -60 C, (b) addition of 2d at -60 C, (c) -40C , (d) -20 C, (e, lower spectrum) 25 C. The solvent signal is denoted by an asterisk. provided evidence for the identity of 24d. In accordance with Cp*2LaC(Ph)=C(H)CCPh (24a) discussed previously (Chapter 4), the thienyl derivative 24d exhibited two singlets in the 1H NMR spectrum at 1.98 and 7.35 ppm in a 30:1 ratio, assigned to the Cp* and the =CH group, respectively. When 2-ethynylthiophene (2d) was condensed onto a solution of a stoichiometric amount of 5D in toluene at -78 C, the colorless solution changed into a suspension containing a bright yellow solid. Upon slowly warming at room temperature the bright yellow color of the suspension changed into dark red. Isolation of the isolated dark red crystalline material led to its identification as the butatrienediyl derivative {(Cp*2La)2[-2:2(2-C4H3S)C4-(2-C4H3S)]} (22d). Analogous low temperature 1H NMR experiments in toluene-d8 revealed the presence of two new organometallic species at -60 C, as indicated by two Cp* 1H NMR resonance at 2.22 and 2.00 ppm. The major species exhibiting a Cp* 1H NMR resonance at 2.22 ppm, was identified as the dimeric alkynyl derivative {Cp*2La[(-CC(2-C4H3S)]}2 (21d), while the minor species exhibiting a Cp* 1H NMR resonance at 2.00 ppm, was identified as the butenynyl derivative Cp*2LaC(2-C4H3S)=CHCC(2-C4H3S) (24d). Upon standing at -60 C, 21d converted gradually into the butatrienediyl derivative 22d, exhibiting a Cp* 1H NMR resonance at 2.11 ppm at -60 C (eq. 5.2). The conversion of 21d into 22d was also observed for the phenyl-substituted analogue, albeit less rapidly (Chapter 4). When the reaction mixture was allowed to warm up to room temperature, the Cp* 1H NMR resonances of 22d and 24d shifted upfield (Figure 5-10). These shifts were found to be reversible, as previously observed for the phenyl-substituted analogues 22a and 24a (Chapter 4). Similar to reactions with phenylacetylene, reactions of 24D and [Cp*2La(-H)]2 (25D) with a small excess of 2-ethynylthiophene (1-2 equiv.) at low temperatures resulted in the concomitant formation of a small amount of an unidentified organometallic by-product (~2-5% relative to starting material), tentatively assigned to a oligomeric alkynyl derivative. Concluding remarks These results indicate that the rate of 1-alkyne insertion into the metal-carbon bond of the monomeric, alkynyl derivative relative to protonolysis of 5D by 1-alkyne is more rapid for reactions of 5D with 2-ethynylthiophene (2d) than for reactions of 5D with phenylacetylene (2a). In fact, quantitative conversion of 5D was observed with stoichiometric amounts of phenylacetylene in conjunction with vigorous stirring, while amounts of unreacted 5D were present in the reaction mixtures of 5D with 2d under identical reaction conditions and after addition of small excesses of 2d. It is believed that this difference in behavior can plausibly be ascribed to the smaller size of the thienyl ring versus the phenyl ring, the higher (kinetic) acidity of the acetylenic proton in 2d versus 2a and precoordination of the heteroatom to the metal in the reaction of 5D with 2d (Section 5.5.3).

5.4.2.

The reactivity of the dimeric alkynyl derivatives

Introduction Another substrate effect was observed for the rate of conversion of the dimeric, alkynyl derivative into the corresponding butatrienediyl derivative. Whereas [Cp*2La(-CCPh)]2 (21a) was found to be stable for several hours at -50 C (Chapter 4), {Cp*2La[(-CC(2-C4H3S)]}2 (21d) underwent quantitative C-C coupling

160

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

0.015

Concentration (mol/L)

0.010

21a

21d

21e

0.005

0.000 0 100 200 time (min) 300 400 500

Figure 5-11. Plot of the concentration of the dimeric alkynyl complexes 21 versus time during C-C coupling forming the butatrienediyl complexes 22 in toluene-d8 at -50 C, as monitored by in situ 1H NMR spectroscopy. Curves connecting the data points represent fitted first-order exponentials (see text for details). into the corresponding butatrienediyl derivative within 1 hour at -50 C (eq. 5.2). This difference in behavior may arguably be attributed to a combination of steric (the ring size of the alkyne substituent) and electronic (viz. the inductive/field effect of the alkyne substituent) effects. In an attempt to determine the relative importance of these effects, a kinetic study was performed using 1H NMR spectroscopy at low temperature to establish the rate of C-C coupling for [Cp*2La(-CCR)]2 [R = Ph (21a), 2-C4H3S (21d), 3-C4H3S (21e)] complexes, prepared in situ from Cp*2LaCH(SiMe3)2 (5D) and the appropriate 1-alkyne. The 3-thienyl derivative was included, as the 3thienyl group represents an alkyne substituent having steric properties similar to those of the 2-thienyl group and electronic properties similar to those of the phenyl group (Section 5.5.1). Kinetic study of C-C coupling As noted before, an accurate kinetic analysis was impeded by the low solubility of the dimeric, alkynyl derivative 21 and overlapping 1H NMR resonances of the butatrienediyl derivative with the residual protio resonances of toluene-d8. Nonetheless, the time-dependent conversion of 21 into 22 could be determined by monitoring the decrease of the normalized, Cp* 1H NMR intensity of 21 formed in situ from Cp*2LaCH(SiMe3)2 (5D) and 1-alkyne. Reasonable kinetic data was obtained by normalization against CH2(SiMe3)2 formed in situ from rapid and quantitative protonolysis of 5D by the 1-alkyne. In all cases, the rate of C-C coupling was found to be first-order in 21, consistent with previous results (Chapter 4). Kinetic analysis indicated that the 2-thienyl derivative (21d, R2 = 0.9988, t1/2 = 8.63(12) min) undergoes C-C coupling ~5 times more rapid than the 3-thienyl derivative (21e, R2 = 0.9967, t1/2 = 40.6(4) min), while the 3-thienyl derivative reacts ~3 times more rapid than the phenyl derivative (21a, R2 = 0.9937, t1/2 = 132(2) min). These findings reveal therefore that the rate of C-C coupling in the dimeric alkynyl complexes is dominated by the steric effects of the alkyne substituent rather than the electronic effects.

5.4.3.

The reactivity of amide derivative towards phenylacetylene

In the absence of Lewis bases In marked contrast to the observed reactivity of the alkyl 5D and hydride 25D derivatives towards phenylacetylene, the amide derivative Cp*2LaN(SiMe3)2 (26D) displayed no significant reactivity towards stoichiometric amounts of phenylacetylene at ambient temperatures. Increasing the reaction temperature led to complete substrate conversion into (E)- (12a) and (Z)-1,4-diphenylbut-1-en-3-yne (13a), 1,3,6-triphenylhexa1,5-diyne (15a) and 1,3,6-triphenylhexa-1,2-diene-5-yne (16a), accompanied by 4% conversion of 26D. Clearly, metalation is rate-limiting in the catalytic conversion of substrate. It seems natural to ascribe the low reactivity of the La-N bond in Cp*2LaN(SiMe3)2 (26D) versus the La-C bond in Cp*2LaCH(SiMe3)2 (1D) towards phenylacetylene to the donation of electron density from the nitrogen lone pair into empty metal

161

Chapter 5 orbitals,24 thereby accounting for the higher thermodynamic stability of amide ligands relative to alkyl ligands in rare-earth metal chemistry. An alternative view involves the lower basicity [N(SiMe3)2]- as compared to [CH(SiMe3)2]-.25 Relative strength of Lewis base coordination In order to asses the relative strengths of the Lewis base coordination to the metal center of Cp*2LaCCPh (20a), reactions of Cp*2LaN(SiMe3)2 (26D) with phenylacetylene (1 equiv.) were performed in benzene-d6 at 80 C in the presence of a small excesses (5 equiv.) of thiophene, THF and pyridine. The reactions were monitored after appropriate intervals with 1H NMR spectroscopy. When Cp*2LaN(SiMe3)2 (26D) was heated to 80 in the presence of phenylacetylene (1 equiv.) and pyridine (5 equiv.), complete conversion of phenylacetylene was observed within three days, accompanied by 78% conversion of 26D (eq. 5.3). The reaction mixture consisted of dimers of phenylacetylene, i.e. (E)- (12a) and (Z)-1,4-diphenylbut-1-en-3-yne (13a), and one new lanthanum species, identified as the Lewis base adduct Cp*2LaCCPhNC5H5 (20aNC5H5) on the basis of its independent synthesis from the reaction of Cp*2LaCCPhTHF (20aTHF) and pyridine (vide infra, eq. 5.7).

(5.3)

Cp*2LaN(SiMe3)2 + Ph + N 5

Ph 80 C C 6D 6 -NH(SiMe3)2 Cp*2La 78% N + phenylacetylene dimers

In the presence of THF, complete conversion of phenylacetylene was also observed within three days. In this case, the reaction mixture consisted not only of small amounts of dimers, (Z)- and (E)-1,4-diphenylbut-1en-3-yne, but trimers, 1,3,6-triphenylhexa-1,5-diyne (15a) and 1,3,6-triphenylhexa-1,2-diene-5-yne (16a), as well. Concomitantly, 62% of 26D was converted into several species of which the major components were identified as Cp*2LaCCPh(THF) (20aTHF, 9%) and [(Cp*2La)2(-3:3-PhC4Ph)] (22a, 12%). The presence of significant amounts of Cp*H (18%) suggested that the formed species are thermally not stable at 80 C. In fact, similar Cp* 1H NMR resonances were observed upon heating a benzene-d6 solution of 20aTHF to 80 C and it seems therefore reasonable to ascribe the observed reactivity, including the formation of 22a, to thermolysis of the initially formed 20aTHF (eq. 5.4). The analogous reaction of Cp*2LaN(SiMe3)2 (26D) with phenylacetylene (1 equiv.) in the presence of thiophene (5 equiv.) was identical to the reaction of 26D with phenylacetylene (1 equiv.) at 80 C. Complete conversion of phenylacetylene into 12a, 13a, 15a and 16a was achieved within 1 day at 80 C, accompanied by 4% conversion of 26D.

(5.4)

Cp*2LaN(SiMe3)2 + Ph + O 5

Ph 80 C Cp*2La O

C6D6 -NH(SiMe3)2

phenylacetylene oligomers + decomposition products

62%

Ph

(5.5)

S Cp*2La S LaCp*2 + 2 Ph Cp*2La

LaCp*2 + 2 Ph

162

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Ph

(5.6)

Cp*2La O

The weak interaction between Cp*2LaCCPh and thiophene was furthermore demonstrated by the reaction of [Cp*2La(-2-C4H3S)]226 with phenylacetylene (1 equiv.) and the reaction of Cp*2LaCCPh(THF) (20aTHF) with thiophene. In the former reaction, [(Cp*2La)2(-2:2-PhC4Ph)] (22a) and thiophene were produced cleanly (eq. 5.5), whereas no reactivity was observed between 20aTHF and an excess of thiophene at room temperature (eq. 5.6). Thus, thiophene fails to protect the monomeric alkynyl derivative Cp*2LaCCPh from dimerization and subsequent C-C coupling. The facile displacement of coordinated THF by pyridine in the reaction of Cp*2LaCCPh (20aTHF) with pyridine (1 equiv.) clearly established the preference for pyridine coordination over THF coordination and the kinetic lability of the base adduct (Eq. 5.7). The thermal stability of Cp*2LaCCPh(NC5H5) (20aNC5H5) relative to the analogous THF adduct 20aTHF provides evidence for a strong interaction between pyridine and the lanthanum metal center. This feature is also apparent from the increased tendency to block catalytic activity relative to THF for the reactions of Cp*2LaN(SiMe3)2 (26D) and phenylacetylene in the presence of small excess of Lewis base.

Ph

Ph N + Cp*2La N + O

(5.7)

Cp*2La O

These results clearly indicate that pyridine coordinates stronger to Cp*2LaCCPh than THF and that the interaction between thiophene and Cp*2LaCCPh is relatively weak. Considerable literature precedent exists for this relative order of the coordinating ability of nitrogen-, oxygen-, and sulfur-containing compounds towards electrophilic, hard metal centers.27 More specifically, pyridine is well-known to displace coordinated ether ligands, such as THF and diethyl ether, in rare-earth metal chemistry28 and the interaction of yttrocene metal centers with the soft sulfur atom of thiophene has been shown to be weaker than those with the relatively hard oxygen and nitrogen atoms.29,30

5.4.4.

The reactivity of amide derivatives towards other 1-alkynes

In the absence of exogenous bases The low reactivity of 26D towards 1-alkynes versus alkyl and hydride derivatives allows for the evaluation of substrate effects on the rate of substrate conversion in the absence of large substrate-to-catalyst ratios. In analogy to phenylacetylene, no reactivity was observed for reaction mixtures in benzene-d6 containing Cp*2LaN(SiMe3)2 (26D) and stoichiometric amounts of 2-ethynyltoluene (2b) after several days at room temperature. The analogous reactions of 26D with the heteroaromatic 1-alkynes were considerably more rapid. For example, complete substrate conversion of 2-ethynylanisole was achieved after 4 days, while 1-methyl-2ethynylpyrole (2g) and 2-ethynylthiophene (2d) were completely consumed within 3 and 2 days, respectively. In all cases, ~1% of 26D was consumed. The reaction of 26D with 2-ethynylpyridine (2f) was found to be even more rapid, as 79% substrate conversion was achieved after 70 min. Monitoring the decrease of substrate with in situ 1H NMR spectroscopy revealed that the rate of substrate conversion was zero-order dependent on substrate concentration during the first 53 min (kobs = 13.4(6) M-1min-1, R2 = 0.9909), corresponding to 69% substrate conversion. In spite of a

163

Chapter 5

0 0

25

50

75

100

125

150

2a -1 ln([S]t/[S]0)

2d

2e

-2

-3 time (min)

Figure 5-12. Integrated rate plot of the concentration of 26D versus time for the reaction of 26D with a 5-fold molar excess of 2a, 2d and 2e in THF-d8 at 80 C, as monitored by in situ 1H NMR spectroscopy. Lines connecting the data points represent fitted linear plots (see text for details). dramatic rate decrease after 70 min, complete substrate conversion was apparent within 20 h, accompanied by 1% conversion of 26D and the formation of a multitude of Cp* 1H NMR resonances. This kinetic behavior is analogous to that described above for reactions of Cp*2LnCH(SiMe3)2 (Ln = Y, La) and 2-ethynylpyridine (2f) and was rationalized in terms of product inhibition (Section 5.3.5). Although metalation may arguably be rate-limiting in the catalytic conversion of 1-alkyne mediated by 26D, the large differences of the substrate conversion rates cannot be rationalized by considering the differences in (kinetic) acidity alone. In fact, the (kinetic) acidity of the acetylenic proton of 2-ethynylthiophene is higher than that of 2-ethynylpyridine, while that of 2-ethynylanisole is similar to that of phenylacetylene (Section 5.5.1). Because neither steric nor electronic effects of the 1-alkyne substituent account for the observed differences, it seems that the rate of substrate conversion is dominated by interactions between metal center and substrate instead. In the case of phenylacetylene, substrate reactivity most likely involves an initial (relatively weak) interaction between the metal center and the carbon-carbon triple bond and/or the phenyl moiety.31,32 2Ethynylthiophene and -pyridine, on the other hand, possess a heteroatom, which is capable of interacting stronger with the metal center of the catalyst. It is interesting to note that metal heteroatom interactions seem to promote catalytic substrate conversion in the oligomerization reactions mediated by 26D, whereas catalytic rate depression was observed in the oligomerization reactions of (hetero)aromatic 1-alkynes mediated by Cp*2LaCH(SiMe3)2 (5D) where protonolysis of the catalyst precursor is not rate-limiting.

(5.8)

Cp*2LaN(SiMe3)2 +

N C6D 6

Cp*2La

SiMe3 N SiMe 3 N

Substrate coordination The high observed reactivity of 2-ethynylpyridine (7f) towards Cp*2LaN(SiMe3)2 (26D) relative to other (hetero)aromatic 1-alkynes is remarkable (vide supra) and was rationalized in terms of metal-heteroatom interactions. Evidence for the coordination of 2-ethynylpyridine to the metal center in 26D in benzene-d6 solution is provided by the upfield shifts of the proton NMR resonances of 2f in the presence of 26D relative to

164

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Scheme 5-5. The proposed benifical effect of heteroatom precoordination in the metalation reaction of 2ethynylthiophene by Cp*2LaN(SiMe3)2 (26D) as compared to the analogous reaction with 3ethynylthiophene.
SiMe3 N SiMe 3

Cp*2La S

Cp*2La S

SiMe3 N SiMe 3

Cp*2La + S

SiMe3 N SiMe 3

Cp*2La S

SiMe3 N SiMe 3 Cp*2La -HN(SiMe3)2 THF

Cp*2La S

SiMe3 N SiMe 3

Cp*2La + S

SiMe3 N SiMe 3

Cp*2La S

SiMe3 N SiMe 3 Cp*2La -HN(SiMe3)2 THF O

free 2f ( -0.01, -0.03, -0.09 and -0.09 ppm for -, -, - and -CH, respectively). The proton NMR resonances of pyridine (1 equiv.) were also shifted upfield relative to free pyridine in the presence of 26D ( 0.00, -0.07 and -0.06 ppm for -, - and -CH, respectively). The relatively small observed upfield shifts of the pyridine 1H NMR resonances relative to metal-coordinated pyridine in Cp*2LaCCPhNC5H5 ( -0.02, -0.23 and -0.28 ppm for -, - and -CH, respectively) suggest that the equilibrium between metal-coordinated 2-ethynylpyridine (26DC5H5N) and free 2-ethynylpyridine lies considerably to the side of free 2-ethynylpyridine (eq. 5.8). No evidence for a significant interaction between the metal center in 26D and other heteroaromatic 1-alkynes or heterocycles such as THF and thiophene was provided by 1H NMR spectroscopy. Literature precedent for Lewis base coordination in sterically congested Cp*2LnR environments is supplied by the Lewis adducts Cp*2CeNHtBuNH2tBu,33 Cp*2CeCH(SiMe3)2NCtBu33 and Cp*2YCH(SiMe3)2NC5H5.5b, a Metalation reactions Because the reaction of Cp*2LaN(SiMe3)2 (26D) with phenylacetylene in THF afforded Cp*2LaCCPh(THF), quantitatively, after 5 h at 80 C, it was decided to study the effect of some representative substrates on the metalation rate under comparable reaction conditions. Monitoring the conversion of 26D in the presence of a 5-fold molar excess of substrate in THF-d8 at 80 C by normalized, in situ 1H NMR spectroscopy revealed that the rate of metalation was first-order in 26D for the reaction with phenylacetylene (2a, R2 = 0.9969, kobs = 0.405(4) M-1min-1, t1/2 = 25.4(4) min), 2-ethynylthiophene (2d, R2 = 0.9948, kobs = 0.55(1) M-1min-1, t1/2 = 19.5(4) min) and 3-ethynylthiophene (2e, R2 = 0.9972, kobs = 0.288(3) M-1min-1, t1/2 = 35.3(4) min). Hence, the metalation rate was found to increase in the following order: 2-ethynylthiophene (1.0) < phenylacetylene (1.4) < 3-ethynylthiophene (1.8). The analogous reaction with 2-ethynylpyridine (2f) in THF took place rapidly at room temperature and was not clean. Complete conversion of a 5-fold molar excess of 2f (relative to 26D) was observed into the corresponding trans-head-to-head dimer after 1 h at room temperature, accompanied by 71% conversion of 26D. Smaller excesses of 2f led to lower degrees of conversion for 26D. In contrast to reactions in noncoordinating solvent at room temperature (i.e. benzene, vide supra), the differences in metalation rates for different acetylenic substrates revealed that the effects of substrate-metal interaction and the (kinetic) acidity are relatively modest in THF at 80 C. Even though 2-ethynylthiophene (2d) possesses an acetylenic proton of considerably higher (kinetic) acidity than that of phenylacetylene (Appendix), metalation of 2d by 26D takes place only ~1.3 times more rapid than that of phenylacetylene. The rate of metalation of 3-ethynylthiophene by 26D is ~1.4 times slower than that of phenylacetylene, but the (kinetic)

165

Chapter 5

Scheme 5-6. The preparation of alkynyl THF adducts Cp*2LaCCR(THF).

80 C Cp*2LaN(SiMe3)2 + S R THF , S R Cp*2La O + HN(SiMe3)2

R = 20a.THF 95%

20d.THF 20e.THF 93% 86%

acidity of these 1-alkynes are comparable (Appendix). As a consequence, the relative order of metalation rate cannot be rationalized in terms of (kinetic) acidity alone. It seems natural to explain the high reactivity of 2-ethynylpyridine (2f) and 2-ethynylthiophene (2d) relative to phenylacetylene to heteroatom metal interactions. The interaction of the metal center with the nitrogen or sulfur atom is stronger than that with the phenyl group or the carbon-carbon triple bond and is likely to compete more efficiently with THF for coordination to the metal center. The close proximity of the ethynyl group to the coordinating heteroatom of 2d and 2f may plausibly accelerate the rate of metalation as well (Scheme 5-5). Literature precedent for precomplexation of the metal center to the heteroatom exists for metalation reactions of heterocycles by rare-earth metallocene derivatives.34 In addition, the heteroatom-directed metalation of (hetero)aromatics by lithium35 and aluminum compounds36 is well-established and also believed to involve precomplexation of the metal center to the heteroatom.37,38 The unfavorable position of the ethynyl group relative to the coordinating sulfur atom in 3-ethynylthiophene may account for the lower reactivity of 3ethynylthiophene (2e) relative to phenylacetylene, as sulfur coordination may arguably compete with metalation (proceeding with or without initial coordination to the ethynyl group). Further discussion of this aspect termed heteroatom-assisted reactivity is deferred to a later section (Section 5.5.3).

5.4.5.

The reactivity of Lewis base adducts of monomeric alkynyl derivatives

Introduction Analogous to Cp*2LaCCPh(THF) (20aTHF), the THF adducts Cp*2LaCCR(THF) (20THF) of most of the studied 1-alkynes (except for R = NC5H4-2, vide infra) were accessible through the reactions of Cp*2LaCH(SiMe3)2 (5D) or Cp*2LaN(SiMe3)2 (26D) with the corresponding 1-alkynes HCCR (2) in the presence of excess THF. The interaction between thiophene and lanthanum was found to be too weak to stabilize the metal center electronically in the Cp*2LaCCPh derivative (Section 5.4.3). As a consequence, no thiophene base adducts 20aSC4H4 could be prepared. The high preference for pyridine coordination relative to THF coordination allowed for the preparation of pyridine adducts Cp*2LaCCR(NC5H5) (20NC5H5) from reactions of 20THF and pyridine, but these compounds were also accessible through reactions of 5D or 26D with the corresponding 1-alkynes RCCH in the presence of a small molar excess of pyridine. The kinetic lability of the present lanthanocene derivatives, typical of rare-earth metal complexes, in combination with the high preference for pyridine coordination and the high reactivity of 2-ethynylpyridine (2f), thwarted all attempts to prepare base adducts of Cp*2LaCC(2-NC5H5) (20f) from Cp*2LaN(SiMe3)2 (26D) or Cp*2LaCCPhNC5H5 (20aNC5H5). For example, reactions of 26D with equimolar amounts of 2f in THF afforded product mixtures, containing considerable amounts of the trans-head-to-head dimer of 2f and unreacted 26D (10-30% conversion). Also the presence of pyridine (10-50 equiv.) in these reactions did not inhibit catalytic substrate conversion. The synthesis of THF adducts In this study, the THF adducts Cp*2LaCCPhTHF (20aTHF), Cp*2LaCC(2-C4H3S)THF (20dTHF) and Cp*2LaCC(3-C4H3S)THF (20eTHF) were prepared from Cp*2LaN(SiMe3)2 and the appropriate 1-alkyne (1 equiv.) in THF after heating for 1 day to 80 C under reflux (Scheme 5-6). The adducts were characterized by multinuclear 1D and 2D NMR spectroscopy, but the relatively large nuclear spin and quadrupole moment of 139 La (I = 7/2, 99.91%) did not allow for the observation of the acetylenic carbon NMR resonances. In all cases,

166

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes off-white solids were obtained in virtually quantitative yield after in vacuo removal of volatiles. They were soluble in aromatic solvents and THF, but highly insoluble in aliphatic solvents. Attempts to obtain singlecrystals suitable for X-ray analysis have not been successful. Upon standing at room temperature in the solid state, these compounds rearranged within several days into the corresponding butatrienediyl derivatives. The reactivity of THF adducts When the above adducts 20aTHF, 20dTHF and 20eTHF were allowed to react with an excess of the corresponding 1-alkyne (50 equiv.) at ambient temperatures, no significant reactivity was observed after several hours. However, small amounts of oligomerization products formed upon increasing the reaction temperature, as observed with 1H NMR spectroscopy. The rate of catalytic substrate conversion was found to decrease in the order: 20dTHF > 20aTHF > 20eTHF and seems related to the effects of heteroatom metal interactions, as discussed for the metalation reactions of the corresponding 1-alkynes by Cp*2LaN(SiMe3)2 (26D) (Section 5.4.4). Even though catalytic 1-alkyne conversion was relatively slow for Cp*2LaCCPhTHF (20aTHF) in non-coordinating solvents, transmetallation was found to take place rapidly at ambient temperatures. The addition of HCCR (2) to a benzene-d6 solution of 20aTHF yielded equilibrium mixtures of 20THF, phenylacetylene (2a), Cp*2LaCCRTHF (20aTHF) and 2. No change in composition was observed ~10 min after addition of 2, as seen with 1H NMR spectroscopy. Gratifyingly, 1H NMR spectroscopy could be used to determine the relative amount of species present in the equilibrium after reaction of 20aTHF with a stoichiometric amount of 2-ethynylthiophene (2d) and 3-ethynylthiophene (2e). The integral data provided a value for the equilibrium K which is likely to represent the difference in (equilibrium) acid strength of the particular 1-alkyne versus phenylacetylene towards the active catalyst of the present 1-alkyne oligomerization reaction (eqs. 5.9-5.10).39 Because the relationship between equilibrium acidity and kinetic acidity is commonly not linear in nonpolar solvents, the relative equilibrium acidities can not be used to estimate the kinetic acidities of the 1-alkynes in the present study (Appendix).

(5.9)

Cp*2La O

S + K = 0.08(5) Cp*2La O +

(5.10)

Cp*2La O

Cp*2La K = 0.80(4) O

5.4.6.

The molecular structures of butatrienediyl derivatives

In analogy to [(Cp*2La)2(-PhC4Ph)] (22a) (Chapter 4), representative (hetero)aromatic analogues [(Cp*2La)2(-ArC4Ar)] (C6H4Me-2 (22b) and 2-C4H3S (22d)) were prepared conveniently in high isolated yields (80-90%) from reactions of [Cp*2La(-H)]2 and stoichiometric amounts of the appropriate 1-alkyne in hexane at room temperature. Crystallization at low temperature afforded dark red crystalline solids for the reaction with phenylacetylene (2a), 2-ethynyltoluene (2b) and 2-ethynylthiophene (2d). The red color seems to be typical of these compounds and independent of both metal and alkyne substituent. To investigate the substrate effects on the molecular structure of the butatrienediyl derivatives 22 both in solution and in the solid state, these complexes were characterized by NMR spectroscopy and singlecrystal X-ray crystallography. The 1H and 13C NMR spectral parameters of the lanthanocene butatrienediyl

167

Chapter 5

Table 5-6. NMR spectral parameters of the [(Cp*2La)2(-RC4R)] [R = Ph (22a), C6H4Me-2 (22b) and 2C4H3S (22d)]complexes.a Ph C6H4Me-2 2-C4H3S (22a) (22b) (22d) 1 H NMR 2.06 C5Me5 1.97 C5Me5 2.08 C5Me5 2.33 CH3 6.79 o-CH 6.82 o-CH 6.78 -CH 7.02 p-CH 7.07 CH 6.82 -CH 7.25 m-CH 7.21 CH 7.03 -CH 7.24 m-CH 13 C NMR 11.23 C5Me5 11.69 C5Me5 11.51 C5Me5 21.76 CCH3 120.28 C5Me5 120.43 C5Me5 120.55 C5Me5 126.84 p-CH 126.63 CH 123.50 -CH 129.30 m-CH 126.90 CH 129.00 -CH 131.58 o-CH 131.22 CH 130.07 i-C 131.59 CH 130.88 -CH 141.39 i-C 137.67 i-C 140.18 CCH3 159.07 LaCC 153.78 LaCC 150.09 LaCC 217.31 LaCC 214.24 LaCC 190.84 LaCC a Chemical shifts in ppm and spectra measured in C6D6 at 25 C. Assignments based on 1H-1H and 1H-13C coupling and 2D NMR experiments (for details, see Experimental Section). derivatives [(Cp*2La)2(-RC4R)] [R = Ph (22a), C6H4Me-2 (22b) and 2-C4H3S (22d)] are shown in Table 5-6. The complexes 22a, 22b and 22d exhibit equivalent Cp* resonances in the 1H and 13C NMR spectrum and the 13 C NMR resonances of the butatrienediyl ligand are in the same range as previously observed for [(Cp*2La)2(RC4R)] (R = Me, tBu).48a No evidence for fluxional behavior was obtained with 1H and 13C NMR spectroscopy for the compounds 22a and 22d in the temperature range from -80 C to 100 C, in both aliphatic (i.e. methylcyclohexane-d14) and aromatic (i.e. benzene-d6, toluene-d8) solvents. The complex 22b exhibited no fluxional behavior in toluene-d8 in the temperature range form -80 C to 100 C, but displayed broad 1H NMR resonances at room temperature at 6.70 and 2.21 ppm in methylcyclohexane-d14 solution, assigned to the o-CH and the CH3 protons of the ortho-tolyl substituent, respectively. When the temperature was increased to 85 C, the 1H NMR resonances coalesced, forming a doublet and singlet at 6.70 and 2.23 ppm, respectively. This behavior can reasonably be ascribed to hindered rotation of the ortho-tolyl group. Single crystals of [(Cp*2La)2(-3:3-1,4-Ph2C4)] (22a), {(Cp*2La)2[-2:2-1,4-(C6H4Me-2)2C4] (22b) and {(Cp*2La)2[-2:2-1,4-(2-C4H3S)2C4]} (22d) were obtained by cooling saturated toluene solutions of the corresponding compounds to -40 C. In all cases, toluene solvates were obtained. Although the molecular structure of 22a2C7H8 was already published,48d the X-ray structure determination in this study allowed for a more accurate determination of bond distances and angles corresponding to the butatrienediyl moiety. In addition, a different space group was found as well. It is believed that these arguments justify the use of the present data for comparison with other butatrienediyl derivatives. Complexes 22a2C7H8, 22b2C7H8 and 22d2C7H8 crystallize in the monoclinic space groups P21/m (Z = 2), C2/m (Z = 2) and C2 (Z = 4), respectively. The molecular structures are shown in Figures 5-13, 5-14 and 5-15 and relevant bond lengths and angles are given in Table 5-7. The solid state molecular structures of complexes 22a, 22b and 22d are analogous to those of other butatrienediyl permethyllanthanidocenes (Cp*2Sm)2(-2:2-RC4R) (R = Ph48b 22Ea, CH2CH2Ph48c 22Ej), (Cp*2Ce2)(-2:2-RC4R)48a (R = Me 22Ci, tBu 22Ch), and (Cp*2La)2(-2:2-tBuC4tBu) 22h. 48d Most La-C bond distances are larger and reflect the differences in La3+ versus Sm3+ (r = 0.081 ) and Ce3+ (r = 0.017 ) for eight-coordination ionic radii. The La-C bonds corresponding to the butatriendiyl fragment are not longer, however, probably as a result of the more open La3+ coordination sphere which allows closer approach. Even though complexes 22a40 and 22d41 exhibit rotational disorder in the conformation of the Cp* rings, the Cp* rings were found to be eclipsed for almost all conformations. The Cp* ligands in 22b are in a perfectly eclipsed conformation with a twist angle of 0.0(5) .42 It

168

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Figure 5-13. An ORTEP representation of 22a2C7H8 with thermal ellipsoids drawn at the 50% probability level. The hydrogen atoms and solvate toluene molecules are omitted for clarity. The conformation shown is one of the four conformations found.40

Figure 5-14. ORTEP plot of 22b2C7H8 with thermal ellipsoids drawn at the 50% probability level. The hydrogen atoms and solvate toluene molecules are omitted for clarity.

169

Chapter 5

Figure 5-15. ORTEP plot of 22d2C7H8 with thermal ellipsoids drawn at the 50% probability level. The hydrogen atoms and solvate toluene molecules are omitted for clarity.

Figure 5-16. ORTEP plot of 28DC7H8 with thermal ellipsoids drawn at the 50% probability level. The hydrogen atoms and solvate toluene molecules are omitted for clarity.

170

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Table 5-7. Selected bond lengths () and angles () of the [(Cp*2La)2(-RC4R)] (R = Ph (22a), C6H4Me-2 (22b) and 2-C4H3S (22d)) complexes.a Ph C6H4Me-2 2-C4H3S (22a) (22b) (22d) Bond lengths La1-C27 2.566(1) La-C18 2.548(4) La-C25 2.598(3) La2-C30 2.707(1) La1-C28 2.801(1) La-C19 2.820(4) La-C26 2.790(3) La2-C29 2.819(1) La1-C29 2.912(1) La-C19a 2.9968(4) La-C26a 2.993(3) La2-C28 2.928(1) La1-Cp 2.84(1) La-Cp 2.834(9) La-Cp 2.78(6); 2.80(7) La2-Cp 2.84(3) 2.84(2); 2.80(2) La1-Ct 2.066(2) La-Ct 2.075(2) La-Ct 2.02(1); 2.04(1) La2-Ct 2.063(4) 2.00(1); 2.04(1) C27-C28 1.338(2) C18-C19 1.312(7) C25-C26 1.319(4) C29-C30 1.327(2) C28-C29 1.332(1) C19-C19a 1.349(7) C26-C26a 1.313(4) C21-C27 1.464(2) C18-C17 1.471(7) C24-C25 1.446(4) C30-C31 1.464(1) Bond angles Ct-La1-Ct 146.2(1) Ct-La-Ct 120.4(1) Ct-La-Ct 127.1(2); 128.8(3) Ct-La2-Ct 137.3(3) 127.0(3); 128.8(3) C28-C29-C30 147.2(1) C18-C19148.5(4) C25-C26153.5(3) C19a C26a C27-C28-C29 148.3(1) C29-C30-C31 122.7(1) C19-C18-C17 132.8(4) C24-C25122.8(3) C26 C21-C27-C28 127.0(1) a The average distance between the centroid of the cyclopentadienyl ligand and the metal is denoted by Ct, while the average distance of the metal to the cyclopentadienyl carbons is denoted by La-Cp (see text for details). has been suggested that the presence of eclipsed Cp* ligands is the result of steric crowding,43 but examples of X-ray structures with eclipsed rings in which steric crowding is not evident have been reported as well.44 The overall geometry of {(Cp*2La)2[-2:2-(2-C4H3S)2C4]} (22d) is similar to that of 22a, but differences are found in the butatrienediyl ligand. The 3-bonding of the lanthanum atom in 22a (i.e. La1C30/La2-C29 = 2.578(2) , La1-C29/La2-C28 = 2.810(2) and La1-C28/La2-C29 = 2.920(2) ) is shifted to 2 in 22a (i.e. La-C25 = 2.598(3) , La-C26 = 2.790 and La-C26a = 2.993(2) ). The butatrienediyl fragment is also more linear and symmetrical in 22d than in 22a and the carbon-carbon bond distances therein are shorter in 22d than in 22a. It may be argued that the linearity in the butatrienediyl moiety and the shift from 3 to 2 coordination is brought about by coordination of the sulfur atom in the thienyl group towards the metal center, but the La-S bond length of 3.6267(11) seems too large for a significant interaction. La-S bond distances in lanthanocenes are lacking in literature, but reported La-S bond distances in non-metallocene lanthanum complexes range from 2.89 to 3.24 .45 To discriminate between an interaction between metal center and sulfur atom in {(Cp*2La)2[-2:2-(2-C4H3S)2C4]} (22d) and a substituent effect, involving the electronic and/or steric properties of the 2-thienyl substituent, an X-ray structure determination of [Cp*2La(-2-C4H3S)] (28D) was undertaken. The dimeric, 2-thienyl derivative 28D was prepared from the reaction of [Cp*2La(-H)]2 and thiophene (2 equiv.), as described previously26, and suitable single crystals were obtained by allowing a warm toluene solution of 28D to cool slowly to room temperature. The complex 28D crystallized as a toluene solvate 28DC7H8 in the monoclinic space group C2/c with Z = 4. The molecular structure of 28DC7H8 is shown in Figure 5-16.46 The observed La-S distance of 3.1134(7) confirms the above hypothesis that the La-S distance of 3.6267(11) in 22d2C7H8 is too large for a significant interaction. The molecular solid state structure of {(Cp*2La)2[-2:2-(C6H4Me-2)2C4]} (22b) is similar to those of 22a and 22b. Although the Ct-La-Ct angle of 120.43(8) is considerably smaller than the values previously observed for Cp*2La(X)(Y) complexes (129.6-140.8), the other metrical parameters of the Cp*2La fragment of 22d are normal.47 Similar to 22d, a change from trihapto bonding of the metal center in 22a (i.e. La1-C30/La2-

171

Chapter 5

Scheme 5-7. Selected bond angles () and distances () of the lanthanum butatrienediyl derivatives 22 as a function of the alkyne substituent R.
Cp*2La C R
1 3

C C
2

LaCp*2 22

t Pha 2-MeC6H4 2-C4H3S Bua (22h) (22a) (22b) (22d) La-C1 2.587(2) 2.548(4) 2.598(3) 2.642(3) La-C2 2.810(2) 2.820(4) 2.790(3) 2.761(3) La-C3 2.920(2) 2.9968(2) 2.993(3) 2.912(3) 148.3(1) 148.5(4) 153.5(3) 153.7(3) C1-C2 1.333(2) 1.312(7) 1.319(4) 1.310(4) C2-C3 1.332(1) 1.349(7) 1.313(4) 1.338(4) a For sake of simplicity average values are given (see Table 5-7). b Values taken from Ref. 16d.

C29 = 2.578(2) , La1-C29/La2-C28 = 2.810(2) and La1-C28/La2-C29 = 2.920(2) ) to dihapto bonding in 22d (i.e. La-C18 = 2.548(4) , La-C19 = 2.820(4) and La-C26a = 2.9968(2) ) is observed relative to the parent compound 22a, accompanied by a transition towards a more linear and symmetrical butatrienediyl ligand (i.e. C19-C18-C17 = 132.8(4) in 22b relative to C29-C30-C31/C21-C27-C28 = 124.8(1) in 22a). However, contrary to 22d, the carbon-carbon bond distances in the butatrienediyl ligand are not shorter in 22d as compared to 22a. More specifically, the carbon-carbon bond distance of the former acetylenic carbon-carbon triple bond is considerably longer in 22b (i.e. C17-C18 = 1.471(7) ) than in 22a (i.e. C27-C28/C29-C30 = 1.333(2) ), while the distances of the central carbon-carbon bond in the butatrienediyl ligand are comparable in both complexes. In view of the observed hindered rotation of the ortho-tolyl substituent in solution and the anticipated mild electronic effects upon ortho-methyl substitution (Appendix), it seems natural to ascribe the differences between the molecular structures of 22a and 22b mainly to the increased steric interactions, due to the presence of the ortho-methyl group. In fact, the distance of the carbon atom of the ortho-methyl group to the metal (i.e. La-C11a = 3.189(7) ) and the closest methyl carbon atoms of the Cp* ligand (i.e. C11-C7 = 3.573(8) , C11C6 = 3.938(7) ) and the butatrienediyl ligand (i.e. C11-C19 = 3.024(8) ) fall in the range of 3.0-3.9 and seem too large for significant interactions. Even so, the notion of relatively large steric interactions within the complex is supported by the eclipsed conformation of the Cp* ligands and the exceptionally large Ct-La-Ct angle. The metric parameters of 22d as compared to 22a suggest that butatrienediyl derivatives with a small, -electron-withdrawing alkyne substituent exhibit characteristics associated with a more advanced stage of C-C coupling towards a 1,4-dimetalated butatriene structure, i.e. relatively short C2-C3 and La1-C1 bond distances and a relatively large La1-C3 bond distance and angle (Scheme 5-7). A similar analysis of the molecular structures corresponding to the other structurally analyzed butatrienediyl derivatives, i.e. [(Cp*2La)2(-RC4R)] (R = Ph 22a, 2-MeC4H4 22b, and tBu 22h), is less clear-cut, however.

5.5.

Discussion of substrate effects

5.5.1.

Introduction

Before discussing the substrate effects on the rate and selectivity of the present oligomerization reaction, the properties of the studied 1-alkynes have to be considered first. The oligomerization reactions involve both insertion and protonolysis reactions of 1-alkynes and the intrinsic rate of these reaction sequences is governed by the electronic and steric properties of the 1-alkyne. Several studies have demonstrated that the kinetic acidity of 1-alkynes is (mainly) determined by the inductive/field effects of the substituent and the

172

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes following order of kinetic acidity has presently been inferred from NMR spectroscopy for the 1-alkynes used in this study: 2-ethynylthiophene (2d) > 2-ethynylpyridine (2f) 1-methyl-2-ethynylpyrrole (2g) > 3ethynylthiophene (2b) phenylacetylene (2a) 2-ethynylanisole (2c) 2-ethynyltoluene (Appendix). In Section 5.5.2, it is argued that a combination of both steric and electronic substituent effects determine the kinetics of the reaction of dimeric, bridged alkynyl derivatives [Cp2Ln(-CCR)]2 into butatrienediyl derivatives [(Cp2Ln)2(-RC4R)]. The present study did not provide evidence that the butatrienediyl derivatives play a role in the catalytic oligomerization reactions, however. The catalytic rate and selectivity in the oligomerization reactions were found to be governed by an interplay of steric and electronic 1alkyne substituent effects, substrate and product inhibition and catalyst deactivation. The relative importance of these effects depends on the specific properties of the substrate and is discussed in Section 5.5.3.

5.5.2.

Substrate effects in the reactivity of the dimeric alkynyl derivatives

Introduction The experimental data reported so far indicate that the rearrangement of dimeric alkynyl lanthanidocene derivatives [Cp2Ln(-C CR)]2 (21) into the corresponding butatrienediyl derivatives [(Cp2Ln)2(-RC=C=C=CR)] (22) is determined predominantly by the steric properties of the Cp ligand, the metal Ln and the alkyne substituent R.48 No rearrangements into butatrienediyls have been reported for rare-earth metallocene alkynyls having sterically less demanding cyclopentadienyl groups, such as the unsubstituted or monoalkyl-substituted cyclopentadienyl ligands.49 In marked constrast, rearrangements of dimeric alkynyls into butatrienediyls are commonly observed for pentamethylcyclopentadienyl complexes. These observations have led to the common belief that the bridged alkynyl dimers of pentamethylcyclopentadienyl complexes are sterically too crowded to be stable and that C-C coupling of the alkynyl ligand takes place in order to relieve steric congestion. The decreased steric interaction between the Cp ligands of the two Cp2Ln moieties in the butatrienediyl complexes relative to the dimeric alkynyl complexes is shown in Scheme 5-8. When steric crowding is minimal, stable bridged dimeric alkynyl structures form and rearrangement into butatrienediyl derivatives does not take place. Scheme 5-8. Steric interactions between Cp2Ln moieties in the dimeric alkynyl [Cp2Ln(-CCR)] (21) and butatrienediyl complexes [(Cp2Ln)2(-3:3-RC4R)] (22).
R Ln Ln R Ln R

Ln R

21

22

Alkyne substituent effects on the rate of C-C coupling The present study provides evidence that the alkyne substituent plays also an important role in determining the rate of C-C coupling in rare-earth metallocene alkynyl derivatives. The following observations have been made within a series of lanthanum derivatives [Cp*2La(-C CR)]2 (21) as a function of the alkyne substituent: (i) the methyl analogue [Cp*2La(-C CMe)]2 (22i) rearranged at room temperature (t1/2 = 3.2 h, 25 C) to produce an equilibrium mixture of 21i and [(Cp*2La)2(-MeC4Me)] (22i),48a (ii) the tert-butyl analogue [Cp*2La(-C CtBu)]2 (22h) rearranged much slower (t1/2 = 2.2 h, 50 C), but irreversibly, into 22h and has been isolated at 0 C,48d (iii) the phenyl analogue [Cp*2La(-C CPh)]2 (21a) could only be observed at low temperatures and rearranged both quantitatively and irreversibly into 22a within 13 h at -50 C (t1/2 = 2.2 h, -50 C) (Chapter 4), (iv) the 3-thienyl analogue {Cp*2La[-C C(C4H3S-3)]}2 (21e) rearranged completely and irreversibly into 22e within 3.8 h at -50 C (t1/2 = 0.7 h, -50 C) (Section 5.4.4), and (v) the 2-thienyl analogue {Cp*2La[-C C(C4H3S-2)]}2 (21d) rearranged completely and irreversibly into 22d within 52 min at -50 C (t1/2 = 8.6 min, -50 C) (Section 5.4.4). Because the electronic properties of the 3-thienyl and phenyl substituent are similar (Appendix), the rate of rearrangement in 21 seems to be determined by both the steric and electronic properties of the alkyne substituent.

173

Chapter 5

Scheme 5-9. The proposed transition state of the C-C coupling reaction of a dimeric alkynyl lanthanidocene (21) into a butatrienediyl lanthanidocene (22).
R C C Cp'2Ln C R C

C C Cp'2Ln C R C

R C C Cp'2Ln C C R

LnCp'2

LnCp'2

LnCp'2

21

22

The electronic effects of the methyl and tert-butyl substituent do not differ significantly, as they both are electron-donating via a resonance and an inductive/field mechanism (Appendix).5051 The difference in behavior between Cp*2La(-CCtBu)]2 (22h) and Cp*2La(-CCMe)]2 (22i) can thus plausibly be ascribed to the larger steric requirements of the tert-butyl group relative to the methyl group. The larger size of the methyl group versus the tert-butyl group seems evident and this reasonable assumption is indeed supported by several quantitative scales of steric substituent effects. In spite of extensive research efforts on the steric effects of substituents, a reliable and general method to quantify steric effects is lacking.52 Among the scales proposed, the steric substituent constants ES and are the most widely used and listings of constants are readily available. Taft defined ES as Es = log(k/k0) on the basis of acid-catalyzed hydrolysis of aliphatic esters with k0 for a methyl substituent,53 while Charton calculated the steric parameter from bond lengths and volumes, as defined by Van der Waals radii.54 It seems therefore that sterically hindered groups, such as the tert-butyl group ( = 1.24), retard the rate of rearrangement in dimeric alkynyls 21 as compared to sterically less hindered groups, such as the methyl group ( = 0.52). The steric requirements of the planar phenyl group are difficult to compare with those of the non-planar tert-butyl group in the present molecular environment.55 Attempts to rationalize the observed higher reactivity of [Cp*2La(-C CPh)]2 (21a) relative to [Cp*2La(-C CtBu)]2 (21h) in terms of electronic substituents depend on the relative importance of inductive/field and resonance effects on the rate of C-C coupling in the dimeric alkynyl derivatives. When only field and inductive effects are considered, the phenyl group has a relative strong -electron-withdrawing effect (F = 0.12, I = 0.12), whereas the tert-butyl group is mildly -electron-donating (F = -0.02, I = -0.01). The phenyl group (R = -0.13) is, on the other hand, less electron-donating via resonance effects than the tert-buty group (R = -0.18). Insight into the relative importance of inductive/field and resonance effects on the rate of C-C coupling in the dimeric alkynyl derivatives [Cp*2La(-C CR)]2 (21) can be obtained by taking the 2- and 3thienyl derivatives into account. The 2-thienyl and 3-thienyl substituent have comparable steric sizes and resonance effects, but they differ in their inductive/field effects (Appendix). The high reactivity of {Cp*2La[C C(C4H3S-2)]}2 (21d) as compared to {Cp*2La[-CC(C4H3S-2)]}2 (21d) can thus plausibly be ascribed to the stronger inductive/field effect of the 2-thienyl substituent, thereby revealing that inductive/field effects dominate the electronic substituent effects on the rate of C-C coupling in the dimeric alkynyl derivatives [Cp*2La(-C CR)]2 (21). The higher reactivity of 21d relative to 21a can be attributed to both a smaller and more -electron withdrawing alkyne substituent.56 The view that small, -electron-withdrawing alkyne substituents accelerate the rate of C-C coupling in [Cp*2Ln(-C CR)]2 is also supported by reported data on decamethylsamarocene derivatives. The following order of decreasing stability of [Cp*2Sm(-C CR)]2 was obtained as a function of the alkyne substituent48d: tBu > iPr > CH2CH2Ph CH2CH2iPr > Ph.57 In addition, the metric parameters of 22d as compared to 22a exhibit characteristics associated with a more advanced stage of C-C coupling towards a 1,4-dimetalated butatriene structure, i.e. relatively short C2-C3 and La1-C1 bond distances and a relatively large La1-C3 bond distance and angle (Scheme 5-7). Although mechanistically different,58 evidence has been presented that electronwithdrawing alkyne substituents also accelerate the oxidative C-C coupling of alkynyl ligands in group 4 metal alkynyl derivatives.59 A simple model can be devised which rationalizes the structures of the bridged alkynyls (21) and butatrienediyls (22) in terms of an interaction between the lanthanide metal and the carbon atoms of the alkynyl ligand (Scheme 5-9). The monomeric alkynyl species is unstable and gains electron density by forming a twoelectron, three-atom bond with the -carbon of an alkynyl ligand of another monomeric alkynyl derivative. The

174

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Scheme 5-10. The general mechanistic scenario proposed for the rare-earth metallocene-catalyzed oligomerization of 1-alkynes.
R Cp'2Ln R R R R Cp'2LnC CR R R R R R . R R Cp'2Ln R Cp'2LnC CR 1,2 R Cp'H Cp'Ln(C CR)2 R Cp'2Ln R R Cp'2Ln R R RC CH 2,1 Cp'2Ln R R R R R + R R R LnCp'2 R LnCp'2 Cp'2Ln R R R

asymmetrically bridging, -bonded acetylide dimers thus formed seem to represent a compromise between a symmetric bridge (no interaction) and a ,-bonded structure.49 In the case of permethyllanthanidocenes, the electrophilic lanthanide metal center finds additional stabilization by interacting with the electron-density of the C C bond and/or the -carbon of the alkynide ligand. If this Ln-C interaction is strong enough and not impeded by steric constraints, C-C coupling occurs and 22 forms. Although the electronic effects of the insertion reaction in lanthanide complexes have not been addressed directly, there exists ample indirect evidence that electronwithdrawing substituents at the -position of the concerted, four-center transition state accelerate the insertion process.60 These considerations support the above notion that -electron withdrawing substituent at C promote C-C coupling of the alkynyl ligands in dinuclear, bridged alkynyl complexes to form butatrienediyl complexes. The available experimental data does not warrant detailed speculation concerning the transition state (concerted or stepwise) or the effect of the resonance electronic effects induced by the alkyne substituent.

5.5.3.

