The Existence of Secondary Orbital Interactions

CHAITANYA S. WANNERE,1 ANKAN PAUL,1 RAINER HERGES,2 K. N. HOUK,3 HENRY F. SCHAEFER III,1 SCHLEYER1 PAUL VON RAGUE

The Center for Computational Chemistry and The Department of Chemistry, University of Georgia, Athens, Georgia 30602-2525 2 Institut fu r Organische Chemie, Universita t Kiel, Otto-Hahn-Platz 4, D-24098 Kiel, Germany 3 Department of Chemistry and Biochemistry, University of California, Los Angeles, Los Angeles, California 90024-1569 Received 8 May 2006; Accepted 23 May 2006 DOI 10.1002/jcc.20532 Published online 15 November 2006 in Wiley InterScience (www.interscience.wiley.com).

Abstract: B3LYP/6-311+G** (and MP2/6-311+G**) computations, performed for a series of Diels-Alder (DA) reactions, conrm that the endo transition states (TS) and the related Cope-TSs are favored energetically over the respective exo-TSs. Likewise, the computed magnetic properties (nucleus-independent chemical shifts and magnetic susceptibililties) of the endo- (as well as the Cope) TSs reveal their greater electron delocalization and greater aromaticity than the exo-TSs. However, Woodward and Hoffmanns original example is an exception: their endo-TS model, involving the DA reaction of a syn- with an anti-butadiene (BD), actually is disfavored energetically over the corresponding exo-TS; magnetic criteria also do not indicate the existence of SOI delocalization in either case. Instead, a strong energetic preference for endo-TSs due to SOI is found when both BDs are in the syn conformations. This is in accord with Alder and Steins rule of maximum accumulation of double bonds: both the dienophile and the diene should have syn conformations. Plots along the IRCs show that the magnetic properties typically are most strongly exalted close to the energetic TS. Because of SOI, all the points along the endo reaction coordinates are more diatropic than along the corresponding exo pathways. We nd weak SOI effects to be operative in the endo-TSs involved in the cycloadditions of cyclic alkenes, cyclopropene, aziridine, cyclobutene, and cyclopentene, with cyclopentadiene. While the endo-TSs are only slightly lower in energy than the respective exo-TSs, the magnetic properties of the endo-TSs are signicantly exalted over those for the exo-TSs and the Natural Bond Orbitals indicate small stabilizing interactions between the methylene cycloalkene hydrogen orbitals (and lone pairs in case of aziridine) with -character and the diene  MOs.
q 2006 Wiley Periodicals, Inc. J Comput Chem 28: 344361, 2007 Key words: pericyclic reactions; secondary orbital interactions; magnetic properties; Diels-Alder and Cope transition states; aromaticity

Introduction
Alder and Steins endo rule1 based on the maximum accumulation of double bonds rationalized the selectivity of Diels-Alder (DA) additions of substituted dienophiles to dienes. As part of their seminal collaboration on pericyclic reactions, Woodward and Hoffmann (WH) attributed the endo preference to attractive secondary orbital interactions (SOI) of p orbitals not involved directly in the formation of new  bonds (Scheme 1).2 In contrast, Garcia et al.s recent review,3 Do Secondary Orbital Interactions Really Exist? concluded that the hypothesis of SOI is not necessary to explain the stereoselectivity results found in the pericyclic reactions. The endo preference might also be due, inter alia, to van der Waals attractions,4 to more favorable charge transfer interactions between dienes and dieno-

philes,5 as well as to differential volumes of activation6 and polarities of the transition states (TS).7,8 Although SOI has often been invoked8e,9an to explain the endo preference of cycloadditions, its existence has been criticized,3,9os and the issue is not settled. Apeloig and Matzner have presented the most convincing theoretical evidence for SOI.10 The degree of SOI involving the
Schleyer; e-mail: schleyer@chem. Correspondence to: P. von Rague uga.edu Contract/grant sponsors: The University of Georgia and National Science Foundation; contract/grant number: CHE-0209857 Contract/grant sponsor: PRF; contract/grant number: 41888-AC4 This article contains supplementary material available via the Internet at http://www.interscience.wiley.com/jpages/0192-8651/suppmat

q 2006 Wiley Periodicals, Inc.

Existence of Secondary Orbital Interactions

345

Scheme 1. The SOI (shown by dotted line) models proposed by WH in the endo-TS (center) formed in the [4 + 2]cycloaddition reactions of s-BD with a-BD. The gure on the right compares the WH versus SalemHouk (SH) type SOIs. Note that these interactions are absent in the exo-TS (extreme left).

pseudo- CH2 group of cyclopropene was found to correlate with the endo preference in the DA reactions of various dienes. On the basis of a detailed analysis of the cyclopentadiene (CPD)maleic anhydride cycloaddition, Cossio et al. also concluded that the stereochemistry is controlled by SOI, even though its magnitude is small (11.5 kcal/mol).11 According to Caramella et al.,12 the endo TS for the CPD dimerization is more stable than the exo-TS due to large Salem/Houk (WH attributed the endo-TS preference of NT (Scheme 1) to a postulated bonding interaction between C(2) and C(7) (WH type SOI). Note that a C2C7 interaction (Scheme 1) also is possible in NC. However, in cases similar to NC, Birney and Houk 9b noted that the Mulliken overlap population between C2 and C7 is negative (unfavorable); instead there is positive overlap between C3 and C8. This Salem-Houk type SOI was rst considered by Salem 8a in a discussion of the endo-TS in acrolein dimerization. Note that NC type TSs also may have C2 symmetry (if diene and dienophile are the same) in which the primary and the secondary interactions are equal in magnitude. Such symmetrical TSs then bifurcate and lead to two enantiomeric products. Hence, SalemHouk type SOIs are extremely important in NC TSs) (and not WoodwardHoffmann type) SOIs (Scheme 1). The close relationship between this endo dimerization TS and the Cope-TS for the degenerate isomerization of the di-CPD product also was demonstrated. Recently Caramella and coworkers extended their SOI investigations to the dimeriza-

tion of butadiene (BD)9l and to cycloadditions of cyclopentadienone.9n The TS for the cycloaddition of anti-butadiene (a-BD) with syn-butadiene (s-BD) was found to be higher in energy than that involving two s-BDs.9l Evans and Warhurst rst recognized the analogy between the -electron sextet of benzene and the six delocalized electrons involved in the cyclic DA TSs (e.g. for ethene and BD).13 Evans14 noted that the greater mobility of the  electrons in the TS the greater will be the lowering of activation energy. Houk et al. showed similarities between the length of the partial bonds for a whole range of pericyclic reactions, and indicated that these distances are consistent with aromatic character of the TSs.9m Schleyer et al. analyzed the aromaticity of pericyclic TSs systematically on the basis of geometric, energetic, and magnetic criteria.1517 In addition to the geometric and energetic evidence that aromatic TSs have cyclically delocalized electronic structures and large energies of concert, they found that that TSs involved in pericyclic reactions generally exhibit strongly enhanced magnetic susceptibilities as well as abnormal 1H NMR chemical shifts. Nucleus-independent chemical shifts (NICS, computed absolute chemical shieldings with their signs reversed) also characterize the aromaticity of various TSs.16 Hence, the magnetic properties of the endo- and the exo-DA TSs might also be employed to differentiate between SOI and alternative effects.17f The latter might inuence the energies but DOI 10.1002/jcc

Journal of Computational Chemistry

346

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

, of the TSs obtained in the cycloaddiFigure 1. The B3LYP/6-311+G** optimized bond lengths, in A tion reactions of various dienes and dienophiles. s-BD refers to syn-butadiene; a-BD is anti-butadiene. The point group symmetry is given in italics in the parenthesis. The thick solid dotted lines show primary interactions (where new  CC bonds are formed) while the thin dotted lines indicate the site of the SOIs.

not the cyclic electron delocalization.17f In contrast, SOI should increase the degree of cyclic electron delocalization and enhance the magnetic effects. The properties of the closely related CopeTSs would afford further comparisons. Hence, we report here the relevant DFT geometries (Figs. 18) of a series TSs of endoand exo-DA additions and the related Cope isomerizations. Energies as well as the CSGT magnetic susceptibility (MS) exaltations (L),18 GIAO NICS, and LMO-IGLO NICS dissections are summarized in Table 1. The Ls were based on the difference in the MS between a TS and the sum of the reactant MSs in their most stable conformations. Note that the computed magnetic susceptibilities of syn- (25.7) and anti-BD (24.7) were similar (Table 1). As the lengths of the CC bonds being formed in exo- and endo-TSs are somewhat different, it can be argued that the greater magnitude of the magnetic properties of the endo-TSs might be

due to differences in geometry, asynchronicity, or the degree of earliness/lateness of the TSs along the reaction paths. Consequently, we established that the endo-TSs really are preferred due to SOI by evaluating the magnetic properties at many points along the intrinsic reaction coordinates (IRC) of both exo- and endo reactions. These comparisons also demonstrate the utility of the application of magnetic criteria to pericyclic reactions.

