This article was published as part of the

2009 Metalorganic frameworks issue

Reviewing the latest developments across the interdisciplinary area of
metalorganic frameworks from an academic and industrial perspective
Guest Editors Jeffrey Long and Omar Yaghi
Please take a look at the issue 5 table of contents to access
the other reviews.
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online / Journal Homepage / Table of Contents for this issue
Thin lms of metalorganic frameworksw
Denise Zacher,
a
Osama Shekhah,
b
Christof Wo ll
b
and Roland A. Fischer*
a
Received 6th October 2008
First published as an Advance Article on the web 9th March 2009
DOI: 10.1039/b805038b
The fabrication of thin lm coatings of metalorganic frameworks (MOFs) on various substrates is
discussed in this critical review. Interestingly, the relatively few studies on MOF lms that have
appeared in the literature are limited to the following cases: [Zn
4
O(bdc)
3
] (MOF-5;
bdc = 1,4-benzenedicarboxylate), [Cu
3
(btc)
2
] (HKUST-1; btc = 1,3,5-benzenetricarboxylate),
[Zn
2
(bdc)
2
(dabco)] (dabco = 1,4-diazabicyclo[2.2.2]octane), [Mn(HCOO)], [Cu
2
(pzdc)
2
(pyz)] (CPL-1;
pzdc = pyrazine-2,3-dicarboxylate, pyz = pyrazine), [Fe(OH)(bdc)] (MIL-53(Fe)) and
[Fe
3
O(bdc)
3
(Ac)] (MIL-88B; Ac = CH
3
COO

). Various substrates and support materials have been

used, including silica, porous alumina, graphite and organic surfaces, i.e. self-assembled monolayers
(SAMs) on gold, as well as silica surfaces. Most of the MOF lms were grown by immersion of the selected
substrates into specically pre-treated solvothermal mother liquors of the particular MOF material. This
results in more or less densely packed lms of intergrown primary crystallites of sizes ranging up to several
mm, leading to corresponding lm thicknesses. Alternatively, almost atomically at and very homogenous
lms, with thicknesses of up to ca. 100 nm, were grown in a novel stepwise layer-by-layer method. The
individual growth steps are separated by removing unreacted components via rinsing the substrate with the
solvent. The layer-by-layer method oers the possibility to study the kinetics of lm formation in more
detail using surface plasmon resonance. In some cases, particularly on SAM-modied substrates, a highly
oriented growth was observed, and in the case of the MIL-53/MIL-88B system, a phase selective deposition
of MIL-88B, rather than MIL-53(Fe), was reported. The growth of MOF thin lms is important for smart
membranes, catalytic coatings, chemical sensors and related nanodevices (63 references).
1. Introduction
Smart membranes, catalytic coatings, chemical sensors,
and many other related nanotechnological devices and
applications depend on the fabrication of thin lms and
coatings of dened porosity, combined with tuneable chemical
functionality. Zeolites, organic polymers, metal oxides
and activated carbon are the typical materials used for this
purpose. Zeolites and related siliceous materials have interest-
ing properties in comparison with the others, namely very well
dened, highly regular pore structures, and often exceptionally
high chemical and thermal stabilities. However, the range of
control of functionality on a molecular level is nevertheless
limited with zeolites and its inorganic congeners. Coordination
a
Anorganische Chemie II-Organometallics and Materials,
Ruhr-Universitat Bochum, D-44780 Bochum, Germany.
E-mail: roland.scher@rub.de; Fax: +49 234 321 4174;
Tel: +49 234 322 4174
b
Physikalische Chemie 1, Ruhr-Universitat Bochum,
D-44780 Bochum, Germany
w Part of the metalorganic frameworks themed issue.
Denise Zacher
Denise Zacher, born 1981,
obtained her BSc at the
Universiat Duisburg-Essen
and received her MSc in
chemistry at the Ruhr-
Universitat Bochum in 2006.
She is currently doing her
doctoral thesis under the
supervision of Roland A.
Fischer on the growth of
MOF thin lms and nano-
crystals, within the frame of
the Priority Programm 1362
Metal Organic Frameworks
of the German Research Foun-
dation (DFG). Her work is
also associated with the specic targeted research project
SURMOF of the European Union (6th FP).