Substrate effects in the catalytic 1-alkyne oligomerization reactions

Introduction A general mechanistic scenario for the rare-earth metallocene-catalyzed oligomerization of 1-alkynes has been proposed and is shown in Scheme 5-10 (Chapter 4). It is believed that the substrate effects in the present oligomerization reactions of (hetero)aromatic 1-alkynes can be explained in terms of steric substituent effects, electronic substituent effects, substrate and product inhibition and heteroatom-assisted C-C bond formation. The relative importance of these effects varies with the properties of the 1-alkyne, thereby altering the relative rates of the catalytic reaction sequences in a specific manner. Steric substituent effects The importance of steric effects in both the regioselectivity of dimerization and the degree of oligomerization (i.e. dimerization versus trimerization) of the rare-earth metallocene-catalyzed oligomerization has previously been discussed for reactions of different precatalyst with phenylacetylene (Chapter 4). The dependence of the regioselectivity of dimerization on the metal coordination space was rationalized by a balance

175

Chapter 5

Scheme 5-11. The polarization of the transition state of 1-alkyne insertion into the Ln-C bond of an alkynyl derivative. + R''' Ln
+

R'

R''

between the electronically preferred 2,1-insertion and the sterically preferred 1,2-insertion into the active catalyst Cp*2LaCCPh, affording the trans-head-to-head and the head-to-tail dimer, respectively, after protonolysis by 1alkyne (Scheme 5-10). The relative increase of trimerization with increasing coordination space around the metal center was explained by the reactivity of the reaction intermediate Cp*2LaC(Ph)=C(H)CCPh towards the substrate. Because the steric requirements for protonolysis by 1-alkyne are less demanding than that of insertion of 1-alkyne, protonolysis to form dimers is favored over insertion to form trimers in relatively sterically congested metal centers.15 The results obtained with the sterically more hindered ortho-substituted phenylacetylenes, 2ethynyltoluene (2b) and 2-ethynylanisole (2c), are in accord with the above view. Reactions of Cp*2LaCH(SiMe3)2 (5D) with these substrates afforded less trimer and more head-to-tail dimer as compared to analogous reactions with phenylacetylene (Table 5-1). The reaction of 2c yielded less trimer and more head-totail dimer than that of 2b, most likely due to the larger steric size of 2c.61 In both cases, the rate of reaction was found to be first-order in substrate. The Cp* 1H NMR resonances during the catalytic conversion of 2b were obscured by product resonances, but only two Cp* 1H NMR resonances at 1.98 and 1.87 ppm were observed in the catalytic conversion of 2c, tentatively assigned to the regioisomeric but-1-en-3-yn-1-yl derivatives Cp*2LaC(R)=C(H)CCR and Cp*2LaC(H)=C(R)CCR (R = 2-MeOC6H4). Thus, both the kinetic behavior and the observed reaction intermediates suggest that the oligomerization reactions of 2b and 2c resembles more that of the reaction of Cp*2YCH(SiMe3)2 (5A) with phenylacetylene than that of 5D with phenylacetylene. This finding implies the presence of ortho-substituents affects the relative rates of the catalytic reaction sequences in such a manner, that the intermolecular protonolysis reaction of the butenynyl derivatives with substrate becomes ratelimiting. Electronic substituent effects Electronic effects can be transmitted either through the framework (inductive and field effects) or the system (resonance effects). In principle, electronic 1-alkyne substituent effects may play a role in determining both the rate and selectivity of the present oligomerization reaction. Evidence has been presented that the kinetic acidity of the present 1-alkynes increases with the -electron-withdrawing character of the alkyne substituent. It can thus be anticipated that the rate of protonolysis reaction sequences by the 1-alkyne is accelerated by the presence of more -electron-withdrawing alkyne substituents. However, the kinetic acidity of the 1-alkyne is only expected to influence the oligomerization reaction rate, when protonolysis by 1-alkyne is rate-limiting. Rate-limiting protonolysis was implicated for the reactions of 2-ethynyltoluene and 2ethynylanisole. Both substrates have comparable kinetic acidities and do not provide evidence for the above hypothesis, since the slower reaction rate of 2-ethynylanisole relative to 2-ethynyltoluene can arguably be attributed to metal-oxygen interactions, resulting in product and/or substrate inhibition (Section 5.3.6). Because the extent of oligomerization (i.e. dimerization versus trimerization) is determined by the relative rate of protonolysis by 1-alkyne versus 1-alkyne insertion (Scheme 5-10), dimerization may be promoted relative to trimerization by a more -electron-withdrawing 1-alkyne substituent. The observed behavior of 2ethynylthiophene and 2-ethynylpyridine in the present catalytic reactions supports this view, as they both undergo less trimerization than other substrates of similar or larger steric requirements and lower kinetic acidity (e.g. phenylacetylene, 2-ethynyltoluene and 2-ethynylanisole). Conversely, 3-ethynylthiophene undergoes more trimerization than the more acidic 2-ethynylthiophene, in spite of the comparable steric requirements of their substituents. Because the electronic demands of the transition state for 1-alkyne insertion into the alkynyl derivative favor an electron-withdrawing group at the -position, the regioselectivity of dimerization may, in principle, also be affected by the -electron withdrawing character of the 1-alkyne substituent (Scheme 5-11).60 An unambiguous correlation between the regioselectivity of dimerization and the -electron-withdrawing character of the alkyne substituent is difficult to discern, as other effects, such as heteroatom-assisted C-C bond formation (vide infra), may also influence the observed regioselectivity. Another consequence of a relatively

176

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Scheme 5-12. Electrophilic metal species in which secondary intramolecular interactions are present.

+ Cp2Zr 29 B(C6F5)3 + Cp2Zr 30 R B(4-C6H4F)4 B(C6F5)3

+ Cp2Zr

high substrate acidity concerns catalyst deactivation via Cp*H abstraction and this aspect is discussed in a following section. Substrate and product inhibition The presence of heteroatoms is well-known to decrease catalytic turnover in rare-earth metallocene catalyzed reactions, most likely due to competition for vacant Cp2LnR coordination sites.17 In the present reactions, both intra- and intermolecular heteroatom metal interactions can be envisaged. The catalytic consequences of intermolecular metal heteroatom coordination are believed to be twofold. Firstly, catalyst inhibition by substrate and product is expected, when intermolecular metal heteroatom coordination is competitive with 1-alkyne coordination and/or protonolysis by 1-alkyne. Secondly, heteroatom-assisted C-C bond formation may take place, when the coordination of a properly placed heteroatom with the metal center preceeds substrate reactivity (vide infra). Evidence for catalyst inhibition by substrate and product coordination was provided by the reactions of Cp*2LaCH(SiMe3)2 (5D) with 2-ethynylthiophene (2d) and 2-ethynylpyridine (2f) in the presence of thiophene and pyridine, respectively (Sections 5.3.4 and 5.3.5). In both cases, catalytic rate depression was observed, accompanied by a lower attainable degree of substrate conversion. Kinetic analysis of these reactions revealed also a deviation from first-order rate dependence on substrate concentration at lower degrees of substrate conversion than observed for analogous reactions in the absence of thiophene and pyridine. Deviation from first-order rate dependence on substrate concentration at lower degrees of substrate conversion, accompanied by a lower degree of catalytic substrate conversion, was also observed for reactions of 5D with 3ethynylthiophene (2e) (Sections 5.3.3), 2-ethynylpyridine (2f) (Section 5.3.5) and 2-ethynylanisole (2c) (Section 5.3.6) upon increasing the absolute substrate concentration. Clearly, nonproductive product and substrate coordination to the catalyst compete more effectively with catalytic reaction sequences upon increasing the absolute substrate concentration, thereby lowering both the catalytic rate and the attainable degree of substrate conversion (Scheme 5-5). The extent of catalyst inhibition via metal heteroatom coordination depends also on the relative strength of these noncovalent interactions. Deviation from first-order behavior was observed after 1.2 half-lives (57% substrate conversion) for 2f, after 1.9 half-lives (72% substrate conversion) for 2c and after 5.6 half-lives (97% substrate conversion) for 2e under similar reaction conditions. This order agrees with the well-known coordinating ability of the nitrogen-, oxygen-, and sulfur-containing substrates and products towards the electrophilic metal center, as unambiguously demonstrated by the reactions of Cp*2LaN(SiMe3)2 (26D) with phenylacetylene in the presence of thiophene, THF and pyridine (Section 5.4.4). Substrate and product inhibition was also suggested by the presence of a multitude of Cp* 1H NMR resonances during substrate conversion in the reaction of Cp*2LaCH(SiMe3)2 (5D) with 3-ethynylthiophene (2e) (Sections 5.3.2) and 2-ethynylpyridine (2f) (Section 5.3.6). The absence of products other than expected for the present oligomerization reactions argues for the formation of various Lewis base adducts of the reaction intermediates (Scheme 5-5). This view is supported by the fact that the number of Cp* 1H NMR resonances increased at higher degrees of substrate conversion, when the low substrate concentration can compete less effectively with base adduct formation of the reaction intermediates. In marked contrast, no multitude of Cp* 1H NMR resonances was observed during and after substrate conversion in the 5D-catalyzed reactions of 2ethynylthiophene (2d) and 1-methyl-2-ethynylpyrrole (2g). The former observation can arguably be explained by

177

Chapter 5

Scheme 5-13. The insertion products of styrene M and N into the M-H bond.
M M R M N R

heteroatom-assisted C-C bond formation (rendering nonproductive sulfur metal interactions during catalytic conversion less important) and/or the stability of the but-1-en-3-yn-1-yl derivative (vide infra). Because the nitrogen lone pair in 1-methylpyrrole is delocalized over the five-membered ring, it is not available for interaction with electrophiles.19 The absence of substrate and product inhibition in the oligomerization reactions of 2g is therefore most likely the result of the absence of nonproductive nitrogen metal interactions. This view is supported by similarity of the observations in the 5D-catalyzed reactions of 1-methyl2-ethynylpyrrole (2g) and phenylacetylene (Sections 5.3.2). Also, the observed zero-order rate dependence on the substrate concentration suggests that the kinetics and mechanism of the reaction of 2g is analogous to that of phenylacetylene. Several types of intramolecular interactions can be envisaged in the but-1-en-3-yn-1-yl derivatives formed from insertion into the alkynyl species (Scheme 5-10). The first type of intramolecular metal interaction in the but-1-en-3-yn-1-yl intermediate involves interaction of the metal with the C C bond bond. Analogous interactions in d0 metal complexes have been reported, such as intramolecular alkyne coordination65b (30) and insertion in but-1-en-3-yn-1-yl metallocenes of cationic group 4 metals65c-g (29). Intramolecular alkyne coordination in a pent-1-en-4-yn-1-yl (Chapter 3) and an ynyl-enolate derivative of yttrocene are known as well.62 This intramolecular coordination results in an increased steric hindrance around the metal center and a competition for a vacant coordination site between the intramolecular carbon triple bond and the incoming substrate. As a consequence, protonolysis by 1-alkyne is favored over 1-alkyne insertion in these derivatives for both steric and electronic reasons.63 The influence of the alkyne substituent on the intramolecular coordination of the carbon-carbon triple bond interaction of the but-1-en-3-yn-1-yl derivatives is ambiguous at present.64 The second type of intramolecular metal interaction in the but-1-en-3-yn-1-yl intermediate is associated with the proximal (hetero)aromatic moiety of the alkenyl species. In the case of phenylacetylene, the formed but-1-en-3-yn-1-yl species 24a is likely to be stabilized by an interaction with the -phenyl ring in analogy with -phenylvinyl zirconocene cations65a and -arylethyl rare-earth metal complexes (M) described in literature.66c,d The lower reactivity of the 3-benzyls (M, the 2,1-insertion product) relative to -arylethyl species (N, the 1,2-insertion product) has been explained in terms of a sterically unfavored insertion (Scheme 5-13).67 The stability of these -arylethyl species has been reported to be influenced by both steric and electronic effects. Ortho-substitution has, for example, been shown to result in a destabilization of -arylethyl species, as indicated by the observed shift towards an 1-benzyl in the case of yttrium half-sandwich complexes66d and an increased reactivity (via -H elimination) in the case of a permethyllanthanocene system.66c The rate of styrene hydrosilylation (in which PhSiH3 protonolysis of the -arylethyl species is rate-determining) is, furthermore, decreased by p-F and increased by p- and o-MeO substitution of styrene.42a Attempts to prepare and isolate the but-1-en-3-yn-1-yl derivatives present as catalytic intermediates in the reaction of 5D and the studied (hetero)aromatic 1-alkynes have not been successful. The above considerations are therefore deemed plausible, but tentative at present. Supporting evidence is, however, provided by Cp*2LaC(2-C4H3S)=C(H)CC(2-C4H3S) (24d) as the only observable organometallic species during and after substrate conversion in the reaction of 5D with excess 2-ethynylthiophene (2d) (Sections 5.3.3). The relative stability of the but-1-en-3-yn-1-yl derivative 24d is surprising, considering the relatively high kinetic acidity of 2d in combination with its smaller size and its arguably stronger interaction with the catalyst as compared to phenylacetylene. It seems therefore not unreasonable to ascribe the relative stability of 24d to an interaction of the metal center and the proximal sulfur atom in 24d. The relative importance of the above intramolecular interactions in the but-1-en-3-yn-1-yl intermediates on the observed reactivity is difficult to establish. Even when the but-1-en-3-yn-1-yl derivatives represent the resting state of the catalyst, it is difficult to determine whether this relative stability is due to the above interactions or due to different effects on the relative stability in other catalytic intermediates. Even so, the reactivity of the but-1-en-3-yn-1-yl derivative has been found to be rate-limiting in the catalytic reactions of Cp*2LaCH(SiMe3)2 with 2-ethynyltoluene (2b), 2-ethynylanisole (2d) and 2-ethynylthiophene (2d), all reactions where the above intramolecular interactions in the but-1-en-3-yn-1-yl intermediate can be expected.

178

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Scheme 5-14. The but-1-en-3-yn-1-yl derivatives 24 present during 5D-catalyzed reactions of selected (hetero)aromatic 1-alkynes.
S Cp*2La S O 24a 24c 24d 24f Cp*2La N N

O Cp*2La Cp*2La

Heteroatom-directed C-C bond formation The position of the sulfur atom relative to that of the ethynyl group appears to be pivotal in the observed selectivity and activity in the catalytic reactions of Cp*2LaCH(SiMe3)2 (5D) with 2-ethynylthiophene (2d) and 3-ethynylthiophene (2e). Whereas 2d is consumed relatively rapidly and selectively (98% for the transhead-to-head dimer 12), 2e is converted 6 times slower and less selectively (88% for 12) (Sections 5.3.3). The origin of this behavior is presently believed to be twofold. Firstly, the 3-thienyl group is less electronwithdrawing than the 2-thienyl group (Appendix), thereby accounting for a possibly lower tendency to undergo trans-head-to-head dimerization relative to head-to-tail dimerization and trimerization (vide supra). And secondly, heteroatom precoordination in the reaction of 2d may have benifical effects on both the rate and selectivity of catalytic trans-head-to-head dimerization. Under the reasonable assumption that alkyne insertion into the active catalyst is preceded by coordination to the ethynyl group, precoordination of the catalyst to the sulfur atom of the substrate is likely to favor trans-head-to-head dimerization in the catalytic conversion of 2-ethynylthiophene, due to stabilization of the transition state leading to coordination of the proximal ethynyl group and stabilization of the transition state leading to 2,1-insertion (Scheme 5-15). When precoordination of the active catalyst to the sulfur atom of the substrate takes place in the oligomerization reaction of 3-ethynylthiophene, however, coordination of the distant ethynyl group has to compete with coordination to the sulfur atom and the transition state leading to 2,1-metal insertion will not be stabilized by a heteroatom metal interaction. Literature precedent for precomplexation of the metal center to the heteroatom exists for metalation reactions of heterocycles by rare-earth metallocene derivatives34 and metalation reactions of aromatics by lithium and aluminum compounds.35,36 In fact, the high selectivity for -metalation was proposed to be the result of the stabilization of the corresponding transition state by a heteroatom metal interaction. Precoordination to the sulfur atom in the 5D-catalyzed reactions of 2d and 2e is supported by the observation of several base adducts during and after substrate conversion in the 5D-catalyzed reaction of 2e, whereas only one lanthanocene derivative was observed in the analogous reaction with 2d. Other observations supporting heteroatom-assisted C-C bond formation are (i) the slower metalation of 2e by Cp*2LaN(SiMe3)2 (26D) relative to the metalation of phenylacetylene by 26D, in spite of a comparable (kinetic) acidity (Appendix), and (ii) the higher preference for trans-head-to-head dimerization in the reactions of 2ethynylpyridine (2f) mediated by Cp*2LaCH(SiMe3)2 (5D) and Cp*2YCH(SiMe3)2 (5A), as compared to the analogous reactions with phenylacetylene (Table 5-2). Most significantly, exclusive trans-head-to-head dimerization was observed for the reaction of 5A with 2f (Sections 5.3.7), whereas selective head-to-tail dimerization was observed for the analogous reaction with phenylacetylene (11a:12a:15a:16a = 93.7:5.2:0.4:0.7). Catalyst deactivation Catalyst deactivation leading to the observed formation of Cp*H in the present 1-alkyne oligomerization reactions has been proposed to involve the formation of a dialkynyl species Cp*La(C CR)2 from the reaction of Cp*2LaCCPh and 1-alkyne, possibly followed by ligand rearrangement of Cp*La(C CR)2 into Cp*2La(C CR)2 and La(C CR)3 (Chapter 4). The observed substrate effects on the formation of Cp*H support the view that catalyst deactivation via Cp* abstraction is promoted by increasing acidity of the acetylenic proton of the substrate (Table 5-1).

179

Chapter 5

Scheme 5-15. Proposed catalytic consequences of heteroatom metal coordination in the catalytic oligomerization reactions of 2d and 2e mediated by 5D.
S Cp*2La S Cp*2La S S Cp*2La S S

S Cp*2La S

S Cp*2La S

Cp*2La

S S Cp*2La Cp*2La Cp*2La + S S S S Cp*2La S Cp*2La

S

Cp*2La S S Cp*2La S S

S Cp*2La S

Concluding remarks The present results indicate that the catalytic reactions of Cp*2LaCH(SiMe3)2 (5D) with phenylacetylene (2a) and 1-methyl-2-ethynylpyrrole (2g) involve a rate-limiting pre-equilibrium of a monomeric, alkynyl derivative Cp*2LaCCR (20) with its Lewis base adduct Cp*2LaCCRRCCH (202). As a consequence, the rate of 1-alkyne oligomerization exhibits saturation kinetics. In both cases, high substrate-tocatalyst molar ratios increase the degree of catalyst deactivation via Cp*H abstraction and the rate of 1-alkyne trimerization relative to that of 1-alkyne dimerization. The higher selectivity for trans-head-to-head dimerization of 2g relative to 2a is attributed to the increased -electron-withdrawing character of the 1-methyl-2-pyrrolyl group relative to the phenyl group. The catalytic reactions of Cp*2LaCH(SiMe3)2 (5D) with 2-ethynyltoluene (2b), 2-ethynylanisole (2c) and 2-ethynylthiophene (2d) proceed via rate-limiting protonolysis of the but-1-en-3-yn-1-yl derivative(s). The rate of these oligomerization reactions is first-order in substrate. Catalyst inhibition via heteroatom metal interactions determine the observed substrate reactivity of 2c and 2d, but the catalytic consequences, such as catalytic rate depression at relatively high substrate-to-catalyst molar ratios, deviation from first-order kinetic behavior at relatively high substrate conversion and a low degree of catalytic substrate conversion at relatively

180

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes high substrate-to-catalyst molar ratios, are more pronounced in the former. The higher selectivity for trans-headto-head dimerization of 2d relative to phenylacetylene is attributed to the increased -electron-withdrawing character of the 2-thienyl group relative to the phenyl group. Steric effects dominate the activity and selectivity of the catalytic reactions with 2b and 2c. Hence, a lowered catalytic turnover, accompanied by a lower selectivity for trans-head-to-head dimerization and trimerization, is observed as compared to the analogous reactions with phenylacetylene. The absence of C-O cleavage products in the reactions of 5D with 2c reveal that catalyst reactivity towards the methoxy group cannot compete with the catalytic reaction sequences.68 The 5D-catalyzed reactions of 3-ethynylthiophene (2e) are dominated by substrate and product inhibition. Even though the rate of reaction is initially first-order in substrate, Lewis base adducts of the catalytic intermediates are observed throughout the course of reaction and increasingly so at relatively high substrate conversion. The catalytic consequences of substrate and product inhibition (i.e. catalytic rate depression, deviation from first-order kinetic behavior at higher substrate conversion and a low degree of catalytic substrate conversion) become increasingy important at higher substrate-to-catalyst molar ratios. The lower selectivity for trans-head-to-head dimerization and the higher selectivity for trimerization relative to the analogous reactions with 2-ethynylthiophene and phenylacetylene are attributed to the lower -electron-withdrawing character of the 3-thienyl group relative to the 2-thienyl and phenyl group. The catalytic reactions of Cp*2LaCH(SiMe3)2 (5D) with 2-ethynylpyridine (2g) are characterized by a competition between highly selective trans-head-to-head dimerization and catalyst inhibition via metal-nitrogen interactions. The catalytic consequences of substrate and product inhibition are more pronounced than observed for other heteroaromatic 1-alkynes, due to the well-recognized strong coordination of nitrogen Lewis bases to rare-earth metals. The absence of products other than those expected from trans-head-to-head dimerization argues that the relatively facile ortho-metalation of pyridyl moieties, typical for rare-earth metallocene derivatives,69 cannot compete with the catalytic reaction sequences. In spite of a pronounced catalyst inhibition by substrate and product, catalytic dimerization of 2g is of considerable practical value at relatively low substrate-to-catalyst molar ratios. For example, exclusive trans-head-to-head dimerization and 95% substrate conversion is achieved within 30 min, when a 50-fold molar excess of 2f relative to 5D is used. This finding is deemed quite remarkable, considering the reported failure to prepare an alkynyl decamethylsamarocene derivative of 2-ethynylpyridine16c and the lack of catalytic reactivity of 4-vinylpyridine in the lanthanidocenecatalyzed intermolecular hydroamination reaction.18e

5.6.

Conclusions

A detailed study of he application of selected (hetero)aromatic 1-alkynes in the rare-earth metallocene-catalyzed oligomerization reaction revealed that the catalytic rate and selectivity is governed by an interplay of steric and electronic 1-alkyne substituent effects, substrate and product inhibition and catalyst deactivation. The relative importance of these effects is blieved to vary with the properties of the 1-alkyne. Improved catalytic activity and selectivity for trans-head-to-head dimerization relative to phenylacetylene were observed for 1-alkynes substituted with 2-thienyl, 1-methyl-2-pyrrolyl and 2-pyridyl groups. For all studied (hetero)aromatic 1-alkynes, the catalytic consequences of catalyst inhibition via metal heteroatom interactions, such as a low catalytic reaction rate and incomplete substrate conversion, can be diminished by lowering the substrate-to-catalyst molar ratios at constant precatalyst concentration. The present study demonstrated that the lanthanocene-catalyzed oligomerization reaction can be used to prepare 1,4-substituted but-1-en-3-ynes from (hetero)aromatic 1-alkynes, both rapidly and selectively. The catalyst tolerates sterically hindered aromatic groups and the presence of heteroatoms. Enynes are valuable intermediates in organic synthesis. The practical utility of the lanthanocene-catalyzed trans-head-to-head dimerization is also demonstrated by its successful application in the preparation of novel conjugated polymers (Chapter 6).

5.7.

Experimental section

General considerations. For general remarks and physical and analytical measurements, see Sections 2.7 and 3.5. The compounds Cp*2LaN(SiMe3)270and [Cp*2La(-2-C4H3S)]2,26 Pd(PPh3)2Cl271 and

181

Chapter 5 2-iodo-1-methylpyrrole10,72 were prepared according to literature procedures. Thiophene (distilled over Na) and pyridine (distilled over KOH) were dried as recommended.73 General purification procedure for 1-alkynes. The colored oils or liquids obtained after synthesis or received after purchase were brought in a flask containing freshly ground CaH2 and stirred at room temperature for at least 24 h.73 Subsequent vacuum transfer and passage through a plug of neutral alumina afforded colorless oils which were stored immediately under nitrogen, in the dark and at -30 C. Preparation of substituted (hetero)aromatic 1-alkynes. (a) General procedure. The following procedure is a modification of a general method.10 A 1-L, three-neck, round-bottom flask, equipped with a magnetic stir bar, cooler, and drop funnel was charged with 2-methyl-3-butyn-2-ol (15.0 ml, 155 mmol) and piperidine (100 mL). The mixture was degassed in vacuo. After Pd(PPh3)2Cl2 (0.379 g, 0.540 mmol) and PPh3 (0.278 g, 1.06 mmol) were added the suspension was stirred for 0.5 h at 60 C. Successive addition of a suspension containing CuI (0.289 g, 1.52 mmol) in THF (20 mL) and the aryl halide (107 mmol) was followed by heating and stirring the reaction mixture was heated and stirred overnight under reflux. Vacuum filtration, rotatory evaporation and flash column chromatography (neutral alumina, petroleum ether/diethyl ether) afforded dark-colored oils. Residual piperidine and 2-methyl-3-butyn-2-ol were removed from the product by vacuum distillation using a 20-cm Vigreux column. Subsequent Kgelrohr distillation provided 2-methyl-4-(2-aryl)but-3yn-2-ol as oils. Flash column chromatography (neutral alumina, petroleum ether/diethyl ether) was followed by crystallization from petroleum ether at low temperature affording white needles in most cases. The 2-methyl-4-(2-aryl)but-3-yn-2-ol (4) was deprotected according to a modified general procedure.10 A 100-mL round-bottom flask was equipped with a magnetic stir bar and connected to a 30-cm Vigreux column, condenser and a single receiver which was cooled with liquid nitrogen. The flask was charged with butynol (7.3 g), freshly powdered KOH (2.3 g) and paraffin oil (20 mL). Gradual heating to 200 C under vacuum and stirring yielded a colorless oil as distillate. Acetone was removed from the distillate by vacuum distillation (40 C/60 mmHg), followed by flash column chromatography (neutral activated alumina, petroleum ether), affording a colorless oil. (b) 2-Ethynyltoluene (2b). Crystalllization from petroleum ether at -40 C provided 2-methyl-4-(2methylphenyl)but-3-yn-2-ol (4b) as white needles. Yield: 25.4 g (93%). Flash column chromatography (neutral activated alumina, petroleum ether) afforded 2-ethynyltoluene (2b) as a colorless oil. Yield: 4.62 g (95%). 2-Methyl-4-(2-methylphenyl)but-3-yn-2-ol (4b): 1H NMR (400 MHz, CDCl3, 25 C): 1.62 (s, CH3, 6 H), 2.15 (s, OH, 1 H), 2.40 (s, CH3, 3 H), 7.10 (dt, 3JHH = 7.7 Hz, 4JHH = 2.3 Hz, CH, 1 H), 7.18 (dd, 3JHH = 6.3 Hz, 4JHH = 1.5 Hz, CH, 1 H), 7.19 (dt, 3JHH = 6.3 Hz, 4JHH = 1.5 Hz, CH, 1 H), 7.36 (dt, 3JHH = 7.7 Hz, 4JHH = 1.5 Hz, CH, 1 H). 13C-{1H} NMR (100 MHz, CDCl3, 25 C): 20.56 (CH3), 31.57 (CH3), 65.71 (CMe2), 80.98 (C C), 97.83 (C C), 122.38 (CC C), 125.45 (CH), 128.24, 129.33, 131.78, 140.10 (CCH3). GC-MS, m/z (relative intensity): 174 (M+; 24), 160 (14), 159 (M+ - CH3; 100), 115 (32), 43 (44). IR (KBr, [cm-1]): 3245 (br s), 3064 (m), 2979 (m), 295 (m), 2861 (m), 2221 (m), 1916 (m), 1597 (m), 1483 (m), 1455 (m), 1375 (m), 1359 (m), 1288 (m), 1265 (m), 1199 (m), 1161 (s), 1042 (m), 964 (s), 910 (m). Anal. Calcd. for C12H14O (174.24): C, 82.72%; H, 8.10%. Found: C, 82.85%; H, 8.22%. 2-Ethynyltoluene (2b)74: 1H NMR (300 MHz, CDCl3, 25 C): 2.42 (s, CH3, 3 H), 3.23 (s, CCH, 1 H), 7.1-7.4 (m, CH, 4 H). 1H NMR (300 MHz, C6D6, 25 C): 2.32 (s, CH3, 3 H), 2.92 (s, CCH, 1 H), 6.95-6.80, 7.40-7.48 (m, CH, 4 H). 13C-{1H} NMR (125 MHz, CDCl3, 25 C): 20.55 (CH3), 80.92 (CCH), 82.50 (CCH), 121.91 (CC CH), 125.49, 128.69, 129.42, 132.50 (CH), 140.72 (CCH3). 13C-{H} NMR (125 MHz, C6D6, 25 C): 20.60 (CH3), 81.53 (CCH), 82.79 (CCH), 122.82 (CC CH), 125.82, 128.86, 129.69, 132.83 (CH), 140.85 (CCH3). GC-MS, m/z (relative intensity): 116 (M+; 62), 115 (M+ - H; 100), 89 (12), 74 (4), 63 (12), 51 (5), 50 (5), 39 (6). IR (neat, [cm-1]): 3296 (s), 3061 (m), 3020 (m), 2948 (m), 2920 (m), 2862 (m), 2735 (w), 2103 (m), 1919 (w), 1806 (w), 1666 (w), 1598 (m), 1482 (s), 1455 (s), 1378 (m), 1264 (m), 1107 (m), 1042 (m), 943 (m). (c) 2-Ethynylanisole (2c). Kgelrohr distillation (120-160 C, ~11 mTorr) provided 2-methyl-4-(2methoxyphenyl)but-3-yn-2-ol (4c) as a light-orange oil. Yield: 20.0 g (99%). No suitable crystallization conditions were found. Flash column chromatography (neutral alumina, petroleum ether) afforded 2ethynylanisole (2c) as a light-orange oil. Yield: 7.33 g (92%). 2-Methyl-4-(2-methoxyphenyl)but-3-yn-2-ol (4c): 1H NMR (300 MHz, CDCl3, 25 C): 1.58 (s, CH3, 6 H), 2.34 (s, OH, 1 H), 3.81 (s, CH3, 3 H), 6.83 (dd, 3JHH = 8.4 Hz, 4JHH = 0.9 Hz, CH, 1 H), 6.86 (dt, 3JHH = 6.4 Hz, 4JHH = 1.1 Hz, CH, 1 H), 7.25 (ddd, 3JHH = 8.4 Hz, 3JHH = 7.0 Hz, 4JHH = 1.3 Hz, CH, 1 H), 7.35 (dd, 3 JHH = 7.5 Hz, 4JHH = 1.8 Hz, CH, 1 H). 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 31.49 (CH3), 55.77 (OCH3), 65.69 (CMe2), 78.35 (C C), 97.96 (C C), 110.68 (CH), 111.91 (CC C), 120.36 (CH), 129.65, 133.58, 159.89 (COCH3). GC-MS, m/z (relative intensity): 190 (M+; 33), 175 (M+ - CH3; 52), 159 (M+ - CH3 - O; 27),

182

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes 131 (15), 115 (12), 105 (14), 91 (13), 77 (14), 43 (100). IR (KBr, [cm-1]): 3395 (br s), 2977 (m), 2932 (m), 2834 (m), 2226 (m), 2038 (w), 1595 (m), 1574 (m), 1491 (s), 1462 (m), 1433 (m), 1361 (m), 1293 (m), 1268 (s), 1241 (m), 1161 (m), 1114 (m), 1047 (m), 1023 (m), 961 (m), 907 (m). 2-Ethynylanisole (2c)75: 1H NMR (500 MHz, CDCl3, 25 C): 3.35 (s, CCH, 1H), 3.92 (s, OCH3, 3H), 6.90 (dd, 3JHH = 8.5 Hz, 4JHH = 0.7 Hz, CH, 1H), 6.94 (dt, 3JHH = 7.6 Hz, 4JHH = 0.7 Hz, CH, 1H), 7.34 (dt, 3 JHH = 8.5 Hz, 4JHH = 0.5 Hz, CH, 1H), 7.50 (dd, 3JHH = 7.6 Hz, 4JHH = 1.5 Hz, CH, 1H). 13C{1H} NMR (75 MHz, CDCl3, 25 C): 55.69 (OCH3), 79.98 (C CH), 81.05 (C CH), 110.43 (CH), 110.96 (CC C), 120.24, 130.09, 133.93 (CH), 160.32 (COCH3). 13C-{1H} NMR (75 MHz, C6D6, 25 C): 55.10 (OCH3), 80.52 (CCH), 81.85 (CCH), 110.95 (CH), 112.24 (CC CH), 120.45 (CH), 130.14, 134.35, 161.24 (COCH3). GC-MS, m/z (relative intensity): 132 (M+; 100), 131 (M+ - H; 97), 117 (M+ - CH3; 4), 103 (M+ - H - CO; 21), 89 (M+ - CH3 - CO; 49), 78 (12), 77 (12), 63 (27), 39 (8). IR (neat, [cm-1]): 3281 (s), 3072 (m), 3003 (m), 2960 (m), 2834 (m), 2104 (m), 1595 (s), 1575 (m), 1489 (s), 1464 (m), 1433 (m), 1290 (m), 1278 (m), 1252 (s), 1178 (m), 1162 (m), 1110 (m), 1046 (m), 1023 (m), 937 (m). (d) 2-Ethynylthiophene (2d). Flash column chromatography (neutral alumina, petroleum ether/diethyl ether) was followed by crystallization from petroleum ether at low temperature affording 2-Methyl4-(2-thienyl)but-3-yn-2-ol (4d) as white needles. Yield: 32.84 g (99%). Flash column chromatography (neutral alumina, pentane) afforded 2-ethynylthiophene (2d) as a light-yellow oil. Yield: 7.24 g (79%). 2-Methyl-4-(2-thienyl)but-3-yn-2-ol (4d)76: 1H NMR (300 MHz, CDCl3, 25 C): 1.59 (s, CH3, 6 H), 2.31 (s, OH, 1 H), 6.93 (dd, 3JHH = 3.6 Hz, 3JHH = 5.2 Hz, CH, 1 H), 7.15 (dd, 3JHH = 3.6 Hz, 4JHH = 1.2 Hz, CH, 1 H), 7.21 (dd, 3JHH = 5.2 Hz, 4JHH = 1.2 Hz, CH, 1 H). 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 31.27 (CH3), 65.69 (CMe2), 75.45 (C C), 97.43 (C C), 122.64 (CC C), 126.85 (CH), 126.98, 131.91. GC-MS, m/z (relative intensity): 166 (M+; 35), 152 (10), 151 (M+ - CH3; 100), 123 (12), 108 (10), 45 (10), 43 (59). IR (KBr, [cm-1]): 3237 (br s), 2976 (m), 2924 (m), 2851 (m), 2216 (m), 1625 (m), 1515 (m), 1424 (m), 1400 (m), 1372 (m), 1361 (m), 1255 (m), 1216 (s), 1166 (m), 1154 (m), 1146 (m), 1040 (m), 960 (s), 887 (m), 851 (m), 827 (m), 765 (m). Anal. Calcd. for C9H10OS (166.24): C, 65.03%; H, 6.06%. Found: C, 64.50%; H, 5.95%. 2-Ethynylthiophene (2d)77: 1H NMR (300 MHz, CDCl3, 25 C): 3.22 (s, CCH, 1H), 6.85 (dd, 3JHH = 3 3.7 Hz, JHH = 5.2 Hz, -CH, 1 H), 7.13 (dd, 4JHH = 1.1 Hz, 3JHH = 5.2 Hz, -CH, 1 H), 7.16 (dd, 4J HH = 1.1 Hz, 3 J HH = 3.7 Hz, -CH, 1 H). 1H NMR (500 MHz, C6D6, 25 C): 2.87 (s, CCH, 1 H), 6.42 (dd, 3JHH = 3.7 Hz, 3 JHH = 5.3 Hz, -CH, 1 H), 6.56 (dd, 4JHH = 0.9 Hz, 3JHH = 5.3 Hz, -CH, 1 H), 7.00 (dd, 4JHH = 0.9 Hz, 3JHH = 3.6 Hz, -CH, 1 H). 13C-{1H} NMR (75 MHz, C6D6, 25 C): 77.22 (CCH), 81.88 (CCH), 122.60 (CCCH), 126.96 (-CH), 127.47 (-CH), 133.12 (-CH). GC-MS, m/z (relative intensity): 108 (M+; 100), 82 (M+ - CCH - H; 6), 69 (12), 63 (7), 62 (4), 61 (3), 58 (8), 45 (6). IR (neat, [cm-1]): 3296 (s), 3061 (m), 3020 (m), 2948 (m), 2920 (m), 2862 (m), 2735 (w), 2103 (m), 1919 (w), 1806 (w), 1666 (w), 1598 (w), 1482 (s), 1455 (m), 1378 (m), 1234 (m), 1107 (m), 1042 (m), 943 (m). (e) 3-Ethynylthiophene (2e). Recrystallization from petroleum ether afforded 2-methyl-4-(3thienyl)but-3-yn-2-ol (4e) as light-yellow needles. Yield: 23.50 g (92%). Flash chromatography (neutral alumina, pentane) afforded 3-ethynylthiophene (2e) as a light-yellow oil. Yield: 5.26 g (81%). 2-Methyl-4-(3-thienyl)but-3-yn-2-ol (4e)78: 1H NMR (400 MHz, CDCl3, 25 C): 1.58 (s, CH3, 6 H), 2.16 (s, OH, 1 H), 7.14 (dd, 3JHH = 5.0 Hz, 4JHH = 1.3 Hz, CH, 1 H), 7.22 (dd, 3JHH = 5.0 Hz, 3JHH = 3.0 Hz, CH, 1 H), 7.39 (dd, 3JHH = 3.0 Hz, 4JHH = 1.3 Hz, CH, 1 H). 13C-{1H} NMR (100 MHz, CDCl3, 25 C): 31.42 (CH3), 65.60 (CMe2), 77.25 (C C), 93.30 (C C), 121.66 (CC C), 125.23 (CH), 128.61, 129.82. GC-MS, m/z (relative intensity): 166 (M+; 33), 152 (11), 151 (M+ - CH3; 100), 123 (11), 45 (10), 43 (52). IR (neat, [cm-1]): 3216 (br s), 3101 (m), 2975 (m), 2928 (m), 2500 (m), 2223 (m), 1766 (m), 1570 (m), 1518 (m), 1455 (m), 1405 (m), 1378 (m), 1361 (m), 1259 (m0, 1241 (m), 1205 (m), 1168 (m), 1154 (m), 1077 (m), 1006 (m), 975 (m), 945 (s), 881 (m0, 860 (m). Anal. Calcd. for C9H10OS (166.24): C, 65.03%; H, 6.06%. Found: C, 65.60%; H, 6.50%. 3-Ethynylthiophene (2e)79: 1H NMR (300 MHz, C6D6, 25 C): 2.70 (s, CCH, 1H), 6.62 (dd, 3JHH = 5.0 Hz, 3JHH = 3.0 Hz, -H, 1H), 6.91 (dd, 4JHH = 1.1 Hz, 3JHH = 5.0 Hz, -H, 1H), 7.10 (dd, 4J HH = 1.1 Hz, 3J HH = 3.0 Hz, -H, 1H). 13C-{1H} NMR (75 MHz, C6D6, 25 C): 77.54 (CCH), 79.10 (CCH), 121.69 (CCCH), 125.49 (-CH), 130.07 (/-CH), 130.14 (/-CH). GC-MS, m/z (relative intensity): 110 (5), 109 (9), 108 (M+; 100), 82 (12), 81 (7), 69 (15), 63 (13), 62 (7), 61 (4), 58 (11), 57 (3), 45 (14), 38 (3), 37 (3). IR (neat, [cm-1]): 3290 (s), 3105 (m), 2921 (m), 2851 (m), 2591 (w), 2437 (w), 2292 (w), 2228 (w), 2110 (m), 1762 (m), 1665 (w), 1571 (m), 1513 (m), 1476 (m), 1404 (m), 1357 (m), 1216 (m), 1148 (m), 1078 (m), 1017 (m), 927 (s), 871 (s). (f) 2-Ethynylpyridine (2f). Crystallization from THF/petroleum ether at room temperature afforded 2-methyl-4-(2-pyridyl)but-3-yn-2-ol (4f) as light-yellow crystals. Yield: 28.73 g (81%). Flash chromatography (neutral alumina, hexanes) yielded 2-ethynylpyridine (2f) as a light-yellow oil. Yield: 5.94 g (93%).

183

Chapter 5 2-Methyl-4-(2-pyridyl)but-3-yn-2-ol (4f)80: 1H NMR (400 MHz, CDCl3, 25 C): 1.61 (s, CH3, 6 H), 3.44 (s, OH, 1 H), 7.17 (ddd, 3JHH = 7.6 Hz, 3JHH = 4.9 Hz, 4JHH = 1.2 Hz, -CH, 1 H), 7.35 (ddd, 3JHH = 7.8 Hz, 4 JHH = 1.9 Hz, 5JHH = 1.0 Hz, -CH, 1 H), 7.59 (dd, 3JHH = 7.8 Hz, 3JHH = 7.6 Hz, 4JHH = 1.8 Hz, -CH, 1 H), 8.51 (ddd, 3JHH = 4.9 Hz, 4JHH = 1.8 Hz, 5JHH = 1.0 Hz, -CH, 1 H). 13C-{1H} NMR (100 MHz, CDCl3, 25 C): 31.16 (CH3), 65.09 (CMe2), 81.15 (C C), 94.43 (C C), 122.78 (CH), 127.01 (CH), 136.20 (CH), 142.89 (CC C), 149.64 (CH). IR (neat, [cm-1]): 3205 (br s), 3062 (m), 2975 (m), 2924 (m), 2231 (m), 1873 (m), 1584 (s), 1562 (m), 1466 (s), 1433 (m), 1406 (m), 1370 (m), 1356 (m), 1290 (m), 1275 (m), 1195 (m), 1174 (m), 1152 (m), 1137 (m), 1096 (m), 997 (m), 968 (m). Anal. Calcd. for C10H11NO (161.20): C, 74.51%; H, 6.88%. Found: C, 74.15%; H, 6.93%. 2-Ethynylpyridine (2f)81: 1H NMR (400 MHz, C6D6, 25 C): 3.10 (s, CCH, 1 H), 6.51 (ddd, 3JHH = 3 7.6 Hz, JHH = 4.9 Hz, 4JHH = 1.2 Hz, -CH, 1 H), 6.85 (ddd, , 3JHH = 7.7 Hz, 3JHH = 7.6 Hz, 4JHH = 1.8 Hz, -CH, 1 H), 7.07 (ddd, 3JHH = 7.7 Hz, 4JHH = 1.2 Hz, 5JHH = 1.0 Hz, -CH, 1 H), 8.32 (ddd, 3JHH = 4.9 Hz, 4JHH = 1.8 Hz, , 5JHH = 1.0 Hz, -CH, 1 H). 13C-{1H} NMR (100 MHz, C6D6, 25 C): 77.22 (CCH), 83.60(CCH), 123.10 (CH), 127.44, 135.58, 143.10 (CCCH), 150.27 (CH). GC-MS, m/z (relative intensity): 103 (M+; 100), 76 (42), 75 (9), 74 (10), 52 (11), 51 (14), 50 (30). IR (neat, [cm-1]): 3290 (s, C-H stretching), 3216 (m), 3050 (m), 3002 (m), 2913 (w), 2851 (w), 2108 (m, C C stretching), 1581 (s), 1559 (m), 1460 (s), 1427 (s), 1285 (m), 1243 (m), 1212 (m), 1150 (m), 1090 (m), 1044 (m), 990 (m), 892 (w). (g) 1-Methyl-2-ethynylpyrrole (2g): Crystallization from petroleum ether at low temperature afforded 2-methyl-4-(1-methyl-2-pyrryl)but-3-yn-2-ol (4g) as white needles. Yield: 5.79 g (74%). Flash chromatography (neutral alumina, pentane) yielded 1-methyl-2-ethynylpyrrole (2g) as a light-yellow oil. Yield: 1.02 g (27%). Complete removal of acetone from the thermolabile product was found to be deleterious to the yield. 2-Methyl-4-(1-methyl-2-pyrryl)but-3-yn-2-ol (4g): 1H NMR (400 MHz, CDCl3, 25 C): 1.60 (s, CH3, 6 H), 2.08 (s, OH, 1 H), 6.04 (dd, 3JHH = 3.6 Hz, 3JHH = 2.5 Hz, CH, 1 H), 6.35 (dd, 3JHH = 3.7 Hz, 3JHH = 1.7 Hz, CH, 1 H), 6.60 (dd, 3JHH = 2.5 Hz, 3JHH = 1.7 Hz, CH, 1 H). 13C-{1H} NMR (100 MHz, CDCl3, 25 C): 31.51 (CH3), 34.33 (NCH3), 65.74 (CMe2), 74.14 (C C), 97.59 (C C), 107.83 (CH), 114.48 (CH), 114.95 (CC C), 123.42 (CH). GC-MS, m/z (relative intensity): 163 (M+; 87), 149 (11), 148 (M+ - CH3; 100), 146 (21), 145 (13), 130 (15), 106 (57), 105 (20), 104 (23), 78 (11), 77 (20), 74 (12), 63 (10), 51 (11), 43 (55), 42 (12). IR (KBr, [cm-1]): 3293 (br s), 3106 (m), 2980 (m), 2930 (m), 2805 (w), 2728 (w), 2507 (m), 2214 (s), 1729 (m), 1656 (m), 1532 (m), 1477 (s), 1411 (m), 1378 (m), 1312 (s), 1245 (s), 1215 (s), 1157 (s), 1085 (m), 1051 (m), 1006 (m). Anal. Calcd. for C10H13NO (163.22): C, 73.59%; H, 8.03%. Found: C, 73.50%; H, 8.18%. 1-Methyl-2-ethynylpyrrole (2g)82: 1H NMR (400 MHz, C6D6, 25 C): 3.01 (s, CCH, 1 H), 3.05 (s, NCH3, 3 H), 6.00 (dd, 3JHH = 3.7, 4JHH = 2.7 Hz, -CH, 1 H), 6.16 (dd, 3JHH = 2.7 Hz, 4JHH = 1.7 Hz, -CH, 1 H), 6.56, (dd, 3JHH = 3.7 Hz, 4JHH = 1.7 Hz, -CH, 1 H). 13C-{1H} NMR (100 MHz, C6D6, 25 C): 33.88 (NCH3), 76.37 (CCH), 81.50 (CCH), 108.22 (CH), 115.02 (CCCH), 116.07 (CH), 123.72 (CH). GC-MS, m/z (relative intensity): 106, (9), 105 (M+; 100), 104 (M+ - H; 38), 78 (M+ - HCN; 15), 77 (M+ - HCN - H; 15), 63 (13), 52 (9), 51 (10), 50 (7), 39 (6). IR (neat, [cm-1]): 3290 (s), 3104 (m), 2936 (m), 2729 (w), 2530 (w), 2463 (w), 2099 (s), 1724 (w), 1656 (w), 1585 (w), 1525 (m), 1472 (m), 1419 (m), 1369 (m), 1306 (s), 1230 (m), 1166 (m), 1087 (m), 1051 (m), 1008 (m), 891 (m), 863 (m), 794 (m). Typical NMR-Scale Catalytic 1-Alkyne Oligomerization Reactions Mediated by Cp*2LaCH(SiMe3)2. (a) General procedure. A catalyst stock solution was prepared in benzene-d6. The amount of (pre)catalyst was weighted, while the volume of the solvent was determined using of volumetric glassware. In cases of relatively high substrate-to-catalyst ratios, a specified amount of cyclooctane was added with a microsyringe to the catalyst solution to serve as an internal standard. The catalyst stock solution was transferred into a vial with screw-cap and stored after use at -40C in the glovebox. The density for each substrate was determined experimentally and the volume of the 1-alkyne needed was added with a microsyringe to the catalyst solution in the NMR tube. After substrate addition, the sample was inserted within 5 min (at room temperature) into the probe of the spectrometer and the reaction was followed by single-pulse, in situ 1H NMR spectroscopy, using appropriate long pulse delays (at least 300 s for the acetylenic proton of the substrate) to avoid signal saturation under anaerobic conditions. As soon as the substrate was completely consumed, the reaction mixture was analyzed with quantitative 1H NMR spectroscopy (appropriate long pulse delays and long experiment times so as to obtain reliable proton intensities and signal-to-noise ratios, respectively). Finally, the reaction mixture was quenched with methanol-d4, methanol, H2O or D2O and the final organic products were identified by 1H, 13C, 13C{1H} and 2D NMR, GC, GC-MS and high-resolution mass spectroscopy. In most cases, the products were characterized in

184

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes situ, but in some cases the major product could be isolated by performing several purification steps (i.e. filtration over a plug of neutral alumina using hexanes as eluent to remove inorganic solids, evaporation of volatiles, sublimation of dimers, fractional crystallization, vide infra). (b) Phenylacetylene (2a). As described above, 100.0 L (910.5 mol, 1071 equiv.) of phenylacetylene was converted by Cp*2LaCH(SiMe3)2 (0.85 mol in 500.0 L of benzene-d6) into a mixture of 2,4-diphenylbut-1-en-3-yne (11a), trans-1,4-diphenylbut-1-en-3-yne (12a), 1,3,6-triphenyl-1,5-hexadiyne (15a) and 1,3,6-triphenylhexa-1,2-5-yne (16a). (c) 2-Ethynyltoluene (2b). As described above, 121.1 mg (1043 mol, 519 equiv.) of 2ethynyltoluene was oligomerized by Cp*2LaCH(SiMe3)2 (2.01 mol in 500.0 L of benzene-d6) into a mixture of 2,4-di(2-methylphenyl)-1-en-3-yne (11b), trans-di(2-methylphenyl)but-1-en-3-yne (12b) and 1,3,6-tri(2methylphenyl)-1,5-hexadiyne (15b). 2,4-Di(2-methylphenyll)-1-en-3-yne (11b): 1H NMR (400 MHz, C6D6, 25 C) (CH3 signals not assigned), 5.34 (d, 2JHH = 2.1 Hz, HC(H)=C, 1H), 5.78 (d, 2JHH = 1.1 Hz, HC(H)=C, 1H). 13C{1H} NMR (75 MHz, C6D6, 25 C): 14.33 (CH3), 22.68 (CH3), (C C, CH=CH and Ph not observed). GC-MS, m/z (relative intensity): 232 (M+), 217 (M+ - CH3; 100), 216 (M+ - CH3 - H), 215 (M+ - CH3 - 2H) , 202 (M+ - 2CH3; 100), 115 (M+ - CH3 - C6H4C=CH2). (E)-1,4-Di(2-methylphenyl)but-1-en-3-yne (12b): 1H NMR (400 MHz, C6D6, 25 C): 2.45 (s, CH3, 3H), 2.00 (s, CH3, 3H), 6.31 (d, 2J HH = 16.1 Hz, 1 H), 7.33 (d, 2J HH = 16.1 Hz, 1H), 7.48-7.29 (m, Ar, CH); 13 C{1H} NMR (75 MHz, C6D6, 25 C): 19.53 (CH3), 20.84 (CH3), 91.12 (C CAr), 93.81 (C CAr), 109.78 (=CHCC), 123.92 (Ar, CC C), 125.25, 126.00, 126.51, 128.48, 128.65, 129.80, 130.77, 132.26 (Ar, CH), 135.59 (Ar, CCH3), 135.95 (Ar, CCH3), 138.94 (ArCH=CH), 140.25 (Ar, CCH=CH). GC-MS, m/z (relative intensity): 232 (M+; 100%), 217 (M+ - CH3), 216 (M+ - CH3 - H), 215 (M+ - CH3 - 2H) , 202 (M+ - 2CH3), 115 (M+ - CH3 - C6H4CH=CH or -CH3C6H4CH=CH). HR-MS: C18H16, calc.: 232.12520, found: 206.12453. 1,3,6-Tri(2-methylphenyl)-1,5-hexadiyne (15b): 1H NMR (400 MHz, C6D6, 25 C): (CH3 signals not assigned), 2.87 (dd, 2J HH = 2.8 Hz, 3JHH = 6.9 Hz, CHCH(H), 1H), 4.36 (t, 3J HH = 6.9 Hz, CHCH2, 1H). 13 C{1H} NMR (75 MHz, C6D6, 25 C): 19.21 (CH3), 21.49 (CH3), 22.68 (CH3), 28.33 (CH2CH), 38.57 (CH2CH, C C, and phenyl carbon signals not observed). GC-MS, m/z (relative intensity): 348 (M+), 333 (M+ - CH3), 219 (M+ - CH3C6H4C CCH2; 100%), 203 (M+ - CH3C6H4C CCH2 -CH3 - H), 128 (M+ - CH3C6H4C CCH2 CH3C6H4). (d) 2-Ethynylanisole (2c). As described above, 137.3 mg (1039 mol, 517 equiv.) of 2ethynyllanisole was dimerized by Cp*2LaCH(SiMe3)2 (2.01 mol in 500.0 L of benzene-d6) to a mixture of 2,4di(2-methoxyphenyl)but-1-en-3-yne (11c) and trans-1,4-di(2-methoxyphenyl)but-1-en-3-yne (12c). 2,4-Di(o-methoxyphenyl)but-1-en-3-yne (11c): 1H NMR (400 MHz, C6D6, 25 C) 3.26 (s, OCH3, 3H), 3.32 (s, OCH3, 3H), 5.56 (d, 2JHH = 1.1 Hz, HC(H)=C, 1H), 5.60 (d, 2JHH = 1.1 Hz, HC(H)=C, 1H). 13C{1H} NMR (75 MHz, C6D6, 25 C): 55.00 (OCH3), 55.22 (OCH3), 93.36 (C CAr), 110.62, 110.91, 120.52, 129.52 133.50 (Ar, CH), 157.39 (Ar, COCH3), 160.59 (Ar, COCH3), (other signals unidentified). 13C{1H} NMR (75 MHz, C6D6, 25 C): 55.70 (OCH3), 55.40 (OCH3), 92.15 (C CAr), 92.59 (C CAr), 110.32, 120.07, 120.63 133.23 (Ar, CH), 156.87 (Ar, COCH3), 160.03 (Ar, COCH3), (other signals obscured). GC-MS, m/z (relative intensity): 264 (M+; 100), 249 (M+ - CH3; 6), 234 (M+ - 2CH3; 13), 218 (M+ - 2 CH3 - O; 19), 205 (M+ - 2CH3 O - CH; 22), 202 (15), 189 (M+ - 2 CH3 2 O - CH), 178 (15), 176 (16), 158 (12), 151 (10), 131 (M+ CH3OC6H4C=CH2; 30), 119 (M+ - CH3OC6H4C=CH2 - CH2; 51), 115 (21), 91 (M+ - CH3OC6H4C=CH2 - CH2 CH - CH3; 54%), 76 (12). Scheme 5-16. Numbering scheme of (E)-1,4-di(2-methoxyphenyl)but-1-en-3-yne (12c).
O
2 7

A
1 8F 11 9

B3 C4 D
5

H
10 18 12

13

14

I
15

17

16

O L

(E)-1,4-Di(o-methoxyphenyl)but-1-en-3-yne (12c): 1H NMR (500 MHz, C6D6, 25 C): 3.20 (s, 3 H, A), 3.33 (s, 3 H, L), 6.42 (dd, 3JHH = 8.4 Hz, 4JHH = 0.8 Hz, 1 H, B), 6.45 (dd, 3JHH = 8.4 Hz, 4JHH = 0.8 Hz, 1

185

Chapter 5 H, K), 6.63 (d, 3JHH= 16.4 Hz, 4JHH = 0.5 Hz, 1 H, F), 6.71 (ddd, 3JHH = 7.5 Hz, 3JHH = 7.5 Hz, 4JHH = 1.7 Hz, 1 H, D), 6.71 (ddd, 3JHH = 7.5 Hz, 3JHH = 7.5 Hz, 4JHH = 1.7 Hz, 1 H, I), 6.98 (ddd, 3JHH = 7.4 Hz, 3JHH = 7.4 Hz, 4JHH = 1.6 Hz, 1 H, C), 6.98 (ddd, 3JHH = 7.4 Hz, 3JHH = 7.4 Hz, 4JHH = 1.6 Hz, 1 H, J), 7.21 (d, 3JHH = 7.7 Hz, 1 H, E), 7.50 (d, 3JHH = 7.6 Hz, 1 H, H), 7.65 (d, 3JHH = 16.4 Hz, 1 H, G). 13C{1H} NMR (125.7 MHz, C6D6, 25 C): 54.86 (1), 55.17 (18), 88.92 (11), 94.52 (10), 109.82 (8), 110.96 (16), 111.17 (3), 113.79 (12), 120.61 (14), 120.88 (5), 125.78 (7), 127.10 (6), 129.61 (4, 15), 133.68 (13), 136.58 (9), 157 (2), 160.53 (17). 1H-1H gCOSY (500-500 MHz, C6D6, 25 C): AB, BC, CBD, ED, FG, HI, JIK, KL. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A1, B3, C4, D5, E6, F8, G9, H13, I14, J15, K16, L18. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): B2,4,5,7, C2,3,6,7, D2-4,6,7, E2-5, F2,5-7,10-13,17, G2,7,8,10-13, H11,15,16, I12,13,15-17, J12,13,16,17, K11-14,17. GC-MS, m/z (relative intensity): 264 (M+; 100), 249 (M+ - CH3; 6), 234 (M+ - 2 CH3; 12), 231 (10), 218 (M+ - 2 CH3 O; 16), 205 (19), 202 (12), 189 (20), 178 (13), 176 (13), 131 (M+ - CH3OC6H4C=CH2; 23), 119 (M+ - CH3OC6H4C=CH2 - CH2; 38), 91 (M+ CH3OC6H4C=CH2 - CH2 - CH - CH3; 39). HR-MS: C18H16O2, calc.: 264.11502, found: 264.11402. Scheme 5-17. Numbering scheme of (E)-1,4-di(2-thienyl)but-1-en-3-yne (12d).