Calculations Methods
Reactants, products, and transition structures of illustrative DA reactions were optimized at B3LYP/6-31G* and then at the B3LYP/6-311+G** DFT levels using the Gaussian-98 program.19 Unscaled zero-point energy corrections (ZPE), based on the B3LYP/6-31G* vibrational frequencies, were applied to the DOI 10.1002/jcc

Journal of Computational Chemistry

Existence of Secondary Orbital Interactions

347

, of the TSs obtained in the dimerizaFigure 2. The B3LYP/6-311+G** optimized bond lengths, in A tion of CPD. CPD refers to cyclopentadiene.

B3LYP/6-311+G** energies. As B3LYP underestimates long range attractions (dispersion), MP2/6-311+G** single point energy computations were performed on the B3LYP/6-311+G** optimized geometries to validate our endoexo TS energy difference (Table 2). The activation energies, DEact, were computed relative to the most stable conformations of the reactants. The energies discussed in the text, unless noted otherwise, refer to B3LYP/ 6-311+G**//B3LYP/6-311+G** computations. The magnetic susceptibilities and exaltations, L, of the TSs were evaluated at the CSGT-B3LYP/6-31+G* level20 while the NICS (at the geometric center of the six heavy atoms involved in the reorganization of the CC-bonds) were computed at GIAO21-B3LYP/6-31+G* using the B3LYP/6-311+G** rened geometries. The localized MO (LMO) NICS dissection employed the deMon-Master program22 at the SOS-DFPT23 level with the Perdew-Wang-91 (PW91)24a,b

exchange correlation functional and the recommended IGLO-III TZ2P25 basis set. Note that the NICS() refers to the total contributions of, e.g., the four LMOs involving all eight electrons of the diene DA dimerization. Plots show these LMOs to have predominant  character, which justies the nomenclature we employ. Points along the B3LYP/6-31G* IRC pathways for two DA reactions were used to compute the magnetic properties. The TSs had a single imaginary vibrational frequency, with one exception. The corresponding stationary point for endodimerization of cycloBD, TS(CBD2)endo, had two degenerate imaginary frequencies at HF/6-31G*. But further optimization at B3LYP/6-31G* and B3LYP/6-311+G** led to the closely related Cope-TS, TS(CBD2)cope. Consequently, the single point B3LYP/6-311+G** energy employed for TS(CBD2)endo was computed using the HF/6-31G* geometry.

, of the TSs obtained in dimerization Figure 3. The B3LYP/6-311+G** optimized bond lengths, in A of CBD. CBD refers to cyclobutadiene. Note that the endo-TS, TS(CBD2)endo, is a second-order saddle point optimized at HF/6-31G*.

Journal of Computational Chemistry

DOI 10.1002/jcc

348

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

, of the TSs obtained in cycloaddition Figure 4. The B3LYP/6-311+G** optimized bond lengths, in A of CPD and BD.

Results and Discussion

Geometries

Dimerization of BD

The B3LYP/6-311+G** optimized structures of all the reactants are reported in the supporting information, but the TS geometries are displayed in Figures 18. Note that the ca. 2.0 and 2.5 lengths of some of the partial C. . .C bonds in the asynchroA in Cope-TSs are different nous exo-, endo-DA TSs, and )2.7 A 26 from those (2.25 A) in the prototype ethene-BD cycloaddition. The central CC bonds of all the dienes shorten in both the exoand endo-TSs to lengths intermediate between a single and a double bond. The separations between carbons involved in SOIs (represented by thin dotted lines in Figs. 18) are typically ca. 3 in all the endo- and the related Cope-TSs. A

WH illustrated their SOI concept using an assumed endo-TS model for BD dimerization involving one s-BD and one a-BD conformation.2 Unfortunately, WHs assumed TS structure, which has frequently been reproduced in discussions of SOI,27 actually is the least favorable compared to other TS alternatives. Moreover, Stephensons experiments showed no signicant endo preference for this dimerization.28 Our results (Table 1) agree with those of Caramella,9l who also nd the energy of the exoTS [TS(s-BD a-BD)exo] actually to be 1.1 kcal/mol lower than WHs endo-TS [TS(s-BD a-BD)endo]! The best TS is indeed endo, but involves the dimerization of two s-BDs (rather than WHs syn and anti); this all syn TS [TS(s-BD2)endo] is only 0.5 kcal/mol more favorable than the exo-TS, TS(s-BD2)exo (in DOI 10.1002/jcc

Journal of Computational Chemistry

Existence of Secondary Orbital Interactions

349

, of the TSs obtained in cycloaddition Figure 5. The B3LYP/6-311+G** optimized bond lengths, in A of CPD and CBD. The 2 in the subscript refers to dienophile while 4 refers to diene moiety.

rough agreement with Stephensons experiment28). The Erel at the MP2/6-311+G** are slightly larger as compared to those at B3LYP/6-311+G** and also are in disagreement with the Stephensons experimental results; TS(s-BD a-BD) exo 2.2, TS(sBD a-BD) endo 3.1, TS(s-BD2) exo 2.4, TS(s-BD2) endo 0.0, and TS(s-BD2) cope 11.1 kcal/mol. However, the endo-syn, syn-TS, is 1.3 kcal/mol lower in energy than WHs assumed syn, anti-TS, upon which SOI was originally based. The TS(s-BD2)endo preference is small because of the energy (ca. 3.6 kcal/mol) required to convert the more stable a-BD conformation to the syn form for both the diene and the dienophile in the TS. It has been noted many times before that this conformational penalty is overcome in CPD, with its restricted syn conformation. Consequently, DA reactions of CPD are more facile than those of BD and show a much greater endo preference. The MS and its exaltation, L, Table 1, for both TS (s-BD aBD)exo (66.2 and 10.8) and TS (s-BD a-BD)endo (65.4 and

11.5) are smaller than those for the syn-endo-TS TS(s-BD2)endo (76.5 and 17.2). The greater cyclic electron delocalization (aromaticity) of TS(s-BD2)endo also is shown by its more negative NICS(0) (18.3) than TS(s-BD a-BD)exo (16.0) and TS(sBD a-BD)endo (17.0). Likewise, NICS() value in TS(sBD2)endo (13.0) also is larger as compared to that in TS(s-BD a-BD)exo (10.3), TS(s-BD a-BD)endo (11.8), or TS(s-BD2)exo (11.1). The lower energy and enhanced magnetic properties exhibited by TS(s-BD2)endo support that SOI effects are larger that those in TS(s-BD a-BD)endo. The computed energies and magnetic properties of the anti-endo-TS model chosen by WH (Scheme 1), TS(s-BD a-BD)endo, does not exhibit appreciable SOI effects. Evidently, SOI in the TS only is signicant if the dienophile has a syn rather than an anti-conformation. A referee stated, Since the SOI is occurring outside the region involving the six delocalized electrons in the cyclic part of the transition structure, the usefulness of the NICS numbers DOI 10.1002/jcc

Journal of Computational Chemistry

350

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

, of the TSs obtained in cycloaddition Figure 6. The B3LYP/6-311+G** optimized bond lengths, in A of BD and CBD.