Osama Shekhah
Osama Shekhah, born 1973,
studied chemistry in Jordan.
He received his PhD in the
eld of physical and surface
chemistry in 2001 from the
Freie Universitat Berlin under
the supervision of Prof. R.
Schlogl. He then worked for
one year as a postdoc at the
Fritz-Haber institute in Berlin.
In 2005 he joined the physical
chemistry group of Prof. C.
Woll at the Ruhr-Universitat
Bochum as a sta scientist.
His research interests include
surface chemistry, self-
assembled monolayers, and metalorganic frameworks and their
growth on organic surfaces.
1418 | Chem. Soc. Rev., 2009, 38, 14181429 This journal is c The Royal Society of Chemistry 2009
CRITICAL REVIEW www.rsc.org/csr | Chemical Society Reviews
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
polymers with permanent porosity (PCPs), or so-called metal
organic frameworks (MOFs), are hybrid inorganicorganic
solid state compounds with zeolite-like structures and
properties, but transgressing the limitations of zeolites in terms
of materials chemistry, at least in principle.
1
A large body of
research on MOFs is directed to unravel the rules of reticular
synthesis and to develop the tool box needed for a design of
MOFs with desired properties.
25
Consequently, the work in
this area concentrates on the synthesis of novel MOFs and
structure/property/function correlations, in particular looking
at responsive, exible frameworks and introducing interesting
chemical functionality.
68
This is also reected by the
collection of articles in this themed issue of Chem. Soc. Rev.
on metalorganic frameworks. Recently, Kitagawa and
Matsuda pointed out: Researchers control the size, shape,
and distribution of pores and will establish this engineering in
the near future. However, even when they have nanosized
channels or cavities, compounds are at least mm-sized micro-
crystals, insoluble in any solvent and therefore hard to prepare
in a thin layer form. . . The ultimate goal is the ability to
control the arrangements of channels of porous modules for
various nanodevices.
8
Thus, there are two challenges: the
deposition or growth of MOF thin lms on substrates, ideally
in a dense, homogeneous and oriented fashion, and also the
preparation of size-, shape- and surface-functionalized MOF
nanocrystals, which can act as wells, wires, rods and dots.
Both areas are, however, connected and the general principles
of how to approach these problems are similar to the related
research on zeolite thin lms
912
and zeolite nanocrystals.
13,14
This review will discuss the most recent progress in the
fabrication of MOF thin lms. It should be noted that the
specic topic of MOF thin lms is closely related to the more
general eld of bottom-up synthesis of functional coatings on
surfaces. In particular, the controlled growth of non-porous
coordination polymers or related inorganicorganic hybrid
structures on surfaces aimed at specic functions is an
important area, but cannot be covered in this review here.
Some reference is, however, given in sections 4 and 5 below.
There are three dierent concepts of MOF thin lm fabrica-
tion: (A) the direct growth/deposition from solvothermal
mother solutions, (B) the assembly of preformed, ideally size
and shape selected, nanocrystals and (C) the stepwise layer-
by-layer growth onto the substrate. To the best of our
knowledge, MOF thin lms have only been manufactured by
employing methods of type A and type C so far. Some very
limited research has been carried out on MOF nanocrystals
and/or MOF colloids, of which the results are reviewed by
Spokoyny et al. in this themed issue of Chem. Soc. Rev..
15
The
inuence of the surface chemistry (functionality) of the chosen
substrate and the use of MOF seeds on the nucleation,
orientation, as well as on the adhesion of the MOF lms,
are important factors to be addressed. In particular,
self-assembled organic monolayers (SAMs)
16
have been used
to direct the nucleation, orientation and structure of the
deposited MOFs. The control of crystallisation of inorganic
solid state materials by the inuence of organic macro-
molecules is the underlying principle of biomineralisation,
17
which can be transferred to MOFs as inorganicorganic
hybrid polymers. SAMs may be used as a well-dened
articial organic interface that mimics the structure-directing
power of the complex biointerfaces, and the oriented
growth of zeolites on substrates has been demonstrated by
using SAMs.