A
1

S C
3

4 5

D
6 8 7

B2

10 9

G
11 12

(e) 2-Ethynylthiophene (2d). As described above, 112.0 mg (1036 mol, 515 equiv.) of 2ethynylthiophene was oligomerized by Cp*2LaCH(SiMe3)2 (2.01 mol in 500.0 L of benzene-d6) into a mixture of 2,4-di(2-thienyl)but-1-en-3-yne (11d), trans-di(2-thienyl)but-1-en-3-yne (12d), 1,3,6-tri(2-thienyl)-1,5hexadiyne (15d) and 1,3,6-tri(2-thienyl)-1,2-diene-5-yne (16d). 2,4-Di(2-thienyl)but-1-en-3-yne (11d): 1H NMR (300 MHz, C6D6, 25 C): 5.49 (d, 2JHH = 0.5 Hz, HCH=C, 1H), 5.53 (d, 2JHH = 0.5 Hz, HCH=C, 1H). EI-MS: 216 (M+; 100), 184 (M+ - S; 21), 171 (M+ - S - CH; 70), 139 (M+ - 2S - CH; 12), 108 (M+ - C4H3SC C - H; 11). (E)-1,4-Di(2-thienyl)but-1-en-3-yne (12d): 1H NMR (500 MHz, C6D6, 25 C): 6.08 (d, 3JHH = 15.9 Hz, 1H, E), 6.55 (dd, 3JHH = 3.6 Hz, 4JHH = 1.7 Hz, 1H, C), 6.57 (dd, 3JHH = 4.8 Hz, 3JHH = 3.6 Hz, 1H, B), 6.57 (dd, 3JHH = 4.8 Hz, 3JHH = 3.6 Hz, 1H, G), 6.67 (dd, 3JHH = 4.8 Hz, 4JHH =1.8 Hz, 1H, A), 6.70 (dd, 3JHH = 5.2 Hz, 4 JHH = 1.2 Hz, 1H, H), 6.83 (d, 3JHH = 15.9 Hz, 1H, D), 7.07 (dd, 3JHH = 3.6 Hz, 4JHH = 1.1 Hz, 1H, F). 13C NMR (125 MHz, C6D6, 25 C): 86.00 (8), 93.37 (7), 107.10 (6), 124.11 (9), 125.74 (1), 127.38 (11,2), 127.50 (3), 127.84 (12), 131.89 (10), 134.26 (5), 141.56 (4). 1H-1H gCOSY (500-500 MHz, C6D6, 25 C): AB, BC, DE, FG, GH. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A1, B2, C3, D5, E6, F10, G11, H12. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A2-5, B1,3-5, C1,2,4,9,10, D2-4,6,7, E4,5,8, F8,9,11, G9,10,12, H9-11. HR-MS: C12H8S2, calc.: 216.00674, found: 216.00681. GC-MS, m/z (relative intensity): 216 (M+; 100), 184 (M+ - S; 21), 171 (M+ - S - CH; 6), 139 (M+ - 2 S - CH; 11), 108 (M+ - C4H3SC C - H; 10), 69 (C3HS; 8). 1,3,6-Tri(2-thienyl)-1,5-hexadiyne (15d): 1H NMR (300 MHz, C6D6, 25 C): 2.72 (dd, 3JHH = 6.6 Hz, 2JHH = 2.8 Hz, HCHCH, 1H), 2.72 (dd, 3JHH = 6.6 Hz, 2JHH = 2.8 Hz, HCHCH, 1H), 4.11 (t, 3JHH = 6.6 Hz, HCHCH, 1H). GC-MS, m/z (relative intensity): 324 (M+; 1), 203 (M+ - C4H3SC CCH2; 100), 171 (M+ C4H3SC CCH2 - S; 5), 158 (M+ - C4H3SC CCH2 - S - CH; 5), 145 (M+ - C4H3SC CCH2 - S 2 CH; 3), 121 (C4H3SC CCH2; 7). 1,3,6-Tri(2-thienyl)-1,2-diene-5-yne (16d): 1H NMR (300 MHz, C6D6, 25 C): 3.39 (t, 5J HH = 2.1 Hz, CH=C=CCH2, 2H), 6.29 (m, CH=C=CCH2, 1H). GC-MS, m/z (relative intensity): 324 (M+; 94), 323 (M+ H; 49), 291 (M+ - S - H; 82), 290 (M+ - S 2 H; 100), 277 (M+ - S - 2H - CH; 35), 264 (M+ - S - 2H 2 CH; 27), 258 (M+ - 2 S- 2 H - CH; 68), 245 (M+ - 2 S - 2H - CH; 37), 240 (M+ - C4H3S - H; 59), 208 (M+ - C4H3S - S CH; 29), 171 (24), 145 (29), 69 (C3HS; 27), 45 (HCS; 36). (g) 3-Ethynylthiophene (2e). As described above, 112.8 mg (1043 mol, 519 equiv.) of 3ethynylthiophene was oligomerized by Cp*2LaCH(SiMe3)2 (2.01 mol in 500.0 L of benzene-d6) into a mixture of 2,4-di(3-thienyl)but-1-en-3-yne (11e), trans-1,4-di(3-thienyl)but-1-en-3-yne (12e), 1,3,6-tri(3-thienyl)-1,5hexadiyne (15e) and 1,3,6-tri(3-thienyl)-1,2-diene-5-yne (16e). 2,4-Di(3-thienyl)but-1-en-3-yne (11e): 1H NMR (300 MHz, C6D6, 25 C): 5.49 (d, 2J HH = 0.5 Hz, HCH=C, 1H), 5.53 (d, 2J HH = 0.5 Hz, HCH=C, 1H). GC-MS, m/z (relative intensity): 216 (M+; 100), 184 (M+ S; 21), 171 (M+ - S - CH; 70), 139 (M+ - 2 S - CH; 12), 108 (M+ - C4H3SC C - H; 11).

186

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes (E)-1,4-Di(3-thienyl)but-1-en-3-yne (12e): 1H NMR (300 MHz, C6D6, 25 C): 6.08 (dt, 2JHH = 16.0 Hz, JHH = 0.5 Hz, 7-H, 1H), 6.65 (dq, 4JHH = 2.9 Hz, nJHH = 0.6 Hz, 10-H, 1H), 6.70 (ddd, 3JHH= 5.0 Hz, 4JHH = 3.0 Hz, 7JHH = 0.5 Hz, 4-H, 1H), 6.72 (ddt, 3JHH= 5.1 Hz, 4JHH = 2.9 Hz, nJHH = 0.6 Hz, 12-H, 1H), 6.80 (ddt, 3JHH = 5.1 Hz, 4JHH = 1.3 Hz, nJHH = 0.5 Hz, 11-H, 1H), 6.86 (dq, 3JHH = 16.1 Hz, nJHH = 0.6 Hz, 8-H, 1H), 7.02 (dd, 3 JHH = 5.0 Hz, 4JHH = 1.2 Hz, 2-H, 1H), 7.17 (dd, 3JHH = 3.1 Hz, 4JHH = 1.2 Hz, 4-H, 1H). 13C{1H} NMR (125 MHz, C6D6, 25 C): 87.54 (5-C), 93.35 (6-C), 108.13 (7-C), 123.16 (3-C), 123.88, 124.56, 125.62, 126.45, 128.54, 129.95 (CH), 135.42 (8-C), 139.40 (9-C). GC-MS, m/z (relative intensity): 216 (M+; 100), 184 (M+ - S; 21), 171 (M+ - S - CH; 6), 139 (M+ - 2 S - CH; 11), 108 (M+ - C4H3SC C - H; 10), 69 (C3HS; 8). HR-MS: C12H8S2, calc.: 216.00674, found: 204.00714.
5

Scheme 5-18. Numbering scheme of (E)-1,4-di(3-thienyl)but-1-en-3-yne (12e).

12 11 8 9 10 7 6 5 3 4 2 1

1,3,6-Tri(3-thienyl)-1,5-hexadiyne (15e). 1H NMR (300 MHz, C6D6, 25 C): 2.75 (dd, 3JHH = 6.8 Hz, JHH = 4.5 Hz, HCHCH, 2H), 4.03 (t, 3JHH = 6.6 Hz, HCHCH, 1H). 13C{1H} NMR (75 MHz, C6D6, 25 C): 23.06 (CH2CH), 33.82 (CH2CH), 86.32, 87.18, 92.29, 92.34, 96.47 (C C). GC-MS, m/z (relative intensity): 324 (M+; 1), 203 (M+ - C4H3SC CCH2; 100), 171 (M+ - C4H3SC CCH2 - S; 5), 158 (M+ - C4H3SC CCH2 - S - CH; 5), 145 (M+ - C4H3SC CCH2 - S 2 CH; 3), 121 (C4H3SC CCH2; 7). 1,3,6-Tri(3-thienyl)hexa-1,2-diene-5-yne (16e): 1H NMR (300 MHz, C6D6, 25 C): 3.40 (t, 5JHH = 2.7 Hz, CH=C=CCH2, 2H), 6.42 (tq, 5JHH = 2.7, nJHH = 0.6 Hz, CH=C=CCH2, 1H). 13C{1H} NMR (75 MHz, C6D6, 25 C): 28.63 (CH2), 86.52 (C=C=CH), 87.18, 90.07 (C C), 108.03 (C=C=CH), 207.02 (C=C=CH). GC-MS, m/z (relative intensity): 324 (M+; 94), 323 (M+ - H; 49), 291 (M+ - S - H; 82), 290 (M+ - S - 2H; 100), 277 (M+ - S 2 H - CH; 35), 264 (M+ - S 2 H - 2CH; 27), 258 (M+ - 2 S- 2H - CH; 68), 245 (M+ - 2 S 2 H CH; 37), 240 (M+ - C4H3S - H; 59), 208 (M+ - C4H3S - S - CH; 29), 171 (24), 145 (29), 69 (C3HS; 27), 45 (HCS; 36). (h) 2-Ethynylpyridine (2f). As described above, 36.3 mg (352 mol, 20 equiv.) of 2ethynylpyridine was dimerized by Cp*2LaCH(SiMe3)2 (17.6 mol in 500.0 L of benzene-d6) into trans-1,4di(2-pyridyl)but-1-en-3-yne (12f).
2

Scheme 5-19. Numbering scheme of (E)-1,4-di(2-pyridyl)but-1-en-3-yne (12f).

12 13 14 11 10 9 7 6 5 4 3 2

(E)-1,4-Di(2-pyridyl)but-1-en-3-yne (12f): 1H NMR (300 MHz, C6D6, 25 C): 8.42 (ddd, 3JHH = 4.9 Hz, JHH = 1.8 Hz, 5JHH = 1.0 Hz, 1-H, 1H), 8.35 (dddd, 3JHH = 4.7 Hz, 4JHH = 1.9 Hz, 5JHH = 1.0 Hz, 14-H, 5JHH = 0.4 Hz, 14-H, 1H), 7.33 (ddd, 3JHH = 15.8 Hz, 4JHH = 0.8 Hz, 5JHH = 0.6 Hz, 9-H, 1H), 7.14 (ddd, 3JHH = 7.8 Hz, 4 JHH = 1.1 Hz, 5JHH = 1.0 Hz, 4-H, 1H), 6.99 (ddd, 3JHH = 15.8 Hz, 5JHH = 0.6 Hz, 6JHH = 0.4 Hz, 8-H, 1H), 6.94, 6.92 (ddd, 3JHH = 7.7 Hz, 3JHH = 7.7 Hz, 4JHH = 1.2 Hz, 3/12-H, 1H), 6.60 (ddd, 3JHH = 7.7 Hz, 4JHH = 1.0 Hz, 4JHH = 1.2 Hz, 11-H, 1H), 6.54, 6.53 (ddd, 3JHH= 4.9 Hz, 3JHH = 7.7 Hz, 4JHH = 1.2 Hz, 2/13-H, 1H). 13C{1H} NMR (125 MHz, C6D6, 25 C): 88.77 (6-C), 93.93 (7-C), 112.14 (8-C), 122.55, 122.74, 123.00, 127.29, 135.55, 136.14, 142.16 (py, CH), 144.20 (10-C), 150.03 (py, CH), 150.47 (9-C), 154.08 (5-C). GC-MS, m/z (relative intensity): 206 (M+; 56), 205 (M+ - H; 100), 178 (M+ - H - HCN; 12), 152 (5), 128 (10), 103 (4), 89 (6), 78 (11), 63 (3), 52 (6), 51 (7). HR-MS: C14H10N2, calc.: 206.08439, found: 206.08342. (i) 1-Methyl-2-ethynylpyrrole (2g). As described above, 108.3 mg (1030 mol, 1200 equiv.) of 1methyl-2-ethynylpyrrole was oligomerized by Cp*2LaCH(SiMe3)2 (0.85 mol in 500.0 L of benzene-d6) into a mixture of 2,4-di(1-methyl-2-pyrrolyl)but-1-en-3-yne (11g), trans-1,4-di(1-methyl-2-pyrrolyl)but-1-en-3-yne
4

187

Chapter 5 (12g), 1,3,6-tri(1-methyl-2-pyrrolyl)-1,5-hexa-diyne (15g) and 1,3,6-(1-methyl-2-pyrrolyl)hexa-1,2-diene-5-yne (16g). 2,4-Di(1-methyl-2-pyrrolyl)but-1-en-3-yne (11g): 1H NMR (400 MHz, C6D6, 25 C): 5.65 (s, HCH=C, 1H), 5.62 (s, HCH=C, 1H). GC-MS, m/z (relative intensity): 210 (M+; 100), 209 (M+ - H; 39), 194 (M+ - H - CH3; 17), 193 (M+ - 2H - CH3; 18), 168 (M+ - CN - CH3 - H or - CH3 - HCN; 32), 167 (M+ - 2H - CH3 - CN or - H - CH3 - HCN; 31). (E)-1,4-Di(1-methyl-2-pyrrolyl)but-1-en-3-yne (12g): 1H NMR (400 MHz, C6D6, 25 C): 2.67 (s, A), 3.15 (s, J), 6.14 (ddd, 3JHH = 3.8 Hz, 3JHH = 2.7 Hz, 5JHH = 0.7 Hz, C), 6.14 (dd, 3JHH = 3.8 Hz, 3JHH = 2.7 Hz, H), 6.18 (d, 3JHH= 15.8 Hz, F), 6.18 (dd, 3JHH = 2.7 Hz, 3JHH = 1.6 Hz, B), 6.27 (ddd, 3JHH = 2.7 Hz, 4JHH = 1.6 Hz, 6JHH = 0.5 Hz, I), 6.49 (dd, 3JHH = 3.8 Hz, 4JHH = 1.6 Hz, 4JHH = 0.7 Hz, D), 6.70 (dd, 3JHH = 3.8 Hz, 4JHH = 1.6 Hz, G), 6.77 (ddd, 3JHH = 15.8 Hz, 4JHH = 0.7 Hz, 5JHH = 0.7 Hz, E). 13C{1H} NMR (100 MHz, C6D6, 25 C): 33.16 (1), 34.02 (14), 83.69 (8), 94.11 (9), 103.97 (7), 108.32 (4), 108.79 (3), 109.06 (12), 115.27 (11), 116.92 (10), 123.65 (13), 124.57 (2), 128.54 (6), 131.30 (5). 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): A1, B2, C3, D5, E6, F10, G11, H12. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): A2,5,7, B3-5, C2,4, D2,3,5, E4,7,9, F5,6,9,10, G10,12,13, H11,13, I10-12, J7,10,13. GC-MS, m/z (relative intensity): 210 (M+; 100), 209 (M+ - H; 40), 194 (M+ - H CH3; 16), 193 (M+ - 2H - CH3; 16), 168 (M+ - CN - CH3 - H or - CH3 - HCN; 24), 167 (M+ - 2 H - CH3 - CN or H - CH3 - HCN; 22). HR-MS: C14H14N2, calc.: 210.11569, found: 210.11574. Scheme 5-20. Numbering scheme of (E)-1,4-di(1-methyl-2-pyrrolyl)but-1-en-3-yne (12g).
1

A E
6 7 9 8 10 14
1

B
2

N D
4

C3

11

H
12 13

N J

1,3,6-Tri(1-methyl-2-pyrrolyl)-1,5-hexadiyne (15g): H NMR (400 MHz, C6D6, 25 C): 3.02 (s, NCH3, 3H), 3.10 (s, NCH3, 3H), 3.15 (s, NCH3, 3H), (CH2CH not observed), 3.98 (ddd, 3JHH = 6.8 Hz, 3JHH = 6.7 Hz, 4JHH = 1.1 Hz, HCHCH, 1H). GC-MS, m/z (relative intensity): 315 (M+; 3), 197 (M+ - C4H3N(CH3)C CCH2; 100), 181 (M+ - C4H3N(CH3)C CCH2 - CH3 - H; 6), 154 (M+ - C4H3N(CH3)C CCH2 - CH3 - H - HCN or - C4H3N(CH3)C CCH2 - CH3 2 H - CN; 8), 118 (C4H3N(CH3)C CCH2; 7), 117 (C4H3N(CH3)C CCH2 H; 5). 1,3,6-Tri(1-methyl-2-pyrrolyl)hexa-1,2-diene-5-yne (16g): 1H NMR (400 MHz, C6D6, 25 C): 3.54 5 (dd, JHH = 4.0 Hz, 4JHH = 1.2 Hz, CH=C=CC(H)H, 1H), 3.54 (dd, 5JHH = 4.0 Hz, 4JHH = 1.2 Hz, CH=C=CC(H)H, 1H), (CH=C=CCH2 not observed). GC-MS, m/z (relative intensity): 315 (M+; 31), 314 (M+ - H; 19), 281 (9), 271 (14), 258 (10), 234 (20), 207 (24), 154 (M+ - C4H3N(CH3)C CCH2 - CH3 - H - HCN or - C4H3N(CH3)C CCH2 - CH3 - 2H - CN; 3), 117 (C4H3N(CH3)C CCH2 - H; 4), 42 (10), 32 (26), 28 (10). Typical NMR-Scale Catalytic 1-Alkyne Oligomerization Reactions Mediated by Cp*2YCH(SiMe3)2. (a) General procedure. As described above for Cp*2LaCH(SiMe3)2-catalyzed 1-alkyne oligomerization reactions. A catalyst solution (18.6 mM) was prepared by dissolving 48.3 mg (93.1 mol) in 5.00 mL of benzene-d6. (b) Phenylacetylene. As described above, 52.4 mg (513 mol, 55 equiv.) of phenylacetylene was oligomerized by Cp*2YCH(SiMe3)2 (9.32 mol in 500.0 L of benzene-d6) into a mixture of 2,4-diphenylbut-1en-3-yne (11a), trans-1,4-diphenylbut-1-en-3-yne (12a), 1,3,6-triphenyl-1,5-hexadiyne (15a) and 1,3,6-triphenylhexa-1,2-diene-5-yne (16a). (b) 2-Ethynylthiophene. As described above, 55.2 mg (510 mol, 55 equiv.) of 2-ethynylthiophene was oligomerized by Cp*2YCH(SiMe3)2 (9.32 mol in 500.0 L of benzene-d6) into a mixture of trans-1,4-di(2thienyl)but-1-en-3-yne (9d), an unidentified oligomer and an unidentified heptamer (on the basis of a 1H-1H gCOSY experiment which indicated seven mutually coupled proton resonances). Attempts to isolate or purify the unidentified oligomers on a preparative scale by sublimation, column chromatography (neutral alumina, hexanes) and crystallization failed under an inert atmosphere.

188

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes (c) 2-Ethynylpyridine. As described above, 53.0 mg (514 mol, 55 equiv.) of phenylacetylene was dimerized by Cp*2YCH(SiMe3)2 (9.32 mol in 500.0 L of benzene-d6) into trans-1,4-di(2-pyridyl)but-1-en-3yne (12f). Kinetic Studies of 1-Alkyne Oligomerization Reactions. A catalyst stock solution was prepared by weighing the amount of precatalyst and dissolving the solid in a specified volume of benzene-d6, as determined by volumetric glassware. After preparation, the catalyst solution was transferred into a pre-weighted vial with screw-cap and weighted. For experiments with relatively high molar substrate-to-catalyst ratios, a specified amount of internal standard (cyclooctane) was also added with a microsyringe. After use, the catalyst stock solution was stored at -40 C in the glovebox. A 1H NMR experiment of the sample containing the catalyst solution (prior to substrate addition) ensured the presence of the prerequisite amount of catalyst after long-term storage or handling. In a typical experiment, an NMR tube was charged with 500.0 L of a catalyst stock solution using a 500.0-L microsyringe. The volume of substrate needed for the kinetic experiment was calculated from the density which was determined experimentally and this amount was added with a microsyringe to the catalyst solution. The time period between substrate addition and the start of an arrayed NMR experiment was measured for each experiment and found to be ususally in the range of 6-10 min. The time needed to transfer the sample tube from the glove box into the probe of the Inova-500 or Unity-400 spectrometer was ~5 min after substrate addition. Prior to sample insertion, the probe had been set to the appropriate temperature (T 0.2 C; checked with a methanol temperature standard). Data were acquired using one scan per time interval. Long time intervals (at least 300 s) were used to avoid signal saturation under anaerobic conditions. In most cases, the reaction kinetics were monitored from the intensity changes in the substrate resonance (the acetylenic proton) and in the product resonances over 3 or more half-lives on the basis of acetylene consumption. For experiments involving relatively low substrate-to-catalyst ratios, the substrate concentration was measured from the normalized integral of the acetylenic proton relative to that of CH2(SiMe3)2. The CH2(SiMe3)2 is present as a result of rapid and quantitative protonolytic ligand cleavage during catalyst generation. For experiments involving relatively high substrate-to-catalyst ratios, hydride or butatrienediyl catalyst precursors, the substrate concentration was measured from normalization of the acetylenic substrate signal against that of cyclooctane. The reproducibility of kinetic data, using different batches of substrate and catalyst stock solutions, was within 5%. Typical NMR tube reaction of Cp*2LaNH(SiMe3)2 (26D) with phenylacetylene. Phenylacetylene (2.89 L, 26.3 mol) was added with a microsyringe to an NMR tube containing Cp*2LaN(SiMe3)2 (15.0 mg, 26.3 ml) in benzene-d6. The solution was followed by 1H NMR spectroscopy. No reaction was observed after 10 days at room temperature. When the reaction temperature was increased to 80 C, 1H NMR spectroscopy indicated that phenylacetylene was completely converted within 12 h into trans-1,4-diphenylbut-1-en-3-yne (12a), cis-1,4-diphenylbut-1-en-3-yne (13a) 1,3,6-triphenyl-1,5-hexadiyne (15a) and 1,3,6-triphenylhexa-1,2-5yne (16a). Concomitantly, Cp*2LaN(SiMe3)2 was converted for 4% into HNSi(Me3)2 and unidentified organometallic species. After quenching with H2O, the presence of 12a, 13a, 15a, 16a, Cp*H and HN(SiMe3)2 was indicated by GC/GC-MS. Preparative-scale 1-alkyne oligomerization reactions catalyzed by Cp*2LaCH(SiMe3)2. (a) General procedure. Substrates (0.88-1.76 mmol, molar substrate-to-catalyst ratios of 50-100) were weighted and/or added with a microsyringe to a stirred solution of Cp*2LaCH(SiMe3)2 (10.0 mg, 17.6 mol) in hexane (5.0 mL) in a Schlenk vessel. After stirring for 2 h at room temperature, the reactions were quenched by exposing the reaction mixtures to air. The crude product mixtures were filtered through a plug of neutral alumina (hexanes as eluent) to remove inorganic residues. After solvent removal by rotatory evaporation, the trans-headto-head dimers were conveniently isolated by vacuum sublimation or fractional crystallization. (b) 2-Ethynylanisole. Substrate (1.86 g, 14.1 mmol, 800 equiv.) was added to a hexane solution of Cp*2LaCH(SiMe3)2 and stirred for 1 day at room temperature. After above work-up, trans-1,4-di(2methoxyphenyl)but-1-en-3-yne (12c) was isolated as light-yellow crystals from a methanol/benzene solution at low temperature. Yield: 781.3 mg (42%). (c) 2-Ethynylthiophene. Substrate (158.0 mg, 1.46 mmol, 83 equiv.) was added to a hexane solution of Cp*2LaCH(SiMe3)2. After above work-up, trans-1,4-di(2-thienyl)but-1-en-3-yne (12d) was isolated as offwhite solid by vacuum sublimation (80 C, 1 mmHg). Yield: 115.3 mg (73%) (d) 3-Ethynylthiophene. Substrate (100.5 mg, 929 mol, 53 equiv.) was added to a hexane solution of Cp*2LaCH(SiMe3)2. After above work-up, trans-1,4-di(3-thienyl)but-1-en-3-yne (12e) was isolated as colorless crystals by slow evaporation of methanol onto a benzene solution at room temperature. Yield: 55.4 mg (55%)

189

Chapter 5 NMR tube reaction of [(Cp*2La)2(-3:3-PhC4Ph)] (22a) with THF. THF (0.50 L, 6.2 mol) was added with a microsyringe to an NMR tube containing [(Cp*2La)2(-3:3-PhC4Ph)] (6.2 mg, 6.1 mol) in benzene-d6. The reaction was followed for 16 h at room temperature, but no change was observed. Similarly, the addition of excess THF (4.5 L, 56 mol) did not result in a change in the reaction mixture as observed with 1H NMR spectropy after 1 day at room temperature. When the reaction mixture was increased to 80 C, a color change from dark-red to light-yellow was observed after 1 h. After 11 h, 1H NMR spectroscopy indicated the presence of Cp*2LaCCPh(THF) (by comparison with an authentic sample, vide infra) and unidentified Cp* 1H NMR resonances. D2O was added to reaction mixture and GC/GC-MS analyis indicated the presence of phenylacetylene-d1, Cp*D and small amounts of unidentified compounds (e.g. m/z = 107 and 205). Thermolysis of Cp*2LaCCPh(THF). A solution of Cp*2LaCCPh(THF) (22.6 mg, 38.3 mol) in benzene-d6 was heated to 100 C and the reaction was followed in situ with 1H NMR spectroscopy. After 8 min Cp*2LaCCPh(THF) was converted for 56% into [(Cp*2La)2(-3:3-PhC4Ph)] and for 64% after 138 min. Further heating to 100 C did not led to a significant change, but additional unidentified Cp* 1H NMR resonances were observed after 6 h at 100 C. Aftter 12 h the mixture was allowed to cool to room temperature and red crystal formed upon standing at room temperature. Decantation and NMR analysis in benzene-d6 led to the identification of the red crystals as [(Cp*2La)2(-3:3-PhC4Ph)]. D2O was added to remaining part of the reaction mixture and GC/GC-MS analyis indicated the presence of phenylacetylene-d1, Cp*D and small amounts of unidentified compounds (e.g. m/z = 107 and 205) similar to those observed in the reaction of [(Cp*2La)2(3:3-PhC4Ph)] with excess THF at 80 C (vide supra). Kinetics of the C-C coupling reaction in [Cp*2La(-CCR)]2 complexes. A stock solution was prepared by dissolving Cp*2LaCH(SiMe3)2 (80.0 mg, 141 mol) in toluene-d8 (4.00 mL). The solution was transferred into a pre-weighted vial with screw-cap and weighted. In a typical experiment, an NMR tube was charged with 500.0 L of the stock solution using a 500.0-L microsyringe. The NMR tube was connected to the high-vacuum line and a specified volume of substrate was condended onto the solution at -196 C. After substrate addition, the tube was sealed off in vacuo and kept at -100 C in a cooling bath of ethanol and liquid nitrogen. Prior to sample insertion, the tube was shaken and warmed quickly. When a homogeneous liquid formed, the sample was inserted into the probe of the Inova-500 spectrometer which had previously been set to 50 C (T 0.2 C; checked with a methanol temperature standard). Data were acquired using one scan per time interval. Long time intervals (at least 300 s) were used to avoid signal saturation under anaerobic conditions. The reaction kinetics were monitored from the normalized intensity changes in the Cp* resonance of the dimeric, alkynyl derivative [Cp*2La(-CCR)]2. In all cases, instantaneous formation of CH2(SiMe3)2 was observed and the intensity of its SiMe3 1H NMR resonance was used as an internal reference. The reactions were performed at least two times and the reproducibility of the kinetic data was within 5%. NMR tube reaction of [(Cp*2La(-2-C4H3S)]2 and phenylacetylene. Phenylacetylene (2.26 L, 20.6 mol) was added with a microsyringe to an NMR tube containing a suspension of [(Cp*2La)2(-2-C4H3S)] (20.3 mg, 20.6 mol) in benzene-d6. Upon addition of phenylacetylene, the sample was shaken quickly and the off-white suspension changed instantaneously into a dark-red solution. The resulting mixture was shown to consist of [(Cp*2La)2(-3:3-PhC4Ph)] and thiophene by NMR spectroscopy and dark-red crystals formed upon standing at room temperature. Upon addition of excess H2O (2.0 L, 0.1 mmol), the dark-red suspension turned into a light-yellow solution and GC/GC-MS analyis indicated the presence of trans-1,4-diphenylbutariene (27a), thiophene and Cp*H. Transmetalation reactions of Cp*2LaCCPh(THF) with 2- and 3-ethynylthiophene. (a) General procedure. A stock solution was prepared by dissolving Cp*2LaCCPh(THF) (151.8 mg, 262.4 mol) and hexamethyldisiloxane (85.0 L, 400 mol) in benzene-d6 (3.00 mL). The solution was transferred into a vial with screw-cap and stored at -30 C after use. A Teflon-capped NMR tube was charged with 500.0 L of stock solution using a 500.0-L microsyringe. The prerequisite amount of 1-alkyne was added with a microsyringe to the solution of Cp*2LaCCPh(THF). The resulting mixture was analyzed with NMR spectroscopy and the relative amount of species present could be determined by quantitative 1H NMR spectroscopy (pulse delays of 50 s to avoid signal saturation under anaerobic conditions and long experiment times in order to obtain good signal-tonoise ratios) using hexamethyldisolxane as an internal standard. The progress of reaction was followed in situ with 1H NMR spectroscopy for several hours, but no changes in the relative amounts of the species present were observed after 8 min at room temperature. The reaction was performed at least two times and the integral data thus obtained provided average values and the experimental error. (b) 2-Ethynylthiophene. 2-Ethynylthiophene (4.51 L, 42.5 mol) was added with a microsyringe to an NMR tube containing 500.0 L of the prepared stock solution of Cp*2LaCCPh(THF) (42.5 mol). 1H NMR spectroscopy indicated the presence of Cp*2LaCCPh(THF), Cp*2LaCC(2-C4H3S)THF (by comparison with an

190

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes authentic sample, vide infra), phenylacetylene and 2-ethynylthiophene in a 0.065:0.315:0.275:0.100 ratio, respectively. (b) 3-Ethynylthiophene. 3-Ethynylthiophene (4.26 L, 42.5 mol) was added with a microsyringe to an NMR tube containing 500.0 L of the prepared stock solution of Cp*2LaCCPh(THF) (42.5 mol). 1H NMR spectroscopy indicated the presence of Cp*2LaCCPh(THF), Cp*2LaCC(3-C4H3S)THF (by comparison with an authentic sample, vide infra), phenylacetylene and 3-ethynylthiophene in a 0.54:0.71:0.69:0.71 ratio, respectively. NMR tube reaction of Cp*2LaCCPh(THF) with pyridine. Pyridine (3.5 L, 43 mol) was added with a microsyringe to a solution of Cp*2LaCCPh(THF) (24.6 mg, 42.5 mol) in benzene-d6. The clean formation of Cp*2LaCCPh(NC5H5) and free THF was observed. The reaction was followed for several hours with 1H NMR spectroscopy, but no change was observed after 8 min at room temperature. The addition of excess THF (42.5 L, 425 mol, 10 equiv.) did not result in shifted proton NMR resonances for THF and pyridine. Evaporation to dryness and dissolution in benzene-d6 allowed for its characterization. Cp*2LaCCPh(NC5H4): 1H NMR (500 MHz, C6D6, 25 C): 2.10 (s, C5Me5, 30 H), 6.48 (ddd, 3JCH = 3 7.7 Hz, JCH = 4.6 Hz, 4JCH = 1.4 Hz, -CH, 2 H), 6.76 (tt, 3JCH = 7.7 Hz, 4JCH = 1.7 Hz, -CH, 1 H), 7.00 (tt, 3JCH = 7.4 Hz, 4JCH = 1.5 Hz, p-CH, 1 H), 7.12 (dt, 3JCH = 7.7 Hz, 4JCH = 1.6 Hz, m-CH, 2 H), 7.73 (dt, 3JCH = 7.7 Hz, 4 JCH = 1.6 Hz, o-CH, 2 H), 8.54 (dt, 3JCH = 5.5 Hz, 4JCH = 1.5 Hz, -CH, 2 H). 13C NMR (125.7 MHz, C6D6, 25 C): 11.26 (q, 1JCH = 125.0 Hz, C5Me5), 106.08 (s, i-C), 119.22 ( C5Me5), 124.68 (dt, 1JCH = 166.6 Hz , 3JCH = 6.8 Hz, -CH), 125.77 (dt, 1JCH = 160.6 Hz , 3JCH = 7.8 Hz , p-CH), 128.35 (dd, 1JCH = 158.6 Hz, 3JCH = 9.2 Hz, m-CH), 131.93 (dt, 1JCH = 160.2 Hz , 3JCH = 7.1 Hz, o-CH), 138.69 (dt, 1JCH = 163.8 Hz , 3JCH = 6.18 Hz, -CH), 148.25 (d, 1JCH = 180.1 Hz, -CH). 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): 2.09 11.26, 6.46 124.5, 6.74 138.7, 6.99 125.8, 7.12 128.3, 7.71 131.7, 8.48 148.26. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): 2.09 119.22, 7.12 131.94-128.37, 7.73 131.94-125.78-106.09. Synthesis of {Cp*2La[-2:2-(2-MeC6H4)C4(2-MeC6H4)]} (22b). An aliquot of 2-ethynyltoluene (17.0 mg, 146 mol) was added with a microsyringe to a stirred suspension of [Cp*2La(-H)]2 (60 mg, 73 mol) in hexane (10 mL). The pale yellow suspension turned dark red immediately. After being stirred for 1 h at room temperature, the resulting dark red solution was concentrated in vacuo and low-temperature crystallization afforded dark-red crystals. Yield: 56 mg (73%). 1 H NMR (500 MHz, C6D6, 25 C): 1.97 (s, C5Me5, 30 H), 2.33 (s, CH3, 3 H), 6.82 (d, 3JCH = 7.4 Hz, o-CH, 1 H), 7.07 (t, 3JCH = 7.4 Hz, CH, 1 H), 7.21 (t, 3JCH = 7.6 Hz, CH, 1 H), 7.24 (t, 3JCH = 7.5 Hz, m-CH, 1 H). 13C{1H} NMR (125.7 MHz, C6D6, 25 C): 11.23 (C5Me5), 21.76 (CH3), 120.28 (C5Me5), 126.63 (CH), 126.90 (CH), 131.22 (CH), 131.59 (CH), 137.67 (i-C), 140.18 (CMe), 153.78 (LaCC), 214.24 (LaCC). 1H NMR (500 MHz, C7D14, 25 C): 1.80 (s, C5Me5, 30 H), 2.21 (s, CH3, 3 H), 6.70 (br. s, o-CH, 1 H), 7.04 (t, 3JCH = 7.1 Hz, m-CH, 1 H), 7.15 (t, 3JCH = 7.1 Hz, p-CH, 1 H), 7.25 (d, 3JHH = 7.3 Hz, m-CH, 1 H). The broad 1H NMR resonance at 6.70 ppm became sharper at higher temperatures. At 85 C a doublet was observed (see text for more details). 13C{1H} NMR (125.7 MHz, C7D14, 25 C): 11.10 (C5Me5), 119.78 (C5Me5), 126.50 (m-CH), 126.63 (p-CH), 130.09 (m-CH), 139.70 (CMe). Neither the 13C{1H} NMR resonances of the i-C and o-CH groups were observed, nor those of the LaCC fragment. The 13C NMR resonance of the CH3 group overlaps with that of the residual protio-solvent, as seen with a 1H-13C correlation experiment. 1H-13C gHSQC (500-125.7 MHz, C7D14, 25 C): 1.80 11.08, 2.22 21.39, 7.04 126.63, 7.15 126.50, 7.25 130.09. 1H-13C gHMBC (500-125.7 MHz, C7D14, 25 C): 1.80 119.78, 2.21 139.70-130.09, 7.15 139.70-130.09, 7.25 139.70-126.50. Anal. Calcd. for C38H74La2 (808.82): C, 56.43%; H, 9.22%. Found: C, 56.23%; H, 9.09%. Synthesis of {Cp*2La[-2:2-(2-C4H3S)C4(2-C4H3S)]} (22d). An aliquot of 2-ethynylthiophene (18.5 mg, 171 mol) was added with a microsyringe to a stirred suspension of [Cp*2La(-H)]2 (70 mg, 85 mol) in hexane (15 mL). The pale yellow suspension turned dark red immediately. After being stirred for 1 h at room temperature, the resulting dark red solution was concentrated in vacuo and low-temperature crystallization afforded dark-red crystals. Yield: 71 mg (81%). 1 H NMR (500 MHz, C6D6, 25 C): 2.08 (s, C5Me5, 60 H), 6.78 (dd, 3JCH = 5.1 Hz, 3JCH = 3.7 Hz , CH, 2 H), 6.82 (dd, 3JCH = 5.1 Hz, 4JCH = 1.1 Hz, -CH, 1 H), 7.03 (dd, 3JCH = 3.7 Hz, 4JCH = 1.1 Hz, -CH, 1 H). 13 C{1H} NMR (125.7 MHz, C6D6, 25 C): 11.69 (C5Me5), 120.43 (C5Me5), 123.50 (-CH), 129.00 (-CH), 130.07 (i-C), 130.88 (-CH), 150.09 (LaCC), 190.84 (LaCC). 1H-1H gCOSY (500-500 MHz, C6D6, 25 C): 6.79 7.25, 7.02 7.25, 7.25 6.79-7.03. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): 2.08 11.69, 6.78 130.88, 6.82 123.50, 7.04 129.00. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): 2.08 120.43, 6.78 129.00-123.50, 6.82 130.88-130.07-129.00, 7.03 190.84-150.23-130.88-123.50. Anal. Calcd. for C32H66La2S2 (792.81): C, 48.48%; H, 8.39%. Found: C, 48.36%; H, 8.32%.

191

Chapter 5 Synthesis of Cp*2LaCCPhTHF (20aTHF). An aliquot of phenylacetylene (9.0 mg. 88 mol) was added with a microsyringe to a solution of Cp*2LaN(SiMe3)2 (50 mg, 88 mol) in THF (2 mL). After stirring for 12 h at 80 C, the resulting solution was evaporated to dryness and an off-white solid was obtained. Yield: 48.6 mg (95%). 1 H NMR (500 MHz, C4D8O, 25 C): 2.02 (s, C5Me5, 30 H), 7.04 (tdd, 3JCH = 6.6 Hz, 4JCH = 1.1 Hz, 4 JCH = 1.2 Hz, p-CH, 1 H), 7.13 (tt, 3JCH = 7.6 Hz, 4JCH = 1.7 Hz, m-CH, 2 H), 7.27 (dt, 3JCH = 8.0 Hz, 4JCH = 1.1 Hz, o-CH, 2 H). 13C{1H} NMR (125.7 MHz, C4D8O, 25 C): 11.42 (C5Me5), 105.51 (i-C), 119.23 (C5Me5), 125.93 (CH), 128.60 (CH), 132.14 (CH). The 13C NMR resonances of the LaCC fragment were not observed. 1H NMR (500 MHz, C6D6, 25 C): 1.20 (m, -THF, 4 H), 2.13 (s, C5Me5, 30 H), 3.57 (m, -THF, 4 H), 6.98 (tt, 3 JCH = 7.4 Hz, 4JCH = 1.5 Hz, p-CH, 1 H), 7.10 (tt, 3JCH = 7.7 Hz, 4JCH = 1.7 Hz, m-CH, 2 H), 7.67 (dt, 3JCH = 7.2 Hz, 4JCH = 1.5 Hz, o-CH, 2 H). 13C NMR (125.7 MHz, C6D6, 25 C): 11.25 (q, 1JCH = 125.0 Hz, C5Me5), 24.25 (t, 1JCH = 133.5 Hz, -THF), 68.71 (t, 148.6 Hz, -THF), 105.52 (s, i-C), 119.23 (s, C5Me5), 125.71 (dt, 1JCH = 160.9 Hz , 3JCH = 7.4 Hz, p-CH), 128.33 (dd, 1JCH = 158.9 Hz , 3JCH = 7.7 Hz , m-CH), 131.86 (dt, 1JCH = 158.4 Hz, 3JCH = 7.0 Hz, o-CH). The 13C NMR resonances of the LaCC fragment were not observed. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): 2.13 11.25, 6.97 125.86, 7.10 128.41, 7.66 131.80. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): 2.13 119.24, 6.98 132.2, 7.10 132.2-125.9, 7.68 132.2125.9-105.3. Synthesis of Cp*2LaCC(2-C4H3S)THF (20dTHF). An aliquot of 2-ethynylthiophene (9.5 mg. 88 mol) was added with a microsyringe to a solution of Cp*2LaN(SiMe3)2 (50 mg, 88 mol) in THF (2 mL). After stirring for 12 h at 80 C, the resulting solution was evaporated to dryness and an off-white solid was obtained. Yield: 48.0 mg (93%). 1 H NMR (500 MHz, C4D8O, 25 C): 2.00 (s, C5Me5, 30H), 6.79 (dd, 3JCH = 5.1 Hz, 3JCH = 3.5 Hz, CH), 6.83 (dd, 3JCH = 3.5 Hz, 4JCH = 1.2 Hz, -CH), 6.98 (dd, 3JCH = 5.1 Hz, 4JCH = 1.2 Hz, -CH). 13C{1H} NMR (125.7 MHz, C4D8O, 25 C): 11.40 (C5Me5), 97.98 (i-C), 119.72 (C5Me5), 123.78 (CH), 127.13 (CH), 128.68 (CH). The 13C NMR resonances of the LaCC fragment were not observed. 1H NMR (500 MHz, C6D6, 25 C): 1.20 (m, -THF, 4 H), 2.10 (s, C5Me5), 3.57 (m, -THF, 4 H), 6.64 (m, -CH), 6.64 (m, -CH), 7.14 (m, -CH). 13 C NMR (125.7 MHz, C6D6, 25 C): 11.27 (q, 1JCH = 125.0 Hz, C5Me5), 24.21 (t, 1JCH = 133.2 Hz, -THF), 68.65 (t, 148.3 Hz, -THF), 98.00 (s, i-C), 119.37 (s, C5Me5), 123.43 (ddd, 1JCH = 185.3 Hz, 3JCH = 10.7 Hz, 2JCH = 7.4 Hz, -CH), 126.84 (ddd, 1JCH = 166.3 Hz , 2JCH = 6.0 Hz, 2JCH = 3.7 Hz, -CH), 128.65 (ddd, 1JCH = 165.8 Hz, 2JCH = 6.6 Hz, 3JCH = 8.0 Hz, -CH). The 13C NMR resonances of the LaCC fragment were not observed. 1H13 C gHSQC (500-125.7 MHz, C6D6, 25 C): 2.10 11.23, 6.64 123.47-126.89, 7.14 128.65. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): 2.10 119.37, 6.64 128.65, 7.14 128.65. Synthesis of Cp*2LaCC(3-C4H3S)THF (20eTHF). An aliquot of 3-ethynylthiophene (9.5 mg. 88 mol) was added with a microsyringe to a solution of Cp*2LaN(SiMe3)2 (50 mg, 88 mol) in THF (2 mL). After stirring for 12 h at 80 C, the resulting solution was evaporated to dryness and an off-white solid was obtained. Yield: 44.5 mg (93%). 1 H NMR (500 MHz, C4D8O, 25 C): 2.01 (s, C5Me5), 6.93 (dd, 3JCH = 4.9 Hz, 4JCH = 1.2 Hz, -CH), 7.07 (dd, 4JCH = 3.0 Hz, 4JCH = 1.2 Hz, -CH), 7.11 (dd, 3JCH = 4.9 Hz, 4JCH = 3.0 Hz, -CH). 13C{1H} NMR (125.7 MHz, C4D8O, 25 C): 11.38 (C5Me5), (m, CD2, THF-d8), (m, OCD2, THF-d8), 99.75 (i-C), 119.55 (C5Me5), 122.74 (CH), 124.47 (CH), 124.80 (CH). The 13C NMR resonances of the LaCC fragment were not observed. 1H NMR (500 MHz, C6D6, 25 C): 1.20 (m, -THF, 4 H), 2.12 (s, C5Me5), 3.57 (m, -THF, 4 H), 6.77 (dd, 3JCH = 5.0 Hz, 4JCH = 3.1 Hz, -CH), 7.19 (dd, 3JCH = 5.0 Hz, 4JCH = 1.1 Hz, -CH), 7.11 (dd, 4JCH = 3.1 Hz, 4JCH = 1.1 Hz, -CH). 13C NMR (125.7 MHz, C6D6, 25 C): 11.27 (q, 1JCH = 125.0 Hz, C5Me5), 24.32 (t, 1 JCH = 133.4 Hz, -THF), 69.07 (t, 148.7 Hz, -THF), 99.76 (s, i-C), 119.24 (s, C5Me5), 124.23 (ddd, 1JCH = 185.4 Hz, 2JCH = 7.7 Hz, 3JCH = 5.9 Hz, -CH), 124.88 (ddd, 1JCH = 186.1 Hz , 3JCH = 8.6 Hz, 3JCH = 4.3 Hz, CH), 131.03 (ddd, 1JCH = 168.4 Hz, 2JCH = 4.6 Hz, 3JCH = 8.7 Hz, -CH). The 13C NMR resonances of the LaCC fragment were not observed. 1H-13C gHSQC (500-125.7 MHz, C6D6, 25 C): 2.13 11.27, 6.77 124.28, 7.19 131.07, 7.20 124.94. 1H-13C gHMBC (500-125.7 MHz, C6D6, 25 C): 2.12 119.24, 7.19 124.2.