in evaluating the SOI seems limited at best. To address this point, we have carried out a detailed LMO analysis of the endoand exo-TSs involved in the dimerization. In all the three TSs, which do not follow the Alder-Stein rule, TS(s-BD a-BD)exo, TS(s-BD a-BD)endo, and TS(s-BD2)exo, the -bond not actively involved in the DA reaction (C7C8) contributes paratropically (ca. +1.0 ppm) to the total NICS(). In contrast, the TS that does follow the rule, TS(s-BD2)endo, the C7C8 bond makes a diatropic contribution (0.8 ppm) to the NICS(). This analysis clearly indicates the reliability of NICS() to help establish the presence (or absence) of SOI. TS(s-BD2)cope for the degenerate Cope rearrangement of the vinylcyclohexene DA product also has C2 symmetry, but is 3.2 kcal/mol lower in energy than the best DA TS [TS(s-BD2)endo]. The enhanced magnetic properties (also given in Table 1) indi-

cate the existence of SOI in the Cope-TS as well. As in TS(sBD2)endo, the dienophile in TS(s-BD2)cope also has syn conformation.
Dimerization of CPD

Our energies (Table 1) and geometries (Fig. 2) for the endo- and the exo-TSs involved in the dimerization of CPD and of the related Cope-TS are similar to those of Caramella and co-workers.12 The endo-TS, TS(CPD2)endo, is 3.4 kcal/mol (Table 1) more stable than the exo-TS, TS(CPD2)exo, but is 0.6 kcal/mol higher in energy than the related Cope-TS, TS(CPD2)cope. CPD dimerization is known experimentally to give only the endo-product.29 The C4 , mimics a C C5 distance in TS(CPD2)cope, 1.65 A C single bond ) and in and is much shorter than that in TS(CPD2)endo (1.92 A DOI 10.1002/jcc

Journal of Computational Chemistry

Existence of Secondary Orbital Interactions

351

, of the TSs obtained in the cycloadFigure 7. The B3LYP/6-311+G** optimized bond lengths, in A dition reactions of CPD with various cycloalkenes and with propene. Cyclopentadiene, cyclopropene, cyclobutene and cyclopentene are denoted as CPD, cyprop, cybut, and cypent, respectively. The symmetry point groups of TSs are indicated in italics in the parenthesis.

Journal of Computational Chemistry

DOI 10.1002/jcc

352

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

, of the endo- and exo-TSs obtained in Figure 8. The B3LYP/6-311+G** optimized bond lengths, in A the DA reactions of CPD with aziridine. The HOMO of the endo-TS shows an orbital interaction of the lone-pair of nitrogen with the C2C3 -bond. The symmetry point groups of TSs are indicated in italics in the parenthesis.

) (Fig. 2). The equivalent C3C8 and C1C6 TS(CPD2)exo (2.01 A ) in the Cope-TS are shorter than those in the distances (2.71 A ), whereas the C1C6 distance in the exo-TS is endo-TS (2.91 A , indicating a greater degree of asynchronous charsmaller, 2.53 A acter in TS(CPD2)cope. While NICS(0) (at the centers of the six carbons forming the TS) values are large for both the endo-TS (22.9) and the Cope-TS (22.3), the sum of the localized mobile electron contributions, NICS(), is greater for TS(CPD2)endo (15.0) than that for TS(CPD2)cope (10.3) and for TS(CPD2)exo (13.6). Note that the NICS(0) value for TS(CPD2)exo also is lower, 20.4, than the respective endo-TS. The LMO-analysis indicates that the -bond (C7C8), which is not actively involved in the DA reaction, in TS(CPD2)exo contributes paratropically (+0.8 ppm) to the total NICS(), while the C7C8 bond makes a diatropic contribution of 0.7 ppm in TS(CPD2)endo. This analysis advocates that the C7C8 -bond of TS(s-BD2)endo participates in the SOI. The computed magnetic susceptibilities and susceptibility exaltations (L) for the endo-TS (90.4, 17.4) and the Cope-TS (91.2, 18.2) are appreciably larger than that for the exo-TS (85.5, 12.5). The computed magnetic properties of these TSs conrm their aromaticity. The energies, and the smaller distances between C3C8, as well as the enhanced magnetic properties of TS(CPD2)endo relative to that of TS(CPD2)exo show that the endo-TS is stabilized due to SOIs.
Dimerization of CBD

point of second order. After following the vector corresponding to either degenerate imaginary frequency, geometry optimizations with no symmetry constraints leads to TS(CBD2)cope. As illustrated in ), typical of Figure 3, the C4C5 distance in the Cope-TS (1.54 A primary interactions, is much shorter than that in the endo-TS (2.42 ) or in the exo-TSs (2.67 A ). Note that the C3C8 distance in A TS(CBD2)endo (2.72 A), typical of SOIs, is shorter than that in TS(CBD2)exo. NICS(0) and NICS() at the center of C1C2C3C4C5C6 in TS(CBD2)endo (32.7 and 24.5) are larger than that for TS(CBD2)exo (2.3 and 4.9). These values for TS(CBD2)cope are 23.9 and 16.7. Remarkably, NICS(0) and NICS() for the endo structure are much more diatropic (negative) than those for CBD (27.5 and 0.2).31 These NICS changes reect the decrease in antiaromatic character of the four-membered rings in the TS. The decrease is less for the exo-TS due to absence of SOI. Because of the strong delocalized ring current in the endo- and the Cope-TS, the magnetic susceptibilities and L for TS(CBD2)endo, 65.6 and 50.7, and TS(CBD2)cope, 65.6 and 50.8, are much larger than that for TS(CBD2)exo, 25.4 and 10.4.
Cycloaddition of CPD and BD

Li and Houk have studied the dimerization of CBD extensively.30 Features are pertinent. The CBD dimerization has no enthalpy barrier. The exo-TS (Cs), TS(CBD2)exo, and the related Cope-TS (C2v), TS(CBD2)cope, are stationary points with only one imaginary frequency. In contrast, the endo-TS (Cs), TS(CBD2)endo, is a saddle

There is no strong stereochemical preference in this DA reaction, where CPD functions as the diene component (Fig. 4 and Table 1). The energy of the syn-endo-TS [TS(CPD4 sBD2)endo] is only 0.5 kcal/mol lower than both the anti-endoTS [TS(CPD4 a-BD2)endo] and the anti-exo-TS, TS(CPD4 aBD2)exo. Note that BD in the anti-conformation (a-BD) (as assumed in WHs model) would result in no stereochemical preference, as both anti-TSs have the same energy. However, BD in the syn conformation does give a stereochemical preference: syn-endo-TS [TS(CPD4 s-BD2)endo] is 1.5 kcal/mol DOI 10.1002/jcc

Journal of Computational Chemistry

Existence of Secondary Orbital Interactions

353

Table 1. The Relative Energies (Erel, kcal/mol), Activation Energies (Eact, kcal/mol), NICS(0) Computed at the Center of the TS and NICS(), Which is the Mobile Electron Contributions to the Total NICS Value, as well as the Magnetic Susceptibility () and Its Exaltations (L, cgs ppm) for the Transition States Involved in Various Cycloaddition Reactions.

TSa TS(s-BD a-BD)exo TS(s-BD a-BD)endo TS(s-BD2)exo TS(s-BD2)endo TS(s-BD2)cope TS(CPD2)exo TS(CPD2)endo TS(CPD2)cope TS(CBD2)exo TS(CBD2)endo TS(CBD2)cope TS(CPD4 a-BD2)exo TS(CPD4 a-BD2)endo TS(CPD4 s-BD2)exo TS(CPD4 s-BD2)endo TS(CPD4 s-BD2)cope TS(CPD2 CBD4)exo TS(CPD4 CBD2)exo TS(CPD4 CBD2)endo TS(CPD4 CBD2)cope TS(CBD4 a-BD2)exo TS(CBD4 a-BD2)endo TS(CBD4 s-BD2)exo TS(CBD2 s-BD4)exo TS(CBD2 s-BD4)endo TS(CBD2 s-BD4)cope Benzene CPD CBD s-BD a-BD
a

PGb C1 C1 C1 C2 C2 C1 C2 C2 Cs Cs C2v C1 C1 C1 C1 C1 C1 Cs Cs C1 C1 C1 C1 Cs Cs C1 D6h C2v D2h C2v C2h

Erelc 0.2 1.3 0.5 0.0 3.2 3.4 0.0 0.6 15.3 0.0 43.7 0.5 0.5 1.5 0.0 0.5 6.2 5.5 0.0 21.5 3.5 3.4 5.0 4.8 0.0 24.4