18
Interestingly, the relatively few studies on MOF lms
are limited to the following: [Zn
4
O(bdc)
3
] (MOF-5; bdc =
1,4-benzenedicarboxylate),
1921
[Cu
3
(btc)
2
] (HKUST-1;
btc = 1,3,5-benzenetricarboxylate),
2225
[Zn
2
(bdc)
2
(dabco)]
(dabco = 1,4-diazabicyclo[2.2.2]octane),
22
[Mn(HCOO)],
26
[Fe(OH)(bdc)] (MIL-53(Fe)) and [Fe
3
O(bdc)
3
(Ac)] (MIL-88B;
Ac = CH
3
COO

).
27
Some of the most recent new reports,
Christof Wo ll
Christof Woll studied Physics
in Gottingen and received his
PhD in 1987 under the
guidance of Peter Toennies in
the eld of surface science.
After a postdoctoral stay
at the IBM Research Lab,
Almaden, USA (19881989),
he joined Heidelberg Univer-
sity, where he obtained
his Habilitation in 1992
and was then appointed
Hochschulassistent (Assistant
Prof.). In 1994 he was
awarded a Heisenberg Fellow-
ship from the German DFG.
Since 1997 he has held a chair in Physical Chemistry at Ruhr
Universitat, Bochum. His current research focuses on the
chemistry of oxide surfaces, heterogeneous catalysis, organic
molecular beam deposition, organic surfaces exposed by self-
assembled monolayers, the nucleation of metalorganic frame-
works, as well as photoelectron spectroscopy and scanning
tunnelling microscopy.
Roland A. Fischer
Roland A. Fischer studied
Chemistry at the Technische
Universitat Munchen (TUM)
and received his Dr rer. nat. in
1989 under the guidance of
Wolfgang A. Herrmann. After
a postdoc with Herb Kaesz at
UCLA, he returned to TUM
in 1990, obtained his Habilita-
tion in 1995 and was appointed
Associate Prof. at the
Ruprecht-Karls Universitat,
Heidelberg in 1996. He moved
to Ruhr Universitat, Bochum
in 1998 for a chair in Inorganic
Chemistry. Currently he is
Dean of the Ruhr Universitat Research School. His research
interests focus on group 13/transition metal bonds and clusters,
precursor chemistry for inorganic materials, chemical vapour
deposition (CVD), thin lms, nanoparticles, colloids and host
guest chemistry of porous coordination polymers (MOFs).
This journal is c The Royal Society of Chemistry 2009 Chem. Soc. Rev., 2009, 38, 14181429 | 1419
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
which the authors became aware of during editing the
manuscript, are cited in section 5 (Conclusions and perspec-
tives). The deposition of non-porous coordination polymers in
the form of coatings or free-standing thin lms are not
included in this review.
28
We will start our discussion with a
summary of the growth of MOFs on bare oxidic substrates,
such as alumina, silica and carbon.
2. Deposition of MOF coatings on alumina, silica
and carbon substrates
Hermes et al. described the deposition of MOF-5 micro-
crystals (510 mm) on c-plane sapphire at 25 1C over 24 h
using a cooled, aged solvothermal mother solution (in N,N-
diethylformamide (DEF)).
19
The density of nucleation and/or
anchoring sites for the seeds and growing MOF crystals is low
on very smooth (ideally atomically at) and defect-free
surfaces. The deposition of an amorphous Al
2
O
3
-type buer
layer onto the c-plane sapphire substrate by means of ALD
(atomic layer deposition)
20
leads to a denser coating with
intergrown MOF-5 crystals (Fig. 1). Interestingly, it was not
possible to deposit MOF-5 on silica substrates (e.g. the native
SiO
2
coating on a silicon wafer) by using the same
conditions.
20,21
A related study by Zacher et al. on the
deposition of HKUST-1 at 120 1C directly from the solvo-
thermal mother solution (waterethanol) gave quite similar
results (Fig. 1).