192

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

Table 5-8. Summary of crystallographic data of the [(Cp*2La)(-RC4R)] [R = Ph (22a), 2-MeC6H3 (22b), R = 2-C4H3S (22d)] and [Cp*2La(-2-C4H3S)] (28D) complexes. 22a 22b 22c 28D C58H74La2 C52H66La2S2 (C24H33LaS)2 Molecular formula (C28H35La)2 2(C7H8) 2(C7H8) 2(C7H8) C7H8 FW 1205.26 1233.32 1217.32 1077.14 T, K Crystal system Space group a, b, c, () V, 3 Z calc, gcm-1 F(000) (Mo K), cm-1 range () wR(F2) reflections parameters R(F) for Fo > 4.0 (Fo) GooF Weighting (a, b) 100(1) monoclinic P21/m 15.0408(7) 14.1013(7) 15.4340(7) 114.511(1) 2978.5(2) 2 1.344 1240 14.54 2.47, 29.71 0.1131 8177 380 0.0413 1.078 0.0622,5.7395 230(1) monoclinic C2/m 15.528(2) 14.313(2) 15.641(2) 116.666(1) 3106.5(7) 2 1.318 1272 13.96 2.63, 26.73 0.0847 3436 194 0.03555 1.036 0.0445,2.30 100(1) monoclinic C2 15.6428(7) 13.9054(7) 15.0648(7) 115.427(1) 2959.5(2) 4 1.366 1248 15.32 2.62, 28.28 0.0726 7121 518 0.0289 0.982 0.0422,1.82 100(1) monoclinic C2/c 15.943(1) 14.796(1) 21.968(1) 107.550(1) 4940.9(5) 4 1.448 2200 18.250 2.33, 28.28 0.0677 6113 422 0.0276 1.067 0.0346, 6.1702

X-ray analyses. In a glovebox, suitable crystals were mounted on top of a glass fiber. The crystals were aligned in a cold nitrogen stream on a Bruker83 SMART APEX CCD diffractometer. The unit cells were determined as listed in Table 5-8. The structures were solved by Patterson methods and extension of the model was accomplished by direct methods applied to difference structure factors using the program DIRDIF.84

Appendix. The acidity of the studied 1-alkynes

Introduction The term acidity or acid strength comprises both a thermodynamic and kinetic aspect.59 Kinetic acidity is related to the rate of proton transfer, while thermodynamic acidity refers to the position of equilibria between acids and their conjugate bases. The rates of proton transfer have been measured for many oxygen and nitrogen acids (termed normal acids) and found to be diffusion-controlled in aqueous solution. Proton transfers tend to be much slower in non-aqueous solvents than in water, especially in non-protic media. In general, carbon acids (termed pseudo acids) undergo proton transfer reactions at a much lower rate than oxygen and nitrogen acids. It is commonly believed that the intrinsically slow rate of proton transfer in carbon acids is the result of the delocalized nature of the proton-accepting electron pairs in the carbon bases, whereas the proton-accepting electron pairs in oxygen and nitrogen acids are localized on single atoms. The proton transfer rates of 1-alkynes have been found to be diffusion-controlled in aqueous solution, on the basis of unit-slope Brnsted relationships (vide infra).85 These findings suggested that the carbanionic electron pair of the acetylide ion resides in a localized sp hybrid orbital and that 1-alkynes behave as normal acids rather than pseudo acids. More recently, doubt was cast on this common belief by the observation of primary kinetic isotope effects in proton-transfer reactions of certain 1-alkynes, indicating rate-limiting proton transfer.86 Kinetic acidity and inductive/field effects The acidities of most of the 1-alkynes employed in this study are not known in literature, but it is believed that the relative order of acidity can be approximated by reported empirical relationships. The kinetic acidity of a number of 1-alkynes have been determined experimentally by measuring the rate of base-catayzed hydrogen-isotope exchange. Even though many of the early investigations in this field have dealt with the effect

193

Chapter 5

Table 5-9. Hammett substituent constants of the 1-alkyne substituents R and spectral data of the corresponding 1-alkynes HCCR.a R I CCH CCH R F R CCH 2 (2JCH) (1JCH) 0.12 -0.13 0.12 -0.13 77.83 83.89 2.73 a (251.0) (50.9) 0.10 -0.13 0.12 -0.15 81.53 82.78 2.92 b (250.0) (50.2)
O

0.11

-0.19

0.13

-0.21

81.83 (250.2) 81.88 (253.4) 77.54 (251.4) 81.51 (252.5) 77.33 (252.8)

80.52 (50.3) 77.22 (51.6) 79.10 (50.9) 76.38 (50.5) 83.56 (50.3)

3.04

d e f

0.19 0.10 0.17

-0.14 -0.12 0.20

0.13 0.08 0.50

-0.08 -0.10 -0.13

2.87 2.70 3.01

0.20

-0.03

0.40

-0.23

3.10

The values for R were calculated from the relationship R = p I. The values for I, p, F and R were obtained from Ref. 51a. The values for 2b, 2c and 2f were approximated with those of C6H4Me-p, C6H4OMe-p and 2-pyrryl, respectively. NMR experiments were conducted in C6D6 under nitrogen at 25 C. Chemical shifts are reported in ppm relative to TMS and 1JCH coupling constant (in brackets) are reportd in Hz. Repeated experiments suggested that 13C chemical shifts were accurate to 0.02 ppm, 1H chemical shifts to 0.01 ppm and 1JCH coupling constants to 0.2 Hz. of the substituent on the rates of isotopic exchange and empirical relationships between substituent parameters and the rate of exchange were found, the number and structural diversity of the 1-alkynes considered in each study has often been somewhat limited.87 A study of 17 substituted phenylacetylenes of which the rate of detritiation were measured in aqueous buffer solution at 25 C revealed a linear correlation for meta- and para-substituents and their corresponding Hammett constants (vide infra).87e Interestingly, the relative rates of ortho-methyl and ortho-methoxy substituted phenylacetylenes were identical and both 0.78-fold lower than that of phenylacetylene. The absence of significant steric effects in this system was demonstrated by the observed rate of mesitylacetylene which differed only 14% from that calculated by assuming additivity of the effects of three methyl groups. A more recent study in this field, based on the rate of base-catalyzed hydrogen exchange of 13 substituted 1-alkynes in aqueous buffer solution, revealed a good linear correlation between the rate of hydrogen-exchange and inductive or field substituent constants I (vide infra).87j The observation that the addition of resonance substituent constants did not improve the correlation implied that resonance interactions of the 1-alkyne substituent are of minor importance in determining the kinetic acidity and that the kinetic acidity of 1-alkynes can be estimated from the inductive or field constants I alone. Even though this conclusion was in contradiction with some early studies in this field, it did support the common belief that the carbanionic electron pair of the acetylide ion resides in an sp hybrid orbital which is orthogonal to the acetylenic -system. Because delocalization of this electron pair by conjugation with the -system is not possible, the electrons are localized on a single atom and 1-alkynes behave as normal acids (vide supra). Other more recent studies also confirmed the notion that the substituent effect in 1-alkynes is mainly a field/inductive effect.88 A number of important relationships between substituent groups and chemical properties have been developed over the last decades.89 In many cases, such relationships can be expressed quantitatively and are valuable for the interpretation of reaction mechanisms and for the prediction of reaction rates and equilibria. The most widely used of these relationships is the Hammett equation, which referred originally to the acidity of mand p-substituted benzoic acids. On the basis of linear free energy relationships, substituent constants were derived that can be used as a measure of the substituents electronic effect at the reaction site. The recognization that substituent effects are transmitted either through the framework (induction and field effects) or a

194

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

measured C NMR chemical shift (ppm)

85 84 83 82 81 80 79 78 77 76 75 74 75 76 77 calculated 78
13

C-1

C-2

13

79

80

81

82

83

84

C NMR chemical shift (ppm)

Figure 5-17. Plot of the measured 13C NMR chemical shifts of the present 1-alkynes versus the corresponding 13C NMR chemical shifts, as calculated from empirical relationships involving the substituent constants F and R. The lines connecting the data points represent fitted linear plots. delocalized system (resonance) led to a number of methods to split values into inductive and resonance contributions.90 Among these dual-parameter correlations, the substituent constants I and R, on the one hand, and, F and R, on the other hand, represent the most widely used indicators for the ability of a substituent to interact via an inductive/field and resonance mechanism, respectively.51 Kinetic acidity and NMR spectroscopy A second method to estimate 1-alkyne acidity makes use of 13C NMR spectroscopy. NMR shielding in alkynes is presently not well understood and studies have been performed on simple models only.91 Even so, linear free energy relationships are routinely used to correlate NMR data with Hammett constants in order to determine the transmission of inductive and resonance electronic effects in conjugated systems.92 This approach is based on the assumption that substituent-induced chemical shifts in NMR spectra primarily depend on the electronic density of the probe nucleus.93 Substituent effects on the chemical shifts of alkyne carbon atoms have received considerably less attention than analogous effects in other aromatic systems. Nonetheless, a study of 18 para-substituted phenylacetylenes revealed that the substituent-induced changes in carbon and proton chemical shifts correlate well with Hammett substituent constants, when NMR experiments are conducted under certain conditions (i.e. measurements in an inert solvent, such as in cyclohexane-d12, and under an inert atmosphere, the use of a relatively large number of analogues and the extrapolation of the NMR parameters to infinite dilution).94 It was concluded that both inductive/field and resonance effects are important in determining the electronic distribution in phenylacetylenes. The authors argued that the observed correlations provided evidence that substituent-induced changes in the proton and carbon chemical shifts in phenylacetylene reflect intramolecular electronic effects with insignificant contributions from solvent and magnetic anisotropy effects. In accord with this view, similar, but poorer correlations were found in a recent study of 13 para- and meta-substituted phenylacetylenes that were measured in chloroform-d1 in air.95 The 1-alkynes used in this study were measured in benzene-d6 under a nitrogen atmosphere at relatively low concentrations (0.5-1 mol%) and the NMR parameters acquired are shown in Table 5-9. The fair correlation between the measured carbon chemical shifts and the carbon chemical shifts calculated from the empirically established relationships with F and R provided assurance that the present NMR data represent intramolecular electronic effects to a reasonable degree (Figure 5-19).96 The discrepancies are most likely due to the magnetic anisotropy effects arising from ortho-substituents, such as in 2-ethynylanisole (2c) and 1-methyl-2ethynylpyrrole (2g).97 The fair correlation suggest, in addition, that Hammett substituent constants can be used to investigate the electronic effects of the present (hetero)aromatic 1-alkyne substituents. Other reported

195

Chapter 5

256 255 254 J CH (Hz) 253 252

S
S

N
N

251 250 249 248 0.06

O

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

Figure 5-18. Plot of the measured coupling constant (1JCH) of the terminal carbon in the present 1-alkynes versus the inductive/field constant I of the corresponding 1-alkyne substituent (Table 5-9). The line drawn represents a fitted linear plot (see text for details). correlations of the proton chemical shift (CCH) with either the terminal carbon chemical shift (CCH) or its firstorder carbon hydrogen coupling constant (1JCH) were not found, most likely due to the larger solvent effects in the 1H NMR spectral parameters. Many studies reported in literature have demonstrated the validity of the well-known linear relationship between the rate of proton transfer from a carbon acid and the first-order carbon-hydrogen coupling constant of the corresponding carbon (1JCH).98 The rationale for this behavior is found in the relative amount of scharacter of the hybrid orbital on the carbon atom. As the s-character increases, the hybrid orbital is more tightly bound to the carbon atom and the C-H bond becomes more polar, thereby increasing the acidity of the hydrogen atom. By the same reasoning, the formed anion is the most stable in the orbital with the highest s-character. In hydrocarbons, the most important factor affecting couplings is the hybridization of the carbon atom. The effects of polar substituents on 1JCH are, however, much larger. Because polar groups affect the electrons of the carbon nucleus via their inductive effects without significantly altering the hybridization of the carbon nucleus, the relationship between 1JCH and s-character is generally not suitable for estimating the hybridization of the carbon bonding orbital in heteroatom-substituted compounds. A plot of the coupling constant 1JCH of the terminal carbon in the present 1-alkynes against the inductive/field substituent constant I of the corresponding 1-alkyne substituent reveals a linear dependence of 1 JCH on I (Figure 5-18). Only 3-ethynylthiophene (2e) and 2-ethynylpyridine (2f) show a significant deviation for which no obvious explanation can be given at present. Even though it cannot be ruled out that the observed correlation between 1JCH and I reflects a correlation of 1JCH with the inductive/field effect of the (hetero)aromatic 1-alkyne substituents rather than a correlation of 1JCH with the hybridization of the terminal alkyne carbon orbital, the importance of inductive/field effects in determining the rate of proton transfer in 1alkynes has already been recognized (vide supra). Supporting evidence for the use of the present 1JCH values as a measure for intramolecular electronic effects comes from the observed correlations of 13C NMR parameters with Hammett substituent constants (vide supra). Relationship between kinetic and equilibrium acidity Transmetalation equilibria have widely been used to determine the relative acidities of hydrocarbons.59 Conversely, kinetic acidities have been measured to determine proton-transfer equilibria. The validity of these methods is based on the commonly observed unit-slope Brnsted relationships. The Brnsted relationship relates the kinetic acidity with equilibrium acidity and, generally, they vary in parallel with each other. Linear relationships between equilibrium and kinetic acidty are frequently found for oxygen and nitrogen acids that display diffusion-controlled proton transfer rates. In contrast, Brnsted plots for proton transfer at

196

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes carbon acids and bases are more complex, due to the need for rehybridization and changes in the geometry of the acid or base upon proton transfer. Unit-slope Brnsted plots have been reported for proton transfer reactions of a number of 1-alkynes in aqueous solution. However, the reactions in this study were performed in non-polar, aprotic solvents, such as benzene and THF. Because proton transfers are well-known to decrease with decreasing solvent polarity and deviations from unit-slope Brnsted plots are well-known to occur, when proton transfers take place at less than diffusion-controlled rates, it seems not unreasonable to assume that Brnsted plots will not be linear for 1alkynes in non-polar, aprotic solvents. Moreover, equilibrium constants in nonpolar media do not generally reflect the free ion acidity, but rather refer to ion pairs.99 In media of low polarity extensive ion-pairing and formation of aggregates between ions and neutral molecules takes place and the extent of ion-pairing depends on the solvent, the size of the ions, and the charge distribution in ions. The transmetalation reaction of Cp*2LaCCPh(THF) with 2-ethynylthiophene (2d) and 3ethynylthiophene (2e) afforded equilibrium constants of K = 0.08(5) and K = 0.80(4), respectively (Section 5.4.5). These values suggest that 2d and 2e are 1.10 and 0.10 pK units, respectively, more acidic than phenylacetylene, thereby contradicting the relative order of kinetic acidities, based on substituent constants (vide supra). In particular, the 3-thienyl group is less electron-donating than the phenyl group, both via a field/inductive and a resonance mechanism, according to the well-established substituent constants I, R, F and R. The reason for this discrepancy is not known, but this observation seems to support the above assumption that the kinetic acidity of the present 1-alkynes do not vary linearly with the corresponding equilibrium acidity in benzene. Estimating the acidities of the present 1-alkynes by measuring the equilibrium concentrations of the species present in the transmetallation reaction of Cp*2LaCCPh(THF) with 1-alkynes is therefore considered to be inappropriate. Concluding remarks It is believed that the agreement between two independent empirical relationships, i.e. the relationship between the kinetic acidity of 1-alkynes with I and the relationship between the rate of proton transfer of carbon acids with 1JCH, provides reasonable assurance that the order of kinetic acidity for the present 1-alkynes can be estimated on the basis of 1JCH or I values. In view of the smaller experimental errors in the 1JCH values relative to those in the calculated detritiation rates, the former approach is presently favored. As a consequence, the following relative order of decreasing kinetic acidity based on measured 1JCH values is obtained for the present 1alkynes: 2-ethynylthiophene (2d) > 2-ethynylpyridine (2f) 1-methyl-2-ethynylpyrrole (2g) > 3ethynylthiophene (2b) phenylacetylene (2a) 2-ethynylanisole (2c) 2-ethynyltoluene.

5.8.
1

References and notes

Other examples of catalysts that combine high activity with high selectivity for trans-1,4diphenylbut-1-en-3-yne, see (a) Ru: Yi, C.S.; Liu, N. Synlett 1999, 281. (b) Ru: Baratta, W.; Herrmann, W. A.; Rigo, P.; Schwartz, J. J. Organomet. Chem. 2000, 593-594, 489. (c) Rh: Werner, H.; Schwab, P.; Heinemann, A. Steinert, P. J. Organomet. Chem. 1995, 496, 207. (d) Pd: Yang, C.; Nolan, S. P. J. Org. Chem. 2003, 67, 591. (a) Brandsma, L. Preparative Acetylenic Chemistry; Springer: Amsterdam, 1971. (b) Vasilevsky, S. F.; Tretyakov, E. V.; Elguero, J. Adv. Heterocycl. Chem. 2001, 82, 1. For a review, see: Negishi, E.-I.; Anastasia, L. Chem. Rev. 2003, 103, 1979. For recent reviews, see: (a) Sonogashira, K. In Metal-Catalyzed Cross-coupling Reactions; Diederich, F., Stang, P.J., Eds.; Wiley-VCH: New York, 1998; Chapter 5. (b) Brandsma, L.; Vasilevsky, S.F.; Verkruijsse, H. D. Application of Transition Metal Catalysts in Organic Synthesis, Springer: Berlin, 1998; Chapter 10. (a) Onopchenko, A.; Sabourin, E. T.; Selwitz, C. M. J. Org. Chem. 1979, 44, 1233. (b) Sabourin, E. T.; Onopchenko, A. J. Org. Chem. 1983, 48, 5135. (c) Havens, S. J.; Hergenrother, P. M. J. Org. Chem. 1985, 50, 1763. The reaction of Cp*2LnCH(SiMe3)2 (Ln = La, Ce) with 2,6-di-tert-butyl-4-methylphenol forming the corresponding aryloxide has been reported, see: (a) Heeres, H. J. Ph. D. Thesis, University of Groningen, 1990; Chapter 3. (b) Heeres, H. J.; Teuben, J. H. Recl. Trav. Chim. Pays-Bas 1990, 109, 226.

2 3 4

197

Chapter 5

10 11

12

13

14

15

16

Alkyl and aryl halides undergo metal-halogen exchange reactions with alkyl derivatives of rare-earth permethylmetallocenes. For examples, see: (a) Finke, R. G.; Keenan, S. R.; Schiraldi, D. A.; Watson, P. L. Organometallics 1986, 5, 598. (b) Ref. 5a. (c) Finke, R. G.; Keenan, S. R.; Schiraldi, D. A.; Watson, P. L. Organometallics 1987, 6, 1356. (d) Finke, R. G.; Keenan, S. R.; Watson, P. L. Organometallics 1989, 8, 263. (e) Ref. 14a. (f) Booij, M. ; Deelman, B.-J. ; Duchateau, R. ; Postma, D. S.; Meetsma, A. ; Teuben, J. H. Organometallics, 1993, 12, 3531. (g) Qian, C.; Zhu, C.; Zhu, D. Appl. Organomet. Chem. 1995, 9, 457. The oxidative copper-catalyzed 1-alkyne dimerization is believed to be catalyzed by adventitious molecular oxygen. For sterically hindered 1-alkynes, diacetylene formation is known to be significant, even when molecular oxygen is rigorously excluded. For examples, see: (a) Xu, Z.; Moore, J. S. Angew. Chem. Int. Engl. 1993, 32, 246. (b) Zhou, Q.; Swager, T. M. J. Am. Chem. Soc. 1995, 117, 12593. (c) Elangovan, A.; Wang, Y.-H.; Ho, T.-I. Org. Let. 2003, 5, 1841. (d) Khan, A.; Hecht, S. Chem. Commun. 2004, 300. (d) Mry, D.; Heuz, K.; Astruc, D. Chem. Commun. 2003, 1934. (e) Tykwinski, R. R. Angew. Chem. Int. Ed. 2003, 42, 1566. 2001, 626, 100. (f) Shultz, D.; Gwaltney, K. P.; Lee, H. J. Org. Chem. 1998, 63, 4034. (g) Chow, H.-F.; Wan, C.-W.; Low, K.-H.; Yeung, Y.-Y. J. Org. Chem. 2001, 66, 1910. Insertions of a 1,4-bis(trimethylsilyl)butadiyne into a Ti-C acetylide bond have been reported, see: (a) Pellny, P.-M., Kirchbauer, F. G.; Burlakov, V. V.; Spannenberg, A.; Mach, K.; Rosenthal, U. Chem. Commun. 1999, 2505. (b) Horek, M.; Csaov, I.; Kubita, J.; Spannenberg, A.; Dallmann, K.; Rosenthal, U.; Mach, K. J. Organomet. Chem. 2004, 689, 4592. Malkina, A. G.; Brandsma, L.; Vasilevsky, S. F.; Trofimov, B. A. Synthesis 1996, 589 1-Alkynes undergo anionic, cationic, radical and thermal polymerization reactions and are quite reactive materials in the pure state, especially when substituted with electron-withdrawing groups. For examples, see: (a) Neenan, T. X.; Whitesides, G. M. J. Org. Chem. 1988, 53, 2489. (b) Ref. f. (c) Stiegman, A. E.; Graham, E.; Perry, K. J.; Khundkar, L. R.; Cheng, L.-T.; Perry, J. W. J. Am. Chem. Soc. 1991, 113, 7658. (d) Rutherford, D. R.; Stille, J. K.; Elliott, C. M.; Reichert, V. R. Macromolecules 1992, 25, 2294. (e) Trost, B. M.; Sorum, M. T.; Chan, C.; Harms, A. E.; Rhter, G. J. Am. Chem. Soc. 1997, 119, 698. (f) Li, J.; Pang, Y. Macromolecules 1997, 30, 7487. (g) Altamura, P.; Giardina, G.; Lo Sterzo, C.; Vittoria Russo, M. Organometallics 2001, 20, 4360 and references therein. The catalyst of the head-to-tail dimerization of 1-alkynes catalyzed by [Cp*2Ti(2-Me3SiC CSiMe3)] is reported to be rapidly deactivated by products deriving from thermal decomposition of the 1-alkyne, see: Varga, V.; Petrusov, L.; Cejka, J.; Mach, K. J. Organomet. Chem. 1997, 532, 251 It should be noted that the acetylenic proton of the substrate requires ~500 s for complete relaxation under the present anaerobic conditions (Chapter 4). The intensity of the acetylenic proton versus delay time under the present reaction conditions was fitted convincingly to a first-order exponential function, thereby revealing that 99.90(16)% of the intensity (relative to the maximum observed intensity) was observed for an experiment with a delay time of 500 s, 99.61(16)% for 400 s, 98.43(16) for 300 s and 93.72(15)% for 200 s. Bond enthalpy data: (a) Nolan, S. P.; Stern, D.; Marks, T. J. J. Am. Chem. Soc. 1989, 111, 7844. (b) Nolan, S. P.; Stern, D.; Hedden, D.; Marks, T. J. Bonding Energetics in Organometallic Compounds; Marks, T. J., Ed.; ACS Symposium Series 428; American Chemical Society: Washington, D. C.; 1990; pp. 159-174. The transition state for protonolysis is sterically favored over that for insertion, as the former places the aromatic moiety of the incoming alkyne molecule further away from the metal center. Protonolysis may also be electronically favored, since it involves a sp carbon and a hydrogen atom orbital which provide more overlap in the four-center transition state than two sp carbon orbitals. In general, the stability of metal alkynyl derivatives MCCR decreases in the order R = aryl < H < alkyl. This order parallels the increasing ability of the alkyne substituent to act as an electronreleasing substituent, thereby suggesting that electron-releasing substituent groups stabilize alkynyl complexes. For general reviews on alkynyl complexes, see: (a) R. Nast, Coord. Chem. Rev. 1982, 47, 89. (b) W. Beck, B. Niemer, M. Wieser, Angew. Chem. Int. Ed. 1993, 32, 923. (c) S. Lotz, P. H. Van Rooyen, R. Meyer, Adv. Organomet. Chem. 1995, 37, 219. (d) J. Manna, K. D. John, M. D. Hopkins, Adv. Organomet. Chem. 1995, 38, 79. (e) H. Lang, M. Weinmann, Synlett 1996, 1. (f) M. I. Bruce, Coord. Chem. Rev. 1997, 166, 91. (g) F. Paul, C. Lapinte, Coord. Chem. Rev. 1998, 178-180, 431. (h)

198

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

17

18 19 20

21

R. Choukroun, P. Cassoux, Acc. Chem. Res. 1999, 32, 494. (i) H. Lang, G. Rheinwald, J. Prakt. Chem. 1999, 341, 1. (j) H. Lang, D. S. A. George, G. Rheinwald, Coord. Chem. Rev. 2000, 206-207, 101. (k) N. J. Long, C. K. Williams, Angew. Chem. Int. Ed. 2003, 42, 2586. (l) P. J. Low, M. I. Bruce, Adv. Organomet. Chem. 2002, 48, 71. (m) P. J. Low, M. I. Bruce, Adv. Organomet. Chem. 2003, 50, 179. (n) Rosenthal, U. In Acetylene Chemistry. Chemistry, Biology and Materials Science; Diederich, F.; Stang, P. J., Tykwinski, R. R, Eds.; Wiley-VCH: Weinheim, 2005; Chapter 4, p. 139. Heteroatoms, such as oxygen, nitrogen, and sulfur, are known to coordinate to the metal center, lowering the catalytic rates of alkene reactions in rare-earth metal chemistry. For examples, see: (a) Jeske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8111. (b) Haar, C. M.; Stern, C. L.; Marks, T. J. Organometallics 1996, 15, 1765. (c) Roesky, P. W.; Stern, C. L.; Marks, T. J. Organometallics 1997, 16, 4705. (d) Molander, G. A.; Dowdy, E. D.; Schumann, H. J. Org. Chem. 1998, 63, 3386. (e) Ref. 66b. (f) Molander, G. A.; Dowdy, E. D.; Schumann, H. J. Org. Chem. 1999, 64, 9697. (g) Gagn, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (h) Hong, S.; Kawaoka, A. M.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 15878. (i) Douglass, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 2001, 123, 10221. The pyrrole system was found also to be particularly reactive in the organolanthanide-catalyzed hydrosilylation, see: Molander, G.A.; Knight, E. E. J. Org. Chem. 1998, 63, 7009. Bianchini, C.; Meli, A.; Vizza, F. Eur. J. Inorg. Chem. 2001, 43. (a) Wilkins, R. G. Kinetics and Mechanism of Reactions of Transition Metal Complexes, VCH, Weinheim: 1991; 2nd ed.; Chapter 1. (b) Espenson, J. H. Chemical Kinetics and Reaction Mechanism, McGraw-Hill, Inc., New York: 1995; 2nd ed.; Chapter 4. (c) Segel, I. H. Enzyme Kinetics, Wiley-Interscience, New York; 1993. (d) Marangoni, A. G. Enzyme Kinetics, WileyInterscience, New York; 2003. In a simple Henri-Michaelis-Menten description (Eqs. 5.10 and 5.11; E = enzyme, S = substrate, P = product), the Michaelis constant KM = (k-1 + kp)/k1 refers to the overall effectiveness of substrate capture summed over all catalyst species ([ET] = [E] + [ES]) and Vmax = kp[ET] refers to the maximum reaction rate. In the present situation, substrate capture is rapid relative to the substrate-binding step (i.e. kp > k1) and the reaction rate is first-order with respect to [S] (KM > [S]). When the effects of reversible, nonproductive substrate and product coordination to the catalyst are included (Eqs. 5.125.13), standard rapid equilibrium analysis yields a modified Henri-Michelis-Menten equation which reproduces the observed kinetic behavior (Eq. 5.14). 20 (5.10)
k1 E + S k-1 ES kp E + P

(5.11)

d[S] dt =

kp [ET][S] KM + [S]
KS

Vmax [S] = KM + [S]

(5.12) (5.13)

E
E

+
+

S
P KP

ES
EP

(5.14)

d[S] dt = KM 1 +

kp[ET][S] [S] + KM KS [S] + KP [P] = KM 1 +

kp[ET][S] (KS + KM)[S] KMKS [P] + KP

22

Determination of initial rates may shed more light on the possibility of substrate self-inhibition, but the rapid rate of substrate conversion, the occurrence of catalyst deactivation at very low precatalyst concentration (due to traces of O2 and H2O) and the long relaxation times of the acetylenic proton do

199

Chapter 5

23

24

25 26 27

28 29

30 31

32

33 34

35

not allow such an analysis by means of the present methodology. For a more detailed discussion of the limitations of the present methodology, see Chapter 4. (a) Okamoto, Y.; Movsovicius, A. G. F.; Hellman, H.; Brenner, W. Chem. Ind. 1961, 2004. (b) Kern, R. J. J. Polym. Sci.: Part A-1: Polym. Chem. 1969, 7, 621. (c) Chauser, M. G.; Rodionov, Y. M.; Cherkasin, M. I. J. Macromol. Sci.: Chem. 1977, A11, 1113. (d) Amdur, S.; Cheng, A. T. Y.; Wong, C. J.; Ehrlich, P.; Allendoerfer, R. D. J. Polym. Sci.: Polym. Chem. Ed. 1978, 16, 407. (e) Pickard, J. M.; Jones, E. G.; Goldfarb, I. J. Macromolecules 1979, 12, 895. (f) Sekiguchi, H.; Kang, H.-C.; Tersac, G.; Sillion, B. Makromol. Chem. Macromol. Symp. 1991, 47, 317. (g) Gandon, S.; Mison, P.; Bartholin, M.; Mercier, R.; Silion, B.; Geneve, B.; Grenier, P.; Grenier-Loustalot, M.-F. Polymer 1997, 38, 1439. (h) Gandon, S.; Mison, P.; Silion, B. Polymer 1997, 38, 1449. For examples where the -donating ability of amide (Ln-NR2) or azomethine (Ln-N=CR2) ligands has been implicated by the observed reactivity or structure in rare-earth metal chemistry, see: (a) Bercaw, J. E.; Davies, D. L.; Wolczanski, P. T. Organometallics 1986, 5, 443. (b) Heeres, H. J; Meetsma, A.; Teuben, J. H. Angew. Chem. 1990, 102, 449. (c) Knight, L. K.; Piers, W. E.; Lessard-Fleurat, P.; Parvez, M.; McDonald, R. Organometallics 2004, 23, 2087. For comparison, the pKa of Ph2NH (24.95) in an aprotic, polar solvent, DMSO, is considerably lower at 25 C than that of Ph2CH2 (32.2), see: Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456. Ringelberg, S. N.; Meetsma, A.; Hessen, B.; Teuben, J. H. J. Am. Chem. Soc. 1999, 121, 6082. Several empirical Lewis basicity scales have been proposed, see: (a) Reichardt, C. Solvent and Solvent Effects in Organic Chemistry, 2 ed.; VCH: Weinheim, 1988. (b) Maria, P.-C.; Gal, J.-F. J. Phys. Chem. 1985, 89, 1296. (c) Rauk, A.; Hunt, I. R.; Keay, B. A. J. Org. Chem. 1994, 59, 6808. (a) Kirilov, E.; Lehmann, C. W.; Razavi, A.; Carpentier, J.-F. Eur. J. Inorg. Chem. 2004, 943. (b) Cameron, Th. M.; Gordon, J. C.; Scott, B. L. Organometallics 2004, 23, 2995. THF and pyridine induce adduct formation in their reactions with [Cp*2Y(-2-C4H3S)]2.30 Furan, THF, diethyl ether, tert-butylnitrile and pyridine give adduct formation in their reactions with [Cp*2Y(-2-C4H3O)]2, but thiophene does not.30 The reactions of THF and pyridine with Cp*2Y(-2C5H4N) result in adduct formation, but diethyl ether does not react, see Deelman, B.-J. Ph. D. Thesis, University of Groningen, 1994; Chapter 5. Ringelberg, S. N. Ph. D. Thesis, University of Groningen, 2001; Chapter 3. Because alkyne coordination is weak due to the absence of d-* back-bonding, alkyne complexes of d0 metals are quite rare. For an example of a non-chelated alkyne d0 metal complex, see: (a) Curtis, M. A.; Finn, M. G.; Grimes, R. N. J. Organomet. Chem. 1998, 550, 469. For examples of chelated alkyne d0 metal complexes, see: (b) Temme, B.; Erker, G.; Frhlich, R.; Grehl, M. Angew. Chem., Int. Ed. Engl. 1994, 33, 1480. (c) Venne-Dunker, S.; Ahlers, W.; Erker, G.; Frhlich, R. Eur. J. Inorg. Chem. 2000, 1671. (d) Ahlers, W.; Erker, G.; Frhich, R.; Peuchert, U. J. Organomet. Chem. 1999, 578, 115. (e) Ahlers, W.; Temme, B.; Erker, G.; Frhlich, R.; Fox, T. J. Organomet. Chem. 1997, 527, 191. (f) Erker, G.; Venne-Dunker, S.; Kehr, G.; Kleigrewe, N.; Frhlich, R.; Mck-Lichtenfeld, C.; Grimme, S. Organometallics 2004, 23, 4391. (g) Burlakov, V. V.; Arndt, P.; Baumann, W.; Spannenberg, A.; Rosenthal, U. Organometallics 2004, 23, 5188. Arene complexes have attracted much interest in rare-earth metal chemistry. For examples, see: (a) Mintz, E. A.; Moloy, K. G.; Marks, T. J. J. Am. Chem. Soc. 1982, 104, 4692. (b) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1990, 112, 219. (c) Giardello, M. A.; Conticello, V. P.; Brard, L.; Gagn, M. R.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10241. For a review, see: (d) Bockarev, M. N. Chem. Rev. 2002, 102, 2089. (a) Heeres, H. J. Ph.D. Thesis 1990 ; Chapter 4. (b) Heeres, H. J.; Renkema, J.; Booij, M.; Meetsma, A.; Teuben, J. H. Organometallics 1988, 7, 2495. (a) Ref. 5a. (b) Thompson, M. E.; Bercaw, J. E. Pure Appl. Chem. 1984, 56, 1. (c) Watson, P. L. J. Chem. Soc., Chem. Commun. 1983, 276. (d) Ref. 5b. (e) Evans, W. J.; Meadows, J. H.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1984, 106, 1291. (f) Ref. 7f. (c) Deelman, B.-J. Ph. D. Thesis, University of Groningen, 1994; Chapter 5. (d) Obora, Y.; Ohta, T.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1997, 119, 3745. For reviews, see: (a) B. J. Wakefield, The Chemistry of Organolithium Compounds, Pergamon, Oxford, 1974. (b) Geschwend, H. W.; Rodriguez, H. R. Org. React. 1979, 26, 1. (c) Snieckus, V. Chem. Rev. 1990, 90, 879. (d) M. Gray, M. Tinkl, V. Snieckus in Comprehensive Organometallic Chemistry II, E. W. Abel, F. G. A. Stone, G. Wilkinson, Eds.; Pergamon, Oxford, 1995. (e) Clark, R.

200

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

36

37

38 39 40

41

42 43

D.; Jahangir, A. Org. React. 1995, 47, 1. For examples of mechanistic studies, see: (f) Beak, P.; Meyers, A. I. Acc. Chem. Res. 1986, 19, 356. (g) Bauer, W.; Schleyer, P. v. R. J. Am. Chem. Soc. 1989, 111, 7191. van Eikema Hommes, N. J. R.; Schleyer, P. v. R. Angew. Chem., Int. Ed. Engl. 1992, 31, 755. (h) Sa, J. M.; Martorell, G.; Frontera, A. J. Org. Chem. 1996, 61, 5194. (i) Anderson, D. R.; Faibish, N. C.; Beak, P. J. Am. Chem. Soc. 1999, 121, 7553. The mechanism of ortho-alumination is considered to be more complex than that of the conventional ortho-lithiation,35 but complex-induced proximity effects are believed to play an important role, see: Uchiyama, M.; Naka, H.; Matsumoto, Y.; Ohwada, T. J. Am. Chem. Soc. 2004, 126, 10526. Organolanthanide chemistry resembles that of main-group metals, due to their Lewis acidity and the shielded nature of the 4fn orbitals which renders their chemistry highly ionic and governed more by electrostatic and steric factors than by orbital filling energetics. For general reviews on rare-earth metal chemistry, see: (a) Aspinall, H. C. Chem. Rev. 2002, 102, 1807-1850. (b) Edelmann, F. T.; Freckmann, D. M. M.; Schumann, H, Chem. Rev. 2002, 102, 1851-1896. (c) Arndt, S.; Okuda, J. Chem. Rev. 2002, 102, 1953-1976 (d) Shibasaki, M.; Yoshikawa, N. Chem. Rev. 2002, 102, 21872210. (e) Inanaga, J.; Furuno, H.; Hayano, T. Chem. Rev. 2002, 102, 2211-2226. (f) Molander, G. A. Chemtracts: Org. Chem. 1998, 18, 237-263. (g) Edelmann, F. T. Top. Curr. Chem. 1996, 179, 247276. (h) Edelmann, F. T. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1995; Vol. 4, Chapter 2. (i) Schumann, H.; Meese-Marktscheffel, J. A.; Esser, L. Chem. Rev. 1995, 95, 865-986. (j) Schaverien, C. J. Adv. Organomet. Chem. 1994, 36, 283-362. (k) Evans, W. J. Adv. Organomet. Chem. 1985, 24, 131-177. (l) Marks, T. J.; Ernst, R. D. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1982; Chapter 21. For a recent review on complex-induced proximity effects, see: Whisler, M. C.; MacNeil, S.; Snieckus, V.; Beak, P. Angew. Chem. Int. Ed. 2004, 43, 2206. Experimental errors were determined by statistical analysis of values obtained from at least two experiments. The Cp* ligands in each Cp*2La moiety are crystallographical equivalent, but the two Cp*2La moieties in 22a2C7H8 are not. Two conformations were found for each Cp* ligand, resulting in two different environments for each lanthanum metal in 22a2C7H8. The average twist angles are 0.0(7) for La1 and 0(1) for La2. In one conformation, La1 is coordinated to both the C1A-C5A ring (Cp1) with centroid Ct1 and the C1Aa-C5Aa ring (Cp1a) with centroid Ct1a, while La2 is coordinated to both the C11C-C15C ring (Cp2) with centroid Ct2 and the C11Ca-C15Ca ring (Cp2a) with centroid Ct2a. In the other conformation, La1 is coordinated to the C1B-C5B ring (Cp3) with centroid Ct3 and the C1Ba-C5Ba ring (Cp3a) with centroid Ct3a, while La2 is coordinated to both the C11D-C15D ring (Cp4) with centroid Ct4 and the C11Da-C15Da ring (Cp4a) with centroid Ct4a. The distance average of La1-Cp1 and La1-Cp3 is denoted by La1-Cp, while the distance average of La2-Cp2 and La2-Cp4 is denoted by La2-Cp. The distance average of La1-Ct1 and La1-Ct3 is denoted by La1-Ct, while the distance average of La2-Ct2 and La2-Ct4 is denoted by La2-Ct. The bond average of Ct1La1-Ct1a and Ct3-La1-Ct31 is denoted by Ct-La1-Ct, while the bond average of Ct2-La2-Cp2a and Ct4-La2-Ct4a is denoted by Ct-La2-Ct. The two Cp*2La moieties are crystallographically equivalent, whereas the Cp* ligands in each Cp*2La moiety are not, resulting in four different environments for the metal center in 22d2C7H8. The Cp* ligands are more eclipsed than staggered, as evidenced by twist angles of 24(2), 12(2), 11.3(3) and 12(2) in each conformation (compared to 36 for a perfectly staggered arrangement). In the first conformation La is coordinated to both the C1A-C5A ring (Cp1a) with centroid Ct1a and the C11C-C15C ring (Cp2a) with centroid Ct2a. In the second conformation La is coordinated to both Cp1a and the C11D-C15D ring (Cp2b) with centroid Ct2b. In the third conformation La is coordinated to the C1B-C5B ring (Cp1b) with centroid Ct1b and Cp2a. In the fourth conformation La is coordinated to both Cp1b and Cp2b. The distance averages between metal and Cp carbon in each conformation are denoted by La-Cp. The distance average between metal and centroid in each conformation are denoted by La-Ct. The bond averages of centroid-metal-centroid in each conformation are denoted by Ct-La-Ct. The twist angle is defined as the average of the five smallest dihedral angles formed between the ten planes which consist of a ring carbon and the two centroids. (a) Watson, P. L.; Whitney, J. F.; Harlow, R. L. Inorg. Chem. 1981, 20, 3271. (b) Ref. 16a.

201

Chapter 5

44

45

46

47

48

49

50

51 52

53 54

(a) Evans, W. J.; Drummond, D. K. J. Am. Chem. Soc. 1986, 108, 7440. (b) [(Cp*2Sm)2(-2:2PhC4Ph)]: Evans, W. J.; Keyer, R. A.; Zhang, H.; Atwood, J. L. J. Chem. Soc., Chem. Commun. 1987, 837. (c) Rausch, M. D.; Moriarty, K. J. Organometallics 1986, 5, 1281. (d) {(Cp*2Sm)2[-4PhCH=C(O)-C(O)=CHPh]}: Evans, W. J.; Drummond, D. K. J. Am. Chem. Soc. 1988, 110, 2772. For examples, see: (a) Karmazin, L.; Mazzanti, M.; Pcaut, J. Chem. Commun. 2002, 654. (b) Banerjee, S.; Emge, Th. J.; Brennan, J. G. Inorg. Chem. 2004, 43, 6307. (c) Schumann, H.; Herrmann, K.; Mhle, S. H.; Dechert, S. Z. Anorg. Allg. Chem. 2003, 629, 1184. Ct1 is the centroid of the C1-C5 ring (Cp1) and Ct2 is the centroid of the C1c-C5c ring (Cp2). The average distances of the metal to the cyclopentadienyl carbons of each ring are denoted by La-Cp1 and La-Cp2. Selected bond lengths (): La-C21, 4.351(3); La-S, 3.1134(7); La-Cp1, 2.828(5), LaCp2, 2.804(5); La-Ct1, 2.063(1), La-Ct2, 3.035(1); C21-C22, 1.368(4), C22-C23, 1.421(5); C23-C24, 1.347(5); C24-S, 1.722(3); C21-S, 1.743(2). Selected bond angles (): Ct1-La-Ct2, 133.91(3), C21-SC24, 96.0(1); S-La-C21a, 78.47(6); S-C21-La, 35.98(7); La-C21-C22, 114.8(2); La-S-C24, 102.7(1); La-S-C21, 124.8(1); S-C21-C22, 105.0(2); C22-C23-C24, 113.1(3); C23-C24-S, 108.8(2). For examples, see: (a) Evans, W. J.; Davis, B. L.; Nyce, G. W.; Perotti, J. M.; Ziller, J. W. J. Organomet. Chem. 2003, 677, 89. (b) Heeres, H. J.; Meetsma, A.; Teuben, J. H. Angew. Chem. Int. Ed. 1990, 29, 420. (c) Scholz, A.; Smola, A.; Scholz, J.; Loebel, J.; Schumann, H.; Thiele, K.-H. Angew. Chem. Int. Ed., 1991, 30, 435. (d) Kretschmer, W. P., personal communication (Verslag werkgroep "Organometaalchemie en Homogene Katalyse"; University of Groningen, 1997; Vol. 11, p. 31). (e) Thiele, K.-H.; Bambirra, S.; Sieler, J. Angew. Chem. Int. Ed. 1998, 37, 2886. (f) Gagn, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (g) Evans, W. J.; Foster, S. E. J. Organomet. Chem. 1992, 433, 79. (a) Heeres, H. J.; Nijhoff, J,; Teuben, J. H.; Rogers, J. D. Organometallics 1993, 12, 2609. (b) Evans, W. J.,; Keyer, R. A.; Ziller, J. W. Organometallics 1990, 9, 2628. (c) Evans, W. J.; Keyer, R. A.; Ziller, J. W. Organometallics 1993, 12, 2618. (d) Forsyth, C. M.; Nolan, S. P.; Stern, C. L.; Marks, T. J.; Rheingold, A. L. Organometallics 1993, 12, 3618. (a) Atwood, J. L.; Hunter, W. E.; Wayda, A. L.; Evans, W. J. Inorganic Chemistry 1981, 20, 4115. (b) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. Organometallics 1983, 2, 709. (c) Shen, Q.; Zheng, D.; Lin, L.; Lin, Y. J. Organomet. Chem. 1990, 391, 307. (d) Evans, W. J.; Drummond, D. K.; Hanusa, T. P.; Olofson, J. M. J. Organomet. Chem. 1989, 376, 311. (e) Ren, J.; Hu, J.; Lin, Y.; Xing, Y. Polyhedron 1996, 15, 2165. According to the I parameters, their electronic effects are identical (I = -0.01, R = -0.16), both electron-donating via an inductive/field and resonance mechanism.51b According to the Swain and Lupton parameters F and R, the methyl group (F = 0.01, R = -0.018) is inductively weakly electronwithdrawing and the tert-butyl group (F = -0.02, R = -0.18) is inductively mildly electronwithdrawing. 51a (a) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev., 1991, 91, 165. (b) Charton, M. Prog. Phys. Org. Chem. 1981, 13, 119. For reviews, see: (a) Gallo, R. Prog. Phys. Org. Chem. 1983, 14, 115. (b) Frster, H.; Vgtle, F. Angew. Chem. Int. Ed. Engl. 1977, 16, 429. Numerous methods to quantify the steric effects of substituents have been proposed, see for examples: (c) Hirsch, J. A. Top. Stereochem. 1967, 1, 199. (d) Allinger, N. L.; Hirsch, J. A.; Miller, M. A.; Tyminski, I. J.; Van-Catledge, F. A. J. Am. Chem. Soc. 1968, 90, 1199. (e) Tolman, C. A. Chem. Rev. 1977, 77, 315. DeSanto, J. T.; Mosbo, J. A.; Storhoff, B. N.; Bock, P. L.; Bloss, R. E. Inorg. Chem. 1980, 19, 3086. (f) DeTar, D. F.; Binzet, S.; Darba, P. J. Org. Chem. 1987, 52, 2074. (g) Brown, T. L. Inorg. Chem. 1992, 31, 1286. (h) Komatsuzaki, T.; Akai, I.; Sakakibara, K.; Hirota, M. Tetrahedron 1992, 48, 1539. (i) White, D.; Taverner, C.; Leach, P. G. L.; Coville, J. J. Comput. Chem. 1993, 14, 1042. (j) Hirota, M.; Sakakibara, K.; Yuzuri, T.; Kuroda, S.-I. J. Phys. Org. Chem. 2001, 14, 788. (k) Seil, P. R.; Leal, K. Z.; Yoneda, J. D. J. Phys. Org. Chem. 2002, 15, 801. (a) Taft, R. W. In Steric Effects in Organic Chemistry; Newman, M. S.; Ed.; Wiley: New York. (b) Shorter, J. J. Chem. Soc. Quart. Rev. 1970, 24, 443. The values are defined by a relationship derived from Van der Waals radii (X = rX - rH, where rX and rH are the Van der Waals radii of the X group, respectively). Very good correlations have been obtained between these values and rates of esterification of substituted carboxylic acids with methanol or ethanol, thereby indicating that the values are solely a function of steric effects.

202

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

55

56

57

58

59

60

61

62 63

64

Additional values for unsymmetrical substituents X were obtained from correlation with the rates of esterification and hydrolysis of acids XCO2H and esters XCO2R, see: (a) Charton, M. J. Am. Chem. Soc. 1975, 97, 1552. (b) Charton, M. J. Am. Chem. Soc. 1975, 97, 3691. (c) Charton, M. J. Org. Chem. 1976, 41, 2217. (d) Charton, M. J. Org. Chem. 1977, 42, 3531. (e) Charton, M. Top. Curr. Chem. 1983, 114, 57. According to the ES and scale, the steric requirements of the phenyl group ( = 1.66, ES = -2.55) are larger than of the tert-butyl group ( = 1.24, ES = -1.54). This discrepancy with other scales has been attributed to the appreciable resonance effect of the phenyl group. More recently, Sung et al. proposed the steric substituent constant SA, based on isodesmic reactions and ab initio calculations of substituted adamantane systems. According to the SA scale, the tert-butyl group (SA = -12.45) is considerably larger than the phenyl (SA = -5.10) and the methyl group (SA = -2.02), see: Sung, K.; Chen, F.-L. Org. Lett. 2003, 5, 889. The 2-thienyl group is most likely smaller ( (cC4H7) = 0.51, (cC5H9) = 0.71), more -electronwithdrawing (F = 0.13, I = 0.19) and less -electron-releasing (R = -0.08) than the phenyl group ( = 1.66, F = 0.12, I = 0.12, R = -0.13).52 The CH2CH2Ph ( = 0.70, F = -0.01, R = -0.11) and CH2CH2iPr ( = 0.68, F(CH2iPr) = -0.01, R(CH2iPr) = -0.11) groups have similar steric and electronic properties.52, Although the isopropyl group is more -electron-withdrawing (F = 0.04), its steric bulk is larger ( = 0.76) and it is more electron-releasing (R = -0.19) than the CH2CH2Ph and CH2CH2iPr groups. Extensive research on oxidative C-C coupling of alkynes in group 4 metal alkynyls has demonstrated that the coupling reaction occurs via an uncoupled, asymmetrically acetylide bridged dimeric species. Numerous experimental and theoretical studies have been dedicated to elucidate the role of the metal, the ligand and the alkyne substituents. For some examples, see: (a) Teuben, J. H.; de Liefde Meijer, H. J. J. Organomet. Chem. 1969, 17, 87. (b) Sekutowski, D. G.; Stucky, G. D. J. Am. Chem. Soc. 1976, 98, 1376. (c) Wood, G. L.; Knobler, C. B.; Hawthorne, M. F. Inorg. Chem. 1989, 28, 382. (d) Erker, G.; Frmberg, R.; Benn, R.; Mynott, D.; Angermund, D.; Krger, C. Organometallics 1989, 8, 911. (e) Rosenthal, U.; Grls, H. J. Organomet. Chem. 1992, 439, C36. (f) Cuenca, T.; Gmez, R.; Gmez-Sal, P.; Rodrguez, G. M.; Royo, P. Organometallics 1992, 11, 1229. (g) Varga, V.; Mach, K.; Hiller, J.; Thewalt, U.; Sedmera, P.; Polasek, M. Organometallics 1995, 14, 410. (h) Pellny, P.M.; Peulecke, N.; Burlakov, V. V.; Tillack, A.; Baumann, W.; Spannenberg, A.; Kempe, R.; Rosenthal, U. Angew. Chem., Int. Ed. Engl. 1997, 36, 2615. (a) Kumar, P. N. V. P.; Jemmis, E. D. J. Am. Chem. Soc. 1988, 110, 125. (b) Jemmis, E. D.; Giju, K. T. Angew. Chem. Int. Ed. Engl. 1997, 36, 606. (c) Jemmis, E. D.; Giju, K. T. J. Am. Chem. Soc. 1998, 120, 6952. (d) Jemmis, E. D.; Phukan, A. K.; Giju, K. T. Organometallics 2002, 21, 2254. (a) Burger, B. J.; Santarsiero, B. D.; Trimmer, M. S.; Bercaw, J. E. J. Am. Chem. Soc. 1988, 110, 3134. (b) Doherty, N.; Bercaw, J. E. J. Am. Chem. Soc. 1985, 107, 2670. (c) Halpern, J.; Okamoto, T.; Zakhariev, A. J. Mol. Catal. 1976, 2, 65-68 (d). Lin, Z.; Marks, T. J. J. Am. Chem. Soc. 1990, 112, 5515. Although values indicate that a methyl group ( = 0.52) is larger than a methoxy group ( = 0.36), this scale is more suited to aliphatic molecules.52a To overcome this deficiency, Gallo et al. proposed a scale of ortho-steric parameters (S0) based on the kinetics of quaternization by methyl iodide of 33 substituted pyridines in acetonitrile at 33 C to quantify the steric size of ortho-substituents in (hetero)aromatic compounds. No correlation with electronic substituent effects was found. The S0 scale indicates that the o-methyl group (S0 = -0.73) is smaller than the o-methoxy group (S0 = -1.28), see: (a) Berg, U.; Gallo, R.; Klatte, G.; Metzger, J. J. Chem. Soc., Perkin Trans.2 1980, 1350. (b) Gallo, R.; Rousel, C.; Berg, U. Adv. Heterocycl. Chem. 1988, 43, 173. (a) Ringelberg, S. N., Ph. D. Thesis, University of Groningen, 2001; Chapter 2. (b) Ringelberg, S. N.; Meetsma, A.; Troyanov, S. I.; Hessen, B.; Teuben, J. H. Organometallics 2002, 21, 1759. The transition state for alkyne protonolysis places the aromatic moiety of the incoming alkyne molecule further away from the metal center. C-H activation seems also to be electronically favored, since it involves a sp carbon and a hydrogen atom orbital which provide more overlap in the fourcenter transition state than two sp carbon orbitals. It can be anticipated that the decrease of electron density at the triple bond will weaken the interaction with the electrophilic lanthanide metal. On the other hand, polarization of the triple bond, as to induce

203

Chapter 5

65

66

67

68

69

70 71 72 73 74

75

76

77

a partial positive charge at the -position and a partial negative charge at the -position is well-known to facilitate insertion into electrophilic metal-carbon bonds.60 (a) Horton, A. D.; Orpen, A. G. Organometallics 1991, 10, 3910-3918. (b) Horton, A. D. J. Chem. Soc., Chem. Commun. 1992, 185. (c) Temme, B.; Erker, G.; Frhlich, R.; Grehl, M. Angew. Chem., Int. Ed. Engl. 1994, 33, 1480. (d) Temme B.; Erker, G.; Frhlich, R.; Grehl, M. J. Chem. Soc., Chem. Commun. 1994, 1713. (e) Venne-Dunker, S.; Ahlers, W.; Erker, G.; Frhlich, R. Eur. J. Inorg. Chem. 2000, 1671. (f) Ahlers, W.; Temme, B.; Erker, G.; Frhlich, R.; Fox, T. J. Organomet. Chem. 1997, 527, 191. (g) Erker, G.; Venne-Dunker, S.; Kehr, G.; Kleigrewe, N.; Frhlich, R.; Mck-Lichtenfeld, C.; Grimme, S. Organometallics 2004, 23, 4391. (h) Ahlers, W.; Temme, B.; Erker, G.; Frhlich, R.; Fox, T. J. Organomet. Chem. 1997, 527, 191. (a) Shapiro, P. J.; Cotter, W. D.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1994, 116, 4623. (b) Kretschmer, W. P.; Troyanov, S. I.; Meetsma, A.; Hessen, B.; Teuben, J. H. Organometallics 1998, 17, 284. (c) Kretschmer, W. P. personal communication. (d) Hultzsch, K. C.; Voth, P.; Beckerle, K.; Spaniol, T. P.; Okuda, J. Organometallics 2000, 19, 228. (e) Trifonov, A. A.; Spaniol, T. P.; Okuda, J. Organometallics 2001, 20, 4869. In spite the well-known electronic preference for 2,1-insertion of 1-arylalk-1-enes and the observations based on stoichiometric reactions, products deriving from an initial 2,1-insertion of styrene into Ln-H bonds have been observed.66 This behavior has been rationalized in terms of a reversible insertion and the higher reactivity of the 1,2-insertion product towards insertion or protonolysis.66b,e C-O activation products are typically observed for reactions of rare-earth metallocene hydride derivatives [Cp*2Ln(-H)]2 with dialkyl ethers, vinyl alkyl ethers and polydimethylsiloxanes. For examples, see: (a) den Haan, K. H. Ph. D. Thesis, University of Groningen, 1986; Chapter 6. (b) Watson, P. L. J. Chem. Soc., Chem. Commun. 1983, 276. (c) Deelman, B.-J. Ph. D. Thesis, University of Groningen, 1994; Chapter 5. (d) Deelman, B.-J.; Booij, M.; Meetsma, A.; Teuben, J. H.; Kooijman, H.; Spek, A. L. Organometallics 1995, 14, 2306. (e) Evans, W. J.; Chamberlain, L. R.; Ulibarri, T. A.; Ziller, J. W. Organometallics 1988, 110, 6423. Both alkyl Cp*2LnR and hydride [Cp*2Ln(-H)]2 derivatives undergo facile ortho-metalation of pyridine, possibly followed by C-C coupling reactions. For examples, see: (a) Ringelberg, S. N. Ph. D. Thesis, University of Groningen, 2001; Chapter 4. (b) Ringelberg, S. N. Ph. D. Thesis, University of Groningen, 2001; Chapter 5. Booij, M.; Kiers, N. H.; Heeres, H. J.; Teuben, J. H. J. Organomet. Chem. 1989, 364, 79 Ref. 4b; p. 4. Brandsma, L.; Verkruijsse, H. D. Preparative Polar Organometallic Chemistry; Springer: Berlin, 1986, Vol. 1; p.167. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals; 2nd ed.; Pergamon Press: Oxford, 1980. For 1H NMR data, see: (a) Abraham, R. J.; Reid, M. J. Chem. Soc., Perkin Trans. 2 2001, 1195. (b) Brandsma, L.; Verkruijsse, H. D.; de Jong, R. L. P. Recl. Trav. Chim. Pays-Bas 1985, 226. For 13C NMR data, see: (c) Aitken, R. A.; Horsburg, C. E. R.; McCreadie, J. G.; Seth, S. J. Chem. Soc., Perkin Trans. 1 1994, 1727. For infrared data, see: (c) Cook, C. D.; Danyluk, S. S. Tetrahedron 1963, 19, 177. For MS data, see: (d) Ref. 74c. (e) Occolowitz, White Aust. J. Chem. 1968, 21, 997. For 1H NMR data, see: (a) Negishi, E.; Kotora, M.; Xu, C. J. Org. Chem. 1997, 62, 8957. (b) Padwa, A.; Krumpe, K. E.; Weingarten, M. D. J. Org. Chem. 1995, 60, 5595. (c) Carran, J.; Waschbuesch, R.; Marinetti, A.; Savignac, P. Synthesis 1996, 12, 1494. For 13C NMR data, see: (d) Ref. 75a. (e) Ref. 75c. For infrared data, see: (f) Barton, T.; Groh, B. L. J. Org. Chem. 1985, 50, 158. (g) Schmidt, C.; Thazhuthaveetil, J. Can. J. Chem. 1973, 51, 3620. For MS data, see: (d) Ref. 75f. (e) Ref. 75g. For 1H NMR data, see: (a) Ref. 10. (b) Brandsma, L.; Vasilevsky, S.F.; Verkruijsse, H. D. Application of Transition Metal Catalysts in Organic Synthesis, Springer: Berlin, 1998; Chapter 10. (c) Sarkar, A.; Talwar, S. S. J. Chem. Soc., Perkin Trans. 1 1998, 4141. For 13C NMR data, see: (d) Ref. 76a. (e) Ref. 76c. For infrared data, see: (f) Ref. 76c. For 1H NMR data, see: (a) Ref. 10. (b) Ref. 74c. (c) Stille, J. K.; Simpson, J. H. J. Am. Chem. Soc. 1987, 109, 2138. (d) DAuria, M.; Ferri, T. J. Org. Chem. 1995, 60, 8360. (e) Wu, R.; Schumm, J. S.; Darren, L.; Tour, J. M. J. Org. Chem. 1996, 61, 6906. For 13C NMR data, see: (f) Ref. 10. (g) Ref.