Eactd 29.2 30.3 29.6 29.0 48.1 27.4 24.1 27.9 3.35 11.9 21.3 27.1 27.1 28.1 26.6 29.7 13.4 12.7 7.2 29.9 12.8 12.8 14.4 14.2 9.4 48.8

GIAO (IGLO)-NICS(0)e 15.3 16.2 14.2 17.2 23.1 19.2 21.9 22.3 2.4 29.1 22.4 19.3 19.7 18.5 20.6 20.1 5.6 13.8 20.4 21.7 3.8 3.5 4.1 11.0 18.2 20.8 8.0 2.7 27.5 (16.0) (17.0) (14.9) (18.3) (24.3) (20.4) (22.9) (23.2) (2.3) (32.7) (23.9) (20.5) (21.0) (19.6) (21.8) (21.1) (8.6) (15.4) (22.5) (22.7) (7.1) (7.3) (7.1) (12.4) (20.3) (21.8) (8.8) (4.1) (20.8)

NICS()f 11.1 11.8 10.3 13.0 13.3 13.6 15.0 10.3 4.9 24.5 16.7 13.8 14.0 12.8 14.7 9.5 1.3 9.5 16.2 14.4 0.7 0.5 2.0 9.0 14.6 14.0 20.7 12.2 0.2

g 66.2 65.4 68.6 76.5 79.2 85.5 90.4 91.2 25.4 65.6 65.6 72.0 72.7 74.7 79.5 82.2 58.2 57.6 68.2 75.4 45.3 46.1 48.0 53.2 64.1 69.9 48.2 36.5 7.4 25.7 24.7

Lh 10.8 11.5 12.5 17.2 19.9 12.5 17.4 18.2 10.4 50.7 50.8 10.8 11.5 12.5 17.2 19.9 14.3 13.7 24.3 31.5 13.2 14.0 15.9 21.1 31.9 37.8 13.4 3.5 13.0

0.0 3.7

Transition state (TS) for the cycloaddition reactions. s-BD refers to syn-butadiene; a-BD is anti-butadiene; CPD is cyclopentadiene; CBD is cyclobutadiene. The exo, endo, or cope (in subscript) at the end indicates the mode of approach of the diene and dienophile components (for e.g. TS(CBD4 a-BD2)endo represents the endo-TS formed in the cycloaddition of cyclobutadiene (diene) and a-BD (dienophile) while TS(CPD2)exo refers to the exo-TS formed in the dimerization of CPD). b Refers to the point group symmetry of the TS. c Relative energies computed at the B3LYP/6-311+G** (with ZPE correction at B3LYP/6-31G*) with respect to the energies of most stable endo-TS. d Activation energies computed at the B3LYP/6-311+G** (with ZPE correction at B3LYP/6-31G*) with respect to the energies of most stable conformation of the reactants. Note that a direct comparison of Erel with Eact cannot be made from the information given in the table. The relative energies, for the TSs, are evaluated with respect to the most stable endo-TS, while activation energies are reported with respect to the most stable conformations of the starting reactants for the exo- and the endo-TS and for the Cope-TS Eact are evaluated with respect to the energies of the endo-products. e NICS(0) computed at the center of the TS with GIAO-B3LYP/6-31+G*//B3LYP/6-311+G**. f NICS() is the mobile electron contribution to the total NICS value computed using the deMon-Master NMR program, with the Perdew-Wang-91 (PW91) exchange correlational functional and the IGLO-III TZ2P basis set. g  is the magnetic susceptibility computed at the CSGT-B3LYP/6-31+G*//B3LYP/6-311+G**. h L is the magnetic susceptibility exaltations, which is taken as the difference between the susceptibilities of the TS and those of the starting reactants, at the CSGT-B3LYP/6-31+G*//B3LYP/6-311+G**.

Journal of Computational Chemistry

DOI 10.1002/jcc

354

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

Table 2. The Relative Energies (Erel, kcal/mol), Activation Energies (Eact, kcal/mol), NICS(0) Computed at the Center of the TS and NICS(), Which is the Mobile Electron Contributions to the Total NICS Value, as well as the Magnetic Susceptibility Exaltations (L, cgs ppm) for the Transition States involved in the DA Reaction of Cyclopentadiene with Cycloalkenes (Cyclopropene, Cyclobutene, and Cyclopentene), with Propene, and with Maleic Anhydride.

TSa TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD TS(CPD cyprop)exo cyprop)endo propene)exo propene)endo cybut)exo cybut)endo cypent)exo cypent)endo aziridine)exo aziridine)endo malanhyd)exo malanhyd)endo
a

PGb Cs Cs Cs Cs Cs Cs Cs Cs Cs Cs Cs Cs

B3LYP-Erelc 3.2 0.0 0.9 0.0 1.2 0.0 1.8 0.0 0.8 0.0 1.3 0.0

MP2-Ereld 4.4 0.0 1.3 0.0 2.0 0.0 3.2 0.0 1.7 0.0 2.3 0.0

Eacte 19.2 16.0 29.0 28.1 26.7 25.6 31.0 29.2 12.2 11.4 18.1 16.8

IGLO-NICS(0)f 21.3 26.0 22.2 22.7 21.8 22.6 21.7 21.9 19.9 25.1 21.2 22.8

NICS()f 15.6 17.7 15.5 15.6 15.9 16.6 16.2 15.5 14.4 17.2 15.4 16.2

Lg 15.4 21.0 12.7 13.6 15.4 17.0 11.8 12.0 14.6 18.0 14.0 15.9

Transition state (TS) for the [4+2]cycloaddition reactions. Cyclopentadiene, cyclopropene, cyclobutene, cyclopentene, and maleic anhydride are denoted as CPD, cyprop, cybut, cypent, and malanhyd, respectively. The endo or the exo subscript at the end refers to the approach of the diene and dienophiles in the respective fashion. b Refers to the point group symmetry of the TS. c Relative energies computed at the B3LYP/6-311+G** (with ZPE correction at B3LYP/6-31G*) with respect to the most stable endo-TS. d Relative energies computed at the MP2/6-311+G**//B3LYP/6-311+G** with respect to the most stable endo-TS. e Activation energies computed at the B3LYP/6-311+G** (with ZPE correction at B3LYP/6-31G*)with respect to the energies of the most stable conformation of the reactants. f NICS() is the mobile electron contribution to the total NICS value computed using the deMon-Master NMR program, with the Perdew-Wang-91 (PW91) exchange correlational functional and the IGLO-III TZ2P basis set. g L are the magnetic susceptibility exaltations, which is taken as the difference between the susceptibilities of the TS and those of the starting reactants, at the CSGT-B3LYP/6-31+G*//B3LYP/6-311+G**.

lower in energy than syn-exo-TS [TS(CPD4 s-BD2)exo] and only 0.5 kcal/mol higher in energy than the related syn-Cope-TS [TS(CPD4 s-BD2)cope] involved in the degenerate isomerization of endo-6-vinyl[2.2.1]bicyclohex-2-ene. Although the energy differences are not large, SOI seems necessary to explain the lower energy of the more crowded TS(CPD4 s-BD2)endo relative to TS(CPD4 a-BD2)endo and the exo TSs (see the geometries shown in Fig. 4 and in Fig. A of the SI). SOI is supported by the more diatropic magnetic susceptibilities and NICS (Table 1) computed for TS(CPD4 aBD2)endo and (TS(CPD4 s-BD2)cope (Table 1) with respect to TS(CPD4 s-BD2)exo and TS(CPD4 a-BD2)exo.