22
The striking dierence between SiO
2
and
Al
2
O
3
surfaces was attributed to the isoelectric points of these
materials, and the obvious requirement of a basic, i.e. electro-
statically compatible, surface for anchoring MOF-5
and HKUST-1 under these conditions. Interestingly, the
deposition experiments of [Zn
2
bdc
2
(dabco)], again under
solvothermal conditions at 120 1C in dimethylformamide,
showed no substrate selectivity.
Dense coatings (see Fig. 1) were obtained on SiO
2
and
amorphous Al
2
O
3
as well. The bifunctional nature of this
pillared, layer-based MOF, with acidic bdc linkers and basic
dabco pillars, is favourable for binding to acidic SiO
2
as well
as basic Al
2
O
3
substrates. The chemical and structural identity
of the deposited materials were conrmed by X-ray diraction
studies and comparison with authentic bulk materials. The
adsorption properties of the MOF coatings (after activation)
were qualitatively tested by exposure to a coloured volatile
organometallic compound, as shown in Fig. 2. Up to now,
only three groups have followed these initial studies. Yoo and
Jeong demonstrated the rapid deposition of MOF-5 on
carbon-coated anodic aluminium oxide (AAO) by micro-
wave-induced thermal deposition,
21
Gascon et al. developed
a seeding approach for the deposition of HKUST-1 and
Fig. 1 The deposition of MOF-5, HKUST-1 and [Zn
2
(bdc)
2
(dabco)] on SiO
2
, alumina and COOH/CF
3
-modied surfaces (SAMs). Figures
adapted from ref. 20 with permission, Copyright 2007, American Chemical Society and ref. 22 with permission, Copyright 2007, RSC Publishing.
Fig. 2 A qualitative demonstration of the adsorption properties of
MOF-5 coatings on alumina: optical images (digital photographs) of
an empty 5 mm thick MOF-5 thin lm on sapphire substrates before
(left) and after (middle) exposure to the vapour of the deep red
MOCVD precursor [(Z
5
-C
5
H
5
)Pd(Z
3
-C
3
H
5
)]. Subsequent treatment
of the loaded lm with UV light converts [(Z
5
-C
5
H
5
)Pd(Z
3
-C
3
H
5
)]@
MOF-5 into Pd@MOF-5, visible by the colour change to deep black
(right). Reproduced from ref. 20 with permission, Copyright 2007,
American Chemical Society.
1420 | Chem. Soc. Rev., 2009, 38, 14181429 This journal is c The Royal Society of Chemistry 2009
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
obtained very dense coatings,
25
and Arnold et al. studied the
oriented crystallisation of [Mn(HCOO)
2
] on porous alumina,
as well as on graphite.
26
The microwave method required a
conductive surface; there was little MOF-5 deposition on bare
AAO, but a coating of well-developed cubic crystallites of
14 mm on amorphous carbon/AAO and a dense, smooth
coating of smaller crystallites (7 mm) on graphite/AAO. The
higher nucleation density on graphite/AAO was attributed to
the better absorption of the microwaves (i.e. local heating).
The authors speculated about the origin of the observed
strong adhesion and the preferred orientation with the [10-2]
direction perpendicular to the substrate. The intense micro-
wave Joule heating may cause the formation of
carboxylic acid groups on the graphite surface under the
conditions of the experiment, which is benecial for both
adhesion and orientation. However, the authors provided no
analytical evidence for that idea. At least MOF-5 was shown
to preferentially bind to COOH-functionalized SAMs.
19
The
adhesion of MOF-5 to graphite/AAO was tested by using a
sonication method. About 80% of the coating remained after
sonication for 1 h. It should be noted here that rapid synthesis
of MOFs using rather highly concentrated solutions, and in
particular microwave assisted heating, may be complicated by
the adsorption of unreacted metal species, organic linkers and
precursors of the secondary building units (SBUs) in the
cavities of the growing MOF. A detailed study on microwave
assisted synthesis procedures of bulk samples of MOF-5 was
published by Hazovic et al.,
29
also showing that rapid
synthesis may cause the formation of an interpenetrated phase
of MOF-5. However, Yoo and Jeong did not provide any
porosity or adsorption experiments on their MOF-5 lms.