204

The oligomerization of (hetero)aromatic 1-alkynes catalyzed by rare-earth metallocenes

78 79 80

81

82 83 84 85

86 87

88

89

90 91

92

93 94 95

77c. (h) Ref. 77d. (i) Ref. 77e. For infrared data, see: (j) Ref. 77c. (k) Ref. 77e. For MS data, see: (l) Fuerstner, A.; Nikolakis, K. Liebigs Ann.Org. Bioorg. Chem. 1996, 12, 2107. For 1H NMR data, see: (a) Ref. 10. (b) Ref. 76c. For 13C NMR data, see: (c) Ref. 76c. For infrared data, see: (f) Ref. 76c. For 1H NMR data, see: (a) Ref. 10. (b) Ref. 74c. For 13C NMR data, see: (c) Ref. 74c. For infrared data, see: (f) Beny, J.-P.; Dhawan, S. N.; Kagan, J.; Sundlass, S. J. Org. Chem. 1982, 47, 2201 For 1H NMR data, see: (a) Rodriguez, J. G.; Marin-Villamil, R.; Cano, F. H.; Fonseca, I. J. Chem. Soc. Perkin Trans.1. 1997, 5, 709. (b) Novk, Z.; Szab, A.; Rpas, J.; Kotschy, A. J. Org. Chem. 2003, 68, 3327. For 13C NMR data, see: (c). Ref. 80a. (d) Ref. 80b. For MS data, see: (e) Ref. 80b. For 1H NMR data, see: (a) Woodgate, Paul D.; Sutherland, H. S. J. Organomet. Chem. 2001, 629, 131. (b) Naulty, R. H.; Cifuentes, M. P.; Humprey, M. G.; Houbrechts, S.; Boutton, C. J. Chem. Soc., Dalton Trans. 1997, 4167. (c) Ref. 10. (d) Ref. 80a. (e) Ref. 77l. For 13C NMR data, see: (e) Ref. 81a. (f) Ref. 81b. (g) Ref. 77l. For infrared data, see: (h) Ref. 77l. (i) Ref. 81a. (j) Ref. 81b. (k) Ref. 80a. For MS data, see: (l) Ref. 80a. (m) Ref. 77l. For 1H NMR data, see: (a) Ref. 10. For infrared data, see: (b) Wentrup, C.; Winter, H.-W. Angew. Chem. 1978, 90, 643. SMART, SAINT, SADABS, XPREP and SHELXK/NT. Smart Apex Software Reference Manuals 2000, Bruker AXS Inc., Madison, Wisconsin, USA. Beurskens, P. T.; Beurskens, G.; De Gelder, R.; Garca-Granda, S.; Gould, R. O.; Isral, R.; Smits, J. M. M. The DIRDIF-99 Program System 1999, University of Nijmegen, The Netherlands. (a) Kresge, A. J.; Powell, M. F. J. Org. Chem. 1986, 51, 819. (b) Kresge, A. J.; Powell, M. F. J. Org. Chem. 1986, 51, 822. (c) Lin, A C.; Chiang, Y.; Dahlberg, D. B.; Kresge, A. J. J. Am. Chem. Soc. 1983, 105, 5380. Aroella, T.; Arrowsmith, C. H.; Hojatti, M.; Kresge, A. J.; Powell, M. F.; Tang, Y. S.; Wang, W.-H. J. Am. Chem. Soc. 1987, 109, 7198. (a) Drenth, W.; Loewenstein, A. Recl. Trav. Chim. Pays-Bas 1962, 81, 635. (b) Cram, D. J. Fundamentals of Carbanion Chemistry, Academic Press: New York, 1965. (c) Charman, H. B.; Vinard, D. R.; Kreevoy, M. M. J. Am. Chem. Soc. 1962, 84, 347. (d) Cook, C. D.; Danyluk, S. S. Tetrahedron 1963, 19, 177. (e) Eaborn, C.; Skinner, G. A.; Walton, D. R. M. J. Chem. Soc. (B) 1966, 922. (f) Eaborn, C.; Skinner, G. A.; Walton, D. R. M. J. Chem. Soc. (B) 1966, 989. (g) Queignac, R.; Wojtkowiak, B. Bull. Soc. Chim. Fr. 1970, 860. (h) Charton, M. J. Org. Chem. 1972, 37, 3684. (i) Lin, A. C.; Chiang, Y.; Dahlberg, D. B.; Kresge, A. J. J. Am. Chem. Soc. 1983, 105, 5380. (j) Powell, M. F.; Peterson, M. R.; Csizmadia, I. G. J. Mol. Struct. (THEOCHEM) 1983, 92, 323. (j) Kresge, A. J.; Powell, M. F. J. Org. Chem. 1986, 51, 819. (a) Laurence, C.; Queignec, R. J. Chem. Soc. Perkin Trans. 2 1992, 1915. (b) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Duce, P. P.; Morris, J. J.; Taylor, P. J. J. Chem .Soc. Perkin Trans. 2 1989, 699. (a) Isaacs, N. S. Physical Organic Chemistry, 2nd ed.; Longman: Essex, 1996. (b) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry, 3rd ed.; Plenum Press: New York, 1990. (c) Carroll, F. A. Perspectives on Structure and Mechanism in Organic Chemistry; Brooks/Cole Publishing Company: New York, 1998. (d) Lowry, T. H.; Richardson, K. H. Mechanism and Theory in Organic Chemistry; Harper & Row Publishers: New York, 1987. For a recent review, see: Krygowski, T. M.; Stpie, B. T. Chem. Rev. 2005, 105, 3482. (a) Wiberg, K. B.; Hammer, J. D.; Zilm, K. W.; Keith, T. A.; Cheeseman, J. R.; Duchamp, J. C. J. Org. Chem. 2004, 69, 1086. (b) Gryff-Keller, A. THEOCHEM 2002, 587, 171. (c) Bernard, G. M.; Wasylishen, R. E. Solid State Nucl. Magn. Reson. 2002, 21, 86. (d) Bohmann, J.; Farrar, T. C. J. Phys. Chem. 1996, 100, 2646. (e) Klaus, E.; Sebald, A. Magn. Reson. Chem. 1994, 32, 679. For reviews, see: (a) Hehre, W. J.; Taft, R. W.; Topsom, R. D. Prog. Phys. Org, Chem. 1976, 12, 159. (b) Ewing, D. F. In Correlation Analysis in Chemistry. Recent Advances.; Chapman, N. B.; Shorter, J., Eds.; Plenum, New York, 1978. Nelson, G. L.; Williams, E. A. Progr. Phys. Org. Chem. 1976, 12, 234. Dawson, D. A.; Reynolds, W. F. Can. J. Chem. 1975, 53, 373. Amass, A. J.; Brough, P. E.; Colclough, M. E.; Philbin, I. M.; Perry, M. C. Designed Monomers and Polymers 2004, 7, 413.

205

Chapter 5

96

97

98

99

Dawson et al.94 found the following correlations of 13C chemical shifts in cyclohexane-d12 for 4substituted phenylacetylenes with the F and R values of the corresponding 1-alkyne substituent: (a) C-1 = + 2.86(27)F + 5.23(43)R (R2 = 0.988) and (b) C-2 = + 2.86(27)F + 5.23(43) R (R2 = 0.990), C-1 and C-2 representing the chemical shift of the terminal and internal carbon, respectively, and representing the chemical shift of the corresponding carbon of the unsubstituted phenylacetylene. Recently, DFT calculations of diarylacetylens revealed that chemical shifts concur with effective charges on sp-carbons in para-tolanes, whereas in ortho- tolanes magnetic anisotropy complicates the analysis, making 13C NMR data inapplicable for ascribing triple bond polarization. See: Rubin, M.; Trofimov, A.; Gevorgyan, V. J. Am. Chem. Soc. 2005, 127, 10243. For examples, see: (a) Closs, G. L.; Closs, L. E. J. Am. Chem. Soc. 1963, 85, 2022. (b) Closs, G. L.; Larrabee, R. B. Tetrahedron Lett. 1965, 287. (c) Streitweiser, A., Jr.; Caldwell, R. A.; Young, W. R. J. Am. Chem. Soc. 1969, 91, 529. (d) Maksi, Z. B.; Eckert-Maksi, M. Tetrahedron 1969, 25, 5113. (e) Maksi, Z. B.; Randi, M. J. Am. Chem. Soc. 1973, 95, 6522. (f) Luh, T.-Y.; Stock, L. M. J. Am. Chem. Soc. 1974, 96, 3712. For reviews on acid-base reactions in benzene, see: (a) Ref. 59a. (b) Izutsu, K. Acid-Base Dissociations Constants in Dipolar Aprotic Solvents, Blackwel Scientific Press: Oxford, 1990. For other examples of acid-base reactions in nonpolar media, see: (c) Steigman, J.; Cronkright, W. J. Am. Chem. Soc. 1970, 92, 6729. (d) Hughes, E. D.; Ingold, C. K.; Mok, S. F.; Patai, S.; Pocker, Y. J.Chem.Soc. 1957, 1265. (e) Jarczewski, A.; Hubbard, C. D. J. Mol. Struct. 2003, 649, 287. (f) Leito, I.; Rodima, T.; Koppel, I. A.; Schwesinger, R.; Vlasov, V. M. J. Org. Chem. 1997, 62, 8479. (g) Rodima, T.; Kaljurand, I.; Pihl, A.; Memets, V.; Leito, I.; Koppel, I. J. Org. Chem. 2002, 67, 1873. (h) Antipin, I. S.; Vedernikov, A. N.; Konovalov, A. I. Zh. Org. Khim. 1986, 22, 446. (i) Hirose, K.; Tanaka, M. Bull. Chem. Soc. Jpn. 1977, 50, 608. (j) Bruckenstein, S.; Saito, A. J. Am. Chem. Soc. 1965, 87, 698. (k) Hojatti, M.; Leffek, K. T. Can. J. Chem. 1986, 64, 2365. (l) Poonia, N. S.; Bajaj, A. V. Chem. Rev. 1979, 79, 389.

206

The organolanthanide-catalyzed polymerization of diynes

6.
6.1.

The organolanthanide-catalyzed polymerization of diynes

Introduction

During the last two decades the research area of -conjugated polymers has experienced a tremendous growth.1 With an ever increasing interest from both academia and industry these polymers have been developed as advanced materials for electronic and photonic applications. The immense interest in -conjugated polymers started in 1977, when Shirakawa, MacDiarmid and Heeger discovered that oxidation with chlorine, bromine or iodide vapor made polyacetylene 109 times more conductive.2 The study of -conjugated polymers has since advanced rapidly in various directions. More recently, this field has witnessed not only a plethora of commercialized applications of conductive plastics (e.g. corrosion inhibitors, compact capacitors, antistatic coatings, electromagnetic shielding of computers, smart windows that can vary the amount of light they allow to pass, etc.), but the development of electroluminescent polymers as well. It is mainly their inherent synthetic flexibility, their potential ease of processing and the possibility of tailoring characteristic properties to accomplish a desired function that makes them promising candidates for a variety of applications in material science. Thus, they are used as laser dyes,3 scintillators,3 light-emitting diodes (LEDs),4 piezoelectric and pyroelectric materials,5 photoconductors,6 photovoltaic cells7 and are investigated for optical data storage,8 optical switching and signal processing9 as well as in nonlinear optical applications.10 Scheme 6-1. Well-established classes of -conjugated polymers.

n Carbyne

n Poly(acetylene) PA

n Poly(diacetylene) PDA

n Poly(phenylene) PP PPV

n Poly(phenylenevinylene) PPE

n P oly(phenyleneethynylene)

The class of conjugated polymers which has found the most attention in the last years are undoubtedly the poly(p-phenylenevinylene)s (PPVs) which were applied as an organic polymeric LED in 1990.11 Other well-established classes of conjugated polymers are the poly(diacetylene)s (PDAs),12 poly(phenylene)s (PPP)13 and the aforementioned poly(acetylene)s (Scheme 6-1).14 Poly(aryleneethynylene)s (PAEs) have attracted considerable attention only during the last decade, as applications, such as molecularly wired sensors,15 polarizers for liquid crystalline displays16 and LEDs, were developed in the late 1990s.17 Among the numerous described conjugated polymers containing heterocyclic units in the backbone those containing 2,5-thienyl units represent a particularly interesting class.1 For example, poly(thiophene)s led to materials showing an important environmental and thermal stability with good conductivity when compared to other poly(heteroarylene)s.18 Poly(3-alkylthiophene)s are, moreover, soluble in common organic solvents which facilitates their processability and has widened their field of application. Substituted poly(2,5-thienylvinylene)s19 display high nonlinear optical responses, moderate charge mobilities and good electroluminescent properties and substituted poly(2,5-thienylethynylene)s20 were found to exhibit strong photoluminescence and third-order optical nonlinearity. As a consequence, thiophene-based oligomers and polymers are widely used in active organic electronic and photonic devices, such as organic field effect transistors (OFETs),21 LEDs22 or solar cells.23 Each class of conjugated polymer has its own methodology of synthesis which is characterized by its versatility in modifying the molecular structure of the material, its compatibility with various functional groups

207

Chapter 6

Scheme 6-3. Well-established thienyl-containing -conjugated polymers.

S n Poly(thiophene) PT PTV

S n Poly(2,5-thienylenevinylene)

S n P oly(2,5-thienyleneethynylene) PTE

and its propensity to introduce structural defects into the polymers. The use of organometallic methodologies, widely exploited in the synthesis of well-defined molecules, has proven to be of great importance in the preparation of polydisperse conjugated materials, especially when a high regio- and stereoselectivity is required in building the polymeric backbone.24 For example, polyacetylenes have been prepared from 1-alkynes by Ziegler-Natta type polymerization with MoCl6, WCl6 or TaCl5 based catalysts25 or by using organorhodium(I) complexes,26 the palladium-catalyzed coupling of 1-alkynes with halides (the Cassar-Heck-Sonogashira reaction) constitutes the classic method to prepare PAEs,17b the palladium-catalyzed coupling of halides with alkenes (the Heck reaction),27 boron derivatives (the Suzuki-Miyaura coupling),28 tin derivatives (the Stille coupling),29 organomagnesium (the Kumada-Corriu coupling),30 organozinc (the Negishi coupling)31 has been used successfully in the preparation of PPVs, PAEs and PPPs, the homocoupling of aromatic halides catalyzed by Ni(0) has been employed in the synthesis of many polyarylenes and their copolymers,32 the acyclic diene metathesis (ADMET)33 and acyclic diyne metathesis (ADIMET)34 processes catalyzed by complexes of molybdenum and tungsten were applied in the synthesis of double- and triple-bond containing conjugated polymers and the ring opening metathesis polymerization (ROMP) process catalyzed by organomolybdenum and -tungsten complexes has been exploited as a precursor route to polyacetylenes and PPVs.35 Considering the well-recognized fascinating electrical and optical properties of poydiacetylenes that also contain enyne scaffolds (-CR=CRC C-),12 it is remarkable that only a few reports on the synthesis and properties of poly(aryleneethynylenevinylene)s (PAEVs) exist in literature. To the best of our knowledge, the first report of a PAEV dates back to 1992, when Kane et al. described the synthesis of a series of conjugated poly(enyne)s I via the palladium-catalyzed coupling of aromatic diynes to vinylic bromides (Scheme 6-2).36 In 1996, Ueda et al. reported the analogous palladium-catalyzed preparation of II.37 The single other report involving PAEVs is by Endo et al. in 2000 who prepared insoluble poly(1,4-phenyleneethylenevinylene)s of unknown stereochemistry via a palladium-catalyzed coupling, followed by a Retro-Diels-Alder reaction.38 The metal-catalyzed dimerization of substituted 1-alkynes constitutes a highly attractive and atom efficient method to effect C-C bond formation, but its applications have been limited by the lack of control of Scheme 6-2. The synthesis of poly(aryleneethynylenevinylene)s.
Ar [Pd] R R R = OC10H21 Ar = , , R OR Br + Br RO R = C4H9, C10H21, C16H35 RO II n [Pd] OR I S n R

Br Ar + R

S R

Br

208

The organolanthanide-catalyzed polymerization of diynes stereo-, regio- and chemoselectivity. Catalytic systems that combine both a high activity and a high selectivity have been developed only recently.39 Among these systems, the catalyst precursor Cp*2LaCH(SiMe3)2 was found to be an active and selective catalyst for the trans-head-to-head dimerization of a variety of substituted (hetero)aromatic 1-alkynes (Chapters 4 and 5). Motivated by the potentially interesting properties of PAEVs that represent a relatively unexplored class of conjugated polymers, a study was initiated to investigate the application of the lanthanocene-catalyzed linear dimerization of bifunctional (hetero)aromatic 1-alkynes to the preparation of PAEVs (Scheme 6-4).40 In the course of this study, three independent studies and one patent appeared, describing a similar approach towards the preparation of PAEVs.41 The details of these processes and comparisons with the present study are discussed in a later section. Moreover, an analogous approach towards poly(arylenebutadiynylene)s (PABs) via the copper-catalyzed oxidative dimerization of bifunctional 1-alkyne is well-established (Scheme 6-4) and has led to materials with photochemical and optoelectronic applications, such as nonlinear optical materials42 and photoresists,43 and as precursors for high-carbon materials.44 Scheme 6-4. The preparation of conjugated polymers via metal-catalyzed linear 1-alkyne dimerization.
catalyst Ar Ar n Poly(aryleneethynylenvinylene) [Cu] Ar n Poly(arylenebutadiynylene)
S N

Ar

Ar =

, etc.

In this chapter, the lanthanocene-catalyzed linear dimerization of substituted (hetero)aromatic 1alkynes is investigated as a means to prepare well-defined poly(aryleneethynylenevinylene)s. Four types of diynes, with and without solubilizing substituents, were used to probe the scope and feasibility of this synthetic approach. The formed polymers were characterized spectroscopically and their optical properties were studied as well.

6.2.

Synthesis of the (hetero)aromatic diynes.

Introduction The preparation of a large variety of (hetero)aromatic diynes is reported in literature. In most cases halide (hetero)aryls were subjected to standard Hagihara-Sonogashira coupling reactions with trimethylsilylacetylene, followed by desilylation.45 This method of synthesis was adapted successfully in the present study after the incorporation of some minor modifications and additional purification steps. Attempts to prepared the desired monomers via palladium-catalyzed coupling reactions between iodo (hetero)arenes and zinc reagents (the Negishi coupling) failed, since solvent removal (i.e. THF) was accompanied with significant decomposition of the diethynyl monomers. It is believed that the present modifications of the reported syntheses and the instability of the diethynyl monomers justify a brief discussion of their synthesis and use. Performing catalytic reactions with organolanthanides requires not only the rigorous exclusion of air and moisture, but substrates of high purity as well, since coordination of heteroatoms (e.g. oxygen, nitrogen) to these electrophilic metal complexes is in general both kinetically and thermodynamically favored.46 Coordination of heteroatoms may lead to a decrease in catalytic activity47 or even catalyst deactivation.48 In addition, halides are known to undergo metal-halogen exchange reactions with organolanthanide complexes.49 Obviously, traces of compounds containing acidic protons, halide precursors and heteroatom-containing sideproducts or solvents can have detrimental effects on the observed catalytic activity. 1,4-Diethynylbenzene (1) The synthesis of 1,4-diethynylbenzene (1) was straightforward and performed analogously to literature procedures.50 The palladium-catalyzed coupling of 1,4-diiodobenzene with trimethylsilylacetylene

209

Chapter 6

Scheme 6-5. The synthetic route towards the aromatic diynes 1-3.
(i) I I 87% Me3Si SiMe3 (ii) 97% 1 R Cl Cl (iii) 73% R = nC6H13 R R (iv) 55% R I I (i) 93% R (ii) 92% OR HO OH (v) 81% R = nC6H13 RO OR (iv) 40% RO I I (i) 83% Me3Si RO OR (ii) 86% RO 3 R 2 OR SiMe3 Me3Si R R SiMe3

Reagents: (i): Me3SiCCH (2.2 equiv), THF/HNiPr2, Pd(PPh3)2Cl2, PPh3, CuI; (ii): KOH, MeOH/THF; (iii): RMgBr, diethyl ether, reflux. (iv): I2, KIO4, HOAc/H2SO4/CHCl3, reflux. (TMSA) afforded white crystals in high yield after work-up which were indefinitely stable at room temperature. Subsequent basic hydrolysis afforded the desired diyne as white needles in high yield. Upon standing at room temperature, the needles of 1 became dark within several hours. The instability of terminal diynes in general is well-recognized51 and 1,4-diethynyl aromatics, in particular, are known to be highly light and heat sensitive, forming cross-linked poly(ene)s presumably via a radical-initiated mechanism.52 As a consequence, only small portions of the silyl-protected precursor were desilylated prior to polymerization. Drying the diyne with CaH2 as a pentane solution and subsequent storage of the crystals or oils at -30 C under nitrogen in the dark proved to be the most convenient method to use the diyne monomers in this study. 2,5-Di-n-hexyl-1,4-diethynylbenzene (2) The synthesis of 2,5-di-n-hexyl-1,4-diethynylbenzene (2) started with 1,4-dichlorobenzene.53 As reported by others, iodination of 1,4-dichlorobenzene with KIO3 gave a mixture of mono- and diiodo compounds which was difficult to purify.54c Iodination with KIO4 was found to give the desired diiodide in reasonable yield and high purity as colorless needles after repeated crystallizations from ethanol.54 Nonetheless, careful monitoring of the reaction progress is required to avoid the formation of relatively large amounts of the monoand triiodides. Palladium-catalyzed coupling with TMSA gave a white crystalline powder after purification by means of column chromatography and subsequent crystallization. Application of several desilylation protocols (e.g. KOH in MeOH/THF, K2CO3 in MeOH/THF, nBu4NF in cold THF) afforded in all cases a colorless oil which became yellow upon complete solvent removal. Subsequent column chromatography using pentane and neutral alumina furnished a pale yellow liquid which turned orange on standing for several hours in the dark in air.55 When the product was allowed to stand at room temperature (either in a pure form or in solution), a darkening of the color was observed, ultimately giving rise to a viscous blood-red liquid. The formed viscous dark-red oil gave poorly reproducible results in the polymerization reactions. When the yellow liquid was stored at -30 C under nitrogen in the dark, it remained seemingly indefinitely stable and gave good results in the organolanthanide-catalyzed polymerization reactions.56 2,5-Di-n-hexyloxy-1,4-diethynylbenzene (3) 1,4-Diethynyl-2,5-di-n-hexyloxybenzene (3) was prepared from 1,4-dihydroxybenzene (hydroquinone). Iodination of 1,4-dihexyloxybenzene57 with KIO4 gave the diiodo compound after repeated

210

The organolanthanide-catalyzed polymerization of diynes

Scheme 6-6. The synthetic scheme of 2,5-diethynyl-3-n-hexylthiophene (4).

Me3Si S R (i) 89% R = nC6H13 R S (ii) 73% R (iv) 95% 4 R S R I S I (iii) S SiMe3

Reagents: (i): RMgBr, diethyl ether, reflux. (ii): I2, KIO4, HOAc/H2SO4/CHCl3, reflux. (iii): Me3SiCCH (2.2 equiv), THF/HNiPr2, Pd(PPh3)2Cl2, PPh3, CuI. (iv): KOH, MeOH/THF. crystallizations from diethyl ether as white crystals in moderate yield.54c The palladium-catalyzed ethynylation with TMSA afforded white crystals after work-up in good yield. Protodesilylation furnished the desired diyne as light-yellow crystals which seemed more stable than its n-hexyl analogue. Even so, crystals of 3 were stored at 30 C under nitrogen and in the dark after drying with CaH2. 2,5-Diethynyl-3-n-hexylthiophene (4) The syntheses of several 2,5-diethynyl-3-alkylthiophenes were reported concurrently by the groups of Whitesides and Stille via the palladium-catalyzed coupling of the corresponding 2,5-dibromothiophenes with TMSA, followed by deprotection.58 In view of the higher reactivity of iodides in the palladium-catalyzed crosscoupling reactions and the concomitant advantages (e.g. lower catalyst loadings, lower reaction temperatures and times, less side-reactions),45 a similar procedure involving the corresponding diiodides was adapted in the present study. The iodination method of Barker et al.59 was modified to prepare 2,5-diiodo-3-n-hexylthiophene from 3-n-hexylthiophene.60 Again, a careful monitoring of the progress of the reaction was necessary to avoid the formation of relatively large amounts of mono- and triiodides. After work-up the desired diiodo compound was obtained as a dark oil and appeared to be contaminated by small amounts of the corresponding monoiodide (<1 % by 1H NMR). Complete removal by column chromatography proved to be difficult and attempts to crystallize the desired product failed. Palladium-catalyzed coupling of this diiodide with TMSA furnished 2,5-bis[(trimethylsilyl)ethynyl]3-n-hexylthiophene as an oil in good yield. The monoacetylenic by-product and n-dodecane could be separated from the protected diyne by Kgelrohr distillation (170 C, 5 mmHg). Similarly to 1,4-diethynyl-2,5-di-nhexylbenzene (2), a colorless oil was obtained after performing different desilylation protocols (i.e. KOH in MeOH/THF, K2CO3 in MeOH/THF, nBu4NF in cold THF), but it darkened upon complete solvent removal. Flash column chromatography (neutral alumina, pentane) afforded a yellowish oil in good yield and purity (>99.5% according to 1H NMR, the only impurities being unidentified silanes, presumably side-products originating from the desilylation reaction). Drying this diyne with CaH2 as a pentane solution, evaporation and storage at -30 C under nitrogen and in the dark was found to be a procedure that provided good and reproducible results in subsequent organolanthanide-catalyzed polymerization reactions. The oil darkened within minutes at room temperature both in air and under nitrogen and formed an insoluble solid within days, but in air considerably faster. Properties of the diyne monomers As mentioned above, a vast majority of terminal diynes are highly thermo- and photosensitive molecules prone to rapid decomposition. The formation of insoluble, highly colored solids upon storage, both in the solid state and in solution, is well-documented for oligoethynyl aromatics.52 It is believed that photopolymerization takes place in the solid state, while a radical mechanism has been put forward to explain the cross-linking reactions of terminal ethynyl groups and the concomitant formation of poly(ene) structures in solution. In addition, oligoethynyl (hetero)aromatics should be treated as potentially explosive materials.4e Many of such terminal diynes, triynes, etc. are reported to decompose exothermically upon heating. The high instability of 2,5-diethynyl-3-n-hexylthiophene (4) at ambient conditions was striking as compared to 1,4-diethynyl-2,5-di-n-hexylbenzene (2) in the present study. The rapid decomposition of 2,5diethynyl-3-n-alkylthiophenes and 2,5-diethynylthiophene has been noted previously in literature.43a,58,61 Neat 2,5-diethynylthiophene is reported to undergo a gradual polymerization to insoluble dark brown resins within 5

211

Chapter 6 days at room temperature.61a Decomposition of dilute pentane solutions was apparent by a color change within a few days when stored at 0 C.43a IR and solid 13C NMR of the solid indicated that most acetylene carbons had been converted to olefin carbons.

6.3.

Polymerization reactions with (hetero)aromatic diynes.

6.3.1.

Introduction

The reaction of Cp*2LaCH(SiMe3)2 with bifunctional 1-alkynes can be regarded as a polymerization reaction taking place by a step growth polymerization mechanism.62 In principle, polymerization reactions taking place by a step growth mechanism should give rise to one single macromolecule. In practice, however, the average length of the polymer chain is limited by the purity of the reactants, the presence of side reactions and the viscosity of the reaction medium. High molecular weight polymers are formed only at high conversion of the reactive functionality. As a consequence, step growth polymerization reactions place generally stringent requirements on any reaction to be used for polymerization, such as very high conversions and selectivities. It seems that these requirements are well-met in the present Cp*2LaCH(SiMe3)2-catalyzed polymerization reactions of diynes, because complete substrate conversion was found to be quite rapid in the analogous oligomerization reactions of monoynes and the by-products formed are not likely to result in chain termination. Scheme 6-8. The oligomerization of phenylacetylene by Cp*2LaCH(SiMe3)2.
Ph Cp*2LaCH(SiMe3)2 2 mol% Ph C6D6 25 C Ph + 0.1% Ph 97.8% Ph Ph + Ph 1.6% Ph 0.5% Ph + . Ph Ph

In the absence of side-reactions, trans-head-to-head dimerization is expected to yield a regioregular polymer consisting only of (E)-vinylic linkages. When Cp*2LaCH(SiMe3)2 is allowed to react with 50-fold molar excess of phenylacetylene, a mixture of 2,4-diphenylbut-1-en-3-yne (0.1%), trans-1,4-diphenylbut-1-en-3-yne (97.8%), 1,3,6-triphenylhexa-1,4-diyne (1.6%) and 1,3,6-triphenylhexa-1,2-dien-4-yne (0.5%) is formed (Scheme 6-8). Hence, the reaction of Cp*2LaCH(SiMe3)2 with a bifunctional analogue of phenylacetylene will plausibly afford a polymer containing other types of linkages as well. Linkages originating from head-to-tail dimerization and trimerization are considered to be structural defects, as they interrupt the conjugation63 and, as a result, will impair the development of properties related to conductivity and nonlinear optics.64 Carbon-carbon bond formation via trimerization of ethynyl moieties in the present catalytic process leads also to chain branching. Substitution of the phenyl ring and the presence of heteroaromatic moieties has been shown to influence both the activity and the selectivity of the Cp*2LaCH(SiMe3)2-catalyzed oligomerization reaction of phenylacetylene. It will be demonstrated that the polymers prepared in this study contain predominantly C-C Scheme 6-7. The possible structure of the polymer resulting from the lanthanocene-catalyzed polymerization of (hetero)aromatic diynes.

Ar

Cp*2LaCH(SiMe3)2

Ar Ar x Ar y

212

The organolanthanide-catalyzed polymerization of diynes

100

poly(1) (KBr, air) 100 90 1 (KBr, air) 80 100 75 (E)-1,4-diphenylbut-1-en-3-yne (KBr, air) 3000 2500 2000
(cm )
-1

50 1500 1000 25 500

Figure 6-1. Infrared spectra of the PAEV of 1,4-diethynylbenzene (poly(1), upper spectrum), 1,4diethynylbenzene (1, middle spectrum) and trans-1,4-diphenylbut-1-en-3-yne (lower spectrum). linkages originating from trans-head-to-head dimerization, depending on the type of monomer used. The only defects as detected by 1H NMR and IR spectroscopy are ,-ethylene linkages resulting from head-to-tail dimerization (Scheme 6-7).

6.3.2.

Polymerization reactions

Introduction The reactions of Cp*2LaCH(SiMe3)2 with an excess of diyne were initially studied in Teflon-sealed NMR tubes with benzene-d6 as the solvent (~500 L) and followed in time by normalized, in situ 1H NMR spectroscopy. In some cases, substrate conversion could be monitored by normalization of the aromatic proton resonance against that of an internal standard. The internal standard that were used are hexamethyldisiloxane (HMDSOO) and CH2(SiMe3)2. The latter forms upon rapid and quantitative protonolysis of the catalyst precursor, Cp2LaCH(SiMe3)2, by the substrate. Quantitative 1H NMR spectroscopy required the use of long pulse delays to avoid signal saturation under the present anaerobic conditions. 1,4-Diethynylbenzene (1) Reactions of Cp*2LaCH(SiMe3)2 with excess of 1,4-diethynylbenzene (1) (10-50 equiv.) in benzened6 led to the formation of a yellow-colored material. The colorless catalyst solution turned instantaneously into an orange suspension upon substrate addition, followed by the precipitate of a yellow solid after ~5 min at room temperature. The solid was found to be completely insoluble in common organic solvents (e.g. toluene, chloroform, THF), even after prolonged heating and sonication. The polymerization of 1 by Cp*2LaCH(SiMe3)2 is rapid, consistent with the analogous oligomerization of its monofunctional congener (e.g. a 55-fold molar excess of substrate relative to Cp*2LaCH(SiMe3)2 is completely converted within 10 min at room temperature in benzene-d6, see Chapter 4). The observed precipitation of the product suggests that the preparation of high molecular weight polymer is limited by the solubility of the growing polymer. Although the insolubility of the polymer hampered the determination of its microstructure by means of NMR spectroscopy, infrared spectroscopy confirmed the view that terminal acetylenic groups have been

Transmittance (%)

213

Chapter 6

ArH H H

ArCH2 C CH

8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm Figure 6-2. 500 MHz 1H NMR spectra of the mixture of Cp*2LaCH(SiMe3)2 and a 10-fold excess of monomer 2 in benzene-d6 at 25 C. The top spectrum is before catalyst addition and the consecutive lower ones are taken each 10, 15, 30, 60 and 60 min later, respectively. The solvent signal is indicated by an asterisk (*).

converted so as to form predominantly (E)-CH=CH linkages. The presence of new peaks at 2183 and 941 cm-1 in the polymer relative to the monomer (Figure 6-1) are, in particular, structurally diagnostic and assigned to the stretching vibration of a carbon-carbon triple bond and a C-H bending vibration of a (E)-CH=CH group, respectively.65 These assignements agree well with values previously observed for PAEVs36-38,92 and other polymers containing enyne units.66 No bending vibrations corresponding to C=CH2 groups were observed (895885 cm-1). It can also be seen that the acetylenic C-H and C-C stretch vibrations at 3260 and 2100 cm-1 are less intense in the polymer than in the monomer. No conclusive information could be obtained with respect to the possible presence of allene moieties, however. 2,5-Di-n-hexyl-1,4-diethynylbenzene (2) In order to facilitate the characterization and investigations of the polymer properties, it was decided to increase the solubility of the polymer by substituting 1,4-diethynylbenzene with flexible aliphatic side chains.67 Substituents of different structure and length are known to affect not only the solubility, but also the solid-state structure, the phase behavior and the electronic properties of the polymers.68 The incorporation of substituents onto the diyne monomer are, furthermore, anticipated to influence both the reactivity of the monomer and the regioselectivity of the reaction, in analogy to the observed substrate effects on the rate and selectivity of the Cp*2LaCH(SiMe3)2-catalyzed 1-alkyne oligomerization reactions (Chapter 4). As a result, the choice of the substituent is critical. On the one hand, it can be expected that both solubility and the degree of polymerization will increase with the length of the side-chain.69 On the other hand, the steric size of the substituent will inevitably block metal coordination to a certain extent, thereby diminishing the observed reactivity. The n-hexyl substituent was chosen based on reports that the degree of polymerization and the solubility of 2,5-disubstituted poly(p-phenyleneethynylene)s34d did not increase significantly upon side-chain elongation. It was also shown that the presence of an ortho-methyl group in phenylacetylene resulted in an increased selectivity for dimerization in the reaction catalyzed by Cp*2LaCH(SiMe3)2 (Chapter 4). When Cp*2LaCH(SiMe3)2 was added to a benzene-d6 solution of 10 or 20 equiv of 1,4-diethynyl-2,5di-n-hexylbenzene (2), an insoluble black solid precipitated immediately and the complete reaction mixture solidified into a dark rubber-like solid within 45 min at room temperature. The formed polymer did not dissolve upon addition of benzene or THF, not even after heating overnight or sonication. Infrared spectra of the solid detected no absorbance at ~3300 cm-1 (acetylenic C-H stretch) which is moderately strong in the monomer. The addition of chloroform formed a faint yellow solution containing practically the same amount of dark solid. In order to confirm that the reaction is catalyzed by Cp*2LaCH(SiMe3)2, the same experiment was conducted in the absence of catalyst. The color of the reaction mixture darkened after several hours (suggestive

214

The organolanthanide-catalyzed polymerization of diynes

0 0 -0.2 -0.4 ln([S]0/[S]t) -0.6 -0.8 -1 -1.2 -1.4

30

60

90

120

150

180

210

240

270

300

t(min)

Figure 6-3. Integrated rate plot of the concentration of 2 and 3 vs time for the diyne polymerization catalyzed by Cp*2LaCH(SiMe3)2. Lines represent fitted linear plots. of thermal decomposition, Section 6.2), but no significant monomer conversion was observed after several days with 1H NMR spectroscopy. In situ monitoring of the reaction with 10 equiv of 2 by 1H NMR spectroscopy in the presence of hexamethyldisiloxane (HMDSO) as internal standard revealed the consumption of 2 and the formation of CH2(SiMe3)2 and two presumed Cp* signals (Figure 6-2). As the reaction progressed, the newly formed 1H resonances became increasingly broad and ultimately disappeared. These observations are consistent with the formation of oligomers and a gradual precipitation of the product as the molecular weight increases at higher substrate conversion. In addition, broad vinylic 1H NMR resonances were observed around 6.5 and 5.5 ppm (vide infra) which are associated with the vinylic protons of the trans-head-to-head and head-to-tail linkages. The observation of the vinylic protons of the trans-head-to-head dimers ( 6.29, 6.99) and head-to-tail dimers ( 5.56, 5.60) of phenylacetylene in this region of the 1H NMR spectrum supports this view (Chapter 4). Complete solidification of the reaction mixture could be retarded, accompanied by an increase of the chloroform-soluble fraction, upon increasing the molar excess of diyne monomer relative to the catalyst. For example, performing the analogous reaction with 50 equiv of substrate 2 resulted in complete solidification of the reaction mixture after 2 h. Addition of CHCl3 yielded a yellow, chloroform-soluble solid in 21% yield after filtration and washing with methanol. A dark-brown solid was obtained as a residue in 74% yield which was found to be insoluble in organic solvents (e.g. THF, toluene, acetone), even after prolonged heating or sonication. Attempts to dissolve the chloroform-insoluble portion of the formed polymer under anaerobic conditions at temperatures of 100-150 C (in tetrachloroethane, nitrobenzene, chlorobenzene, o-dichlorobenzene, 1,2,4-trichlorobenzene) were unsuccessful. A kinetic plot for the reaction with 20 equiv of 2 was obtained by monitoring the intensity of the aromatic signal relative to the internal standard (HMDSOO) with in situ 1H NMR spectroscopy. First-order rate dependence on substrate concentration (kobs = 2.33(4) M-1min-1, R2 = 0.9989) was observed for the first 45 min in which 53% of 2 was converted. Deviation from first-order rate dependence on substrate concentration at a higher conversion is probably the result of the lowered solubility of the growing polymer chain, competing reactivity from dimers, trimers, etc. and the increasing viscosity of the reaction medium (Figure 6-3). 1 H NMR spectroscopy of the chloroform-soluble fraction revealed resonances similar to those of the monomer. The ratio of the intensities corresponding to the acetylenic C CH 1H end-groups and the -alkyl-aryl groups ArCH2 is an indication of the number-average molecular weight Mn and in this particular example the soluble fraction contains oligomers having an average of 10(1) repeating units.70 The number-average degree of polymerization (Pn) thus obtained can be verified by integrating the combined vinylic signals or aliphatic signals versus the acetylenic signals or dissolving the sample in CD2Cl2 or THF-d8 and integrating the aromatic signals

215

Chapter 6

H x y H CH=CH CH CH=CH =C(H)H =C(H)H C CH CCH2

*
CH C CH CCH2 H H

7.0

6.0

5.0

4.0

3.0

2.0

1.0

ppm

Figure 6-4. 500 MHz 1H NMR spectra in CDCl3 of monomer 2 (lower spectrum) and the chloroform-soluble fraction of its polymer after work-up (upper spectrum). * denotes the solvent signal. versus the acetylenic signals. A Pn of 10(1) implies that oligomers of higher molecular-weight are insoluble in the reaction mixture and confirms the notion that solubility is the limiting factor in obtaining polymers of high molecular weight. As the -alkyl groups having an ortho-ethynyl and -ethenyl substituent resonate at different chemical shifts (also observed for the linear dimers of 2-methyl- and 2-methoxyphenylacetylene, Chapter 4), two sets of 1H NMR resonances are observed for the present -alkyl groups. By integrating the vinylic 1H NMR resonances corresponding to the head-to-tail (HT) and trans-head-to-head (HH) coupling the relative amount of HT linkages can be determined. In this particular example, the polymer consists for 89(1)% of HH linkages and for 11(1)% of HT linkages.70 The chloroform solution of the soluble polymer also exhibited a strong blue fluorescence upon irradiation ( = 366 nm). The formation of poly(2,5-di-n-hexylphenylethynylenevinylene)s was furthermore supported by 13C NMR and IR spectroscopy (vide infra). It should be noted that a precipitate formed in the chloroform solutions of the soluble polymers upon standing at room temperature. This precipitate was a red solid which was found to be insoluble in common organic solvents, even after prolonged heating or sonication. When this solid was analyzed by IR spectroscopy, the spectra of the insoluble solid and the soluble fraction were practically identical, except that the vibrations corresponding to acetylenic end-groups were less intense in the precipitate than in the soluble fraction. This finding may point to acetylenic cross-linking, as discussed previously for the diethynyl (hetero)aromatic monomers (Section 6.2). Precipitation of the chloroform-soluble fraction was found to be accelerated by heat and light and also took place under an inert nitrogen atmosphere. The instability of these acetylenic end-capped polymers is not unreasonable, considering the instability of the monomers and reports of oligo(enyne)s that were found to become increasingly reactive with increasing oligo(enyne) length.71 The polymer solutions could not be filtered through 0.5-1.0 m filters and as a result the soluble polymer fraction could not be analyzed by gel permeation chromatography (GPC). Whether this is the result of the acetylenic cross-linking or microgel formation is not known at present. Concentrated polymer solutions that were prepared for 13C NMR spectroscopy underwent gelation and the increased viscosity led to difficulties in obtaining a spectrum of good quality. It was attempted to prepare concentrated polymer solutions suitable for 13C NMR analysis by preparing low-molecular weight polymers, as oligo(enyne) instability is known to increase with chain length.71 The synthesis of low-molecular weight oligomers was achieved by quenching reaction mixtures before complete solidification. When the reaction of Cp*2LaCH(SiMe3)2 with 2 (20 equiv) was quenched with methanol after 1.5 h a dark-red, viscous oil was obtained nearly quantitatively which was completely soluble in chloroform. 1H NMR analysis indicated Pn =

216

The organolanthanide-catalyzed polymerization of diynes

R c b' a' e f g R h x R d f c CH CCH CH CCH CH CCH CH CC C R CR CH CC C R 2 CC C CC C g h j y H i

b' b e c

H a

i j

a b d R

a'

CC CH CC C

150

140

130

120

110

100

90

80

ppm

Figure 6-5. 100 MHz 13C{1H} NMR spectra in CDCl3 of monomer 2 (lower spectrum), 1,4-diphenylbut-1en-3-yne (middle spectrum) and the chloroform-soluble fraction of the polymer derived from 2 (upper spectrum). 3.9(4) and HH:HT = 94(1):6(1), thereby suggesting that polymerization of 2 proceeds in the gel-state. It is important to note that the yield of these thermo- and photolabile polymers is strongly dependent on the isolation and purification procedure. The 13C{1H} NMR spectrum of the oligomers can readily be interpreted by comparison with the monomer and the model enyne compound, (E)-1,4-diphenylbut-1-en-3-yne (Figure 6-5). For each type of aromatic resonance several sets are observed depending on the nature of substitution (Scheme 6-9). Also multiple sets are found for each type of vinylic, acetylenic and aliphatic resonance, probably as a result of the different neighboring groups in the oligomeric chain. The exact nature of the solidification of the reaction mixture during reaction is unknown at present. It is generally accepted that gelation takes place via molecular aggregation and that the structures intertwine to create thermo-reversible three-dimensional networks.72 In the present study, thermo-reversible gelation was not observed. When the rubber-like material obtained after complete solidification of the reaction mixture was dissolved in chloroform or THF (dry and under nitrogen in the dark), only a small portion of the solid dissolved (<1-5 wt%), even after prolonged stirring, mild heating and sonication. The possibility that the insoluble polymer fraction consists of polymers having a molecular weight too high to be soluble could neither be discarded nor confirmed by means of IR analysis. Several types of chemical cross-linking processes may, in addition, proposed to take place during Scheme 6-9. Different possible types of substitution in the 2,5-dialkyl-substituted poly(pphenyleneethynylenevinylene)s.
R H R R H R R R R R R R

217

Chapter 6

O O H O x O y H

ArH

ArOCH2

C CH

3 * 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm

Figure 6-6. 500 MHz 1H NMR spectra in CDCl3 of monomer 3 (lower spectrum) and the chloroform-soluble fraction of its polymer after work-up (upper spectrum). reaction or during work-up. Examples of photo- or thermooxidation of terminal ethynyl groups73,74 and internal carbon-carbon double and triple bonds75 are well-known. The infrared spectrum of the insoluble solid after aerobic isolation indicated the presence of acetylenic groups, but the possibility that a small degree of the acetylenic end-groups is responsible for chemical cross-linking cannot be ruled out based on IR spectroscopy alone. Oxidative chemical cross-linking seems unlikely, because irreversible gelation was also observed after work-up under anaerobic conditions. In addition, IR spectra of solids exposed to air for prolonged times did not exhibit the presence of carbonyl or hydroxyl groups. Another possibility is the thermal and light-induced crosslinking via the carbon triple bond which is well-documented for poly(arylenebutadiynylene)s.76 IR spectra of the insoluble solids could neither confirm nor discard this type of chemical cross-linking. Experimentally, more light can be shed on this issue by performing in situ IR and UV-vis spectroscopy during the diyne polymerization reaction or solid-state 13C NMR spectroscopy of the insoluble solid. 2,5-Di-n-hexyloxy-1,4-diethynylbenzene (3) Motivated by the interesting properties exhibited by many diortho-substituted alkoxy derivatives of phenyl containing -conjugated polymers (e.g. PPs, PPVs, PPEs77) and the fact the Cp*2LaCH(SiMe3)2-catalyzed oligomerization of ortho-ethynylanisole was found to be highly selective for linear dimerization, it was decided to conduct polymerization reactions with 2,5-di-n-hexyloxy-1,4-diethynylbenzene (3). When Cp*2LaCH(SiMe3)2 was added to a benzene-d6 solution containing 20 equiv of 3, the clear light-yellow solution turned brown instantaneously and a viscuous, red solution formed within 2 h. The reaction mixture solidified completely into a red solid overnight. Chloroform was added to the reaction mixture and filtration afforded a red solid both as residue (71% yield) and filtrate (22%). The residue did not dissolve in common organic solvents, even after prolonged heating or sonication. The filtrate was soluble in CHCl3 and THF and exhibited green fluorescence upon irradiation ( = 366 nm). A kinetic plot for the reaction with 20 equiv of 3 was obtained by monitoring the intensity of the aromatic signal relative to the internal standard (HMDSO) with in situ 1H NMR spectroscopy (Figure 6-3). Firstorder rate dependence on substrate concentration was observed (kobs = 0.694(9) M-1min-1, R2 = 0.9988) for the first 188 min at which 63% of 3 was converted. The polymerization of 2 is 3.2 times faster than that of 3 which is consistent with the observation that the analogous oligomerization of 2-ethynyltoluene is faster that that of 2ethynylanisole (Chapter 5). Similar to 2, deviation from first-order kinetics at a higher conversion is observed, probably due to the lowered solubility of the growing polymer chain, competing reactivity from higher oligomer and the increasing viscosity of the reaction medium. 1 H NMR spectroscopy of the chloroform-soluble fraction indicated that the polymer consisted of 92(1)% of HH couplings and of 8(1)% of HT couplings. This result is surprising, as 1-alkyne oligomerization

218

The organolanthanide-catalyzed polymerization of diynes

h i

a b g e

c OR d a' j k OR lm x RO a a' d kg l y H m h i cf

RO f j

CCH CH CCH CH CO

CH

O CCH CH O CC C C C C C

OR CH CO CC RO

C CH

C CH OCH2

160

140

120

100

80

ppm

Figure 6-7. 100 MHz 13C{1H} NMR spectra in CDCl3 of monomer 3 (lower spectrum), 1,4-di(2methoxyphenyl)but-1-en-3-yne (middle spectrum) and the chloroform-soluble fraction of the polymer derived from 3 (upper spectrum). reactions demonstrated that the selectivity for the trans-head-to-head dimer was lower for 2-ethynylanisole than for 2-ethynyltoluene at analogous substrate-to-catalyst molar ratios (Chapter 5). Acetylenic 1H resonances were observed in the chloroform-soluble fraction of 3 after work-up and integration indicated a number-average degree of polymerization of 16(1), thereby implying that n-hexyloxy groups confer a higher degree of solubility to the poly(phenylethynylenevinylene)s than n-hexyl groups. Also, 13C NMR and IR spectroscopy of the soluble fraction support the formation of dialkoxy-substituted poly(p-phenyleneethynylevinylene)s (Figure 6-8). Similar to the di-n-hexyl substituted derivatives, the formation of an insoluble solid was observed, when a solution of a chloroform-soluble polymer fraction was allowed to stand at ambient conditions. 2,5-Diethynyl-3-n-hexylthiophene (4) Previous investigations directed to assess the variation in the aromatic moiety of the 1-alkyne in the Cp*2LaCH(SiMe3)2-catalyzed oligomerization reaction indicated that 2-thienylacetylene is converted faster and more selectively to the desired trans-head-to-head dimer than phenylacetylene. This finding, together with the demonstrated enhancement of several desired properties (e.g. solubility, thermal and environmental stability, etc.) upon incorporating a 2,5-thienyl moiety into the main chain of -conjugated polymers (Section 5.1), encouraged us to perform polymerization reactions with diethynylthiophene analogues. Substitution of a n-hexyl substituent was based on reports that the solubility of 3-substituted poly(thiophene-2,5-diyls)s1a and poly(2,5thienyleneethynylene)s43a did not increase significantly upon chain elongation beyond the n-hexyl substituent. Addition of Cp*2LaCH(SiMe3)2 to a solution of 50 equiv of 4 caused the reaction mixture to solidify completely into a dark solid rubber-like material within 8 h at room temperature. The dark solid was found to be insoluble in common organic solvents such as toluene, THF and CHCl3. No 1H NMR resonances indicative of head-to-tail dimerization were observed during the course of reaction as monitored by 1H NMR spectroscopy. Analogous reactions with higher relative amounts of 4 (70-140 equiv) afforded in all cases brown solids which were practically insoluble in chloroform. When a similar reaction mixture containing 140 equiv of monomer was quenched with methanol prior to complete solidification, a dark red viscous oil was obtained which was only partially soluble in chloroform. Filtration yielded a red solid as filtrate (28% yield) and a shiny green solid as residue. 1H and 13C NMR analysis of the chloroform-soluble fraction supported the formation of the expected polymers. The intensities of the acetylenic 1H resonances are low (Figure 6-8), but indicate, nonetheless, that the chloroform-soluble fraction consists of polymers having a number-average degree of polymerization of 100(5). The regioregularity of the polymers with respect to the formed butenyl linkages is very high, as no vinylic

219

Chapter 6

S n

3.80

3.60

3.40 ppm

S CH

CH

4 *

CCH2

7.0

6.0

5.0

4.0

3.0

2.0

1.0

ppm

Figure 6-8. 500 MHz 1H NMR spectra in CDCl3 of monomer 4 (lower spectrum) and the chloroform-soluble fraction of its polymer after work-up (upper spectrum). * denotes the solvent signal. signals of HT couplings were observed. The multitude of aromatic signal suggests that the position of alkyl substituent is regiorandom, however. The proposed structure is supported by 13C{1H} NMR and IR spectroscopy, even though the 13C{1H} NMR spectrum of the polymer is too complex to allow unambiguous assignements. The latter is undoubtedly related to the asymmetrical alkyl substitution of the thiophene moiety which gives rise to twice the amount of resonances as observed for 2 and 3. It is important to note that the precipitate formed in the chloroform-soluble fraction upon standing in solution at ambient conditions took place at a considerably higher rate than observed previously for the di-nhexyl and di-n-hexyloxy derivatives of poly(p-phenylene-ethynylenevinylene)s. After filtration of the reaction mixture and washing with chloroform, a non-negligable amount of the filtrate (5%) did not redissolve in chloroform upon standing for 4 days at ambient conditions. The solid formed was found to be insoluble in common organic solvents and no acetylenic vibrations were observed in its IR spectrum. These observations are suggestive of acetylenic cross-linking affording poly(ene)s, as encountered before. Cross-linking was found to be accelerated by heat and also took place under an inert nitrogen atmosphere. The high reactivity of 4 impeded a kinetic analysis under similar reaction conditions as performed for Table 6-1. The diyne polymerization reactions catalyzed by Cp*2LaCH(SiMe3)2.a Entry Catalyst Monomer Yield soluble t Pnb HH:HTb (mM) (equiv) fraction (%) (h) 1 8.4 1 (20) 0 0.5 2 13 1 (50) 0 0.5 3 32 2 (20) <5 12 4 13 2 (50) 21 12 10(1) 89(1):11(1) 5 20 2 (20) 98 1.5 3.9(4) 94(1):6(1) 6 12 3 (20) 22 12 16(1) 92(1):8(1) 7 13 4 (50) <1 12 8 16 4 (72) <1 12 9 12 4 (120) <5 12 10 11 4 (138) 28 12 100(5) 100:0 a Reaction conditions: C6D6 (0.50 mL), 25 C. b Number-average degree of polymerization of the chloroform-soluble polymer fraction as determined by 1H NMR spectroscopy.