Cycloaddition of CPD with CBD

Since both CBD and CPD have syn C C C C conformations, either might serve as the diene component. However, attempts to locate the endo-TS involving CBD as diene and CPD as dienophile failed at the HF and the B3LYP levels. The TS involving the exo approach of CPD (diene) and CBD (dienophile), TS(CPD4 CBD2)exo, is synchronous with Cs symmetry (Fig. 5). The exo approach with the components reversed is asynchronous and gives TS(CPD2 CBD4)exo (point group C1). The

bond lengths in the four-membered rings of TS(CPD4 CBD2)exo and TS(CPD2 CBD4)exo do not change much from CC lengths of CBD itself, optimized at the 1.333 and 1.577 A the same level. The B3LYP/6-311+G** relative energies (Table 1) of the exo alternatives indicate only a slight preference (0.7 kcal/mol) for TS(CPD4 CBD2)exo over TS(CPD2 CBD4)exo i.e. for CBD acting as the diene component. In sharp contrast, the endoTS, TS(CPD4 CBD2)endo, with CPD as the diene, is ca. 5.5 kcal/mol more stable than the exo-TS due to the SOI stabilization. The closely related Cope-TS, TS(CPD4 CBD2)cope, formed from the endo addition product (tricyclo[4.2.1.02,5]nona3,7-diene), is about 21.5 kcal/mol more favorable even than TS(CPD4 CBD2)endo (Table 1). The greater stability of TS(CPD4 CBD2)cope can be attributed to the loss in antiaromaticity of CBD on going from the endo-TS to the Cope-TS; the NICS values in the center of the CBD32 in the endo- and the Cope-TS are 4.9 and 9.9, respectively. Also note the single , in the TS(CPD4 CBD2)cope; bond-like C4C5 distance, 1.55 A , in TS(CPD4 CBD2)endo. this distance is much larger, 2.51 A The computed magnetic properties document the much greater electron delocalization in the endo- and Cope-TSs than in the exo-TSs. NICS(0) and NICS() for the TS(CPD4 CBD2)endo (20.4 and 16.7) and the TS(CPD4 CBD2)cope DOI 10.1002/jcc

Journal of Computational Chemistry

Existence of Secondary Orbital Interactions

355

Scheme 2. The destabilized exo-TSs formed in the [4 + 2]cycloaddition of propene with s-BD (left) and with CPD (right). The arrows indicate Garcias argument of the steric repulsion between the inner hydrogens.

(21.7 and 14.4) are much larger than that of TS(CPD4 CBD2)exo (13.8 and 12.6 value). Likewise, the computed MS exaltations indicate a more diatropic ring current and the existence of SOI in endo- (24.2) and Cope-TSs (31.5); these values are twice as large as the 13.7 for the exo-TS.

Cycloaddition of CPD with Cyclopropene, Cyclobutene, and Cyclopentene

Cycloaddition of CBD with BD

Similarly, s-BD can participate as a diene or as a dienophile in its cycloaddition with CBD. Moreover, BD also may adopt the anti-conformation when acting as the dienophile. The geometrical features of the TSs (Fig. 6) are similar to those discussed CC distances at the priabove; note, in particular the *2.5 A mary interaction sites and ca. 3.0 A CC separation in the endoand the Cope-TS, where SOI takes place. With an a-BD dienophile, the endo TS(CBD4 a-BD2)endo is no better than the exo TS(CBD4 a-BD2)exo; both these TSs have similar energies (Table 1). Likewise, the computed magnetic properties for both these a-BD endo- and exo-TSs are almost the same. Also the C2C8 distance in TS(CBD4 a , and in TS(s-BD a-BD), 3.36 A , are almost BD2)endo, 3.39 A identical. Hence, SOI is absent in TS(CBD4 a-BD2)endo. (Note that these a-BD TS models are similar to those originally proposed by WH for the DA reaction of syn- with a-BD.) In sharp contrast, the endo-TS, TS(CBD2 s-BD4)endo, involved in the cycloaddition of s-BD with CBD is * 5 kcal/ mol lower in energy than both the exo-TSs, TS(CBD2 sBD4)exo and TS(CBD4 a-BD2)exo. However, TS(CBD2 sBD4)cope is much more stable, 24.4 kcal/mol, than the TS(CBD2 s-BD4)endo due to loss in antiaromaticity of the CBD ring in the Cope-TS. The NICS(0) values in the center of CBD in the endo- and the Cope-TS are 1.4 and 10.4, respectively. The enhanced MS exaltations of the endo- and Cope-TS, 31.9 and 37.8, respectively, indicate stronger diatropic ring currents as compared to those in TS(CBD2 s-BD4)exo (21.1) and TS(CBD4 a-BD2)exo (15.9). Likewise the larger NICS(0) and NICS() values also show the presence of SOI in TS(CBD2 s-BD4)endo and TS(CBD2 s-BD4)cope, Table 1.

Apeloig and Matzner were the rst to support the existence of SOI in their systematic study of the DA reaction of cyclopropene with various dienes.10 They showed that the endo-TSs were preferred due to the favorable orbital interactions of the cyclopropene methylene hydrogens with the -orbitals of the diene. Wiberg and Barley32 obtained only the endo cyclopropeneCPD DA product experimentally; hence, the activation barrier difference favoring the endo- over the exo-TS is more than 3 kcal/mol. Apeloig obtained a 4.9 kcal/mol difference at MP2/6-31G*. We now have extended these computations, also to the DA reactions of CPD with cyclobutene and with cyclopentene and nd more evidence for the involvement of SOI in stabilizing their endo-TSs. The exo-TS [TS(CPD cyprop)exo] for the cycloaddition of cyclopropene with CPD is about 3 kcal/mol higher in energy (B3LYP/6-311+G** and experiment) than the endo-TS [TS(CPD cyprop)endo]; the MP2/6-311+G**//B3LYP/6-311+G** value, 4.4 kcal/mol, is in agreement with Apeloigs10 computations. These TS geometries are shown in Figure 7. Since the inner methylene in Hs of cyclopropene and of CPD are separated by only 1.90 A the exo-TS, Garcia disputed that the resulting van der Waals repulsion (Scheme 2) favors the endo-TS; consequently, SOI explanations are not needed. To verify Garcias steric repulsions argument we repeated the above computations by replacing cyclopropene by aziridine (Fig. 8). Aziridines conformation with the NH on the outside was used (Fig. 8). The B3LYP/6311+G** energies, Table 2, reveal that the endo-TS involved in the cycloaddition of aziridine and CPD is still favored over the exo-TS by 0.8 kcal/mol; the MP2/6-311+G**//B3LYP/6311+G** value is 1.7 kcal/mol while the CCSD[T]/cc-pVTZ// B3LYP/6-311+G** energy difference is 1.5 kcal/mol. This endo exo difference is smaller than that for the CPD-cyclopropene case, but part of the reason may be due to CH. . .N hydro between the N to the neargen bonding. A distance of 2.180 A est CPD H in the exo-TS is quite large for lone-pair repulsion. However, part of the 3 kcal/mol energetic preference10 for the DOI 10.1002/jcc

Journal of Computational Chemistry

356

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

Scheme 3. The extra delocalization due to the SOIs (shown by dotted lines) in the HOMOs of the endo-TSs of the DA reactions of CPD with propene, cyclopropane, cyclobutane, and cyclopentene. Note the -like character of the alkyl moiety orbitals of the dienophiles.

TS(CPD cyprop)endo may be due to unfavorable steric HH repulsions in the TS(CPD cyprop)exo.3 The enhanced magnetic properties of the endo-TSs for both the cycloaddition of CPD with cyclopropene and with aziridine indicate greater electron delocalization than the respective exoTSs. Furthermore, as pointed out by Apeloig and Matzner (see above), the HOMO shows the involvement of SOI in stabilizing the TS(CPD cyprop)endo (note the favorable interaction of the inner methylene hydrogens of cyclopropene with -orbitals of diene in HOMO, Scheme 3). Like cyclopropene, cyclobutene and cyclopentene also favor endo-TSs in their cycloadditions with CPD (by 1.2 and 1.8 kcal/

mol over the respective exo-TSs). The van der Waals repulsion in the exo-TSs decrease considerably in going from cyclopropene to cyclopentene, since the separations of the relevant for TS(CPD cyprop)exo vs. 2.2 and 2.1 A hydrogens (1.9 A TS(CPD-cyclobutene)exo and TS(CPD-cyclopentene)exo, respectively) are much larger (compare the B3LYP/6-311+G** geometries in Fig. 7). Since MP2 are better than the DFT methods in describing the dispersion and van der Waals repulsions, the relative energies predicted by MP2/6-311+G**//B3LYP/6-311+G** are ca. 1 kcal/mol higher than those given by the B3LYP, Table 2. The orbitals depicted in Scheme 3 reveal a stabilizing SOI (indicated by dotted lines), between the endo CH2 hydrogens

, of the endo- and exo-TSs obtained in Figure 9. The B3LYP/6-311+G** optimized bond lengths, in A the DA reactions of CPD with maleic anhydride. CPD and maleic anhydride are denoted as CPD and malanhyd, respectively. The symmetry point groups of TSs are indicated in italics in the parenthesis.