Gascon et al. reported dense coatings of HKUST-1 on
a-alumina supports by a combination of suitable seeding with
low concentration mother liquors. The best results in terms of
thin lm morphology (studied by scanning electron micro-
scopy (SEM), see Fig. 3) were obtained by seeding
(spin coating) with a slurry of the 1D coordination polymer,
catena-triaqua-m-(1,3,5-benzenetricarboxylate)copper(II)
[Cu(Hbtc)(H
2
O)
3
], obtained by modication of the original
HKUST-1 recipe. Using water as the solvent instead of a 1 : 1
ethanolwater mixture, only two of the three carboxylic
groups of the 1,3,5-benzenetricarboxylic acid are deproto-
nated. Dense coatings of small intergrown octahedral micro-
crystals (B2 mm) are formed by immersion of the pre-treated
substrates into a diluted solvothermal mother solution at
110120 1C over a period of 1218 h. Alternatively, slurries
of amorphous so-called proto-HKUST-1 can be used for
seeding purposes to obtain coatings with slightly larger
crystals (B5 mm). This proto form of HKUST-1 precipitates
when concentrated waterethanol solutions of copper(II)-
nitrate and btc are combined at room temperature. Based on
PXRD studies, the authors claimed the coatings consisted of
phase-pure HKUST-1, without preferential orientation. The
authors point out that no traces of the characteristic peaks of
the 1D seeding MOF were detected in the XRD pattern of the
coating, and suggest a rather complete conversion of the
seeding material into the HKUST-1 phase under the
conditions of the experiment. However, the quality of the
published PXRD pattern (broad lines and high background
noise) was surprisingly low compared to the microcrystalline
HKUST-1 reference samples. Also, neither qualitative nor
quantitative data on the adsorption and porosity properties
of the obtained coatings were given in the publication.
The general problem of inhibited nucleation, and thus low
crystal density, on bare alumina and graphite was also
observed in the case of deposition of [Mn(HCOO)
2
] under
solvothermal conditions at 115 1C (1,4-dioxaneDEF).
Arnold et al. reported densities of 10 crystals mm
2
for
alumina and 80 crystals mm
2
for graphite.
26
Electrostatic
repulsion between the substrate (alumina) and the MOF nuclei
and growth species was suggested as the key factor when
[Mn(HCOO)
2
] is synthesized according to the original recipe
by Dybtsev et al.
30
By developing the so-called formate
route (i.e. using sodium formate rather than formic acid) and
oxidized graphite substrates (i.e. creating dierent oxo-
functionalities at the surface), nucleation was enhanced and
rather dense coatings of [Mn(HCOO)
2
] with crystallite sizes
4100 mm were obtained. The important factor for membrane
applications, however, is the orientation of the 1D channels of
the [Mn(HCOO)
2
] structure in a perpendicular, or at least
tilted, fashion with respect to the underlying (meso/macro)
porous substrate. Only in case of the formate route with
functionalized graphite as a substrate were the authors
successful in obtaining a coating with a preferred orientation
and a tilt angle of about 341 of the 1D channel system
(see Fig. 4). The same authors studied the methanol uptake
and anisotropic mass transport on individual single crystals of
Fig. 3 SEM micrographs of HKUST-1 layers obtained under
dierent synthesis conditions according to Gascon et al. Adapted
from ref. 25 with permission, Copyright 2008, Elsevier.
This journal is c The Royal Society of Chemistry 2009 Chem. Soc. Rev., 2009, 38, 14181429 | 1421
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
[Mn(HCOO)
2
] by interference microscopy,
31
but again, data
on the porosity and transport properties of the deposited
MOF coatings were not given.
3. Deposition of MOFs on substrates modied
by self-assembled organic monolayers
The above cited observation of inhibited nucleation of MOFs
on bare metal oxide and graphite supports, together with the
known property of SAMs to direct the nucleation and growth
of zeolites,
18
motivated the rst study of MOF growth at
SAM-modied substrates. Again, MOF-5 was chosen as the
rst test case. By immersion of SAM-modied Au substrates
into a preconditioned, supersaturated mother solution in DEF
at 25 1C, Hermes et al. obtained the selective deposition
of well-shaped MOF-5 cubes of 0.51 mm at the COOH-
terminated sites of the patterned, mixed COOH/CF
3
-
terminated SAM surface.