220

The organolanthanide-catalyzed polymerization of diynes 2 and 3. No 1H resonances of 4 were observed after 5 min, when 20 equiv of 4 were reacted with Cp*2LaCH(SiMe3)2. By means of line-shape analysis a conversion of 85% was found after 15 min, when Cp*2LaCH(SiMe3)2 was reacted with 72 equiv of 4. As the high reactivity of 4 relative to 2 and 3 was clear and in agreement with analogous 1-alkyne oligomerization reactions (i.e. the rate of 2-ethynylthiophene conversion was found to be ~10 times faster than that of 2-ethynyltoluene and ~15 times faster than that of 2-ethynylanisole under identical reaction conditions, Chapter 5), no further attempts were made to quantify this difference in reactivity.

6.3.3.

Polymerization reactions with end-capping agents

Introduction The use of a monofunctional monomer to control the molecular weight of the polymer is wellestablished in step-growth polymerization reactions.28,78 In the present study, the application of a suitable endcapping agent could in principle provide a means to control the molecular weight of the polymers by adjusting the molar ratio of both substrates. A concomitant advantage of an end-capping group is its ability to infer stability upon conjugated oligomers containing yne-, ene- and enyne-scaffolds,79 as arylacetylenic (vide supra) and vinylacetylenic80 end-groups are known to be fairly reactive. As a result, the use of end-capping agents is expected to yield products that are more stable than those of analogous homopolymerization reactions. Another advantage of the application of end-capping groups is their use as a tool to determine the molecular weight by, for example, 1H NMR spectroscopy. Scheme 6-10. The use of a monoacetylenic end-capping group to prepare soluble and more stable poly(aryleneethynylenevinylene)s.
Cp*2LaCH(SiMe3)2 R Ar n R

Ar

Obviously, the success of this method relies on the relative reactivity of both substrates. The reactivity of the monoyne as compared to that of the diyne depends not only on the structural properties of the 1alkyne (such as electronic and steric effects of the 1-alkyne substituent), but also on its concentrations relative to that of the diyne. A detailed study of substituent effects on the catalytic oligomerization reaction mediated by Cp*2LaCH(SiMe3)2 demonstrated that the reactivity of 1-alkynes can be enhanced by substituents with an increasing -electron-withdrawing character or the presence of heteroatoms in proximity of the ethynyl group, while steric effects can be used to retard the reactivity (Chapter 5). Hence, the choice of an appropriate monoacetylenic end-capping group seems a priori restricted to 1-alkynes having (kinetic) acidities similar to those of the aromatic diynes.81 Phenylacetylene (5) Phenylacetylene (5) seemed the most obvious first choice in the pursuit of a suitable end-capping agent for the reactions of Cp*2LaCH(SiMe3)2 with aromatic diynes. Phenylacetylene was found to be more reactive than 1,4-diethynyl-2,5-di-n-hexylbenzene (2), as is clearly demonstrated by the reaction of Cp*2LaCH(SiMe3)2 with 24 equiv of 2 and 25 equiv of phenylacetylene.82 Under these reaction conditions, phenylacetylene was converted completely within several minutes (1H NMR), while the reaction mixture solidified after several hours. When Cp*2LaCH(SiMe3)2 was allowed to react with 44 equiv of 2 and 7 equiv. of phenylacetylene in benzene-d6, no precipitate was observed. Instead, a dark brown solution formed, thereby demonstrating the successful application of phenylacetylene as an end-capping agent to prepare soluble oligo(phenylene-ethynylenevinylene)s. Washing with methanol, filtration and dissolving in CHCl3 yielded a brown solid in practically quantitative yield. Overlapping 1H NMR resonances hampered the determination of the relative rate of consumption of phenylacetylene and 2 by in situ 1H NMR spectroscopy. Based on the observation that the solutions of the phenylacetylene end-capped poly(2) did not yield any observable precipitate after standing for several weeks at room temperature, it can be concluded that the endcapping phenylacetylene groups also confer some stability on the polymer. However, the formation of an

221

Chapter 6

ArCH3

x y

7.0

6.0

5.0

4.0

3.0

2.0

1.0

ppm

Figure 6-9. 500 MHz 1H NMR spectra of the substrate mixture of 2 and 8 (1:1) in benzene-d6 before (lower spectrum) and 4 h after addition of Cp*2LaCH(SiMe3)2 (middle spectrum). The upper spectrum corresponds to the end-capped polymer in CDCl3 after work-up. insoluble solid was observed, when end-capped poly(2) was heated to 80 C or higher temperatures. Infrared spectroscopy of this solid revealed some minor changes, but did not provide conclusive evidence for a distinct chemical change. Unfortunately, the solutions of the end-capped polymer of 2 could not be filtered through 0.51.0 m filters and GPC analysis of the end-capped poly(2) was therefore not possible. Trimethylsilylacetylene (6) Another advantage of the application of end-capping groups is their use as a tool to determine the molecular weight by, for example, 1H NMR spectroscopy. As the 1H NMR resonances of a phenylacetylenic end-group overlap with those of the polymer, phenylacetylene cannot be used in this respect. Thus, other 1-alkynes were investigated to satisfy this objective. In view of its kinetic acidity which is comparable to that of phenylacetylene83 and its relatively high 1H NMR sensitivity, trimethylsilylacetylene (6) was investigated. Surprisingly however, trimethylsilyl-acetylene was found to be substantially more reactive than 2. For example, when Cp*2LaCH(SiMe3)2 was allowed to react with 55 equiv of 2 and 5 equiv of trimethylsilylacetylene, the monoacetylene was completely converted to oligomers before consumption of 2, as monitored by in situ 1H NMR spectroscopy, and solidification of the reaction mixture took place after several hours. (2,6-Dimethyl)phenylacetylene (7) As monoacetylenes applicable both as end-capping agents and tools for quantitative 1H NMR endgroup analysis are quite severely limited by their acidity and availability, alkyl-substituted phenylacetylenes were subsequently explored. The study of (2,6-dimethylphenyl)acetylene (7) as end-capping agent was motivated by the need to mimic the ortho-substituted bifunctional substrates (as phenylacetylene was found to be significantly more reactive than 2 and the addition of small quantities was not considered practical in these small-scale reactions) and its relatively high 1H NMR sensitivity. Unfortunately, the reactivity of (2,6dimethylphenyl)acetylene (7) was found to be much lower than that of the (hetero)aromatic diynes 2 and 3. For example, following the reaction of Cp*2LaCH(SiMe3)2 with 1,4-diethynyl-2,5-di-n-hexyloxybenzene 3 (25 equiv.) and 7 (26 equiv.) by in situ 1H NMR spectroscopy revealed that the amount of 7 did not change significantly during consumption of 3. A dark precipitate formed after several hours, while unreacted 7 was observed in considerable amounts in the soluble polymer fraction after work-up. 2-Ethynyltoluene (8) 2-Ethynyltoluene (8) represents another alkyl-substituted phenylacetylene that was investigated as an end-capping agent capable of controlling the molecular weight of the polymer, possibly providing these oligomers with stability and concomitantly serving as a tool for quantitative end-group analysis by 1H NMR spectroscopy. When Cp*2LaCH(SiMe3)2 was allowed to react with 2 (25 equiv) and 2-methylphenylacetylene (26 equiv) in a 1:1 ratio, a brown solution formed and no precipitate was observed upon standing for several

222

The organolanthanide-catalyzed polymerization of diynes

Table 6-2. The diyne polymerization reactions catalyzed by Cp*2LaCH(SiMe3)2 in the presence of an endcapping agent.a Entry Catalyst Diyne Monoyne Yield soluble Pnb HH:HTb (mM) (equiv) (equiv) fraction (%) 1 2 3 4 5 6
a

12 13 13 13 13 11

2 (20) 2 (55) 3 (25) 2 (50) 3 (25) 4 (92)

5 (7) 6 (5) 7 (26) 8 (26) 8 (5) 8 (80)

b

93 <1 <4 96 95 88 4.7(4) 11(1) 3.9(4) 94(1):6(1) 90(1):10(1) 100:0

Reaction conditions: C6D6 (0.50 mL), 25 C. Number-average degree of polymerization of the soluble polymer fraction as determined by 1H NMR spectroscopy.

hours. In situ 1H NMR spectroscopy indicated the consumption of both substrates at approximately similar rates. By estimating the decrease in intensity of the aromatic ArH 1H resonance of 2 by means of line-shape analysis and integrating the ArCH3 1H resonance of 2-ethynyltoluene versus the internal standard (HMDSO), it was found that 2 and 2-ethynyltoluene were converted for 64% and 73%, respectively, 33 min after addition of the catalyst solution to the substrate mixture. After 63 min 2 and 2-ethynyltoluene were converted for 86% and 92%, respectively, and after 93 min for 91% and 96%, respectively. After 2 h the 1H NMR spectrum of the reaction mixture did not change significantly. The newly formed 1H signals corresponding to ArCH3 of the end-capping group are similar to those of the linear dimers of 2-ethynyltoluene (Chapter 4). Thus, two sets corresponding to vinylic ( 2.06 ppm) and acetylenic ( 2.00 ppm) aromatic groups were observed for the terminal ArCH3 groups. The reaction mixture was quenched with methanol and evaporated to dryness. The polymer was dissolved completely in chloroform, filtered, washed with copious amounts of methanol (filtrate was analyzed and did not contain linear dimers of 2-ethynyl-toluene) and isolated as a brown-yellow powder in a virtually quantitative yield. Subsequent 1H and 13C NMR analysis of the polymer was indicative of oligo(2,5-di-nhexylphenyleneethynylenevinylene)s end-capped with 2-ethynyltoluene (Figure 6-9). No acetylenic 1H NMR resonances were observed. The intensities of the end-capping ArCH3 and internal ArCH2 1H NMR resonances were estimated by line-shape analysis and indicated a number-average degree of polymerization of 4.7(4). The oligomers contained 6(1)% HT linkages as determined by integration of the vinylic 1H NMR resonances. In an attempt to prepare 2-ethynyltoluene end-capped poly(2) of lower molecular weight, the analogous reaction was performed with 2 (10 equiv) and 2-ethynyltoluene (40 equiv) in a 1:4 ratio, respectively. Upon standing a reaction mixture formed that solidified overnight. 1H NMR and GC-MS analysis of the soluble fraction after work-up revealed the presence of head-to-tail and trans-head-to-tail dimers of 2-ethynyltoluene. It appears therefore that the control over the molecular weight of the polymer by adjusting the molar ratio between the di- and monoyne is a rather delicate balance. At both a high and low concentration of monoyne relative to diyne, consumption of the monoyne was found to be more rapid than the consumption of the diyne, thereby allowing the diyne polymerization to take place in the absence of end-capping agent. Obviously, an optimal molar ratio can be found experimentally, but it will most likely be highly dependent on the nature of the monomer and the end-capping agent, the amount of solvent and the reaction temperature. After several attempts a five-fold molar excess of diyne relative to the end-capping agent was found to provide a reaction mixture which eventually did not solidify in the Cp*2LaCH(SiMe3)2-catalyzed polymerization reactions of 1,4-diethynyl-2,5-di-n-hexyloxybenzene 3 (25 equiv) in the presence of 2ethynyltoluene (5 equiv). This five-fold excess instead of a stoichiometric amount as observed for the polymerization reactions of 2 reflects the lower reactivity of the di-n-hexyloxy monomer 3 relative to its di-nhexyl congener 2. Monitoring this reaction with in situ 1H NMR spectroscopy indicated that 2-ethynyltoluene (8) was consumed after conversion of 3. This finding indicates that the rate of conversion of 8 can only compete with that of 3 at relatively high conversion of 3. A red viscous reaction mixture was obtained after 16 h and 1H NMR spectroscopy indicated that both monomers were completely consumed. Addition of chloroform and precipitate with methanol afforded a red suspension which was filtered and washed with methanol and pentane. After

223

Chapter 6 rotatory evaporation the residue was obtained as a bright red viscous oil (68% yield) and the filtrate as a yellow oil (20%). 1H and 13C{1H} NMR analysis of the red oil indicated the presence of the expected resonances for the end-capped poly(2,5-di-n-hexyloxyphenyleneethynylenevinylene)s having a number-averaged degree of polymerization of 11(1) and consisting of 90(1)% of HH couplings and 10(1)% of HT couplings. 1H NMR and GC-MS analysis of the filtrate revealed the presence of linear dimers of 2-ethynyltoluene and some lowmolecular weight oligomers. When a catalyst solution was added to a nearly equimolar amount of 1,4-diethynyl-3-nhexylthiophene 4 (92 equiv) and 2-ethynyltoluene (88 equiv), no precipitate was observed after 1 day at room temperature. Monitoring the reaction in situ with 1H NMR spectroscopy revealed that 4 was consumed nearly completely within 40 min and that the amount of 2-ethynyltoluene had not changed significantly at this point. As observed above, 1H NMR spectroscopy indicated that 2-ethynyltoluene is converted after complete consumption of 4. After 2 h significant conversion of 2-ethynyltoluene is observed concomitant with the formation of its linear dimers. After 8 h no 1H NMR resonances of the monomers were observed and the no changes were observed in the 1H NMR spectrum of the reaction mixture. The oligomers were isolated and purified by dissolving the reaction mixture in chloroform, precipitate by addition of methanol and filtration followed by washing the residue with methanol and pentane. The washings were evaporated to dryness forming an orange solid containing mainly linear dimers of 2ethynyltoluene and low-molecular weight oligomers (1H NMR, GC-MS). The residue is redissolved in chloroform and was cast as a thin film upon rotatory evaporation (88% yield). 1H and 13C NMR analysis support the formation of 2-ethynyltoluene end-capped oligo(3-n-hexyl-2,5-thienylethynylene-vinylene)s. No vinylic 1H NMR resonances attributable to HT couplings were observed. Unfortunately, the resonance of the terminal methylphenyl groups overlapped with those of the -alkylthienyl groups in CDCl3, but in THF-d8 the overlap was smaller and an averaged-number degree of polymerization of 3.9(4) could be estimated by line-shape analysis.

6.3.4.

The effect of reaction temperature

Introduction The effect of reaction temperature on the present polymerization reactions and the formed products is manifold and difficult to predict a priori. Higher reaction temperatures are likely to decrease both the reaction time and selectivity of the coupling reaction, in analogy to Cp*2LaCH(SiMe3)2-catalyzed 1-alkyne oligomerization reactions (Chapter 5). The solubility of the growing polymer and the viscosity of the reaction medium may also be affected by the reaction temperature. Furthermore, it can be expected that the reaction temperature will influence the relative rate of mono- and diyne conversion in the polymerization reactions with end-capping reagents. To determine the precise effect of reaction temperature on the present polymerization reactions and the properties of the formed polymer, several reactions were performed at higher reaction temperature. Polymerization reactions When the reaction of Cp*2LaCH(SiMe3)2 and 1,4-diethynyl-2,5-di-n-hexylbenzene 2 (50 equiv) was performed at 50 C, a decrease in the selectivity for the trans-head-to-head dimerization (i.e. HH:HT = 82(1):18(1) as compared to 89:11 for the analogous reaction at 25 C) was observed. Concomitantly, the amount of the chloroform-soluble polymer fraction (5% rather 21%) was also smaller. In addition, solidification of the reaction mixture took place after 2 h rather than 12 h as observed at 25 C. Remarkably, no decrease in selectivity was observed by increasing the reaction temperature in the reaction with 1,4-diethynyl-3-di-n-hexylthiophene (4). For example, when the reaction of Cp*2LaCH(SiMe3)2 and 4 (100 equiv) was performed at 50 C, the reaction mixture solidified within 4 h. The polymer was isolated as a shiny green, insoluble solid (77%) and a red, chloroform-soluble solid (10%). In situ 1H NMR spectroscopy indicated that 4 was consumed completely within 10 min and that no vinylic 1H NMR resonances, due to HT couplings, formed throughout the course of reaction. 1H NMR analysis of the chloroform-soluble polymer fraction confirmed the absence of HT couplings and indicated a number-averaged degree of polymerization of 72(5).

224

The organolanthanide-catalyzed polymerization of diynes Table 6-3. The diyne polymerization reactions catalyzed by Cp*2LaCH(SiMe3)2 at different temperatures.a Entry Catalyst Diyne Monoyne soluble t T Pnb HH:HTb (mM) (equiv.) (equiv.) fraction (h) (C) (%) 1 13 2 (50) 21 12 25 89(1):11(1) 2 3 4 5 6
a

13 13 12 11 15

2 (50) 3 (25) 3 (25) 4 (138) 4 (100)

8 (5) 8 (5) b

5 95 92 28 10

8 12 2 12 4

50 25 80 25 50 11(1) 5.7(5) 100(5) 72(5)

82(1):11(1) 90(1):10(1) 86(1):14(1) 100:0 100:0

Reaction conditions: C6D6 (0.50 mL), 25 C. Number-average degree of polymerization of the soluble polymer fraction as determined by 1H NMR spectroscopy.

Polymerization reactions with end-capping groups As discussed above, the successful application of 2-ethynyltoluene as an end-capping agent in the present diyne polymerization reactions depends on the relative reactivity of both substrates. It was found that the use of molar ratios that provided soluble oligomers at ambient temperatures afforded reaction mixtures that solidified after several hours, when the analogous reactions were conducted at higher reaction temperatures. Experiments revealed that a five-fold molar excess of 1,4-diethynyl-2,5-di-n-hexyloxybenzene 3 (25 equiv.) relative to 2-ethynyltoluene (5 equiv.) was necessary to impede the formation of insoluble, high molecular weight oligomers at 80 C. In situ 1H NMR spectroscopy revealed that 3 is consumed completely within 10 min at 80 C and 5 within 26 min and that the reaction mixture did not change after complete conversion of these substrates. The reaction mixture was completely soluble and oligomers formed exhibited an averaged-number degree of polymerization of 5.7(5) and consisted of 86(1)% of HH couplings and 14(1)% of HT couplings (determined by 1H NMR). Interestingly, no linear dimers of 2-ethynyltoluene were found. The results obtained for reactions of 1,4-diethynyl-2,5-di-n-hexylbenzene (2) and 1,4-diethynyl-2,5di-n-hexyloxybenzene (3) indicate that increasing the reaction temperature leads both to lower reaction times and selectivities for trans-head-to-head dimerization. The monomer 2,5-diethynyl-3-n-hexylthiophene (4) represents a remarkable exception on this general observation, however. It is believed that the higher activity and selectivity of 4 can plausibly be attributed to a combination of favorable electronic substituent effects and the heteratomdirected nature of 1-alkyne oligomerization, as proposed for analogous Cp*2LaCH(SiMe3)2-catalyzed 1-alkyne oligomerization reactions (Chapter 4).

6.3.5.

The effect of monomer concentration

Introduction The scope and limitations of the present organolanthanide-catalyzed diyne polymerization reactions were mostly identified by monitoring reaction mixtures with NMR spectroscopy. As a result, these exploratory polymerization reactions were performed at a relatively high monomer concentration. In order to investigate the effects of relatively low monomer concentration on the polymerization reaction, experiments were carried out in a larger amount of solvent. It seems reasonable to expect that the molecular weight of the formed polymers will increase under more dilute reaction conditions, because both the precipitation of a growing polymer chain and the gelation of the polymer solution are promoted by high polymer concentrations. When the experiments are conducted at very low monomer concentration, however, cyclic polymers may form, as is well-documented for the oxidative copper-catalyzed reactions of diynes.84 Typical polymerization experiments at relatively high monomer concentration were performed with 50-100 mg in 0.5 mL of benzene-d6, corresponding to polymer concentrations of 100-200 g/L or 11-21 wt%. Typical polymerization experiments at relatively low monomer concentration, on the other hand, were conducted with 50-100 mg of monomer in 2.0 mL of toluene, corresponding to polymer concentrations of 1-6 wt%.

225

Chapter 6 Table 6-4. The diyne polymerization reactions catalyzed by Cp*2LaCH(SiMe3).a Entry Catalyst Diyne Substrate Soluble t T (mM) (mM) fraction (h) (C) (%) 1 2.3 75.4 78 12 25 2 2 3 4
a

Pnb

HH:HT

38(1) 32(1) 13(1) 200(10)

96(1):4(1) 96(1):4(1) 92(1):8(1) 100:0

2.3 1.7 2.0

2 3 4

75.4 69.8 224

b

87 79 33

4 48 1.3

25 25 50

Reaction conditions: toluene (2.0 mL). NMR spectroscopy.

Number-averaged degree of polymerization determined by 1H

1,4-Diethynyl-2,5-di-n-hexylbenzene (2) When Cp*2LaCH(SiMe3)2 was allowed to react under stirring with 40 equiv. of 1,4-diethynyl-2,5-din-hexylbenzene (2) in toluene (2.0 mL) at 25 C, the brown solution solidified completely within 4 h, yielding a brown solid. The reaction mixture was opened to air after 12 h and subjected to three cycles of dissolution in chloroform, precipitation with methanol, filtration and washing with methanol. The polymer was isolated as a yellow solid in reasonable yield (78%). 1H NMR analysis of the chloroform-soluble polymer fraction indicated an average degree of polymerization of 38(2) and the presence of 96(1)% HH couplings and 4(1)% HT couplings. When the same reaction was quenched after 4 h, the chloroform-soluble polymer fraction was isolated in a high yield (87%) and was found to have an number-average degree of polymerization Pn of 32(1) and a HH:HT ratio of 94:6 (1H NMR). 1,4-Diethynyl-2,5-di-n-hexyloxybenzene (3) Remarkably, the analogous reaction with 1,4-diethynyl-2,5-di-n-hexyloxybenzene (3) did not solidify completely after 2 days. Instead, a red viscous solution was formed containing a red solid. The reaction mixture was opened to air and quenched with chloroform. Heating under stirring afforded a clear solution from which the polymer was precipitated by addition of methanol. The work-up procedure as described above produced a red solid in good yield (79%). A number-average degree of polymerization Pn of 13(1) was found and the oligomers contained 92% HH couplings and 8% HT couplings (1H NMR analysis). 2,5-Diethynyl-3-n-hexylthiophene (4) The reaction of Cp*2LaCH(SiMe3)2 with 110 equiv of 2,5-diethynyl-3-n-hexylthiophene (4) in toluene afforded a reaction mixture which solidified completely into a red-brown solid in 1.3 h at 50 C. The above described work-up procedure afforded a green solid as the insoluble polymer fraction and a red solid (33%) as the soluble polymer fraction. 1H NMR analysis of the chloroform-soluble polymer fraction indicated a number-average degree of polymerization of 200(10) and a very high regioregularity (HH couplings >99.8(1)%).

6.4.

Structural characterization and properties of the polymer.

General remarks As discussed above, 1H and 13C{1H} NMR and IR spectroscopy support the identity of the expected poly(aryleneethynylenevinylene)s. Most of the 1H and 13C NMR resonances could be assigned by comparison with the respective monomers and the model compounds 1,4-diphenylbut-1-en-3-yne (9), 1,4-di(2methoxyphenyl)but-1-en-3-yne (10) and 1,4-di(2-thienyl)but-1-en-3-yne (11). All polymers exhibited a C C stretching band at ~2160 cm-1 and a C-H bending vibration of the (E)-CH=CH group at ~945 cm-1 in the IR spectra. Unfortunately, GPC analysis was not possible, as polymer solutions could not be filtered through 0.5-1.0 m filters. The exact reason for this behavior is unknown at present. Optical properties The optical properties of the polymers were studied by UV-Vis absorption and fluorescence (FL) spectroscopy. The results are summarized in Table 6-5. It can be seen that all polymers are fluorescent at excitation wavelengths that correspond to their maximum absorption, whereas the monomers are not. UV-Vis

226

The organolanthanide-catalyzed polymerization of diynes

1.2

1.2

1.0 Emission intensity (normalized)

0.8

3 poly(3) 10

1.0 Absorption intensity (normalized)

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0 300 400 Wavelength (nm) 500 600

0.0

Figure 6-10. UV-vis absorption and fluorescence spectra (CHCl3) of 1,4-diethynyl-2,5-di-n-hexyloxybenzene (3), (E)-1,4-diphenylbut-1-en-3-yne (10) and poly(3) (see for details, Table 6-5).

1.2 poly(2+8) 2 poly(3+8)

1.2

1.0 Emission intensity (normalized)

1.0 Absorption intensity (normalized)

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0 300 400 500 Wavelength (nm) 600

0.0

Figure 6-11. UV-vis absorption and fluorescence spectra (CHCl3) of 1,4-diethynyl-2,5-di-n-hexylbenzene (2) and polymers end-capped with 2-ethynyltoluene (8), poly(2+8) and poly(3+8) (see for details, Table 6-5).

227

Chapter 6

1.2 poly(4)a 11 poly(4)b

1.2

1.0 Emission intensity (normalized)

1.0 Absorption intensity (normalized)

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0 300 400 500 Wavelength (nm) 600

0.0

Figure 6-12. UV-vis absorption and fluorescence spectra (CHCl3) of (E)-di(2-thienyl)but-1-en-3-yne (11), poly(4)a and poly(4)b (see for details, Table 6-5).

spectroscopy offers a qualitative measure of the -orbital overlap of the conjugated polymer. The maximum UVVis absorption max is attributed to the -* transition of the conjugated polymer backbone. The higher the max in the UV-Vis spectrum, the higher the degree of conjugation present in the polymer backbone. It is wellestablished that the band gap increases with the length of the oligomers until a convergence limit is reached which is termed the effective conjugation length.85 The considerable red-shift of the polymers relative to their corresponding monomers is indicative of extended -conjugation. It is difficult to make fair comparisons between the different types of polymers in the present study, since max is also influenced by the oligomeric chain length and the regioregularity (i.e. (E)-head-to-head (HH) Table 6-5. Optical data of the polymers in solutions.a Entry Compound 1 2 3 4 5 6 7 8 9
a

3 10 poly(3) [Pn 16, 92% HH] 2 poly(2+8) [Pn 7, 94% HH] poly(3+8) [Pn 11, 92% HH] 11 poly(4)a [Pn 10, 100% HH] poly(4)b [Pn 100, 100% HH]

Absorpion max (nm) 275, 337 318 450 300, 310, 336 397 462 348 420,454 480

Emission max (nm) 352, 754 514 472, 502 467 385, 403 515 565

UV-vis absorption and fluorescence spectra were recorded in a dilute chloroform solution at room temperature. The wavelength of the absorption maximum was chosen as the excitation wavelength.

228

The organolanthanide-catalyzed polymerization of diynes

Figure 6-13. MALDI-TOF MS for poly(3). and head-to-tail (HT) couplings) of the polymer. Based on the difference of max between polymer and monomer, the present results suggest that the polymers derived from 1,4-diethynyl-2,5-di-n-hexylbenzene, poly(2), exhibit a lower degree of conjugation than those derived from 1,4-diethynyl-2,5-di-n-hexyloxybenzene, poly(3) (Entries 3,5 and 6, Table 6-5), while polymers based on 2,5-diethynyl-3-n-hexylthiophene, poly(4), display a higher degree of conjugation than poly(3) (Entries 3 and 8, Table 6-5). A blue-shift in the absorption maximum is observed for poly(2+8) relative to poly(3+8) ) (Entries 5 and 6, Table 6-5). This behavior has also been observed for 2,5-dialkyl-substituted poly(pphenyleneethynylene)s as compared to their dialkoxy congeners and has been attributed to the less electrondonating nature of the alkyl groups.17b,86 The influence of the oligomeric chain length can be seen clearly by comparing poly(4)a and poly(4)b (Table 6-5, entries 8 and 9). A considerable red shift is observed in both the absorption and fluorescence spectrum, indicative of a higher effective conjugation length. MALDI-TOF spectrometry Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has gained increasing importance in the characterization of synthetic polymers.87 Both the sample preparation and the nature of the polymer determine the success and the quality of the MALDI mass spectrometric analysis. The type of matrix and the relative ratio of matrix to sample are of considerable importance and their selection and optimization is still often a trial-and-error process. It is generally believed that polymers which are not amenable to MALDI analysis lack suitable ionization sites making the creation of intact gas- phase macromolecules difficult. Experimental experience has shown that polar polymers containing heteroatoms show cationization after being mixed with sodium or potassium salts, while unsaturated polymers without heteroatoms (e.g. polystyrene, polybutadiene, polyisoprene) can be successfully ionized after the addition of silver or copper salts which interact with the -electrons of these ion-binding sites. Unfortunately, MALDI-TOF mass spectra of good quality could not be obtained for the polymers derived from 1,4-diethynylbenzene (1). The use of chloroform as solvent and -cyano-4-hydroxycinnaminic acid as matrix produced a MALDI-TOF mass spectrum of good quality for poly(2). However, even though the signals were 294.3 Da apart corresponding exactly to the molecular mass of the repeating unit, the signals were consistently 120 Da too high relative to the masses of the expected oligo(2) species. The cause of this discrepancy is unknown presently. Variation of matrix, solvent and the addition of metal salts gave similar results. Figure 6-13 shows a MALDI-TOF mass spectrum of poly(3) (Mn 16(1) and 92% HH, as determined by 1H NMR spectroscopy) using ditranol as a matrix and chloroform as a solvent. No cationization salt solutions

229

Chapter 6

Figure 6-14. MALDI-TOF MS for poly(4). were needed to obtain ionization and the addition of copper(I) and silver(I) salts did not improve the quality of the spectrum. The repeat unit from this spectrum is found to be 326 Da, which corresponds to the monomer repeat unit formula weight exactly. However, the distribution of molecular weights does not correspond to the polymer having a number-average molecular weight of 16(1). The underrepesentation of high-mass components with respect to the lower mass components in MALDI-TOF mass spectra is commonly observed and is believed to be caused by several factors, including sample preparation, mass-dependent desorption/ionization, ion focusing/transmission and mass-dependent ion detection.88 The origin of the second series of peaks in Figure 6-13 corresponding to masses which are 71-85 Da lower than those of poly(3) is unknown at present. Attempts to obtain a MALDI-TOF mass spectrum of poly(3) end-capped with 2methylphenylacetylene (8) produced a high quality spectrum (dithranol, chloroform) exhibiting signals that were 36 Da lower than those expected for poly(3+8) (Pn = 16(1) and 92% HH, as determined by 1H NMR spectroscopy). This observation can be rationalized by the addition of HCl to the double or triple carbon bond either during the MALDI desorption/ionization process or during polymer dissolution in chloroform under heating, as hydrogen chloride is known to be formed from chloroform upon heating.89 No attempts were undertaken to study this phenomenon more carefully. A MALDI-TOF mass spectrum of moderate quality was obtained of poly(4) (Pn = 10(1) and 100% HH, as determined by 1H NMR spectroscopy) using dithranol and chloroform. Similar results were obtained upon changing the matrix and the amount of solvent and after addition of metal salts (Figure 6-14). The main peaks are equally spaced and 216 Da apart, representing the expected oligomers which differ in mass by the molecular weight of the repeating unit. The lower intensity signals (including the signal at 347.4 Da) cannot be accounted for at present.

6.5.

Discussion

After initiation of this study three independent reports by different research groups appeared in literature describing a similar synthetic approach to the preparation of PAEVs by means of the metal-catalyzed dimerization of terminal acetylenes. Katayama et al. studied the polymerization of 2,7-diethynyl-9,9dioctylfluorene by three different catalysts which were highy regioselective for (E)-head-to-head, (Z)-head-tohead and head-to-tail dimerization.40a Based on the palladium catalyst developed by Nolan et al.,90 a highly regioregular polymer (99% HH couplings, as indicated by 1H NMR) of moderate molecular weight and high

230

The organolanthanide-catalyzed polymerization of diynes

Scheme 6-11. The regio- and stereoselective synthesis of poly(p-phenylene-ethynylenevinylene)s containing (E)-, (Z)- and gem-vinylene units.40b,c
OC8H17 Cp*2PrCH(SiMe3)2 C7H8, 2 h, r.t. H H H17C8O n Mn 16000, Mw/Mn 3.72 99% yield, 98% selectivity

Si N OC8H17

Lu(THF)CH2(SiMe3)2 OC8H17 H Mn 5300, Mw/Mn 5.48 95% yield, 96% selectivity

H17C8O

C7H8, 7 h, 110 C

H17C8O

OC8H17 Cp*2Lu(THF)CH2(SiMe3) C7H8, 159 h, 110 C H H17C8O

H n

Mn 10000, Mw/Mn 4.44 99% yield, 100% selectivity

polydispersity was obtained in good yield. It is interesting to note that GPC analysis in THF based on polystyrene standard could be performed. However, comparison of the number-average molecular weight values obtained from 1H NMR end-group analysis (Pn = 14.1) with those from GPC (Pn, GPC = 31.9, Mw/Mn = 2.35) indicated that GPC overestimates the molecular weight of the PAEVs by a factor of 2.3, probably due to the rigid-rod structure of the polymers.91 Nishiura, Hou and co-workers reported the polymerization of 1,4-diethynyl-2,5-dioctylbenzene by three different organolanthanide catalysts which were regioselective for (E)-head-to-head, (Z)-head-to-head and head-to-tail acetylene dimerization (Scheme 6-11).40b,c The lanthanidocene catalyst Cp*2PrCH(SiMe3)2 was used to prepare the corresponding (E)-vinylene rich poly(arylene-ethynylenevinylene) in high yield and regioselectivity (98% HH as determined by 1H NMR). GPC analysis (based on polystyrene) indicated a high molecular weight and polydispersity (Pn, GPC = 39.0, Mw/Mn 2.35), but no end-group analysis by means of 1H NMR spectroscopy was reported. The actual degree of polymerization is thus unknown. Assuming that the correction factor found by Katayama et al. is also valid for this system, this result points to a polymer of Pn 17. It was shown that the molecular weight could be controlled by varying the polymerization time (i.e. Mn = 2400 after 15 min, Mn = 5400 after 30 min, Mn = 16000 after 1 h at room temperature), but the occurrence of gelation is not mentioned. Comparison with the present results is not possible, because important experimental conditions such as monomer and catalyst concentration and spectral data including NMR and IR spectroscopy were not reported. The (Z)-rich polyenyne was studied with UV-vis spectroscopy and no relationship between Mn and max was observed. This process was later patented by Nishiura et al.40e and both aromatic and heteroaromatic diyne monomers were claimed, but only the preparation of a polymer derived from 1,4-diethynyl-2,5-di-noctyloxybenzene was described. Ueda and Tomita used a low-valent titanocene catalyst generated from Cp*2TiCl2 and i-PrMgBr to Scheme 6-12. Model compound III and enyne-containing polymers IV and V.
R

Ph S

Ph O O H21C10O III OC10H21 IV O n V n

231

Chapter 6

Table 6-6. Comparison of the optical properties of poly(2+8) in solution with similar polymers classes.a Entry Polymer Properties max (nm) [solvent] b 1 = 7 397 [CHCl3] P n C6H13 94% HH 472, 502c [CHCl3]
H13C6 n

of other Ref. present work

2
H H13C6

C6H13 H n

Pn, GPC = 13

309b [cC6H13] 432c [cC6H13]

53

3
I H13C6

C6H13 I n

Pn, GPC = 58 Pn, EA = 31

388b [cC6H13] 428c [cC6H13]

53

4
RMe2Si H17C8

C8H17 SiMe2R n
C6H13 CH3 n H13C6 CH3

Pn = 22-26

384b [C7H8] 426, 448c [C7H8]

91f

384b [CHCl3] 425, 450c [CHCl3]

17b

Abbreviations: R = CH2CH(CH3)CO2CH2CH2OH. Pn, Pn, GPC, Pn, EA are the number-average degrees of polymerization based on 1H NMR, GPC and elemental analysis, respectively. b UV-vis absorption maximum. c Emission maximum upon excitation. d Not reported. polymerize 1,7-octadiyne and 1,4-bis(2-propynyloxy)benzene to gem-vinylene rich PAEVs.40d The reaction of 1,4-diethynylbenzene gave rise to low-molecular weight oligomers, while 1,4-diethynyl-2,5-dialkoxybenzenes provided insoluble cross-linked products. Low conversions of 1,4-bis(2-propynyloxy)benzene were found and the authors believed that this was the result of metal oxygen coordination. When comparing the present results with those of poly(aryleneethynylenevinylene)s prepared either by metal-catalyzed 1-alkyne dimerization or palladium-catalyzed cross-coupling between vinylic halides and diacetylenes, several dissimilarities are apparent. Firstly, the occurence of gelation is not entioned in these studies. This is remarkable, as gelation and aggregation phenomena are common in concentrated solutions (typically 5-20 wt%) of many conjugated polymers having extensive back bone conjugation, including rigid-rod polymers such as polyarylenes72q, poly(p-phenyleneethynylene)s72h-k, poly(p-phenylenebutadiynylene)s43f and poly(thiophene)s72e-g having linear alkyl or alkoxy groups. Only Venkatasan et al. noted that poly(aryleneethynylenevinylene)s II (Scheme 6-2) swelled in a variety of solvents (e.g. N-methylpyrrolidine, dimethylformamide, dimethylsulfoxide, chloroform, THF).37 Secondly, the formation of insoluble material in PAEV solutions upon standing at room temperature has little precedent. The single similar observation is by Kane et al. who reported that the prepared poly(aryleneethynylenevinylene) I (Scheme 6-2) did not redissolve after drying.36 The same authors also prepared the model compound III and found that it decomposed upon standing at room temperature (Scheme 6-12). The instability of I seems not to be heat-induced, as DSC indicated a single exotherm with onsets ranging from 300 to 380C. The only other DSC study performed on PAEVs involves II which was found to undergo acetylenic cross-linking at temperatures above 250 C.37 However, thermal cross-linking of the triple carbon bond of the repeating enyne unit has also been observed at lower temperatures in other enyne-containing polymers such as IV (~100 C)92, V (~80 C)40d and a variety of polyimides93 (200-300 C) (Scheme 6-12). As noted before and in analogy to poly(arylenebutadiynylene)s, both photochemical and thermal acetylene cross-

232

The organolanthanide-catalyzed polymerization of diynes

Table 6-7. Comparison of the optical properties of poly(3) and poly(3+8) in solution with similar polymers of other classes.a Entry Polymer Properties max (nm) Ref. [solvent] OC6H13 present Pn = 16 1 450b [CHCl3] work 92% HH 514c [CHCl3]
H H H13C6O n
OC8H17 H H H17C8O n
OC6H13

Pn, GPC = 14 98% HH (Pn 6.2)d Pn = 11 92% HH

451b

40b,c

462b [CHCl3] 467c [CHCl3]

present work

H13C6O

94

OC10H21 I H21C10O n
OC8H17

Pn, GPC = 2226

430-440b [CHCl3] 490, 520c [CHCl3]

37

Pn = 23
n

449b [CHCl3] 475, 505c [CHCl3]

97

H17C8O

6
H H7C3O
a

OC3H7

Pn = 11
nH

481b [CHCl3]

98

Pn and Pn, GPC are the number-average degrees of polymerization based on 1H NMR and GPC analysis, respectively. b UV-vis absorption maximum. c Emission maximum upon excitation. d Estimated, see text for details.

linking affording insoluble solids are plausible processes which rationalize the observed instability of the present PAEVs. Another difference between the present PAEVs and those reported in literature is that GPC analysis could be performed on II37, poly(2,5-dioctylphenyl-ethynylenevinylene)s,40b,c and poly(9,9-dioctyl fluorenylethynylenevinylene)s.40a This difference in behavior may possibly be attributed to the following rationales: (i) polymer II is most likely to have iodine end-groups94, thereby affecting both stability and gelation behavior, (ii) the fluorenyl moiety might confer an increased stability or a decreased tendency for microgelation relative to the present phenyl or thienyl moieties. Unfortunately, Nishiura, Hou and co-workers did not report experimental details of their GPC analyses, such as solvent or temperature, rendering any attempt to rationalize this difference in behavior as mere speculation. Nishiura, Hou and co-workers also used a lanthanidocene catalyst to prepare (E)-vinylene-rich dialkoxy-substituted PAEVs. The present choice of Cp*2LaCH(SiMe3)2 was based on a systematic study of the substrate effects in the Cp*2LnR-catalyzed 1-alkyne oligomerization reaction (Chapter 4). It was found that the lanthanidocene catalysts have an electronic preference for trans-head-to-head dimerization, but that this preference is counterbalanced by unfavorable steric effects between the ancillary ligand system and the coordinated substrate. Increasing the steric requirements of the substrate by means of ortho-substitution or decreasing the metal ion radius gave rise to larger amounts of the head-to-tail dimers. The observation that the catalyst Cp*2PrCH(SiMe3)2 is more selective for trans-head-to-head dimerization than Cp*2LaCH(SiMe3)2, while Pr is considerably smaller than La,95 is thus surpising and not understood at present. The chemical structure of PAEVs combines those of the well-established conjugated polymer classes of poly(aryleneethynylene)s and poly(arylenevinylene)s, but relationships with polydiacetylenes and poly(arylenebutadiynylene)s are also apparent. It seems therefore interesting to compare the optical properties of

233

Chapter 6

Table 6-8. Comparison of the optical properties of poly(4) in solution with similar polymers of other wellestablished classes.a Entry Polymer Properties max (nm) Ref. [solvent] present Pn = 100, 100% HH 1 480b [CHCl3] H S H work 565c [CHCl3]
C6H13 n

2 3 4 5 6 7 8
H S

S n C6H13

50% HTT 70% HTT 98-99% HTT Pn, GPC = 152, 98-99% HTT Pn, GPC = 43, regioirregular/ cross-conjugated defects Pn, GPC = 29, 100% HTH-HTT Pn = 16, 100% HTT

S C12H25 n

428b 436b 442b 456b [CHCl3] 570c [CHCl3] 550b [CH2Cl2] 577b [CH2Cl2] 440b [CHCl3]

64a 100 60a 64a 88c 88c 101a

SiMe3 n

C6H13

S C6H13 n

Pw, GPC = 105, 100% HTT

10

Pw, GPC = 605, 100% HTH

11

Me3Si

S C6H13

SiMe3 n

Pn, GPC = 19

441b [CHCl3], 440b [THF] 506, 536c [THF] 437b [CHCl3], 436b [THF] 505, 535c [THF] 430b

101b

101b

102a

12

S n

Pn, GPC = 1200-18000

410b [THF] 450c [THF]

102b

Pn and Pn, GPC are the number-average degrees of polymerization based on 1H NMR and GPC analysis, respectively. Pw, GPC is the weight-average degree of polymerization based on GPC analysis b UV-vis absorption maximum. c Emission maximum upon excitation.

the present PAEVs with those of analogous polymers of the well-established classes. The absorption and emission maxima of poly(2+8) and those of similar polymers belonging to the classes of PAEs and PABs are tabulated in Table 6-6. It can be seen that poly(2+8) exhibits both a higher absorption and emission maximum relative to analogous PAEs (Entries 3-5) and PABs (Entry 2), indicative of a higher degree of conjugation length. It is interesting to note that the nature of the end-group and the solvent seem to have little effect on the absorption and emission maxima in PAEs (Entries 3-5).96 Assuming similar behavior in PAEVs, the higher absorption and emission maximum wavelengths may plausibly arise from an intrinsically higher effective conjugation length. The absorption and emission maxima of the hexyloxy derivatives poly(3) and poly(3+8) and those of analogous PAEVs, PPEs and PAVs are shown in Table 6-6. Poly(3) exhibits a significantly lower absorption maximum relative to the other conjugated polymers, while its emission maximum is higher (Entry 1). It seems unlikely that this difference can be attributed to the relatively low regioregularity, because the regioregular gemvinylic dioctyloxy analogue (Pn, GPC = 12.8, 100% HT) showed an absorption maximum of 435 nm.40c Poly(3+8), on the other hand, displays a higher absorption and lower emission maximum relative to its non end-capped dioctyloxy PAEV derivative (Entry 2) and phenyl end-capped dioctyloxy PPE analogue97 (Entry 5), but a lower absorption maximum than its dipropyloxy PPV analogue98 (Entry 6). Assuming that the convergence limit of poly(3+8) is reached at the undecamer stage, these results indicate that the effective conjugation length of the present end-capped PAEVs is higher than that of similar PABs and PAEs, but lower than that of similar PAVs. It

234

The organolanthanide-catalyzed polymerization of diynes is interesting to note in this respect that a systematic study revealed that the absorption maximum of oligo(2,5propyloxy-p-phenylenevinylene) did not increase beyond the undecamer stage. The observation99 that the didecyloxy derivative of PPE exhibits an absorption maximum of 429 in THF suggests, moreover, that the length of the alkoxy side group and the nature of the solvent and end-capping group affect the absorption maximum of the PPEs only to a small extent. The optical properties of poly(4) are compared with those of poly(3-n-hexyl-2,5-thienylene)s60a,64a,100 (P3DTVs), poly(3-n-hexyl-2,5-thienylene(P3HTs), poly(3-n-dodecyl-2,5-thienylenevinylene)s88c 101 ethynylene)s (P3HTEs) and poly(3-n-hexyl-2,5-thienylenebutadiynylene)s102 (P3HTBs) in Table 6-8. It is well-recognized that irregularly substituted poly(thiophene)s have structures where unfavorable head-to-head couplings (HTH) cause a sterically driven twist of thiophene rings, resulting in loss of conjugation (Scheme 6-13).18 Regioregular, head-to-tail (HTT) PTs can easily access a low energy planar conformation, leading to highly conjugated polymers. The effect of this type of regioregularity on the effective conjugtion length is still present in PTVs (Entries 9-10) and PTEs, albeit less pronounced than in PTs (Entries 2-5). It can be anticipitated that the effect of this type of regioregularity plays only a minor role in polymers where the distance between the thienyl moieties is larger, such as PABs and PAEVs, but no systematic studies are known that confirm this hypothesis. Scheme 6-13. Regioregularity in asymmetrically substituted thienyl polymers.
R S S R head-to-head (HTH) R head-to-tail (HTT) S S R

Pearson et al. found that a near saturation of the optical properties of HTT oligo(3-n-ethyl-2,5thienyleneethynylene)s occurred at the octamer stage, based on the observation that doubling the conjugation length to the 16-mer caused little change in the absorbance maximum (Entry 8) of the octamer.101a Wenz et al. reported that poly(diacetylene)s display an effective conjugation length of approximately 5-7 repeating units, as indicated by UV-vis, resonance-Raman and 13C NMR spectroscopy.103 These results suggest that the present poly(4) having Pn = 100(1) must certainly have reached its convergence limit with respect to its absorption and emission maximum. Comparison of the optical properties of poly(4) with similar polymers of other wellestablished classes such as PAs, PAVs, PAEs and PABs confirms previous notion that PAEVs have effective conjugation lengths in solution that are longer than PAs, PAEs and PABs, but shorther than PAVs. Further studies are needed to evaluate the full potential of the present polymers as components in advanced materials. Even though the present synthetic study established that highly regioregular conjugated polymers with relatively large effective conjugation lengths can be prepared, investigations with respect to their solid state properties and the exact nature of their instability, in particular, are required.

6.6.

Conclusions

Soluble conjugated poly(aryleneethynylenevinylene)s have been prepared in good yields by step growth polymerization of (hetero)aromatic diynes using the organolanthanide-catalyzed dimerization of 1alkynes. The molecular weight and the yield of the soluble fraction of the polymers can be controlled by variation of monomer structure, monomer concentration and the ratio between the diyne and an end-capping monoyne. The polymers based on 1,4-diethynyl-2,5-di-n-hexylbenzene (2), 1,4-diethynyl-2,5-di-nhexyloxybenzene (3) and 2,5-diethynyl-3-n-hexylthiophene (4) have been characterized by 1H NMR, 13C NMR, IR spectroscopy and MALDI-TOF mass spectrometry. They are highly fluorescent and their photophysical properties suggest that they represent promising materials for electronic and photonic applications.

235

Chapter 6

6.7.