Journal of Computational Chemistry

DOI 10.1002/jcc

Existence of Secondary Orbital Interactions

357

and the -orbitals of the CPD unit. The exalted magnetic properties of the endo-TSs agree with the energy results, and further demonstrate the greater delocalization attributable to the SOIs. However, the relatively large TS(CPD cyprop)endo preference is due not only to the stabilizing SOI interaction in the endo-TS, but also to the steric repulsion between the inner protons of the exo-TS. We investigated Garcias argument3 further by replacing cyclopropene by propene. Because of the rotational mobility of the CH3 group, propene avoids van der Waals repulsions in its exo-TS with CPD, (TS(CPD-propene)exo. The distance between the inner methylene H of CPD and the closest methyl proton in in TS(CPD-propene)exo, is much larger than propene, 2.25 A (Fig. 7). Garcias steric that in TS(CPD-cyprop)exo, 1.90 A effects are not present. Nevertheless, the exo TS is 0.9 kcal/mol higher in energy than the endo-TS (TS(CPD-propene)endo). The corresponding values at the MP2/6-311+G**//B3LYP/6-311+ G** (1.3 kcal/mol) and CCSD[T]/cc-pVTZ//B3LYP/6-311+G** (1.1 kcal/mol) (CCSD[T]/cc-pVTZ single points were carried out using MOLPRO ab initio package, version 2002.2.) are also in agreement with the B3LYP energy difference (0.9 kcal/mol). Although the NICS and L for the exo and endo TSs only differ modestly (Table 2), they are consistent with the energies in supporting the existence of SOI in TS(CPD-propene)endo. The TS(CPD-propene)endo HOMO (Scheme 3) depicts the SOI interactions. The SOI for Apeloigs TS(CPD cyprop)endo10 does not involve the conventional WH type -orbitals of the dienophile. Gleiter argued that SOI consists of interactions between the frontier orbitals.33 By symmetry, hyperconjugation in cyclopropene can only involve the interaction between the *CH2 (antibonding) orbital and the bonding  C C MO. This enhances the ability of the methylene orbitals to engage in stabilizing SOIs with the p orbitals of CPD (shown by dotted lines in Scheme 3). Both methylene groups adjacent to the double bonds in cyclobutene and cyclopentene participate in SOI and favor the endo-TSs in DA reactions with CPD.
Cycloaddition of CPD with Maleic Anhydride

in WHs TS(s-BD a-BD)endo, how can the stabilization of TS(CPD malanhyd)endo be rationalized? Similar to that in TS(sBD a-BD)endo, the HOMO (and the lower orbitals) of TS(CPD malanhyd)endo does not show any orbital coefcients on the C7 of the dienophile. However, HOMO-3, Figure 9, (and HOMO-1, not shown) in TS(CPD malanhyd)endo indicate a favorable interaction between the lone-pairs of the oxygen of the maleic anhydride and the C2C3 -electron cloud of the CPD (see the dotted line shown in HOMO-3 in Fig. 9).
Changes in Magnetic Properties along IRCs

Experimentally, the DA reaction of CPD with maleic anhydride proceeds with 98.5% preference for the endo product.28 At B3LYP/6-31G*, Cossio et al. found that the endo-TS was 1.0 kcal/mol (in isolation) and 1.12 kcal/mol (in benzene) more stable than the exo-TS.11 We obtained 1.3 kcal/mol (in isolation) at B3LYP/6-311+G**, and our endo-(TS(CPD malanhyd)endo) and the exo-TS (TS(CPD malanhyd)exo) geometries (Fig. 9) are similar to those reported earlier.11,34 According to Cossios detailed analysis using second-order perturbation theory, SOI do exist and are responsible for at least an important part of the observed stereocontrol. The computed NICS and L (Table 2) support the existence of SOI in TS(CPD malanhyd)endo. Although small, the differences in the magnetic properties differences between the endo- and exo-TS are consistent with the modest (1 kcal/mol) SOI effect deduced by Cossio.11 Conventionally SOI in both TS(s-BD a-BD)endo and TS(CPD malanhyd)endo are depicted as interactions between the -orbitals of C7C2. 2,9o Since our analysis show that SOI effect is absent

It can be argued that the more diatropic (favorable) NICS(0) and NICS() values for the endo-TSs might be related to relative earliness/lateness of the TSs. Herges, Jiao, and Schleyer have investigated the changes in energetic and magnetic properties along the reaction pathways of a series of pericyclic processes.15c,17f This work established that such TSs are aromatic due to large ring currents, and have more negative s than those of the reactants or products. We applied their approach and computed the energies, NICS(0), NICS(), and  along the IRCs of two reactions: (1)the CPD dimerization (Fig. 10) and (2) the cycloaddition of s-BD with CBD (Fig. 11). The latter reaction involves loss of the antiaromatic character of CBD and the magnetic property changes in cycloaddition reactions. The results veried that SOI was responsible for the differences in magnetic properties between the endo- and exo-TSs. In accord with Carmella et al.s computed potential energy surface (PES) for CPD dimerization,12 the energy of the endoTS is lower than the exo-TS but the exo-product is more stable than the endo (Fig. 10A). Note that the switch in the exoendo preference occurs quite far away from the TS. In contrast, the exoendo differences in computed magnetic properties along both IRCs (Figs. 10B and 10C) do not show such cross-overs: NICS and s have more diatropic values at all points along the endo IRCs than along the exo IRCs. The enhanced magnetic properties of the endo-TSs provide compelling evidence on the existence of SOIs. Other possible explanations, like earliness, lateness, or asynchronicity of the TSs, are not supported by these plots. The large and negative NICS in Figure 10B documents the larger delocalization of the mobile  electrons on the pathway, leading to the endo-TS. As found earlier,15c,17f the energetic TSs do not correspond exactly to the magnetic minima, since other effects help determine the energies. The exoIRCs in Figures 10B and 10C show that the favorable (most negative)  and NICS values correspond to geometries that are slightly away from the TS. We examined the cycloaddition of s-BD and CBD similarly as a special case. The PESs, Figure 11A, leading to both the exo- and the endo-products, have very low barriers but are qualitatively similar to Figure 10A; note the exoendo crossover. The large exothermicity of cycloaddition of s-BD and CBD (Fig. 11A) is due to the loss of the antiaromatic character of CBD (Fig. B in the SI plots the decreases in NICS(0) of the CBD moiety along the IRC). The behavior of magnetic properties along the IRC (Fig. 11B) for the cycloaddition of s-BD and CBD is complex. The decrease in NICS(0) and NICS() along the IRC (Fig. 11B), as well as the DOI 10.1002/jcc

Journal of Computational Chemistry

358

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

Figure 10. (A) Relative energies (B) NICS(0) and NICS() computed at the geometric center of the heavy atoms forming the TS (C) MS, , along the B3LYP/6-31G* IRC for the TSs (endo- and exo-) involved in the dimerization of CPD. NICS() refers to the total contributions of four LMOs involving the eight electrons.

Journal of Computational Chemistry

DOI 10.1002/jcc

Existence of Secondary Orbital Interactions

359

Figure 11. (A) Relative energies (B) NICS(0) and NICS() (C) MS, , along the B3LYP/6-31G* IRC for the TSs (endo- and exo-) involved in the DA reaction of butadiene with CBD. NICS() refers to the total contributions of four LMOs involving the eight electrons.