19
The thiol-type SAMs used on
Au are not thermally robust; thus, the growth studies were
undertaken at room temperature. Note, that MOF-5 can be
synthesized at room temperature in DMF in the presence of
triethylamine as base.
32
In case of the standard solvothermal
recipe in DEF at 105 1C, however, the situation is more
complex, and the thermal decomposition of DEF is an
important process.
33
The obtained MOF-5/SAM coating was
activated by the known process of washing with DEF and
CHCl
3
, and gentle drying in vacuo at elevated temperatures.
The adsorption of an organometallic Pd-precursor from
the gas phase was used as a qualitative test for the
adsorption properties, similarly to the above cited case of
MOF-5/alumina coatings.
20
It is well known, that MOF-5 deposits as perfect cubes of
several mm in length from the standard solvothermal mother
solution (Fig. 5). This perfect cubic shape persists from the
early stages of homogeneous crystal growth after the so far
still unknown nucleation step, as shown by in situ time
resolved light scattering.
34
Also, the Zn
2+
sites in the bulk
structure of MOF-5 are coordinatively saturated, which
possibly limits the inuence of structure- and orientation-
directing additives being present during the crystal growth
and deposition on the chosen substrates. This is dierent for
HKUST-1, which consists of btc-bridged Cu
2
units. These
paddle-wheel type copper dimers exhibit weakly bound water
molecules in the apical position of the Cu
2+
centres, which can
be fully desorbed in a drying step, as well as being reversibly
exchangeable against other (weak) ligands, such as pyridine.
22
Therefore, Biemmi et al. chose HKUST-1 as candidate for a
study aiming at the oriented growth of HKUST-1 on dier-
ently terminated SAMs.
24
They used preconditioned, aged
(8 d, 75 1C) and ltered standard HKUST-1 mother solutions
cooled down to room temperature (presumably containing
only low concentration of the growth species). The growth of
HKUST-1 microcrystals on Au substrates modied with thiol-
based SAMs using HS(CH
2
)
10
COOH, HS(CH
2
)
10
CH
2
OH and
HS(CH
2
)
10
CH
3
was monitored by SEM over a period of 16 to
100 h. The key nding was a highly oriented growth of
individual crystals over time. Under the conditions of the
Fig. 4 SEM micrographs of [Mn(HCOO)
2
] layers on an oxidized graphite support according to Arnold et al. Adapted from ref. 26 with
permission, Copyright 2007, Wiley-VCH.
Fig. 5 MOF-5 deposition of SAM-modied silica/silicon substrates.
1422 | Chem. Soc. Rev., 2009, 38, 14181429 This journal is c The Royal Society of Chemistry 2009
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
experiment, the COOH-terminated SAM favoured orientation
along the [100] direction, which resulted in the formation
of pyramids. The OH-terminated SAM favoured the [111]
direction, which led to the formation of octahedral crystals
resting on one triangular face (see Fig. 6).
Interestingly, the authors observed a growth on CH
3
-
terminated SAMs as well. An even faster growth process
was reported than on the other two polar and coordinatively
more strongly interacting surfaces. However, a strongly
preferred orientation was absent. The growth on CH
3
-
termination is a somewhat surprising result; however,
dispersive forces between the supposedly organic-terminated
crystal faces and the alkyl-terminated SAM may be dominant
in that case. One should note that in the case of CF
3
-
terminated thiol-based SAMs on Au, as well as silane-based
SAMs on SiO
2
/Si substrates (vide infra), neither MOF-5 nor
HKUST-1 could be grown.
19,20,22
Nevertheless, the protocol
to achieve the highly selective growth developed by Bemmi
et al. appears a bit complicated. The authors argue that the
thermal pre-treatment of the synthesis solution (8 d at 75 1C)
induces the crystallization process. After ltration and
removal of the deposited solid product (see also Gascon
et al.), they suggest the existence of colloidal nanocrystals or
small molecular building blocks of HKUST-1 in the possibly
rather diluted solution.