Experimental Section

General experimental procedures. For general remarks and physical and analytic measurements, see Sections 2.7 and 5.7. The compounds 1,4-bis(trimethylsilylethyne)benzene,50 1,4-di-n-hexylbenzene,104 1,4di-n-hexyloxy-benzene,105 3-n-hexylthiophene,60 2-ethynyltoluene106 and 2,6-dimethylethynylbenzene106 were prepared according to literature procedures. N,N-Diisopropylamine (KOH) and phenylacetylene (CaH2) was purchased from Aldrich and dried as recommended.107 The UV-vis absorption spectra were recorded on a Hewlett-Packard HP 8453 diiode array spectrophotometer and the fluorescence spectra were recorded on a SPF500C spectrofluorometer (SLM Aminco). All emission studies were performed at room temperature in optically dilute solutions. The solutions were freshly prepared by weighing out the polymer (typically 2.00 mg) and dissolution in spectroscopic grade chloroform (typically 100.0 mL). The solutions were diluted until absorption maxima less than 0.1 were obtained to avoid the inner filter effect. General purification procedure for aromatic diynes. The oils or solids obtained after synthesis were dissolved in pentane containing freshly ground CaH2 and stored at 4 C under nitrogen for at least 24 h after several freeze-thaw-pump degassing cycles. Filtration, followed by solvent evaporation, or crystallization afforded the dried monomers which were stored under nitrogen, in the dark and at -30 C as soon as possible. 1,4-Diethynylbenzene. 1,4-Bis(trimethylsilylethynyl)benzene (1.5 g, 5.55 mmol) was dissolved in MeOH (50 mL) and THF (30 mL) in a 200-mL Erlenmeyer equipped with a magnetic stir bar. After addition of 2.0 mL (24 mmol) of an aqueous KOH solution (12 M), the colorless solution was stirred at room temperature for 1 h during which it turned light-yellow. After addition of water (100 mL) and pentane (80 mL), the organic layer was separated, washed with water and rotatory evaporated to yield a white solid with a distinct smell. Cooling a concentrated pentane solution gave white needles. Yield: 0.68 g (97%). 1 H NMR (400 MHz, CDCl3, 18 C): 7.42 (s, CH, 4 H), 3.15 (s, CCH, 2 H). 13C-{1H} NMR (100 MHz, CDCl3, 18 C): 131.98 (CH), 122.52 (CCCH), 82.98 (CCH), 79.06 (CCH). IR (KBr, [cm-1]): 3298 (m), 3260 (s, C-H stretching, C-H), 2956 (m), 2921 (m, C-H stretching, aromatic =C-H), 2851 (m), 2101 (w, C C stretching), 1916 (m), 1628 (m), 1492 (m), 1401 (m), 1249 (m), 1103 (m), 1015 (m), 962 (w), 834 (s), 706 (m), 674 (s), 639 (s), 619 (s), 545 (s). 2,5-Diiodo-1,4-bis(n-hexyl)benzene. The following procedure is a modification of a reported one.54c 1,4-Di-n-hexylbenzene (22.9 g, 92.8 mmol) was brought in a round-bottomed, three-neck flask (500 mL), equipped with a cooler and stir bar, and dissolved in a mixture containing acetic acid (140 mL), CCl4 (38 mL) and concentrated sulfuric acid (24 mL). After addition of iodine (26.30 g, 103.6 mmol) and KIO4 (7.50 g, 32.6 mmol) the reaction mixture was stirred overnight at 90 C, forming a deep-purple colored suspension. Then, the reaction mixture was cooled, neutralized with an aqueous solution of NaHCO3 and an aqueous solution of Na2SO3 to remove the excess of iodine. The crude product was extracted with petroleum ether and rotatory evaporation afforded a brown, viscous oil. Repeated crystallization from ethanol yielded off-white needles. Yield: 25.4 g (55%). 1 H NMR (300 MHz, CDCl3, 25 C): 7.55 (s, CH, 2 H), 2.55 (t, 3JHH = 7.9 Hz, CCH2, 4 H), 1.50 (m, CH2, 4 H), 1.30 (m, CH2, 12 H), 0.86 (t, 3JHH = 6.6 Hz, CH3, 6 H). 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 144.83 (CCH2), 139.28 (CH), 100.32 (CI), 39.83(CH2), 31.59 (CH2), 30.15 (CH2), 28.98 (CH2), 22.57 (CH2), 14.07 (CH3). Anal. Calcd. for C18H28I2 (498.23): C, 43.39%; H, 5.66%. Found: C, 43.47%; H, 5.61%. 2,5-Bis(trimethylsilylethynyl)-1,4-di-n-hexylbenzene.53 N,N-Diisopropylamine (50 mL) was brought into a three-neck, round-bottom, 250-mL flask equipped with a magnetic stir bar and condenser. Addition of 2,5-diiodo-1,4-bis(n-hexyl)benzene (4 g, 8.03 mmol), Pd(PPh3)2Cl2 (0.30 g, 0.43 mmol) and PPh3 (0.22 g, 0.84 mmol). The yellow suspension was heated for 1 h at 60C under stirring. In a Schlenk flask equipped with a magnetic stir bar CuI (0.08 g, 0.42 mmol) is added to THF (50 mL) forming a suspension which was subsequently stirred under heating for 1 h. Then trimethylsilylacetylene (2.70 mL, 19.1 mmol) and the CuI suspension are added to the yellow suspension. The reaction mixture was heated to 90 C and stirred for 12 h. After allowing the mixture to cool to room temperature the suspension was filtered over a glass filter in vacuo. The volatiles were removed from the brown filtrate by rotatory evaporation to yield a brown oil. The crude product was purified by column chromatography (neutral alumina, petroleum ether 40-60 C) to afford a yellow oil. White crystalline material was obtained by cooling a concentrated solution of the yellow oil in aqueous ethanol. Yield: 3.61 g (93%). 1 H NMR (400 MHz, CDCl3, 18 C): 7.22 (s, CH, 4 H), 2.65 (t, 3JHH = 7.7 Hz, CCH2, 4 H), 1.57 (m, CH2, 4 H), 1.30 (m, CH2, 12 H), 0.87 (t, 3JHH = 7.7 Hz, CH2CH3, 6 H), 0.23 (t, 3JHH = 6.8 Hz, SiCH3, 18 H). 13C{1H} NMR (100 MHz, CDCl3, 18 C): 142.66 (CCH2), 132.45 (CH), 122.53 (CC C), 103.93 (CC C), 98.88

236

The organolanthanide-catalyzed polymerization of diynes (C CSi), 34.11 (CCH2), 31.72 (CH2), 30.57, 29.26, 22.62, 14.11 (CH2CH3), -0.04 (SiCH3). Anal. Calcd. for C28H46Si2 (438.84): C, 76.63%; H, 10.57%. Found: C, 76.49%; H, 10.65%. 2,5-Bis(trimethylsilylethynyl)-1,4-di-n-hexyl-benzene 2,5-Diethynyl-1,4-bis(n-hexyl)benzene.53 (4.38 g, 9.05 mmol) is dissolved in MeOH (85 mL) and THF (15 mL) in a 200-mL Erlenmeyer equipped with a magnetic stir bar. After addition of 2.0 mL (24 mmol) of an aqueous KOH solution (12 M), the colorless solution was stirred at room temperature for 1 h during which it turned light-yellow. After addition of water (100 mL) and pentane (80 mL), the organic layer was separated, washed with water, and rotatory evaporated to yield a light-yellow oil. Yield: 2.46 g (92%). 1 H NMR (300 MHz, CDCl3, 25 C): 7.27 (s, CH, 2 H), 3.26 (s, CCH, 2 H), 2.69 (t, 3JHH = 7.7 Hz, CCH2, 4 H), 1.58 (m, CH2, 4 H), 1.30 (m, CH2, 12 H), 0.87 (t, 3JHH = 6.8 Hz, CH3, 6 H). 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 142.70 (CCH2), 132.94 (CH), 121.90 (CCCH), 82.25 (CCH), 81.05 (CCH), 33.76 (CCH2), 31.62 (CH2), 30.41 (CH2), 29.08 (CH2), 22.57 (CH2CH3), 14.08 (CH3). IR (neat, [cm-1]): 3310 (m), 3295 (m), 2955 (m), 2925 (s), 2855 (s), 2105 (w), 1490 (m), 1460 (m), 1395 (w), 1380 (w), 1212 (w), 900 (m), 720 (m). 2,5-Diiodo-1,4-bis(n-hexyloxy)benzene. The following procedure is a modification of a published one.105 1,4-Di-n-hexyloxybenzene (20.1 g, 72.2 mmol), iodine (25.1 g, 98.9 mmol), KIO4 (20.4 g, 88.7 mmol), acetic acid (130 mL), water (10 mL), and H2SO4 (7 mL) were brought in a 500-mL three-necked flask equipped with magnetic stir bar and condenser. The reaction mixture was stirred for 4 h at 70 C. The flask was placed in an ice-water bath and Na2S2O42H2O (20 g, 95 mmol) was added under stirring to remove the excess of iodine upon which the mixture turned first brown and then yellow. The mixture is filtered. The filtrate was washed with water and extracted with ether and dried over MgSO4. The crude product in the residue and extracted filtrate were purified by repeated crystallizations in diethyl ether at low temperature (-30 C) to afford white crystals. Yield: 15.23 g (40%). 1 H NMR (300 MHz, CDCl3, 25 C): 7.13(s, CH, 1 H), 3.88 (t, OCH2, 4 H), 1.76 (m, CH2, 4 H), 1.51.3 (m, CH2, 12 H), 0.87 (t, CH3, 6 H). 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 152.86 (CO), 122.79 (CH), 86.29 (CI), 70.35 (OCH2), 31.46 (CH2), 29.10 (CH2), 25.70 (CH2), 22.58 (CH2), 14.01(CH3). Anal. Calcd. for C18H28I2O2 (530.23): C, 40.77%; H, 5.32%. Found: C, 40.86%; H, 5.40%. 2,5-Bis(trimethylsilylethynyl)-1,4-bis(n-hexyloxy)benzene. N,N-Diisopropylamine (100 mL) is brought into a three-neck, round-bottomed flask equipped with a magnetic stir bar and condenser. Addition of 2,5-diiodo-1,4-bis(n-hexyloxy)benzene (4.02 g, 7.58 mmol), Pd(PPh3)2Cl2 (0.34 g, 0.49 mmol) and CuI (0.13 g, 0.66 mmol). Trimethylsilylacetylene (2.65 mL, 18.8 mmol) was added to the yellow suspension under stirring. The reaction mixture was heated to 70 C and stirred for 2 h. After allowing the mixture to cool to room temperature toluene (50 mL) was added. The suspension was filtered in vacuo over a glass filter and the brown filtrate was rotatory evaporated to dryness yielding a brown oil which solidified upon cooling to room temperature. The brown solid was purified by column chromatography (silica, 230-400 mesh, 60 , petroleum ether 40-60 C) affording a brown solid after rotatory evaporation of the solvent. When a concentrated solution of this solid was cooled -40 C, off-white crystals formed. Yield: 2.95 g (83%). 1 H NMR (400 MHz, CDCl3, 25 C): 6.85 (s, CH, 2 H), 3.90 (t, 3JHH = 6.5 Hz, OCH2, 4 H), 1.74 (m, CH2, 4 H), 1.46 (m, CH2, 4 H), 1.30 (m, CH2, 8 H), 0.86 (t, 3JHH = 6.7 Hz, CH3, 6 H,). 13C-{1H} NMR (100 MHz, CDCl3, 25 C): 154.00 (CO), 117.23 (CH), 113.97 (CC), 101.07 (C C), 100.03 (C C), 69.44 (OCH2), 31.59 (CH2), 29.28 (CH2), 25.67 (CH2), 22.61 (CH2), 14.03 (CH3), -0.08 (SiCH3). Anal. Calcd. for C28H46O2Si2 (470.84): C, 71.43%; H, 9.85%. Found: C, 71.49%; H, 10.01%. 2,5-Diethynyl-1,4-bis(n-hexyloxy)benzene. 2,5-Bis(trimethylsilylethy-nyl)-1,4-bis(n-hexyloxy)benzene (1.52 g, 3.23 mmol) was dissolved in THF (50 mL) and MeOH (50 mL) in a 250-mL Erlenmeyer equipped with magnetic stir bar. Addition of 15 mL (75 mmol) of an aqueous NaOH solution (5 M) upon which the colorless solution changed to light-yellow. After stirring for 2 h at room temperature the reaction mixture was extracted with pentane. The organic layer was rotatory evaporated to dryness to yield a yellow oil. When a concentrated ether solution of this oil was cooled, yellow crystals formed. Yield: 0.91 g (86%). 1 H NMR (500 MHz, CDCl3, 25 C): 6.91 (s, CH, 2 H), 3.93 (t, 3JHH = 6.6 Hz, OCH2, 4 H), 3.29 (s, CCH, 2H), 1.75 (m, CH2, 4 H), 1.42 (m, CH2, 4 H), 1.29 (m, CH2, 8 H), 0.86 (t, 3JHH = 7.0 Hz, CH3, 6 H). 13C{1H} NMR (125 MHz, CDCl3, 25 C): 153.97 (CO), 117.75 (CH), 113.27 (CCCH), 82.37 (CCH), 79.77 (CCH), 69.65 (OCH2), 31.49 (CH2), 29.07 (CH2), 25.55 (CH2), 22.56 (CH2), 13.98 (CH3). IR (KBr, [cm-1]): 3271 (s, C-H stretching, C-H), 2969 (m), 2952 (m), 2922 (m, C-H stretching, aromatic =C-H), 2852 (m), 2103 (w, C C stretching), 1710 (m), 1497 (s), 1464 (m), 1399 (m), 1384 (s), 1271 (m), 1220 (s), 1196 (m), 1048 (m), 996 (m), 943 (m), 861 (m), 692 (m), 659 (m).

237

Chapter 6 2,5-Diiodo-3-n-hexylthiophene.60a 3-n-Hexylthiophene (5.50 g, 32.7 mmol) was dissolved in dry CH2Cl2 in a 250-mL, three-neck, round-bottom flask, equipped with stir bar, cooler and drop funnel. After addition of finely powdered iodine (8.30 g, 32.7 mmol) the reaction mixture was heated to 65 C under reflux. A mixture of concentrated HNO3 (5 mL) and water (5 mL) was added dropwise within 30 min. After 3 h at 65 C the reaction mixture was allowed to cool to room temperature. The excess of iodine was removed by the addition of an aqueous solution of Na2S2O4. The organic layer was extracted with petroleum ether (40-60 C) and dried over MgSO4. Rotatory evaporation afforded a dark-brown, viscous oil. The crude product was purified by column chromatography (silica, petroleum ether) to yield a yellow liquid. Yield: 9.95g (73%). 1 H NMR (300 MHz, CDCl3, 25 C): 6.87 (s, CH, 1 H), 2.49 (t, 3JHH = 7.8 Hz, CCH2, 2 H), 1.53 (m, CH2, 2 H), 1.29 (m, CH2, 6 H), 0.87 (t, 3JHH = 6.9 Hz, 3 H). 13C-{1H} NMR (75 MHz, CDCl3, 25.0 C): 149.49 (CCH2), 137.76 (CH), 130.25, 127.93 (CI), 32.90 (CCH2), 31.57 (CH2), 29.90, 38.87, 22.55, 14.03 (CH3). 2,5-Bis(trimethylsilylethynyl)-3-n-hexylthiophene.43a In a 250-mL, round-bottom flask equipped with a cooler and stir bar dry N,N-diisopropylamine (50 mL) was brought. After addition of 2,5-diiodo-3-nhexylthiophene (4.95 g, 11.8 mmol), PPh3 (0.30 g, 1.14 mmol), Pd(PPh3)2Cl2 (0.43 g, 0.61 mmol) the yellow suspension was heated to 60 C under stirring until the solids were dissolved. Then, a suspension of CuI (0.11 g, 0.58 mmol) in dry THF (25 mL) and trimethylsilylacetylene (4.50 mL, 31.8 mmol) were added successively. The reaction mixture was heated overnight to 90 C on reflux. After allowing the mixture to cool to room temperature toluene (50 mL) was added. The suspension was filtered over a glass filter in vacuo and the brown filtrate was rotatory evaporated to dryness yielding a viscous yellow oil. The oil was purified by column chromatography (silica, 230-400 mesh, 60 , petroleum ether 40-60 C) affording a light-yellow oil after rotatory evaporation of the solvent. The product was found to be contaminated with traces of a mono(trimethylsilylethynyl) analogue (<1%) which was removed completely by Kgelrohr distillation (~250 C, ~1 mmHg). Yield: 4.10 g (95%). 1 H NMR (400 MHz, CDCl3, 25 C): 6.93 (s, CH, 1 H), 2.58 (t, 3JHH = 7.7 Hz, CCH2, 2 H), 1.55 (m, CH2, 2 H), 1.27 (m, CH2, 6 H), 0.86 (t, 3JHH = 6.6 Hz, CH3, 3 H). 13C-{1H} NMR (100 MHz, CDCl3, 25 C): 148.39 (CCH2), 133.47 (CH), 122.76, 119.86 (CC C), 101.80, 99.40, 97.40, 96.86 (C C), 31.54 (CCH2), 29.87, 29.34, 28.77, 22.56, 14.06 (CH3), -0.12, -0.20 (SiCH3). Anal. Calcd. for C20H32SSi2 (360.71): C, 66.60%; H, 8.94%. Found: C, 66.78%; H, 9.11%. 2,5-Diethynyl-3-n-hexylthiophene.43a,58a 2,5-Bis[(trimethylsilyl)ethynyl]-3-n-hexyl-thiophene (3.73 g, 10.3 mmol) was dissolved in MeOH (50 mL) and THF (50 mL) in an erlenmeyer under stirring. After addition of KOH (0.62 g, 11.0 mmol) the mixture was allowed to stir in the dark for 1 h at room temperature. Then, the mixture was washed with water, extracted with petroleum ether (40-60 C) and dried over MgSO4. After rotatory evaporation, the crude product was purified by flash chromatography (neutral alumina, pentane) to afford a lightyellow oil. Yield: 2.12 g (95%). 1 H NMR (400 MHz, CDCl3, 25 C): 6.99 (s, CH, 1 H), 3.42 (s, CCH, 1 H), 3.30 (s, CCH, 1 H), 2.62 (t, 3JHH = 7.7 Hz, CCH2, 2 H), 1.54 (m, CH2, 2 H), 1.27 (m, CH2, 6 H), 0.86 (t, 3JHH = 7.7 Hz, CH3, 3 H). 13 C-{1H} NMR (75 MHz, C6D6, 25 C): 148.83 (CCH2), 134.00 (CH), 122.72, 119.44 (CCCH), 84.46, 82.29 (CCH), 76.77, 76.11 (CCH), 31.81 (CH2), 30.16, 29.51, 29.07, 22.84, 14.22 (CH3). IR (neat, [cm-1]): 3308 (s, CH stretching, C-H), 2956 (m), 2928 (m, C-H stretching, aromatic =C-H), 2858 (m), 2103 (m, C C stretching), 1692 (w), 1529 (w), 1465 (m), 1379 (w), 1199 (m), 1118 (m), 1017 (m), 847 (m), 723 (m), 662 (m), 594 (m). Representative polymerization reaction of diyne catalyzed by Cp*2LaCH(SiMe3)2 at relatively high monomer concentration (NMR scale). To a NMR tube containing a solution of 1,4-diethynyl-2,5-di-nhexyloxybenzene (41.3 mg, 127 mol) in benzene-d6 (400 L) was added a 100 L (6.13 mol) of a stock solution of Cp*2LaCH(SiMe3)2 (61.3 mM) in benzene-d6 with a microsyringe. The mixture was stirred manually during which the light-yellow solution turned brown in color. The NMR tube was brought in a sonicator bath and the reaction mixture competely solified after 12 h at room temperature into a red solid. The product mixture was quenched by opening the tube to air and the addition of chloroform (~2 mL). Addition of chloroform (5 mL) and prolonged stirring led to the formation of a red suspension. The insoluble polymer fraction was separated by filtration and washed with chloroform. Yield: 29.2 mg (71%). The filtrate was precipitated with methanol and collected by filtration. After washing with hexanes and methanol, the soluble polymer fraction was dried in vacuo and obtained as a red solid. Yield: 9.1 mg (22%). Soluble polymer fraction of poly(3): 1H NMR (400 MHz, CDCl3, 25 C): 6.92 (s, Ar), 6.51 (d, J = 16.4 Hz, CH=CH), 5.91 (br. s, =CH2), 3.97 (m, CCH2), 3.32 (s, CCH), 3.31 (s, CCH), 1.8-0.8 (m, CH2 and CH3). Integration revealed HH:HT = 92:8 and Pn = 16. 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 153.6 (CO), 151.0

238

The organolanthanide-catalyzed polymerization of diynes (CO), 136.0 (CCH=CH), 126.7 (CCC), 118.5-116.7 (CCH=CH), 101.6 (CC), 95.6 (CC), 89.0 (CCH), 82.2-82.0 (CCH), 31.8 (CH2), 29.7 (CH2), 29.2 (CH2), 25.9 (CH2), 22.9 (CH2), 14.1 (CH3). Only characteristic resonances are reported due to the complexity of the spectrum. IR (neat, [cm-1]): 3286 (s, C-H stretching, C-H), 2941 (m, C-H stretching, aromatic =C-H), 2854 (m), 2104 (m, C C stretching), 1496 (s), 1389 (s), 1211 (s), 1029 (s), 860 (m). Representative polymerization reaction of diyne catalyzed by Cp*2LaCH(SiMe3)2. at relatively low monomer concentration. To a Schlenk vessel (50 mL) containing a stirred solution of 2,5-diethynyl-3-nhexylthiophene (96.8 mg, 447 mol) in toluene (2.0 mL) was added 100.0 L (4.00 mol) of a stock solution of Cp*2LaCH(SiMe3)2 (40.0 mM) in toluene. The reaction mixture was heated under stirring to 50 C and the dark yellow solution gradually darkened. After 1.5 h, the reaction mixture solidified completely and the product mixture was quenched by opening to air and the addition of chloroform (~10 mL). Addition of more chloroform (10 mL) led to the formation of a red suspension. The insoluble polymer fraction was separated by filtration and washed with chloroform to afford a dark red material that turned shiny green upon drying in vacuo. Yield: 51.3 mg (53%). The filtrate was precipitated with methanol and collected by filtration. After washing with hexanes and methanol, the soluble polymer fraction was dried in vacuo and obtained as a dark red solid. Yield: 31.5 mg (33%). Soluble polymer fraction of poly(4): 1H NMR (500 MHz, CDCl3, 25 C): 6.94 (br. s, Ar), 6.78 (br. s, CH), 6.13 (m, CH=CH), 3.47 (s, CCH), 3.45 (s, CCH), 2.62 (m, CCH2), 2.56 (m, CCH2), 1.58 (m, CH2), 1.30 (m, CH2), 0.88 (m, CH3). Integration revealed HH:HT = 100:0 and Pn = 200. 13C-{1H} NMR (75 MHz, CDCl3, 25 C): 148.7 (CCH2), 147.9 (CCH2), 141.4 (CCH=CH), 123.3 (CCC), 107.8-106.1 (CCH=CH), 97.1 (CC), 94.1 (CC), 89.0 (CCH), 31.6 (CH2), 20.8 (CH2), 30.0 (CH2), 28.9 (CH2), 22.6 (CH2), 14.2 (CH3). Only characteristic resonances are reported due to the complexity of the spectrum. IR (neat, [cm-1]): 2921 (m, C-H stretching, aromatic =C-H), 2851 (m), 2167 (m, C C stretching), 1464 (m), 1259 (s), 1019 (s), 807 (m).

6.8.
1

References and notes

(a) Skotheim, T. A.; Elsenbaumer, R. L.; Reynolds, J. R. Handbook of Conducting Polymers, Dekker: New York, 1997. (b) Brdas, J. L.; Silbey, R. J. Conjugated Polymers, Kluwer Academic Publishers: Dordrecht, 1991. (c) Salaneck, W. R.; Clark, D. T.; Samuelsen, E. J. Science and Application of Conducting Polymers, Adam Hilger: Bristol, 1991. (d) Conjugated Polymeric Materials: Opportunities in Electronics, Optoelectronics, and Molecular Electronics, Brdas, J. L., Chance, R. R., Eds., Kluwer Academic Press: Dordrecht, 1990. (e) Chandrasekhar, P. Conducting Polymers, Fundamentals and Applications. A Practical Approach, Kluwer Academic Publishers: Norwell MA, 1999. (f) Pron, A.; Rannou, P. Prog. Polym. Sci. 2002, 27, 135. Shirakawa, H.; Louis, E. J.; MacDiarmid, A. G.; Chiang, C. K.; Heeger, A. J. J. Chem. Soc., Chem. Commun. 1977, 579. (a) Meier, H., Angew. Chem. Int. Ed. Engl. 1992, 31, 1399. (b) Hide, F.; Diaz-Garcia, M. A.; Schwartz, B. J.; Heeger, A. J. Acc.Chem. Res. 1997, 30, 430. (c) Lemmer, U.; Haugeneder, A.; Kallinger, C.; Feldman, J. in Semiconducting Polymers. Chemistry, Physics and Engineering; Hadziioannou, G.; van Hutten, P. F., Eds.; Wiley-VCH, Weinheim, 2000, p. 309. (a) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew. Chem. Int. Ed. 1998, 37, 403. (b) Segura, J. L. Acta Polym. 1998, 49, 319. (c) Mitschke, U.; Buerle, P. J. Mater. Chem., 2000, 10, 1471. (d) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.; Marks, R. N.; Taliani, C.; Bradley, D. D. C.; Dos Santos, D. A.; Brdas, J. L.; Lgdlund, M.; Salaneck, W. R. Nature 1999, 397, 121. (e) Shim, H.K.; Jin, J.-I. Adv. Polym. Sci. 2002, 158, 193. (f) Akcelrud, L. Prog. Polym. Sci. 2003, 28, 875. (g) Bernius, M. T.; Inbasekaran, M.; OBrien, J.; Wu, W. Adv. Mater. 2000, 12, 1737. Das-Gupta, D. K. in Introduction to Molecular Electronics; Petty, M. C.; Bryce, M. R.; Bloor, D., Eds.; Edward Arnold, London, 1995, p. 47. (a) Bleier, H. in Organic Materials for Photonics; Zerbi, G., Ed.; Elsevier, Amsterdam, 1993, p. 77. (b) Loutfy, R. O.; Hor, A.-M.; Hsiao, C.-K.; Baranyi, G.; Kazmaier, P. Pure. Appl. Chem. 1988, 60, 1047. (c) Perlstein, J. H.; Borsenberger, P. M. in Extended Linear Chain Compounds, Miller, J. S., Ed.; Plenum, New York, 1982, p. 339.

2 3

5 6

239

Chapter 6

10

11 12

13 14 15

16 17 18

19

20

(a) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science, 1995, 270, 1789. (b) Brabec, C. J.; Sariciftci, N. S. in Semiconducting Polymers. Chemistry, Physics and Engineering, Hadziioannou, G.;van Hutten, P. F., Eds.; Wiley-VCH, Weinheim, 2000, p. 515. (a) Feringa, B. L.; Jager, W. G.; de Lange, B. Tetrahedron 1993, 49, 8267. (b) Werumeus Buning, B. H. in Organic Materials for Photonics; Zerbi, G., Ed.; Elsevier, Amsterdam, 1993, 367. (c) Kmpf, G. Ber. Bunsenges. Phys. Chem. 1985, 89, 1179. (a) Lsch, K. Macromol. Symp. 1995, 100, 65. (b) Lehn, J.-M. Supramolecular Chemistry. Concepts and Perspectives, VCH, Weinheim, 1995, p. 124. (c) Martin, P. J. in Introduction to Molecular Electronics; Petty, M. C.; Bryce, M. R.; Bloor, D., Eds., Edward Arnold, London, 1995, p. 112. (d) Kim, J.-J.; Lee, E.-H. Mol. Cryst. Liq. Cryst. 1993, 227, 71. (e) Photochromism, Molecules and Systems, Drr, H.; Bouas-Laurent, H., Eds.;, Elsevier, Amsterdam, 1990. (f) Drr, H. Angew. Chem. Int. Ed. Engl. 1989, 28, 413. (g) Photochromism, Brown, G. H., Ed., Wiley-Interscience, New York, 1971. (a) Heeger, A. J.; Long, J., Jr. Optics & Photonics News 1996, 7, 24. (b) Nalwa, H. S. in Nonlinear Optics of Organic Molecules and Polymers; Nalwa, H. S.; Miyata, S., Eds., CRC, New York, 1997, p. 611. (c) Marder, S. R.; Kippelen, B.; Jen, A. K.-Y.; Peyghambarian, N., Nature 1997, 388, 845. Burroughes, J. H.; Bradey, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.; Friend, R. H.; Burns, P. L.; Holmes, A. Nature 1990, 347, 539. (a) Wegner, G. Z .Naturforsch. 1969, 24B, 824. (b) Enkelmann, V. Adv. Polym. Sci. 1984, 63, 91. (c) Polydiactylenes, Cantow, H.-J., Ed.; Springer, Stuttgart, 1984. (d) Polydiacetylenes, Bloor, D.; Chance, R. R., Eds.; Martinus Nijhoff, Dordrecht, 1985. (e) Wegner, G.; Schott, M. in Nonlinear Optical Properties of Organic Molecules and Crystals, Chemla, D. S.; Zyss, J., Eds., Academic Press, New York, 1987, Vol. 2, p. 3. (f) Ogawa, T. Prog. Polym. Sci. 1995, 20, 943. (g) Sarkar, A.; Okada, S.; Matsuzawa, H.; Matsuda, H.; Nakanishi, H. J. Mater. Chem. 2000, 10, 819. (h) Zuilhof, H.; Barentsen, H. M.; Van Dijk, M.; Sudhlter, E. J. R.; Hoofman, R. J. O. M.; Siebbeles, L. D. A.; De Haas, M. P.; Warman, J. M. In Supramolecular Photosensitive and Electroactive Materials, Nalwa, H. S., Ed.; Academic Press: Los Angeles, 2001, p. 339. (a) Scherf, U. Top. Curr. Chem. 1999, 201, 163. (b) Schlter, A. D.; Wegner, G. Acta Polym. 1993, 44, 59. Schlter, A. D. Adv. Mater. 1991, 3, 282. (a) Gorman, C. B.; Ginsburg, E. J.; Grubbs, R. H. J. Am. Chem. Soc. 1993, 115, 1397. (b) Sailor, M. J.; Ginsburg, E. J.; Gorman, C. B.; Kumar, A.; Grubbs, R. H.; Lewis, N. S. Science 1990, 249, 1146. (a) Bumm, L. A.; Arnold, J. J.; Cygan, M. T.; Dunbar, T. D.; Burgin, T. P.; Jones, L.; Allara, D. L.; Tour, J. M.; Weiss, P. S. Science 1996, 271, 1705. (b) Samori, P.; Francke, V.; Mllen, K.; Rabe, J. P. Chem. Eur. J. 1999, 5, 2312. (c) Samori, P.; Sikharulidze, I.; Francke, V.; Mllen, K.; Rabe, J. P. Nanotechnology 1999, 10, 77. (d) Samori, P.; Francke, V.; Mllen, K.; Rabe, J. P. Thin Solid Films 1998, 336, 13. (e) Samori, P.; Francke, V.; Mangel, T.; Mllen, K.; Rabe, J. P. Opt. Mater. 1998, 9, 390. (f) Mllen, K.; Rabe, J. P. Ann. N.Y. Acad. Sci. 1998, 852, 205. Ng, P. K.; Gong, X.; Wong, W. T.; Chan, W. K. Macromol. Rapid Commun. 1997, 18, 1009. For reviews about PAEs, see: (a) Giesa, R. J. M. S.-Rev. Macromol. Chem. Phys. 1996, 36, 631. (b) Bunz, U. H. F. Chem. Rev. 2000, 100, 1605. (a) Roncali, J. Chem. Rev. 1992, 92, 711. (b) McCullough, R. D. Adv. Mater. 1998, 10, 93. (c) Handbook of Oligo- and Polythiophenes, Fichou, D., Ed.; Wiley-VCH: Weinheim, 1999. (d) Buerle, P. In Electronic Materials: The Oligomer Approach; Mllen, K., Wegner, Gerhard, Eds.; Wiley-VCH: Weinheim, 1998; Chapter 2. (e) Gustafsson, G., Ingans, O., Salaneck, W. R., Laakso, J., Loponen, M., Taka, T., sterholm, J.-E., Stubb, H., Hjertberg, T. In Conjugated Polymers; Brdas, J. L., Silbey, R., Eds.; Kluwer: Dordrecht, 1991, p. 315. (f) Chan, H. S. O.; Choon Ng, S. Prog. Polym. Sci. 1998, 23, 1167. (g) Schopf, G.; Kossmehl, G. Adv. Polym. Sci. 1997, 129, 3. (a) Blohm, M. L.; Pickett, J. E.; Van Dort, P. C. Macromolecules 1993, 26, 2704. (b) Cheng, H.; Elsenbaumer, R. L. J. Chem. Soc., Chem. Commun. 1995, 703. (c) Fu, Y.; Cheng, H.; Elsenbaumer, R. L. J. Chem. Soc., Chem. Commun. 1997, 9, 1720. (d) Reynolds, J. R.; Sotzing, G. A. J. Chem. Soc., Chem. Commun. 1995, 703. (a) Yamamoto, T.; Yamada, W.; Takagi, M.; Kizu, K.; Maruyama, T.; Ooba, N.; Tomaru, S.; Kurihara, T.; Kaino, T.; Kuboto, K. Macromolecules 1994, 27, 6620. (b) Swanson, L. S.; Lane, P. A.; Shinar, J.; Pang, Y.; Barton, T. J. Synth. Met. 1993, 55-57, 293. (c) Ooba, N.; Tomaru, S.; Kurihara, T.; Kaino, T.; Yamada, W.; Takagi, M.; Yamamoto, T. Jpn. J. Appl. Phys. 1995, 34, 3139.

240

The organolanthanide-catalyzed polymerization of diynes

21

22

23 24

25 26 27

28

29

30

31 32

33

34

35

(a) G. Horowitz, D. Fichou, X. Peng, Z. Xu, F. Garnier, Solid State Commun. 1989, 72 ,38. (b) G. Horowitz, X. Peng, D. Fichou, F. Garnier, J. Mol. Electron 1991, 7, 85. (c) A. Dodabalapour, L. Torsi, H.E. Katz, Science 1995, 268, 270. (d) A.J. Levinger, Z. Bao, A. Dodabalapour, A.J. Levinger, Appl. Phys. Lett. 1996, 69, 4110. (e) C.D. Dimitrakopolous, P. Malenfant, Adv. Mater. 2002, 14, 99. (f) H. Sirringhaus, N. Tessler, R.H. Friend, Science 1998, 280, 1741. (a) J. Pei, L. Yu, W. Huang, A.J. Heeger, Macromolecules 2000, 33, 462. (b) A. Kraft, A.C. Grimsdale, A.B. Holmes, Angew. Chem., Int. Ed. 1998, 37, 102. (c) P. Barta, W.R. Salaneck, M. Zagorska, A. Pron, S. Niziol, Adv. Mater. Opt. Electron. 1996, 6, 406. (a) C. Videlot, A. El Kassmi, D. Fichou, Sol. Energy Mater. Sol. Cells 2000, 63, 69. (b) N. Noma, T. Tsuzuki, Y. Shirota, Adv. Mater. 1997, 7, 647. (a) Rusanov, A. L.; Khotina, I. A.; Begretov, M. M. Russ. Chem. Rev. 1997, 66, 1053. (b) Babudri, F.; Farinola, G. M.; Naso, F. J. Mater. Chem. 2004, 14, 11. (c) Cheng, Y.-J.; Luh, T.-Y. J. Organomet. Chem. 2004, 689, 4137. Percec, V. Polym. Bull., 1983, 10, 1. Kishimoto, Y.; Eckerle, P.; Miyatake, T.; Kainosho, M.; Ono, A.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc. 1999, 121, 12035. For examples, see: (a) Wetzel, H. P.; Mllen, K. Makromol. Chem., 1990, 191, 2837. (b) Martelock, H., Greiner, A.; Heitz, W. Makromol. Chem. 1991, 192, 967. (c) Klingelhfer, S.; Schellenberg, C.; Pommerehne, J.; Bssler, H.; Greiner, G.; Heitz, W. Macromol. Chem. Phys. 1997, 198, 1511. (d) Bao, Z.; Cheng, Y.; Cai, R.; Yu, L. Macromolecules 1993, 26, 5281. (e) Bao, Z.; Amundson, K. R.; Lovinger, A. J. Macromolecules 1998, 31, 8647. (f) Mikroyannidis, J. A. Macromolecules 2002, 35, 9289. (g) Wang, Q.; Yu, L. J. Am. Chem. Soc. 2000, 122, 11806. (h) You, W.; Wang, L.; Wang, Q.; Yu, L. Macromolecules 2002, 35, 4636. For examples, see: (a) Goldfinger, M. B.; Swager, T. M. J. Am. Chem. Soc. 1994, 116, 7895. (b) Kaeriyama, K.; Tsukahara, Y.; Negoro, S.; Tanigaki, N.; Masuda, H. Synth. Met. 1997, 84, 263. (c) Izumi, A.; Teraguchi, M.; Nomura, R.; Masuda, T. Macromolecules 2000, 33, 5347. (d) Chen, X.; Liao, J.-L.; Liang, Y.; Ahmed, M. O.; Tseng, H.-E.; Chen, S.-A. J. Am. Chem. Soc. 2003, 125, 636. For examples, see (a) Bao, Z.; Chan, W. K.; Yu, L. J. Am. Chem. Soc. 1995, 117, 12426. (b) Zhang, Q. T.; Tour, J. M. J. Am. Chem. Soc. 1998, 120, 5355. (c) Farina, V. Pure Appl. Chem. 1996, 68, 73. (d) Meng, H.; Tucker, D.; Chaffins, S.; Chen, Y.; Helgeson, R.; Dunn, B.; Wudl, F. Adv. Mater. 2003, 15, 146. (e) Galarini, R.; Musco, A.; Pontellini, R.; Bolognesi, A.; Destri, S.; Catellani, M.; Mascherpa, M.; Zhuo, G. J. Chem. Soc., Chem. Commun. 1991, 364. For examples, see: (a) Yamamoto, T.; Hayashi, Y.; Yamamoto, Y. Bull. Chem. Soc. Jpn., 1978, 51, 2091. (b) Rehahn, M.; Schlter, A.-D.; Wegner, G.; Feast, W. J. Polymer, 1989, 30, 1054. (c) Kobayashi, M.; Chen, J.; Chung, T. C.; Moraes, F.; Heeger, A. J.; Wudl, F. Synth. Met., 1984, 9, 77. (d) McCullough, R. D.; Lowe, R. D. J. Chem. Soc., Chem. Commun., 1992, 70. For examples, see: (a) Chen, T.-A.; Wu, X.; Rieke, R. D. J. Am. Chem. Soc., 1995, 117, 233. For examples, see: (a) Yamamoto, T.; Morita, A.; Miyazaki, Y.; Maruyama, T.; Wakayama, H.; Zhou, Z. H.; Nakamura, Y.; Kanbara, T.; Sasaki, S.; Kubota, K. Macromolecules, 1992, 25, 1214. (b) An, B.-Y.; Kim, Y.-H.; Shin, D.-C.; Park, S. Y.; Yu, H.-S.; Kwon, S.-K. Macromolecules, 2001, 34, 3993. (c) Ego, C.; Marsitzky, D.; Becker, S.; Zhang, J.; Grimsdale, A. C.; Mllen, K.; MacKenzie, J. D.; Silva, C.; Friend, R. J. Am. Chem. Soc., 2003, 125, 437. For a reviews, see: (d) Yamamoto, T. J. Organomet. Chem. 2002, 653, 195. (e) Yamamoto, Y. Macromol. Rapid Commun. 2002, 23, 583. For examples, see: (a) Thorn-Csny, E.; Kraxner, P.; Strachota, A. Macromol. Rapid Commun., 1998, 19, 223. (b) Thorn-Csny, E.; Kraxner, P. Macromol. Chem. Phys., 1997, 198, 3827. (c) Sclick, H.; Stelzer, F.; Tasch, S.; Leising, G. J. Mol. Catal. A: Chem., 2000, 160, 71. For exampes, see: (a) Weiss, K.; Michel, A.; Auth, E.; Bunz, U. H. F.; Mangel, T.; Mllen, K. Angew. Chem., 1997, 36, 506. (b) Mortreux, A.; Blanchard, M. J. Chem. Soc., Chem. Commun., 1974, 786. (c) Mortreux, A.; Dy, N.; Blanchard, M. J. Mol. Catal.,1976, 1, 101. (d) Kloppenburg, L.; Jones, D.; Bunz, U. H. F. Macromolecules, 1999, 32, 4194. (e) Wilson, J. N.; Steffen, W.; McKenzie, T. G.; Lieser, G.; Oda, M.; Neher, D.; Bunz, U. H. F. J. Am. Chem. Soc., 2002, 124, 6830. (f) Brizius, G.; Pschirer, N. G.; Steffen, W.; Stitzer, K.; zur Loye, H.-C.; Bunz, U. H. F. J. Am. Chem. Soc., 2000, 122, 12453. For examples, see: (a) Edwards, J. H.; Feast, W. J. Polym. Commun, 1980, 21, 595. (b) Edwards, J. H.; Feast, W. J.; Bott, D. C. Polymer, 1984, 25, 395. (c) Conticello, V. P.; Gin, D. L.; Grubbs, R. H.

241

Chapter 6

36 37 38 39

40

41

42 43

44

45

46

J. Am. Chem. Soc., 1992, 114, 9708. (d) Pu, L.; Wagaman, M. W.; Grubbs, R. H. Macromolecules, 1996, 29, 1138; Wagaman, M. W.; Grubbs, R. H. Macromolecules, 1997, 30, 3978. (e) Klavetter, F. L.; Grubbs, R. H. J. Am. Chem. Soc, 1988, 110, 7807. (f) Gorman, C. B.; Ginsburg, E. J.; Grubbs, R. H. J. Am. Chem. Soc., 1993, 115, 1397. Kane, J. J.; Gao, F.; Reinhardt, B. A.; Evers, R. C.ACS Polym. Prepr. 1992, 33, 1064. Venkatesan, D.; Yoneda, M.; Ueda, M. React. Funct. Polym. 1996, 30, 341. Choi, C.-K.; Tomita, I.; Endo, T. Macromolecules 2000, 33, 1487. For examples of catalysts with high activity and selectivity for cis-1,4-diphenylbut-1-en-3-yne, see: (a) Ti: Akita, H.; Yasuda, H.; Nakamura, A. Bull. Chem. Soc. Jpn 1984, 57, 480. (b) Zr: Horton, A. D. J. Chem. Soc., Chem. Commun. 1992, 185. (c) Ti: Varga, V.; Petrusov, L.; Cejka, J.; Mach, K. J. Organomet. Chem. 1996, 509, 235. (d) Al: Dash, A. K.; Eisen, M. Org. Lett. 2000, 2, 737. For examples of catalysts have a high activity and selectivity for trans-1,4-diphenylbut-1-en-3-yne, see: (e) Ru: Yi, C.S.; Liu, N. Synlett 1999, 281. (f) Ru: Baratta, W.; Herrmann, W. A.; Rigo, P.; Schwartz, J. J. Organomet. Chem. 2000, 593-594, 489. (g) Rh: Werner, H.; Schwab, P.; Heinemann, A. Steinert, P. J. Organomet. Chem. 1995, 496, 207. (h) Pd: Yang, C.; Nolan, S. P. J. Org. Chem. 2003, 67, 591. In the course of this study three independent similar studies appeared in literature and one patent, see: (a) Katayama, H.; Nakayama, M.; Nakano, T.; Wada, C.; Akamatsu, K.; Ozawa, F. Macromolecules 2004, 37, 13. (b) Nishiura, M.; Hou, Z. J. Mol. Cat. A: Chem. 2004, 213, 101. (c) Nishiura, M.; Tanikawa, M.; Hoshino, M.; Miyamoto, T.; Hou, Z. Kidorui 2003, 42, 54. (d) Ueda, M.; Tomita, I. Polym. Bull. 2004, 51, 359. (e) Nishiura, M.; Ho, Z.-M. JP 2004263072, 2004. (a) Katayama, H.; Nakayama, M.; Nakano, T.; Wada, C.; Akamatsu, K.; Ozawa, F. Macromolecules 2004, 37, 13. (b) Nishiura, M.; Hou, Z. J. Mol. Cat. A: Chem. 2004, 213, 101. (c) Nishiura, M.; Tanikawa, M.; Hoshino, M.; Miyamoto, T.; Hou, Z. Kidorui 2003, 42, 54. (d) Ueda, M.; Tomita, I. Polym. Bull. 2004, 51, 359. (e) Nishiura, M.; Ho, Z.-M. JP 2004263072, 2004. For a review, see: Siemsen, P.; Livingston, R.C.; Diederich, F. Angew.Chem. Int. Ed. 2000, 39, 2632. (a) Rutherford, D. R.; Stille, J. K.; Elliott, C. M.; Reichert, V. R. Macromolecules 1992, 25, 2294. (b) Newkirk, A. E.; Hay, A. S.; McDonald, R. S. J. Polym. Sci., Part A: Polym. Chem. 1964, 2, 2217. (c) Bunten, K. A.; Kakkar, A. K. Macromolecules 1996, 29, 2885. (d) D'Ilario, L.; Eattorre, A.; Ortaggi, G.; Sleiter, G. J. Mater. Sci. 1995, 30, 4273. (e) Kwock, E. W.; Baird, T., Jr.; Miller, T. M. Macromolecules 1993, 26, 2935. (f) Miller, T. M.; Kwock, E. W.; Baird, T., Jr.; Hale, A. Chem. Mater. 1994, 6, 1569. (g) Ogawa, T. Prog. Polym. Sci. 1995, 20, 943. (a) Hay, A. S. J. Org. Chem. 1960, 25, 1275. (b) Hay, A. S., U.S. Pat. 3332916, 1967. (c) Hay, A. S., U.S. Pat. 3300456, 1967. (d) Hay, A. S., U.S. Pat. 3519611, 1970. (e) White, D. M.; Klopfer, H. J., U.S. Pat. 3748305, 1973. (f) Chalk, A. J.; Gilbert, A. R., U.S. Pat. 4108942, 1978. (g) White, D. M., U.S. Pat. 3816374, 1974. (h) White, D. M., U.S. Pat. 4020265, 1977. (i) Dawson, D. J. In Reactive Oligomers; American Chemical Society: Washington, DC, 1985; pp. 63-79. (j) Economy, J.; Flandera, M. A.; Lin, C.-Y., U.S. Pat. 4258079, 1981. (k) Economy, J.; Flandera, M. A., U.S. Pat. 4273906, 1981. (l) Cessna, L. C., Jr., U.S. Pat. 3882073, 1975. (m) Jabloner, H., U.S. Pat. 4070333, 1978. (n) Jabloner, H., U.S. Pat. 4097460, 1978. (o) Korshak, V. V.; Volpin, M. E.; Sergeev, V. A.; Shitikov, V. K.; Kolomnikov, I. S., U.S. Pat. 3705131, 1972. (p) Korshak, V. V.; Gribova, I. A.; Krasnov, A. P.; Sergeev, V. A.; Shitikov, V. K.; Elerdashvili, G. V., U.S. Pat. 375982, 1973. (q) Hergenrother, P. M. In Reactive Oligomers; American Chemical Society: Washington, D.C., 1986; pp. 1-16. (r) Neenan, T. X.; Whitesides, G. M. J. Org. Chem. 1988, 53, 2489. (s) Neenan, T. X.; Callstrom, M. R.; Scarmoutzos, L. M.; Stewart, K. R.; Whitesides, G. M. Macromolecules 1988, 21, 3525. (t) Callstrom, M. R.; Neenan, T. X.; McCreery, R. L.; Alsmeyer, D. C. J. Am. Chem. Soc. 1990, 112, 4954. (a) Heck, R. F. Palladium Reagents in Organic Synthesis, Academic Press: London, 1985. (b) Modern Cross-Coupling Reactions, Stang, P. J., Diederich, F., Eds.;VCH: Weinheim, 1997. (c) Brandsma, L., Vasilevsky, S. F., Verkruijsse, H. D. In Applications of Transition Metal Catalysts in Organic Synthesis; Springer: Berlin, 1998; p 179. (d) Sonogashira,, K. in Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press, Oxford; Vol. 3, p. 521. (e) MetalCatalyzed Cross-Coupling Reactions, Stang, P. J.; Diederich, F, Eds.; Wiley-VCH, Weinheim, 1998, p. 203. For recent organolanthanide reviews, see: (a) Aspinall, H. C. Chem. Rev. 2002, 102, 1807-1850. (b) Edelmann, F. T.; Freckmann, D. M. M.; Schumann, H, Chem. Rev. 2002, 102, 1851-1896. (c) Arndt,

242

The organolanthanide-catalyzed polymerization of diynes

47

48

49

50

51 52

53 54

55 56 57

S.; Okuda, J. Chem. Rev. 2002, 102, 1953-1976 (d) Shibasaki, M.; Yoshikawa, N. Chem. Rev. 2002, 102, 2187-2210. (e) Inanaga, J.; Furuno, H.; Hayano, T. Chem. Rev. 2002, 102, 2211-2226. (f) Molander, G. A. Chemtracts: Org. Chem. 1998, 18, 237-263. (g) Edelmann, F. T. Top. Curr. Chem. 1996, 179, 247-276. (h) Edelmann, F. T. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1995; Vol. 4, Chapter 2. (i) Schumann, H.; Meese-Marktscheffel, J. A.; Esser, L. Chem. Rev. 1995, 95, 865-986. (j) Schaverien, C. J. Adv. Organomet. Chem. 1994, 36, 283-362. (k) Evans, W. J. Adv. Organomet. Chem. 1985, 24, 131-177. (l) Marks, T. J.; Ernst, R. D. In Comprehensive Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon Press: Oxford, U.K., 1982; Chapter 21. For examples, see: (a) Jeske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8111. (b) Haar, C. M.; Stern, C. L.; Marks, T. J. Organometallics 1996, 15, 1765. (c) Roesky, P. W.; Stern, C. L.; Marks, T. J. Organometallics 1997, 16, 4705. (d) Molander, G. A.; Dowdy, E. D.; Schumann, H. J. Org. Chem. 1998, 63, 3386. (e) Kretschmer, W. P.; Troyanov, S. I.; Meetsma, A.; Hessen, B.; Teuben, J. H. Organometallics 1998, 17, 284. (f) Molander, G. A.; Dowdy, E. D.; Schumann, H. J. Org. Chem. 1999, 64, 9697. The permethyllanthanidocene hydrides [Cp*2Ln(-H)]2 are known to affect C-O cleavage of ethers, see: (a) Watson, P. L. J. Chem. Soc., Chem. Commun., 1983, 276. (b) Booij, M. Ph. D. thesis, University of Groningen, 1989. (c) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. Organometallics 1991, 10, 134. (d) Deelman, B.-J., Ph. D. thesis, University of Groningen, 1994. (e) Deelman, B.-J.; Booij, M.; Meetsma, A.; Teuben, J. H.; Kooijman, H.; Spek, A. L. Organometallics 1995, 14, 2306. For examples, see: (a) Finke, R. G.; Keenan, S. R.; Schiraldi, D. A.; Watson, P. L. Organometallics 1986, 5, 598. (b) Finke, R. G.; Keenan, S. R.; Schiraldi, D. A.; Watson, P. L. Organometallics 1987, 6, 1356. (c) Finke, R. G.; Keenan, S. R.; Watson, P. L. Organometallics 1989, 8, 263. (d) Ref. 14a. (e) Booij, M. ; Deelman, B.-J. ; Duchateau, R. ; Postma, D. S.; Meetsma, A. ; Teuben, J. H. Organometallics 1993, 12, 3531. (f) Qian, C.; Zhu, C.; Zhu, D. Appl. Organomet. Chem. 1995, 9, 457. (a) Whitall, I. R.; Cifuentes, M. P.; Humphrey, M. G.; Luther-Davies, B.; Samoc, M.; Houbrechts, S.; Persoons, A.; Heath, G. A.; Hockless, D. C. R. J.Organomet.Chem. 1997, 549, 127. (b) Bodwell, G. J.; Miller, D. O.; Vermeij, R. J. Org.Lett. 2001, 3, 2093. (c) Takahashi, S.; Kuroyama, Y.; Sonogashira, K.; Hagihara, N. Synthesis 1980, 627. Brandsma, L. Preparative Acetylenic Chemistry, Elsevier: Amsterdam, 1988 (a) Ried, W. Angew. Chem. 1964, 76, 933. (b) Ried, W. Angew. Chem. 1964, 76, 973. (c) Ried, W. In Neuere Methoden der prparativen organischen Chemie; Verlag Chemie: Weinheim, 1966; Vol. 4. (d) Ruthledge, T. F. Acetylenic Compounds. Preparation and Substitution Reactions, Reinhold: New York, 1968. (e) Rohde, O.; Wegner, G. Makromol. Chem. 1978, 179, 1999. (f) Bohlmann, F. In Chemistry of Acetylene; Viehe, H. G., Ed.; Marcel Dekker: New York, 1969; Chapter. 14. (g) Hart, H.; Shamouillian, S.; Takehira, Y. J. Org. Chem. 1981, 46, 4427. (h) Hagihara, M.; Yamamoto, Y.; Takahashi, S.; Hayashi, K. Int. J. Radiat. Appl Instrum., Part C 1986, 28, 165. (i) Bleicher, L.; Cosford, N. D. P. Synlett 1995, 1115. (j) Khan, M. S.; Al-Suti, M. K.; Al-Mandhary, M. R. A.; Ahrens, B.; Bjernemose, J. K.; Mahon, M. F.; Male, L.; Raithby, P. R.; Friend, R. H.; Khler, A.; Wilson, J. S. J. Chem. Soc., Dalton Trans. 2003, 65. Mangel, T.; Eberhardt, A.; Scherf, U.; Bunz, U. H. F.; Mllen, K. Macromol. Rapid Commun. 1995, 16, 571. (a) Scherf, U.; Mllen, K. Synthesis 1992, 23. (b) Bao, Z.; Chan, W.; Yu, L. Chem. Mater. 1993, 5, 2. (c) Li, H.; Powell, D. R.; Hayashi, R. K.; West, R. Macromolecules 1998, 31, 52. (d) Kloppenburg, L.; Jones, D.; Bunz, U. H. F. Macromolecules 1999, 32, 4194. (e) Zhang, W.; Moore, J. S. Macromolecules 2004, 37, 3973. (f) Kloppenburg, L.; Song, D.; Bunz, U. H. F. J. Am. Chem. Soc. 1998, 120, 7973. Similar observations are reported for the di-n-butyl analogue, see: Pelter, A.; Jones, D. E. J. Chem. Soc., Perkin Trans. 1 2000, 2289. It is remarkable that Mllen et al. who reported the preparation of this monomer as a colorless oil do not comment on its instability.53 (a) Castro, R.; Nixon, K. R.; Evanseck, J. D.; Kaifer, A. E. J. Org. Chem. 1996, 61, 7298. (b) Meier, H.; Ickenroth, D.; Stalmach, U.; Koynov, K.; Bahtiar, A.; Bubeck, C. Eur. J. Org. Chem. 2001, 4431.