Journal of Computational Chemistry

DOI 10.1002/jcc

360

Wannere et al. Vol. 28, No. 1 Journal of Computational Chemistry

decrease of  (Fig. 11C), are dominated by the loss of the CBD antiaromaticity. The most negative values are far away from the TS towards the product, where the paratropic character of the four-membered ring is signicantly reduced. The plots in Figures 11B and 11C also show that the endo-TS are favored considerably over the exo-TSs at all points along the IRCs. Although the magnetic properties are dependent on structural changes, these results further conrm that the earliness or lateness of endo-TSs is not the cause of the exalted  or NICS. The endo-TS exo-TS energy differences (Table 2) of CPD cycloadditions with alkenes, like those in Table 1, generally are small (<2 kcal/mol). Moreover, the alkene endo-TS preferences (Table 2) may be due, in addition to SOI, to the following: 1. unfavorable steric repulsions in the exo-TSs (Garcias argument3 and also shown as MP2 vs. B3LYP energy difference, Erel in Table 2). 2. attractive dispersion effects in the endo-TSs, between alkyl (CH2) substituents and the CPD  system (note the polarization energy term from Morokuma decomposition analysis in Table C of the SI). These two effects, if operative, will reduce the SOI contribution to Erel in Table 2. However, because of the rotational mobility of the methyl group in the propene TSs, repulsive interactions can be avoided (Fig. 7). The exoendo difference in the van der Waals attraction also is likely to be negligible. Consequently, the 1.1 kcal/mol at the CCSD[T]/cc-pVTZ (0.9 kcal/mol at B3LYP/6-311+G**) energetic preference of TS(CPD propene)endo serves as a lower bound for the magnitude of SOI in the cycloadditions of CPD with cycloalkenes. As is indicated by the visual depiction of the operation of SOI in Scheme 3, propene exhibits the smallest orbital interaction. Note that the energetic and magnetic evidence is even more compelling for the existence of SOI in cycloadditions of unsaturated systems (Table 1 as well as Figs. 17), where SalemHouk orbital interactions are involved.

require the dienophile (as well as the diene) to have a syn-conformations so that the Alder-Stein accumulation of double bonds is maximum. Endo preference and SOI also are manifested when cyclopropene, cyclobutene, cyclopentene, and even propene serve as the dienophiles with CPD. The somewhat larger endo preference of cyclopropene DA reaction may be due, in part, to nonbonded H. . .H repulsion in the competing exo-TS. The endo-TS in the cycloaddition of aziridine (conformation with NH outside) and CPD is still preferred even though HH repulsions are not present in the exo-TS. Consistent with the energies, the magnetic properties are marginally favorable for the endo-TS. The magnetic properties plotted along the reaction coordinates show, as expected, that the TSs are aromatic and have more exalted  and NICS values than the reactants or products. However, these values are more diatropic for endo-TSs than for exo-TSs. This is not only true at the TSs but also at all points on the IRCs we examined. The IRC plots and the endo-TS preferences further establish the existence of SOIs.

Acknowledgments
Dedicated with warmth to Prof. G. Frenking on the occasion of his 60th birthday. We also thank him for helpful discussions. CSW thank Andy Simmonett for his constructive suggestions.

References
1. (a) Alder, K.; Stein, G. Justus Liebigs Ann Chem 1934, 514, 1; (b) Alder, K.; Stein, G. Angew Chem 1937, 50, 510; (c) Alder, K. Ann Chem 1951, 571 87; (d) Martin, J. G.; Hill, R. K. Chem Rev 1961, 61 537. 2. (a) Hoffmann, R.; Woodward, R. B. J Am Chem Soc 1965, 87, 4388; (b) Woodward, R. B.; Hoffmann, R. The Conservation of Orbital Symmetry; Academic Press: New York, 1969. a, J. I.; Mayoral, J. A.; Salvatella, L. Acc Chem Res 2000, 33, 3. Garc 658. 4. (a) Wassermann, A. J Chem Soc 1935, 828; (b) Wassermann, A. J Chem Soc 1935, 1511; (c) Wassermann, A. J Chem Soc 1936, rez, D.; Sordo, J. A. Chem Commun 1998, 385. 432. (d) Sua 5. Woodward, R. B.; Baer, H. J Am Chem Soc 1944, 66, 645. 6. (a) Blokzijl, W.; Blandamer, M. J.; Engberts, J. B. F. N. J Am Chem Soc 1991, 113, 4241; (b) Gajewski, J. J. J Org Chem 1992, 57, 5500. 7. (a) Berson, J. A.; Hamlet, Z.; Mueller, W. A. J Am Chem Soc 1962, pez, M. F.; Assfeld, X.; Garc a, J. I.; Mayoral, J. 84, 297; (b) Ruiz-Lo A.; Salvatella, L. I. J Am Chem Soc 1993, 115, 8780; (c) Sustmann, R.; Sicking, W. Tetrahedron 1992, 48, 10293; (d) Karcher, K.; Sicking, W.; Sauer, J.; Sustmann, R. Tetrahedron Lett 1992, 33, 8027. 8. (a) Salem, L. J Am Chem Soc 1968, 90, 553; (b) Houk, K. N. Tetrahedron Lett 1970, 2621; (c) Kobuke, Y.; Fueno, T.; Furukawa, J. J Am Chem Soc 1970, 92, 6548; (d) Kobuke, Y.; Sugimoto, T.; Furukawa, J.; Fueno, T. J Am Chem Soc 1972, 94, 3633; (e) Houk, K. N.; Strozier, R. W. J Am Chem Soc 1973, 95, 4094; (f) Mellor, J. M.; Webb, C. F. J Chem Soc Perkin Trans 2 1974,17; (g) Cantello, B. C. C.; Mellor, J. M.; Webb, C. F. J Chem Soc Perkin Trans 2 1974, 22; (h) McCarrick, M. A.; Wu, Y.-D.; Houk, K. N. J Am Chem Soc 1992, 114, 1499; (i) McCarrick, M. A.; Wu, Y.-D.; rez, D.; Sordo, Houk, K. N. J Org Chem 1993, 58, 3330; (j) Sua rez, D.; T. L.; Sordo, J. A. J Am Chem Soc 1994, 116, 763; (k) Sua

Conclusions
Alder and Steins endo rule1 based on the maximum accumulation of double bonds implies that attractive effects are responsible for the observed DA stereoselectivity. Woodward and Hoffmanns secondary orbital interaction (SOI) theory provided an appealing explanation; it has received support, but steric hindrance, attractive van der Waals interactions, and other inuences have been proposed as alternatives. SOI should enhance the degree of cyclic electron delocalization in endo, relative to exo TSs, but steric and other effects should not. Consequently, we compared the MS and NICS behavior of these TSs as well as the related Cope TS in order to evaluate the degree of cyclic electron delocalization and thus detecting SOI. Although these magnetic criteria are shown to provide rm support for SOI, the original example employed illustratively by WH (the DA dimerization involving the synand anti-conformations of BD, the latter as dienophile) proves not to exhibit endoexo SOI differences. Manifestations of SOI

Journal of Computational Chemistry

DOI 10.1002/jcc

Existence of Secondary Orbital Interactions

361

9.

10.

11. 12. 13. 14. 15.

16.

17.

18.