243

Chapter 6

58 59

60

61

62 63

64

65

66

67 68

69 70

(c) Lowe, J. L.; Peak, D. A.; Watkins, T. I. J. Org. Chem. 1951, 3286. (d) Ruiz, J. P.; Dharia, J. R.; Reynolds, J. R. Macromolecules 1992, 25, 849. (a) Rutherford, D. R.; Stille, J. K. Macromolecules 1988, 21, 3530. (b) Callstrom, M. R.; Neenan, T. X.; Whitesides, G. M. Macromolecules 1988, 21, 3528. (c) Ref. 43a. (a) Barker, J. M.; Huddleston, P. R.; Wood, M. L. Synth. Commun. 1975, 5, 59. (b) Mao, H.; Xu, B.; Holdcroft, S. Macromolecules 1993, 26, 1163. (c) Antonelli, E.; Rosi, P.; Lo Sterzo, C.; Viola, E. J. Organomet. Chem. 1999, 578, 210. (a) Mao, H.; Xu, B.; Holdcroft, S. Macromolecules 1993, 26, 1163. (b) Li, W.; Maddux, T.; Yu, L. Macromolecules 1996, 29, 7334. (c) Chaloner, P. A.; Gunatunga, S. R.; Hitchcock, P. B. J. Chem. Soc., Perkin Trans. 2 1997, 1597. (a) Li, J.; Pang, Y. Macromolecules 1997, 30, 7487. (b) Williams, V. E.; Swager, T. M. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 4669. (c) Krmer, J.; Rios-Carrera, I.; Fuhrmann, G.; Musch, C.; Wunderlin, M.; Debaerdemaeker, T.; Mena-Osteritz, E.; Buerle, P. Angew. Chem. Int. Ed. 2000, 39, 3481. (d) Higuchi, H.; Ishikura, T.; Mori, K.; Takayama, Y.; Yamamoto, K.; Tani, K.; Miyabayashi, K.; Miyake, M. Bull. Chem. Soc. Jpn. 2001, 74, 889. (e) Altamura, P.; Giardina, G.; Lo Sterzo, C.; Vittoria Russo, M. Organometallics, 2001, 20, 4360. (a) Odian, G. Principles of Polymerization, Wiley: New York, 2004; 4rd ed. (b) Synthetic Methods in Step-Growth Polymers, Rogers, M. E.; Long, T. E. (Eds.), Wiley-Interscience: New Jersey, 2003. Polymer containing linkages that originate from the allenic trimerization reaction will be conjugated, but are anticipated to be unstable in analogy to the trimers (Chapter 4) and may reasonably give rise to other types of structural defects. Generation of structurally homogeneous, regioregular materials allows for efficient solid-state packing, a necessary criterion for optimizing electronic and photonic properties of such materials. For example, fully regioregular, head-to-tail coupled, poly(3-alkylthiophene)s have higher electrical conductivities, nonlinear optical responses, higher charge mobilities for field-effect transistors, and more pronounced chemical sensory responses than regiorandom/regio-irregular analogues, see: (a) Chen, T.-A.; Wu, X.; Rieke, R. D. J. Am. Chem. Soc. 1995, 117, 233. (b) McCullough, R. D.; Lowe, R. D. J. Chem. Soc., Chem. Commun. 1992, 70. (c) Xu, B.; Holdcroft, S. Macromolecules 1993, 26, 4457. (d) Bouman, M.; Meijer, E. W. Adv. Mater. 1995, 7, 385. (e) McCullough, R. D.; TristramNagle, S.; Wiliams, S. P.; Lowe, R. D.; Jayaraman, M. J. Am. Chem. Soc. 1993, 115, 4910. (f) McCullough, R. D.; Williams, S. P. J. Am. Chem. Soc. 1993, 115, 11608. (g) McCullough, R. D.; Lowe, R. D.; Jayaraman, M.; Anderson, D. L. J. Org. Chem. 1993, 58, 904. (h) McCullough, R. D.; Jayaraman, M. J. Chem. Soc., Chem. Commun. 1995, 135. (i) McCullough, R. D., Ewbank, P. C. In Handbook of Conducting Polymers; Skotheim, T. A., Reynolds, John R., Eds.; Marcel Dekker: New York, 1998; p 225. (j) Wu, X.; Chen, T.-A.; Rieke, R. D. Macromolecules 1995, 28, 2101. (a) Lin-Vien, D.; Colthup, N. B.; Fately, W. G.; Grasselli, J. G. The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules; Academic Press: London, 1991. (b) Silverstein, R. M.; Bassler, G. C. ; Morrill, T. C. Spectrometric Identification of Organic Compounds, Wiley: New York, 1981. (c) Siesler, H. W., Holland-Moritz, K. Infrared and Raman Spectroscopy of Polymers, Marcel Dekker: New York, 1980. Carbon-carbon triple bond stretching vibrations due to enyne units in polymers typically are in the range of 2130-2220 cm-1, see: (a) Wudl, F.; Bitler, S. P. iJ. Am. Chem. Soc. 1986, 108, 4685. (b) Carre, F.; Devylder, N.; Dutremez, S. G.; Guerin, C.; Henner, B. J. L.; Jolivet, A.; Tomberli, V.; Dahan, F. Organometallics 2003, 22, 2014. (c) Corriu, R. J. P.; Gerbier, P. Guerin, C.; Henner, B. J. L.; Jean, A.; Mutin, P. H. Organometallics 1992, 11, 2507. Typical examples of this approach include poly(thiophene)s, poly(arylene-vinylene)s and poly(aryleneethynylene)s.17b For examples, see: (a) Rehahn, M.; Schlter, A. D.; Wegner, G. Makromol. Chem. 1990, 191, 1991. (b) Vahlenkamp, T.; Wegner, G. Macromol. Chem. Phys. 1994, 195, 1933. (c) Witteler, H.; Wegner, G.; Schulze, M. Macrmol. Rapid Commun. 1993, 14, 471. Examples in which the degree of polymerization and the solubility of the polymer increased with side-chain elongation include poly(p-phenylene-ethynylene)s34d and poly(thiophene)s.1a The average number and the experimental error were obtained by integration of the same spectrum for at least three times.

244

The organolanthanide-catalyzed polymerization of diynes

71

72

73

74

75

76

77 78

For examples, see: (a) Anthony, J.; Boudon, C.; Diederich, F.; Gisselbrecht, J.-P.; Gralich, V.; Gross, M.; Hobi, M.; Seiler, P. Angew. Chem. Int. Ed. 1994, 33, 763. (b) Polhuis, M.; Hendrikx, C. C. J.; Zuilhof, H.; Sudhlter, E. J. R. Tetrahedron Lett. 2003, 44, 899. (c) Walker, J. A.; Bitler, S. P.; Wudl, F. J. Org. Chem. 1984, 49, 4733. (d) Boldi, A. M.; Anthony, J.; Knobler, C. B.; Diederich, F. Angew. Chem. Int. Ed. Engl. 1992, 31, 1240. For reviews, see: (a) te Nijenhuis, K. Adv. Polym. Sci. 1997, 130, 1. (b) Guenet, J.-M. Thermoreversible Gelation of Polymers and Biopolymers, Academic Press: London, 1992. For examples, see: (c) Semenov, A. N.; Rubinstein, M. Macromolecules 1998, 31, 1373. (d) Cotts, P. M.; Swager, T. M.; Zhou, Q. Macromolecules 1996, 29, 7323. (e) Malik, S.; Nandi, A. K. J. Phys. Chem. B. 2004, 108, 597. (f) Yue, S.; Berry, G. C.; McCullough, R. D. Macromolecules 1996, 29, 933. (g) Malik, S.; Jana, T.; Nandi, A. K. Macromolecules 2001, 34, 275. (h) Huang, W. Y.; Matsuoka, S.; Kwei, T. K.; Okamoto, Y. Macromolecules 2001, 34, 7166. (i) Perahia, D.; Traiphol, R.; Bunz, U. H. F. J. Chem. Phys. 2002, 117, 1827. (j) Chu, Q.; Pang, Y.; Ding, L.; Karasz, F. E. Macromolecules 2002, 35, 7569. (k) Perahia, D.; Jiao, X.; Traiphol, R. J. Polym. Sci., Part B: Polym. Phys. 2004, 42, 3165. (l) Ajayaghosh, A.; George, S. J.; Praveen, V. K. Angew. Chem. Int. Ed. 2003, 42, 332. (m) Hsieh, B. R.; Yu, Y.; VanLaeken, A. C.; Lee, H. Macromolecules 1997, 30, 8094. (n) Yin, C.; Yang, C.-Z. Synth. Met. 2001, 118, 75. (o) Nakaoki, T.; Tashiro, K.; Kobayashi, M. Macromolecules 2000, 33, 4299. (p) George, M.; Weiss, R. G. Chem. Mater. 2003, 15, 2879. (q) Grell, M.; Bradley, D. D. C.; Long, X.; Chamberlain, T.; Inbaskaran, M.; Woo, E. P.; Soliman Acta Polymer. 1998, 49, 439. The polymer with acetylenic end groups obtained by homopolycyclotrimerization of 1,8-nonadiyne became partially insoluble upon storage under ambient conditions, see: (a) Xu, K.; Peng, H.; Sun, Q.; Dong, Y.; Salhi, F.; Luo, J.; Chen, J.; Huang, Y.; Zhang, D.; Xu, Z.; Tang, B. Z. Macromolecules, 2002, 35, 5821. Development of (pre)polymers with triple bond functional ends (acetylene-terminated (pre)polymers) as thermally curable resins for industrial application has been an active area of research. For reviews, see: (a) Hergenrother, P. M. In Concise Encyclopedia of Polymer Science and Engineering; Kroschwitz, J. I., Ed.; Wiley: New York, 1990. (b) Sergeev, V. A.; Chernomordik, Yu. A.; Kurapov, A. S. Rus. Chem. Rev. 1984, 53, 307. For examples, see: (c) Gandon, S.; Mison, P.; Silion, B. Polymer 1997, 38, 1449. (d) McQuilkin, R. M.; Garratt, P. J.; Sondheimer, F. J. Am. Chem. Soc. 1970, 92, 6682. (e) Nakamura, K.; Ando, S.; Takeichi, T. Polymer 2001, 42, 4045. (f) Reghunadhan Nair, C. P.; Bindu, R. L.; Ninan, K. N. Polymer 2002, 43, 2609. (g) Rohde, O.; Wegner, G. Makromol. Chem. 1978, 179, 1999. (h) Badarau, C.; Wang, Z. Y. Macromolecules 2003, 36, 6959 and references therein. (a) Ranby, B.; Rabek, J. F. Photodegradation, Photo-oxidation and Photostabilization of Polymers, Wiley: London, 1000. (b) Handbook of Polymer Degradation, 2nd ed.; Hamid, H. S., Ed.; Marcel Dekker: New York, 2000. (c) Becker, H.; Spreitzer, H.; Ibrom, K.; Kreuder, W. Macromolecules 1999, 32, 4925. (d) Wan, W. C.; Antoniadis, H.; Choong, V. E.; Razafitrimo, H.; Gao, Y.; Feld, W. A.; Hsieh, B. R. Macromolecules 1997, 30, 6567. (e) Hay, A. S. J.Org.Chem. 1960, 637. (f) Hay, A. S. J. Polym. Sci., Part A: Polym. Chem. 1969, 7, 1625. (g) Choi, S.-K.; Gal, Y.-S.; Jin, S.-H.; Kim, H. K. Chem. Rev. 2000, 100, 1645. (h) Ruthledge, T. F. In Acetylenes and Allenes; Reinhold: New York, 1969; Chapter 5. (i) Ginsburg, E. J., Gorman, C. B., Grubbs, Robert H. In Modern Acetylene Chemistry; Stang, P. J., Diederich, Franois, Eds.; VCH: New York, 1995. For examples, see: (a) Kijima, M.; Tanimoto, A.; Oya, A.; Liang, T.-T.; Yamada, Y. Carbon 2001, 39, 297. (b) Kijima, M.; Tanimoto, H.; Shirakawa, H. Synth. Met. 2001, 119, 353. (c) Hay, A. S.; Bolon, D. A.; Leimer, K. R.; Clark, R. F. Polymer Lett. 1970, 8, 97. (d) Liang, R.-C.; Reiser, A. J. Polym. Sci.: Polym. Chem. Ed. 1987, 25, 451. (e) Nallicheri, R. A.; Rubner, M. F. Macromolecules 1991, 24, 517. (f) Stanford, J. L.; Young, R. J.; Day, R. Polymer 1991, 32, 1713. (g) Chance, R. R. Polym. Prepr. 1987, 28, 445. (h) Enkelmann, V. Adv. Polym. Sci. 1984, 63, 91. (i) Ref. 43b. (j) Ref. 44a,s. (k) Ref. 78c. The 2,5-dialkoxy-substituted PPEs are a special class of -conjugated polymers that have exhibited high photoluminescence, some electroluminescence and liquid crystalline properties.17b For examples, see: (a) Anthony, J.; Boudon, C.; Diederich, F.; Gisselbrecht, J.-P.; Gralich, V.; Gross, M.; Hobi, M.; Seiler, P. Angew. Chem. Int. Ed. 1994, 33, 763. (b) Ref. 44a. (c) Miller, T. M.; Kwock, E. W.; Baird, T., Jr.; Hale, A. Chem. Mater. 1994, 6, 1569. (d) Ref. 58b. (e) Ref. 44s. (f) Diederich, F.

245

Chapter 6

79

80 81

82

83

84

85 86 87

88

89

90

Chem. Commun. 2001, 219. (g) Wudl, F.; Bitler, S. P. J. Am. Chem. Soc. 1986, 108, 4685. (h) Lindsell, W. E.; Preston, P. N.; Tomb, P. J. J. Organomet. Chem. 1992, 439, 201. The use of a tert-butyl, methyl and phenyl group has been crucial in the preparation of isolable and characterizable oligo(yne)s, oligo(enyne)s and oligo(enediyne)s. Their main function seems to be the steric inhibition of polymerization at the end groups. Hydrogen-capped oligo(yne)s and oligo(enyne)s78 are normally stable only below room temperature in the dark. For examples of oligo(yne) chemistry, see: (a) Raphael, R. A. Acetylenic Compounds in Organic Synthesis, Academic Press: New York, 1955. (b) Lagow, R. J.; Kampa, J. J.; Wei, H.-C.; Battle, S. L.; Genge, J. W.; Laude, D. A.; Harper, C. J.; Bau, R.; Stevens, R. C.; Haw, J. F.; Munson, E. Science 1995, 267, 362. (c) Kijima, M.; Kinoshita, I.; Hattori, T.; Shirakawa, H. J. Mater. Chem. 1998, 8, 2165. (d) Diederich, F. In Modern Acetylene Chemistry; Stang, P. J., Diederich, F., Eds.; VCH: Weinheim, 1995; p 443. (e) Grsser, T.; Hirsch, A. Angew. Chem. Int. Ed. 1993, 32, 1340. (f) Schermann, G.; Grsser, T.; Hampel, F.; Hirsch, A. Chem. Eur. J. 1997, 3, 1105. (g) Jones, E. R. H.; Lee, H. H.; Whiting, M. C. J. Chem. Soc. 1960, 3483. (h) Bohlmann, F. Chem. Ber. 1953, 86, 657. (i) Johnson, T. R.; Walton, D. R. M. Tetrahedron 1972, 28, 5221. (j) Dembinski, R.; Bartik, T.; Bartik, B.; Jaeger, M.; Gladysz, J. A. J. Am. Chem. Soc. 2000, 122, 810. Vinylacetylene is readily polymerized by heat to form viscous oils and hard resinous solids, see: Nieuwland, J. A.; Calcott, W. S.; Downing, F. B.; Carter, S. J. Am. Chem. Soc. 1931, 53, 4197. The kinetic acidities of several acetylenes have been estimated based on hydrogen-exchange rates. For recent examples, see: (a) Kresge, A. J.; Powell, M. F. J. Org. Chem., 1986, 51, 819. (b) Kresge, A. J.; Pruszynski, P.; Stang, P. J.; Williamson, B. L. J. Org. Chem. 1991, 56, 4808. This difference in reactivity is likely to be mainly steric in origin, as the kinetic acidity of 1,4diethynylbenzene has been found 14 times higher than that of phenylacetylene, based on hydrogenexchange rates. See: Dessy, R. E.; Okuzumi, Y.; Chen, A. J. Am. Chem. Soc. 1962, 84, 2899. The estimates of the pKa values for trimethylsilylacetylene and phenylacetylene are 21.1 and 19.1, respectively, based on detritiation rates in aqueous solution, see: Kresge, A. J.; Pruszynski, P.; Stang, P. J.; Wiliamson, B. L. J. Org. Chem. Soc. 1991, 56, 4808. For reviews, see: (a) Chemistry of Acetylenes, Viehe, H. G. (Ed.), Marcel Dekker: New York, 1969. (b) Cyclic Polymers, Semlyen, J. A. (Ed.). Kluwer Academic Press: New York, 2000. (c) Ref. 42. For recent examples, see: (d) Zhou, Q.; Carroll, P. J.; Swager, T. M. J. Org. Chem. 1994, 59, 1294. (e) Rubin,Y.; Kahr, M.; Knobler, C. B.; Diederich, F.; Wilkins, C. L. J. Am. Chem. Soc. 1991, 113, 495. (f) Pak, J. J.; Weakley, T. J. R.; Haley, M. M. J. Am. Chem. Soc. 1999, 121, 8182. (a) Electronic Materials: The Oligomer Approach, Mllen, K.; Wegner, K. (Eds.), Wiley-VCH, Weinheim, 1998. (b) Martin, R. E.; Diederich, F. Angew. Chem. Int. Ed. 1999, 38, 1350. Yoshino, K.; Tada, K.; Onoda, M. Jpn. J. Appl. Phys. 1994, 33, L1785. For reviews, see: (a) Raeder, H. J.; Schrepp, W. Acta Polym. 1998, 49, 272. (b) Nielen, M. W. F. Mass Spectrom. Rev. 1999, 18, 309. (c) Hanton, S. D. Chem. Rev. 2001, 101, 527. (d) Pasch, H.; Shrepp, W. MALDI-TOF Mass Spectrometry of Synthetic Polymers, Springer: Berlin, 2003. For examples, see: (e) Weber, L.; Barlmeyer, M.; Quasdorff, J.-M.; Sievers, H.; Stammler, H.-G.; Neumann, B. Organometallics 1999, 18, 2497. (f) Lin-Gibson, S.; Brunner, L.; Vanderhart, D. L.; Bauer, B. J.; Fanconi, B. M.; Guttman, C. M.; Wallace, W. E. Macromolecules 2002, 35, 7149. (g) Trimpin, S.; Rouhanipour, A.; Az, R.; Rder, H. J.; Mllen, K. Rapid Commun. Mass Spectrom. 2001, 15, 1364. (h) Ji, H.; Sato, N.; Nakamura, Y.; Wan, Y.; Howell, A.; Thomas, Q. A.; Storey, R. F.; Nonidez, W. K.; Mays, J. W. Macromolecules 2002, 35, 1196. (i) Kki, S.; Dek, G.; Zsuga, M. Rapid Commun. Mass Spectrom. 2001, 15, 675. (j) Chen, H.; He, M.; Pei, J.; He, H. Anal.Chem. 2003, 75, 6531. (a) Montaudo, G.; Montaudo, M. S.; Puglisi, C.; Samperi, F. Rapid Commun. Mass Spectrom. 1995, 9, 453. (b) Montaudo, G.; Garozzo, D.; Montaudo, M. S.; Puglisi, C.; Samperi, F. Macromolecules 1995, 28, 7983. (c) Loewe, R. S.; McCullough, R. D. Chem. Mater. 2000, 12, 3214. (d) Byrd, H. C. M.; McEwen, C. N. Anal. Chem. 2000, 72, 4568 and references therein. Addition of hydrogen bromide to the carbon triple bond of poly(aryleneethynylene)s has been reported, see: (a) Yamaoto, T. Chem. Lett. 1993, 1959. (b) Yamaoto, T.; Yamada, W.; Takagi, M.; Kizu, K.; Maruyamata, T.; Ooba, N.; Tomaru, S.; Kurihara, T.; Kaino, T.; Kubota, K. Macromolecules 1994, 27, 6620. Yang, C.; Nolan, S. P. J. Org. Chem. 2002, 67, 591.

246

The organolanthanide-catalyzed polymerization of diynes

91

92 93

94

95 96

97 98 99 100 101 102 103 104

105 106 107

GPC experiments based on polystyrene calibration are well-known to lead to a significant overestimation of the molecular weights of rigid-rod polymers (such as poly(3-alkylthiophene)s and poly(aryleneethynylene)s17). For examples, see: (a) Francke, V.; Mangel, T.; Mllen, K. Macromol. Rapid Commun. 1998, 31, 2447. (b) Dellsperger, S.; Dtz, F.; Smith, P.; Weder, C. Macromol. Chem. Phys. 2000, 201, 192. (c) Holdcroft, S. J. Polym. Sci., Part B: Polym. Phys. 1991, 29, 1585. (d) Pearson, D. L.; Schumm, J. S.; Tour, J. M. Macromolecules 1994, 27, 2348. (e) Ref. 34d. (f) Huang, W. Y.; Gao, W.; Kwei, T. K.; Okamoto, Y. Macromolecules 2001, 34, 1570. Fomine, S.; Fomina, L.; Quiroz Florentino, H.; Mendez, J. M.; Ogawa, T. Polym. J. 1995, 27, 1085. (a) Arnold, F. E.; Reinhardt, B. A. U. S. Pat. 4,220,750, 1980. (b) Reinhardt, B. A.; Arnold, F. E. J. Appl. Polym. Sci. 1981, 26, 2679. (c) Reinhardt, B. A.; Arnold, F. E. J. Polym. Sci. Polym. Chem. Ed. 1981, 19, 271. (d) Harris, F. W.; Padaki, S. M.; Varaprath, S. Polym. Prepr. Am. Chem. Soc. Div. Polym. Chem. 1980, 21, 3. The nature of the end-group analysis was not reported, but polycondensation based on palladiumcatalyzed cross-coupling reactions between halides and terminal acetylenes are well-known to produce polymers (e.g. poly(aryleneethynylene)s17) having halide end-groups. Ionic radii for eight coordinate complexes: La3+ (1.160 ) and Pr3+ (1.126 ), see: Shannon, R. D Acta Crystallogr., Sect. A 1976, A32, 751. (a) Red-shifting of absorption maxima is observed in oligo(p-phenylene-vinylene)s96b and oligotriacetylenes96c upon introduction of electron-donating end-groups, while the introduction of electron-withdrawing groups in carotenoids96d leads to blue-shifting. (b) Pascal, L.; Van den Eynde, J. J.; Van Haverbeke, Y.; Dubois, P.; Michel, A.; Rant, U.; Zojer, E.; Leising, G.; Van Dorn, L. O.; Gruhn, N. E.; Cornil, J.; Brdas, J. L. J. Phys. Chem. B 2002, 106, 6442. (c) Martin, R. E.; Gubler, U.; Cornil, J.; Balakina, M.; Boudon, C.; Bosshard, C.; Gisselbrecht, J.-P.; Diederich, F.; Gunter, P.; Gross, M.; Brdas, J. L. Chem. Eur. J. 2000, 6, 3622. (d) Deng, Y.; Gao, G.; Kispert, L. D. J. Phys. Chem. B 2000, 104, 5651. Weder, C.; Wrighton, M. S.; Spreiter, R.; Bosshard, C.; Gnter, P. J. Chem. Phys. 1996, 100, 18931. Stalmach, U.; Kolshorn, H.; Brehm, I.; Meier, H. Liebigs Ann. 1996, 1449. Moroni, M.; Le Moigne, J.; Luzzati, S. Macromolecules 1994, 27, 562. McCullough, R. D.; Lowe, R. D.; Jayaraman, M.; Anderson, D. L. J. Org. Chem. 1993, 58, 904. (a) Pearson, D. L.; Schumm, J. S.; Tour, J. M. Macromolecules 1994, 27, 2348. (b) Li, J.; Pang, Y. Macromolecules 1997, 30, 7487. (a) Nishihara, Y.; Kato, T.; Ando, J.; Mori, A.; Hiyama, T. Chem. Lett. 2001, 950. (b) Park, Y. T.; Seo, I. K.; Kim, Y.-R. Bull. Korean Chem. Soc. 1996, 17, 480. Wenz, G.; Mller, M. A.; Schmidt, M.; Wegner, G. Macromolecules 1984, 17, 837. (a) Rehahn, M.; Schlter, A.-D.; Feast, W. J. Synthesis 1988, 386. (b) Rehahn, M.; Schlter, A.-D.; Wegner, G.; Feast, W. J. Polymer 1989, 30, 1054. (c) Rehahn, M.; Schlter, A.-D.; Wegner, G. Makromol. Chem. 1990, 191, 1991. (a) Weder, C.; Wrighton, M. S. Macromolecules 1996, 29, 5157. (b) Swager, T. M.; Gil, C. J.; Wrighton, M. S. J. Phys. Chem. 1995, 99, 4886. Negishi, E.; Kotora, M.; Xu, Caiding J. Org. Chem. 1997, 62, 8957. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals; 2nd ed.; Pergamon Press: Oxford, 1980.

247

Summary

Summary
The permethylcyclopentadienyl ligand (Cp*) has played a very important role in the development of organo rare-earth metal chemistry. The application of the bis-Cp* ligation has allowed for the preparation of well-defined derivatives, exhibiting interesting physical and chemical properties. Motivated by their catalytic properties towards unsaturated substrates (such as polymerization, hydrogenation, hydrosilylation, hydroamination, hydrophosphination, hydroboration of olefins and polyene cyclization reactions), many of the reactive features of these compounds have been developed. The chemistry of rare-earth metallocene derivatives is characterized by a relative ease of C-H bond activation in combination with a high kinetic lability, as compared to early transition metals. This provides a unique possibility to develop catalytic transformations involving C-H activation steps in the presence of Lewis basic functionalities. Compared to the catalytic transformations of alkenes, alkynes have received much less attention as substrates. In this thesis, a detailed look is taken at the catalytic dimerization of 1-methylalk-2-ynes and 1-alkynes. For both of these processes, the scope of reaction, substrate selectivities, the kinetics and mechanism have been investigated. Furthermore, these catalytic dimerization reactions were explored, using bifunctional substrates, as carbon-carbon bond-forming reaction steps in the preparation of new (cross-)conjugated oligomers and polymers. Scheme 7-1. The 3-propargyl/allenyl derivatives (Ln = La, Y; R = H, Me, iPr) of rare-earth metallocenes.

Ln

CH2 C R

C R

In marked contrast to alkyl, hydride and allyl derivatives of rare-earth metallocenes, the chemistry of propargyl and/or allenyl derivatives is largely unknown. Permethyllanthanidocene propargyl derivatives were proposed as the active species in the catalytic cyclodimerization of 1-methylalk-2-ynes to substituted methylenecyclobutenes. In Chapter 2, the synthesis, structure and reactivity of the propargyl/allenyl derivatives of rare-earth metallocenes is described (Scheme 7-1). Evidence for tri-hapto bonding in these compounds was indicated by several spectroscopic techniques and structrural studies, and both steric and electronic substituent effects were observed in the Cp*2LaCH2CCAr (Ar = C6H5, C6H4Me-2, C6H3Me2-2,6, C6H3iPr2-2,6, C6F5) series (Scheme 7-1). The nature of this bonding is reflected in the observed reactivity, as these derivatives are thermally robust and do not insert 1-alkenes (i.e. ethylene, 1-hexene) into the metal-carbon bond. Upon reaction with protic acids, they furnish both acetylenic and allenic quenching products. Sterically hindered 1-methylalk-2ynes CH3CCAr (Ar = C6H3Me2-2,6, C6H3iPr2-2,6) undergo transmetallation with Cp*2LnCH2CCAr (Ln = Y, La; Ar = C6H5, C6H4Me-2, C6H3Me2-2,6, C6H3iPr2-2,6), while sterically less hindered 1-methylalk-2-ynes (Ar = C6H5, C6H4Me-2) undergo catalytic cyclodimerization with Cp*2LaCCAr (Ar = C6H5, C6H4Me-2). The Cp*2LnCH2CCAr complexes react with hard Lewis bases, such as THF and pyridine, affording the corresponding base adducts that are thermally less stable than their parent compounds, but still contain a trihapto-bound propargyl ligand. The cyclodimerization of 1-methylalk-2-ynes CH3CCAr (Ar = C6H5, C6H4Me-2), catalyzed by L2LnCH2CCAr (L2 = Cp*2, Me2Si(C5Me4)2; Ln = La, Ce, Y; Ar = C6H5, C6H4Me-2), is treated in Chapter 3 (Scheme 7-2). Factors governing the rate and selectivity were investigated, such as temperature, ortho-methyl substitution of the substrate, variation in the ancillary ligand system (Cp*2 versus Me2Si(C5Me4)2), precatalyst and metal ion radius (La versus Y). A mechanistic scenario was proposed to account for the observed products. The observed selectivities towards the major product, (E)-3-benzylidene-2-methyl-1-phenylcyclobutene, were modest (73-79 %) and the catalytic turnover rates were low. The major product could, moreover, not be obtained in a pure form and was not stable under ambient conditions. Substituting the lanthanum metal center with the smaller yttrium did not afford an active catalyst for cyclodimerization and ortho-methyl substitution of 1-phenyl-

249

Summary Scheme 7-2. The catalytic cyclodimerization of 1-phenyl-1-propyne.

catalyst +

catalyst =

La

H C SiMe3 , SiMe3

Si

Ce

H C SiMe 3 SiMe3

1-propyne decreased both the rate and selectivity of the reaction. Opening the coordination sphere around the metal center by the use of the Me2Si(C5Me4) ligation increased the rate, but decreased the selectivity of the reaction. As a result, the poor regioselectivity, slow conversion rate and the instability of the resulting products make this reaction not very attractive for the synthesis of (cross-)conjgated polymers (Scheme 7-3). Nevertheless, attempts to polymerize bifunctional substrates were carried out, and the resulting products were characterized spectroscopically. Scheme 7-3. The formation of (cross-)conjugated polymers via catalytic cyclodimerization of 1,4dipropynylbenzenes.

catalyst

x R R
y

The rare-earth metallocene-catalyzed linear dimerization of phenylacetylene is discussed in Chapter 4 (Scheme 7-4). The effect of substrate concentration, ancilllary ligation (Cp*2 versus Me2Si(C5Me4)2), catalyst precursor and metal ion radius (La versus Y) on the rate and selectivity was investigated. An increase of the initial substrate concentration was found to promote the formation of higher oligomers, while the rate of catalyst deactivation was found to parallel that of catalytic dimerization. Replacing the bis-Cp* ligation by Me2Si(C5Me4), on the one hand, and relatively small rare-earth metal centers (Y) by relatively large ones (La, Ce), on the other hand, revealed that the precatalyst Cp*2LaCH(SiMe3)2 performs best in the catalysis, combining a high activity with a high selectivity for trans-head-to-head dimerization. A plausible mechanism is proposed to account for the reaction products, the reaction intermediates and the kinetic behavior. Six representative (hetero)aromatic 1-alkynes were synthesized to probe the scope of the permethyllanthanocene-catalyzed 1-alkyne dimerization reaction. The observed substrate effects are discussed in Chapter 5. Appropriate acetylenic substitution leads to enhanced reaction rates and selectivities for (E)-but-1en-3-yne formation. In particular, substrates containing five-membered heterocycles, such as 2-ethynylthiophene and 2-ethynyl-1-methylpyrrole, were converted both rapidly and selectively to their trans-head-to-tail dimer. Even 2-ethynylpyridine was converted rapidly and exclusively to its trans-head-to-tail dimer, but at relatively low substrate concentrations competing reactions between the catalyst and the pyridyl moiety took place. A

250

Summary

Scheme 7-4. The catalytic linear dimerization of phenylacetylene.

catalyst +

catalyst =

Ln

H C SiMe , 3 SiMe3

Si

H Ln C SiMe 3 SiMe3

Ln = La, Ce, Y

kinetic and mechanistic study in combination with stoichiometric reactions of catalyst precursors revealed that the increased reactivity and selectivity are the result of electronic alkyne substituent effects and metalheteroatom interactions. The presumed active catalyst species Cp*2LaCCR is not stable, as it dimerizes into an alkynyl bridged, dinuclear derivative [Cp*2La(-CCR)]2 that undergoes C-C coupling to form a butatrienyl derivative [(Cp*2La)2(-n:n-RC4R)]. The formation of butatrienyl derivatives from dimeric alkynyl derivatives was found to be governed by a combination of steric and electronic factors. The application of the catalytic 1-alkyne dimerization reaction to the preparation of conjugated polymers is discussed in Chapter 6 (Scheme 7-5). The effect of the (hetero)arene moiety, with and without solubilizing aliphatic substituents, on the regioregularity of the polymers has been investigated. The application of 3-n-hexyl-2,5-diethynylthiophene, in particular, was found to afford conjugated polymers that were highly regioregular with respect to the formed butenynyl linkage. Monoethynyl substrates were used to control the molecular weight of the polymers. The formed conjugated polymers and their physical properties have been studied by a variety of spectroscopic methods. Scheme 7-5. The formation of (cross-)conjugated polymers via catalytic dimerization of 1,4diethynylbenzenes.

catalyst R R

R
y

251

Samenvatting

Samenvatting
Een katalysator is een stof die een chemische reactie versnelt, zonder zelf verbruikt te worden. Een gewenst eindproduct kan met de juiste katalysator niet alleen heel snel gemaakt worden vanuit een bepaalde grondstof of grondstoffen, maar ook heel selectief. Omdat katalytische processen in het algemeen schoner zijn dan gewone chemische reacties, worden zij in de chemische industrie op grote schaal gebruikt voor de productie van bijvoorbeeld kunststoffen, margarine, benzine, geneesmiddelen en kleurstoffen. Verreweg het grootste deel van de vloeibare brandstoffen en 80% van alle chemische producten worden tegenwoordig geproduceerd met behulp van katalytische processen. Economische overwegingen, hogere kwaliteitseisen en een strengere milieuwetgeving zorgen ervoor dat de chemische industrie in toenemende mate op zoek is naar schone processes die zo weinig mogelijk bijproducten opleveren en zo efficint mogelijk gebruik maken van de grondstoffen. Katalytische systemen bestaan uit een katalysator, de stof die omgezet wordt (het substraat) en het medium waarin de katalytische reactie plaatsvindt. Zij kunnen uit n of meerdere fysische fasen bestaan en worden dan respectievelijk homogeen of heterogeen genoemd. Hoewel katalytische processen vaak transformaties van organische moleculen betreffen, is de katalyse een multidisciplinaire wetenschap die zich vooral buiten de organische chemie heeft ontwikkeld (uitgezonderd de zuur-base katalyse en de katalyse met behulp van organische katalysatoren). Dit is deels te wijten aan het feit dat industrieel relevante katalytische processen vaak relatief simpele organische moleculen produceren (bulkchemicalin) met behulp van complexe katalytische anorganische systemen, terwijl de moderne organische chemie meer gericht is op de synthese van complexe organische moleculen (fijnchemicalin). Bovendien vereist de synthese van homogene en heterogene anorganische katalysatoren kennis van de anorganische chemie, de organometaalchemie, de oppervlakte-chemie en de spectroscopie. Andere belangrijke deelgebieden van de katalyse zijn de biokatalyse die betrekking heeft op biologische katalysatoren (enzymen) en de procestechnologie die de industrile toepassing van katalysatoren mogelijk maakt. Diverse belangrijke katalytische processen in de industrie zijn gebaseerd op organometaalkatalysatoren. De belangrijkste voorbeelden zijn de synthese van polyethyleen en polypropyleen met behulp van Ziegler-Natta katalysatoren en de hydroformylering van alkenen met behulp van kobalt en rhodium metaalcomplexen. Ook op kleinere schaal worden vele organometaalkatalysatoren gebruikt om een grote verscheidenheid aan moleculen te maken die zowel van industrieel als van academisch belang zijn. Hiervan vormen de enantioselective organische transformaties met behulp van chirale metaalkatalysatoren een belangrijke groep die de laatste jaren veel aandacht krijgt. Hoewel het katalytische onderzoek in het begin empirisch van aard was, is de kennis van de organometaalchemie en cordinatiechemie de laatste jaren zodanig toegenomen, dat steeds meer homogene organometaalkatalysatoren ontwikkeld kunnen worden door middel van rationeel ontwerp en kennis van het reactiemechanisme. Desalniettemin, worden nog veel belangrijke katalytische processen nog onvoldoende begrepen en zijn er nog genoeg mogelijkheden om meer katalytische processen te ontdekken en/of te ontwikkelen. Chemische reacties waar een binding tussen twee koolstofatomen gevormd wordt, spelen een belangrijke rol in de synthese van organische verbindingen die zowel academisch als industrieel relevant zijn. Katalysatoren die bestaan uit vroege overgangsmetalen en groep 3 metalen en lanthaniden zijn zeer actief gebleken voor dit type reacties. Ondanks hun hoge katalytische activiteit, zijn deze katalysatoren beperkt toepasbaar, omdat de aanwezigheid van heteroatoom-bevattende functionele groepen dikwijls leidt tot inhibitie of zelfs deactivering van de katalysator. Omdat metaalcomplexen van groep 3 en lanthanide metalen kinetisch labieler zijn dan die van vroege overgangsmetalen, biedt de toepassing van katalysatoren die bestaan uit deze eerstegenoemde groep metalen in principe de mogelijkheid om katalytische processen te ontwikkelen die niet alleen een hoge katalytische activiteit en efficintie vertonen voor C-C koppelingsreacties, maar ook bestand zijn tegen de aanwezigheid van Lewis basische functionele groepen. In vergelijking met andere metalen is de organometaalchemie van de groep 3 en lanthanide elementen relatief laat ontwikkeld. Dit is gedeeltelijk het gevolg van het feit dat deze verbindingen zeer zuurstof en watergevoelig zijn en dat de experimentele en analytische technieken die het mogelijk maken om deze verbindingen te synthetiseren en te karakteriseren relatief laat voorhanden waren. Een andere mogelijke reden is de aanvankelijk wijdverbreide overtuiging dat groep 3 en lanthanide elementen slechts driewaardige analogen waren van de meer bekende groep 1 en groep 2 elementen. In de jaren tachtig kwam men echter tot de ontdekking dat de organometaalverbindingen van groep 3 en lanthanide elementen zeer interessante chemische

253

Samenvatting en fysische eigenschappen vertonen die sterk verschillen van gelijksoortige verbindingen op basis van andere metalen. De organometaalchemie van groep 3 en lanthanide elementen is met name sterk ontwikkeld door complexen die het pentamethylcyclopentadienyl ligand, afgekort Cp* (Cp* = 5-C5Me5), bevatten. Omdat de complexen van groep 3 en lanthanide metalen van het type Cp*2LnR (Ln = groep 3 metaal of lanthanide metaal; R = alkyl of hydride groep) een hoge katalytische activiteit vertonen in C-C en C-X (X = heteroatoom) koppelingsreacties van alkenen (zoals polymerisatie, oligomerisatie, hydrogenering, hydrosilylering, hydroaminering, hydrophosphinering en hydroborering), is tamelijk veel onderzoek naar hun gedrag verricht in de laatste twintig jaar en heeft men vast kunnen stellen dat zij relatief makkelijk C-H en C-X (X = N, S, O) bindingen kunnen activeren. De combinatie van C-H activering en C-C koppeling biedt bovendien de mogelijkheid om selectieve katalytische processen te ontwikkelen die geen bijprodukten opleveren. Een inleiding in de eigenschappen en het gedrag van deze groep verbindingen is gegeven in hoofdstuk 1. In dit proefschrift zijn twee C-C koppelingsreacties van alkynen bestudeerd die door Cp*2LnR complexen worden gekatalyseerd, namelijk de cyclodimerisatie van 1-methylalk-2-ynen en de dimerisatie van 1alkynen. In vergelijking tot de katalytische reacties van alkenen, is over de toepassing van Cp*2LnR complexen in de katalytische reacties van alkynen veel minder bekend. Het doel van dit onderzoek was tweeledig en bestond in de eerste plaats uit het bestuderen van het reactiemechanisme van beide bovenstaande alkyn transformaties. Een beter begrip van het reactiemechanisme heeft ertoe bijgedragen dat nieuwe katalytische processen ontwikkeld konden worden. Zo is gebleken dat de katalysator van de oligomerisatie van 1-methylalk2-ynen de aanwezigheid van aromatische groepen tolereert, terwijl de katalysator van de oligomerisatie van 1alkynen ook bestand is tegen de aanwezigheid van functionele groepen die N, S en O atomen bevatten. Vervolgens is bestudeerd of deze processen toegepast konden worden in de synthese van nieuwe materialen met interessante eigenschappen. Figuur 8-1. De moleculaire structuur van de propargyl/allenyl derivaten (Ln = La, Y; R = H, Me, iPr).

Ln

CH2 C R

C R

In hoofdstuk 2 wordt de synthese, structuur en de reactiviteit beschreven van propargyl/allenyl derivaten van Cp*2LnR complexen. In tegenstelling to hydride, alkyl and allyl derivaten, behoren de propargyl/allenyl derivaten tot een klasse verbindingen van de groep 3 en lanthanide metallocenen waar weinig over bekend is. Zij zijn voorgesteld als een intermediair in de katalytische cyclodimerisatie van 1-methylalk-2ynen met kleine alkyl groepen CH3CCR (R = CH3, CH2CH3 en CH2CH2CH3). Propargyl/allenyl derivaten Cp*2LnCH2CCAr (Ar = C6H5, C6H4Me-2, C6H3Me2-2,6, C6H3iPr2-2,6, C6F5) werden gesynthesiseerd via reacties van hydride and alkyl derivaten Cp*2LnR met 1-methylalk-2-ynen CH3CCAr. Spectroscopische en structurele analyse van deze verbindingen heeft aangetoond dat zij een drievoudig onverzadigde koolstof-koolstof binding bezitten die redelijk sterk aan het metaal centrum cordineert (Figuur 8-1). Deze interne binding zorgt ervoor dat de stabiliteit van deze complexen hoog is en de katalytische activiteit beperkt is. Zo werden geen katalytische C-C koppelingsreacties waargenomen met alkenen en alkynen die grote substituenten bezitten. Alkynen met relatief kleine substituenten CH3CCAr (Ar = C6H5, C6H4Me-2) konden daarentegen wel katalytisch omgezet worden onder vorming van onverzadigde cyclodimeren (Figuur 8-2). In hoofdstuk 3 wordt de katalytische cyclodimerisatie van 1-methylalk-2-ynen met aromatische groepen CH3CCAr (Ar = C6H5, C6H4Me-2) beschreven. De katalytische activiteit en selectiviteit van dit proces is bestudeerd door verschillende factoren te variren, zoals het metaal, het ligand, het substraat en de reactieomstandigheden. Een kinetische studie en de spectroscopische identificatie van reactie-intermediairen en reactieproducten heeft ervoor gezorgd dat het mechanisme van dit katalytisch process nu goed begrepen wordt. Helaas is dit katalytisch process moeilijk toepasbaar, omdat de katalytische activiteit laag, de katalytische regioselectiviteit niet optimaal en het hoofdproduct moeilijk te isoleren is. Desalniettemin, is de katalytische cyclodimerisatie toegepast in de synthese van nieuwe (cross-)geconjugeerde polymeren, omdat het hoofdproduct tot een zeldzame groep van organische moleculen behoort die moeilijk te maken is via andere synthetische

254

Samenvatting routes (Figuur 8-3). Deze materialen zijn spectroscopisch gekarakteriseerd, maar meer onderzoek is nodig om hun potentieel kenbaar te maken.

Figuur 8-2. De katalytische cyclodimerisatie van 1-phenyl-1-propyn.

katalysator +

katalysator =

La

H C SiMe3 , SiMe3

Si

Ce

H C SiMe 3 SiMe3

Figuur 8-3. De toepassing van de katalytische cyclodimerisatie van 1-phenyl-1-propyn in de synthese van nieuwe geconjugeerde polymeren.

katalysator

x R R
y

De katalytische lineaire dimerisatie van phenylacetyleen in aanwezigheid van Cp*2LnR complexen wordt besproken in hoofdstuk 4 (Figuur 8-4). De katalytische activiteit en selectiviteit van dit proces is bestudeerd door het metaal, het ligand en de reactie-omstandigheden te veranderen. Ook in dit geval heeft een kinetische studie en de spectroscopische identificatie van reactie-intermediairen en reactie-producten ervoor gezorgd dat het mechanisme van dit katalytisch proces nu goed begrepen wordt. Deze kennis draagt ertoe bij dat substraat-effecten op de katalytische activiteit en selectiviteit ook beter kunnen worden begrepen. Het onderzoek dat in hoofdstuk 5 wordt beschreven was erop gericht om het bereik van substraten in de katalytische lineaire dimerisatie van 1-alkynen uit te breiden naar aromatische substraten met een functionele groep, zodat dit katalytisch process toegepast zou kunnen worden in de synthese van nieuwe geconjugeerde polymeren. Uit deze studie is gebleken dat alle bestudeerde (hetero)aromatische 1-alkynen snel en efficint omgezet kunnen worden in trans-but-1-en-3-ynen. Reacties die dikwijls leiden to inhibitie of deactivering van de katalysator bleken kinetisch niet competitief met de katalytische reacties. De butenynen behoren tot een klasse van organische moleculen die veelvuldig toegepast worden in de organische synthese om bijvoorbeeld biologisch-actieve stoffen te maken, maar ook materialen met interessante optoelectronische eigenschappen. Door gebruik te maken van bifunctionele substraten is de katalytische lineaire dimerisatie van (hetero)aromatische 1-alkynen toegepast in de synthese van nieuwe geconjugeerde polymeren (Figuur 8-5). Dit onderzoek wordt beschreven in hoofdstuk 6 waar de reactie-condities en het type monomeer werden onderzocht

255

Samenvatting waarmee goed gedefinieerde, geconjugeerde polymeren kunnen worden gemaakt. De synthese en karakterisering van een nieuwe klasse van geconjugeerde polymeren is bewerkstelligd en de optische eigenschappen suggereren dat zij veelbelovende materialen kunnen zijn met interessante toepassingen. Geconjugeerde polymeren staan afgelopen 30 jaar onder grote belangstelling binnen het academisch en industrieel wetenschappelijk onderzoek, omdat zij behoren tot een speciale klasse van kunstoffen die zich als halfgeleiders gedragen. Deze eigenschap heeft ertoe bijgedragen dat zij nu toegepast worden in de electronica (LEDs, transistoren) en ook tal van andere (mogelijke) toepassingen hebben (fotovoltasche cellen, batterijen, antistatische coatings, biosensoren).

Figuur 8-4. De katalytische dimerisatie van phenylacetyleen.

katalysator +

katalysator =

Ln

H C SiMe , 3 SiMe3

Si

H Ln C SiMe 3 SiMe3

Ln = La, Ce, Y

Figuur 8-5. De toepassing van de katalytische lineaire dimerisatie van phenylacetyleen in de synthese van nieuwe geconjugeerde polymeren.

katalysator R R

R
y

256

Dankwoord

Dankwoord
Dit proefschrift belichaamt het einde van mijn promotie-onderzoek dat ik heb uitgevoerd binnen de onderzoeksgroep Moleculaire Anorganische Chemie aan de Rijksuniversiteit Groningen. Een groot aantal mensen heeft ervoor gezorgd dat dit zowel een aangename als een leerzame periode is geweest. Enkelen van hen wil ik hier in het bijzonder bedanken. Allereerst wil ik mijn promotor Bart Hessen bedanken voor de mogelijkheid die hij mij heeft geboden om een promotie-onderzoek te doen onder zijn begeleiding. Zijn enthousiasme en deskundigheid, maar ook de grote vrijheid die hij mij heeft gegeven in mijn onderzoek heb ik enorm gewaardeerd. Jan Teuben wil ik ook bedanken. Hoewel hij niet inhoudelijk was betrokken bij mijn onderzoek, was hij de grote roerganger van onze groep gedurende de eerste twee jaar van mijn promotie-tijd en heb ik zijn aanwezigheid en invloed op de onderzoeksgroep erg op prijs gesteld. De leden van de leescommissie bestaande uit de professoren Erik Heeres, Koop Lammertsma en Uwe Rosenthal wil ik bedanken voor de beoordeling van het manuscript. Auke Meetsma wil ik bedanken voor het meten en oplossen van de kristalstructuren in dit proefschrift. Andries Jekel wil ik bedanken voor de GC-MS analyses, Oetze Staal en Jan Helmantel voor de technische assistentie, Wolter Beukema, Wim Kruizinga en Klaas Dijkstra voor hun hulp met de NMR spectrometers en Albert Kievit voor het bepalen van de exacte massas en de MALDI-TOF spectras. The laboratory of the Teuben/Hessen group has always been a dynamic working environment consisting of numerous people who come and go. I remember my early days, when the Hummelen group shared a lab room with the Teuben group. Within a few years, however, the Hummelen group separated and the Teuben group transformed into the Hessen group. Nowadays, the Van Koningsbruggen group is by our side and it is not unthinkable that soon we will be all engulfed by the everexpanding Feringa group. A successful lab life is not possible without a fullfilling social life and I have been fortunate to meet the right people who have made my stay in Groningen a pleasant one. The large amount of people together with the natural limitations of my memory render any attempt to name everyone who has contributed, either directly or indirectly, to the enjoyable work atmosphere within the Teuben/Hessen group, the realization of my thesis and my agreeable stay in Groningen incomplete. Still, I want to name some of the most important people in this respect, such as Alex Wittenaar, Anthony England, Aurora Batinas, Bas ten Brummelhuis, Bodo Richter, Christian Nijhuis, Cindy Visser, Coen van de Brom, Dennis van der Honing, Dirk Beetstra, Edward Brussee, Edwin van der Eide, Edwin Otten, Elena Novarino, Erica Jellema, Esther Vertelman, Geert-Jan Meppelder, Giuseppe Caroli, Jolanda de Vries, Joop Knol, Maaike Wander, Marten Hettinga, Martin Klok, Marco Bouwkamp, Menno Brandsma, Niels Tazelaar, Nicky Philips, Patrick Deckers, Pavel Shutov, Peter Dijkstra, Piet-Jan Sinnema, Rienk Eelkema, Ralph van Calck, Sascha Dorok, Saskia Dekkers, Srgio Bambirra, Shaozhong Ge, Steven Boot, Thomas Koch, Timo Sciarone, Ton Hubregtse, Toon van Zijl, Weidong Li, Winfried Kretschmer, Xiaochun Zhang and the usual suspects at the ACLO. Por fin, agradezco mis padres por su apoyo, inters y todas las otras cosas que son inexpresables.

257

Dankwoord

258

List of abbreviations

List of Abbreviations
Me Et i Pr t Bu R Ph Ar Cp Cp Cp* TMS M Ln X L DMSO HDMSO THF py TMSA BHT NMR gNOESY gCOSY gHSQC IR MS GC FID-GC MALDI-TOF GPC s d t q ppm w m s i p m o methyl group (CH3) ethyl group (CH2CH3) iso-propyl group (CHMe2) tert-butyl group (CMe3) alkyl or organic group phenyl group (C6H5) aromatic group cyclopentadienyl group (5-C5H5) substituted cyclopentadienyl group pentamethylcyclopentadienyl group (5-C5Me5) trimethylsilyl group (Me3Si) metal group 3 or lanthanide metal monoanionic ligand or halide neutral Lewis base ligand dimethylsulfoxide hexamethyldisiloxane tetrahydrofuran pyridine trimethylsilylacetylene 2,6-di-tert-butyl-4-methylphenol nuclear magnetic resonance gradient nuclear overhauser enhancement spectroscopy gradient correlation spectroscopy gradient heteronuclear single quantum correlation infrared mass spectrometry gas chromatography flame ionization detector gas chromatography matrix-assisted laser desorption induced time-of-flight gel permeation chromatography chemical shift singlet doublet triplet quartet parts per million weak medium strong ipso para meta ortho

259

List of abbreviations R2 kobs K Ea [S]t [S]0 Pn t T max PA PAE PAV PAB PAEV PDA PPP PPV PPE PT PTV PTE P3HT P3DTV P3HTE P3HTB ADMET ADIMET ROMP HH HT HTH HTT LED FET OFET LC linear correlation coefficient observed reaction rate constant equilibrium constant activation energy substrate concentration at time t substrate concentration at time t = 0 number-average degree of polymerization time temperature wavelength corresponding to maximum absorption or emission poly(acetylene) poly(aryleneethynylene) poly(arylenevinylene) poly(arylenebutadiynylene) poly(aryleneethynylenevinylene) poly(diacetylene) poly(phenylene) poly(phenylenevinylene) poly(phenyleneethynylene) poly(thiophene) poly(thienylenevinylene) poly(thienyleneethynylene) poly(3-n-hexyl-2,5-thienylene) poly(3-n-dodecyl-2,5-thienylenevinylene) poly(3-n-hexyl-2,5-thienyleneethynylene) poly(3-n-hexyl-2,5-thienylenebutadiynylene) acyclic diene metathesis acyclic diyne metathesis rong-opening metathesis polymerization head-to-head coupling (trans-1,4-disubstituted but-1-en-3-yne linkage) head-to-tail coupling (2,4-disubstituted but-1-en-3-yne linkage) head-to-head configuration head-to-tail configuration light-emitting diodes field-effect transistors organic field-effect transistor liquid crystalline

260