lez, J.; Sordo, T. L.; Sordo, J. A. J Org Chem 1994, 59, 8058; Gonza rez, D.; Lo pez, R.; Gonza lez, J.; Sordo, T. L.; Sordo, J. A. Int (l) Sua J Quantum Chem 1996, 57, 493; (m) Domingo, L. R.; Picher, M. T.; z, J.; Safont, V. S. J Org Chem 1997, 62, 1775; (n) ManoAndre haran, M.; Venuvanalingham, P. Int J Quantum Chem 1998, 66 309. (a) Alston, P. V.; Ottenbrite, R. M.; Cohen, T. J Org Chem 1978, 43 1864; (b) Birney, D. M.; Houk, K. N. J Am Chem Soc 1990, 112, 4127; (c) Ogawa, A.; Fujimoto, H. Tetrahedron Lett 2002, 43, 2055; (d) Toma, L.; Quadrelli, P.; Caramella, P. Tetrahedron Lett 2001, 42, 731; (e) Xidos, J. D.; Gosse, T. L.; Burke, E. D.; Poirier, R. A.; rez, D.; Burnell, D. J. J Am Chem Soc 2001, 123, 5482; (f) Sua Sordo, J. A. Chem Commun 1998, 385; (g) Imade, M.; Hirao, H.; Omoto, K.; Fujimoto, H. J Org Chem 1999, 64, 6697; (h) Gleiter, R.; Bohm, M. C. Pure Appl Chem 1983, 55, 237; (i) Sodupe, M.; Rios, R.; Branchadell, V.; Nocholas, T.; Oliva, A.; Dannenberg, J. J. rez, D. J Am J Am Chem Soc 1997, 119, 4232; (j) Calvo L. S.; Sua Chem Soc 2000, 122, 390; (k) Quadrelli, P.; Romano, S.; Toma, L.; Caramella, P. J Org Chem 2003, 68, 6035; (l) Quadrelli, P.; Romano, S.; Toma, L.; Caramella, P; Tetrahedron Lett 2002, 43, 8785; (m) Houk, K. N.; Li, Y.; Evanseck, J. D. Angew Chem Int Ed Engl 1992, 31, 682; (n) Quadrelli, P.; Romano, S.; Toma, L.; Caramella, P. J Org Chem 2003, 68; 6035; (o) Fleming, I. Pericyclic Reactions; Oxford University Press: Oxford, 1999; (p) Mehta, G.; Uma, R. Acc Chem Res 2000, 33, 278; (q) Ginsburg, D. Tetrahedron 1983, 39, 2095; (r) Sauer, J.; Sustmann, R. Angew Chem 1980, 19, 779; (s) Oppolzer, W. In Comprehensive Organic Synthesis; Paquette, L. A., Ed.; Pergamon Press: Oxford, 1991. (a) Apeloig, Y.; Matzner, E. J Am Chem Soc 1995, 117, 5375; (b) Apeloig, Y.; Arad, D. In 9th IUPAC Conference on Physical Organic Chemistry, Regensburg, Germany, 1988. Arrieta, A.; Cossio, F. P.; Lecea, B. J Org Chem 2001, 66, 6178. Caramella, P.; Quadrelli, P.; Toma, L. J Am Chem Soc 2002, 124, 1130. Evans, M. G.; Warhurst, E. Trans Faraday Soc 1938, 34, 614. Evans, M. G. Trans Faraday Soc 1939, 35, 824. (a) Jiao, H.; Schleyer, P. v. R. Angew Chem Int Ed Engl 1993, 32, 1763; (b) Jiao, H.; Schleyer, P. v. R. J Chem Soc Perkin Trans 2 1994,407; (c) Herges, R.; Jiao, H.; Schleyer, P. v. R. Angew Chem Int Ed Engl 1994, 33, 1376; (d) Jiao, H.; Schleyer, P. v. R. J Chem Soc Faraday Trans 1994, 90, 1559; (e) Jiao, H.; Schleyer, P. v. R. J Am Chem Soc 1995, 117, 11529; (f) Jiao, H.; Schleyer, P. v. R. Angew Chem Int Ed Engl 1994, 34, 334. (a) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.; Hommes, N. J. R. van E. J Am Chem Soc 1996, 118, 6317; (b) Schleyer, P. v. R.; Jiao, H.; Hommes, N. J. R. v. E.; Malkin, V. G.; Malkina, O. L. J Am Chem Soc 1997, 119, 12669; (c) Schleyer, P. v. R.; Manoharan, M.; Wang, Z. X.; Kiran, B.; Jiao, H.; Puchta, R.; Hommes, N. J. R. v. E. Org Lett 2001, 3, 2465; (d) Chen, Z.; Wannere, C. S.; Corminboeuf, C.; Puchta, R.; Schleyer, P. v. R. Chem Rev 2005, 105, 3842. (a) Jiao, H.; Schleyer, P. v. R. Angew Chem Int Ed Engl 1996, 35, 2383; (b) Subramanian, G.; Schleyer, P. v. R.; Jiao, H. Angew Chem Int Ed Engl 1996, 35, 2638; (c) Subramanian, G.; Schleyer, P. v. R.; Jiao, H. Organometallics 1997, 16, 2362; (d) Jiao, H.; Schleyer, P. v. R.; Mo, Y.; Mc Allister, M. A.; Tidwell, T. T. J Am Chem Soc 1997, 119, 6561; (e) Schleyer, P. v. R.; Manoharan, M.; Jiao, H.; Stahl, F. Org Lett 2001, 3, 3643; (f) Jiao, H.; Schleyer, P. v. R. J Phys Org Chem 1998, 11, 655. (a) Dauben, H. J.; Wilson, J. D., Jr.; Laity, J. L. Nonbenzoid Aromaticity, Vol. 2; Snyder, J. P. Ed.; Academic Press: New York, 1971; (b) J Am Chem Soc 1968, 90, 811; (c) J Am Chem Soc 1991, 91, 1969; (d) Jiao,

19.

20.

21. 22.

23.

24.

25. 26.

27.

28. 29. 30. 31. 32. 33. 34.

H.; Schleyer, P. v. R. Angew Chem Int Ed Engl 1993, 32, 1763; (e) Jiao, H.; Schleyer, P. v. R. J Am Chem Soc 1995, 117, 11529; (f) Jiao, H.; Schleyer, P. v. R. Angew Chem Int Ed Engl 1994, 34, 334. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A. Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 1998; Gaussian, Inc.: Pittsburgh PA, 1998. (a) Keith, T. A.; Bader, R. F. W. Chem Phys Lett 1993, 210, 223; (b) Cheeseman, J. R.; Frisch, M. J.; Trucks, G. W.; Keith, T. A. J Chem Phys 1995, 104, 5497. Wolinski, K.; Hilton, J. F.; Pulay, P. J Am Chem Soc 1990, 112, 8251. (a) St-Amant, A.; Salahub, D. R. Chem Phys Lett 1990, 169, 387; (b) Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Can J Chem 1992, 70, 560; (c) Salahub, D. R.; Fournier, R.; Mlynarsky, P.; Papai, I.; St-Amant, A.; Ushio, J.In Density Functional Methods in Chemistry; Labanowsky, J.; Andzelm, J., Eds.; Springer: New York, 1991; p. 77. (a) Malkin, V. G.; Malkina, O. L.; Casida, M. E.; Salahub, D. R. J Am Chem Soc 1994, 116, 5898; (b) Malkin, V. G.; Malkina, O. L.; Eriksson, L. A.; Salahub, D. R. In Modern Density Functional Theory; Seminario, J. M.; Politzer, P., Eds.; Elsevier: Amsterdam, 1995; p. 273. (a) Perdew, J. P.; Wang, Y. Phys Rev B 1992, 45, 13244; (b) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.; Fiolhais, C. Phys Rev B 1992, 46, 6671. Kutzelnigg, W.; Fleisher, U.; Schindler, M. In NMR-Basic Principles and Progress,Vol. 23; Spinger: Heidelberg, 1990; p. 165. (a) Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.; Bartberger, M. D.; Houk, K. N. J Phys Chem A 2003, 107, 11445; (b) Wannere, C. S.; Bansal, R. K.; Schleyer, P. v. R. J Org Chem 2002, 67, 9162. (a) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic Chemistry, 3rd ed.; Plenum Press: New York, 1987; p.925; (b) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry, Part A: Structure and Mechanisms, 3rd ed.; Plenum Press: New York, 1991; p. 851; (c) March, J. Advanced Organic Chemistry-Reaction Mechanisms and Structure, 4th ed.; Wiley: New York, 1992; pp. 851853; (d) Vollhardt, K. P. C.; Schore, N. E. Organic Chemistry-Structure and Function, 3rd ed.; Freeman: New York, 1999; pp. 605607. Stephenson, L. M.; Smith, D. E.; Current, S. P. J Org Chem 1982, 47, 4170. Cristol, S. J.; Seifert, W. K.; Soloway, S. B. J Am Chem Soc 1960, 82, 2351. Li Y.; Houk, K. N. J Am Chem Soc 1996, 118, 880. Wannere, C. S.; Moran, M.; Allinger, N. L.; Hess, B. A., Jr.; Schaad, L. J.; Schleyer, P. v. R. Org Lett 2003, 5, 2983. (a) Wiberg, K. B.; Barley, W. J. J Am Chem Soc 1960, 82, 6375; (b) Baldwin, J. E.; Reddy, V. P. J Org Chem 1989, 54, 5264. hm, M. C. Pure Appl Chem 1983, 55, 237. Gleiter, R.; Bo Birney, D.; Lim, T. K.; Koh, J. H. P.; Pool, B. R.; White, J. M. J Am Chem Soc 2002, 124, 5091.

Journal of Computational Chemistry

DOI 10.1002/jcc