Prolonged exposure of 4100 h of all the SAM-modied
substrates to the growth solution resulted in the formation of
about 600 nm thick lms of intergrown crystals. In parallel,
and independently, Zacher et al. studied the deposition of
HKUST-1 on COOH- and CF
3
-terminated (patterned) silane-
based SAMs on SiO
2
/Si-substrates, as well as on c-plane
sapphire under the usual solvothermal growth conditions of
110120 1C.
22
Silane SAMs are known to be much more
thermally robust than thiol-based SAMs. Again, a highly
selective growth of HKUST-1 on the COOH-terminated sites
of the SAMs was observed. However, the deposited crystals
were orientated along the [111] direction, just the orientation
that Bemmi et al. observed for OH-terminated SAMs, but not
for COOH-termination. Note, that the carboxylic acid termi-
nating groups of the SAM can interact in various ways
with surface-exposed Cu
2
-dimers in order to complete the
paddle-wheel structural motif. In fact, a SAM with COOH-
termination is compatible with both orientations, either along
[100], or along [111] (see Fig. 7). This discrepancy of the two
results on COOH-terminated SAMs may be related to the very
dierent growth conditions; on the one hand the low tempera-
ture growth and possibly rather low concentration of growth
species, and on the other hand, high temperature and higher
concentrations. More detailed studies on the temperature and
concentration dependence of HKUST-1 growth on top of
SAMs are clearly warranted.
In addition to the oriented growth of MOFs on SAMs, the
structure-directing inuence of SAMs on heterogeneous
nucleation was recently demonstrated. Scherb et al. reported
the growth of MIL-88B(Fe) on SAMs of mercaptohexa-
decanoic acid on Au;
27
this is quite similar to the above
outlined study on deposition of HKUST-1/SAM, but now
using a preconditioned mother solution for the MIL(Fe)
synthesis. In the Fe
3+
/bdc system, several dierent MOF
Fig. 6 (a) X-Ray diraction patterns (background corrected) of thin
lms of HKUST-1 on functionalized gold surfaces, compared with a
randomly oriented powder reference sample measurement. Each
pattern is normalized to the most intense reection. Bottom: schematic
illustrations of the oriented growth of HKUST-1 nanocrystals
controlled via surface functionalization; (b) on an 11-mercapto-
undecanoic acid SAM, and (c) on 11-mercaptoundecanol-modied
gold surfaces. The alkanethiol self-assembled monolayers are repre-
sented with a tilt of ca. 301 from the surface normal as reported in the
literature. Figures adapted from ref. 24 with permission, Copyright
2007, American Chemical Society.
Fig. 7 The matching of dierent lattice planes of HKUST-1 with
COOH-terminated SAMs.
22
This journal is c The Royal Society of Chemistry 2009 Chem. Soc. Rev., 2009, 38, 14181429 | 1423
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
structures are known, including the so-called MIL-53(Fe) and
MIL-88B types.
These frameworks are quite exible and the exact cell
parameters strongly depend on the interaction with guest
molecules, which makes these materials particularly interest-
ing candidates for sensors and smart membranes.
35,36
In the
monoclinic structure of [Fe(OH)(bdc)(py)
0.85
], the Fe
3+
version of MIL-53, chains of FeO
6
octahedra are connected
by bdc. Thus, rhombic 1D channels are formed that run along
the a axis of the structure. The hexagonal 3D structure of
MIL-88B is built up from trimers of FeO
6
octahedra linked to
bdc. Thus, the 3D pore system of MIL-88B consists of tunnels
along the c axis connected by bipyramidal cages
(see Fig. 8).
27,37
The SAM-functionalized gold substrates were
placed face-down in a crystallization solution, obtained by the
solvothermal treatment of a synthesis mixture for MIL-53 at
150 1C for 2 d, ltration, and further treatment of the clear
solution at 150 1C for 5 d. The X-ray diraction analysis of the
crystals attached to the SAM revealed a phase-selective
growth of the MIL-88B material, with an orientation along
the [001] direction. During the experiment, however, a
precipitate formed again. This precipitate was identied as
the other phase, pure MIL-53(Fe). The authors discussed this
observation as follows. As indicated in Fig. 8, the structure of
MIL-53(Fe), the product of homogeneous nucleation from the
crystallization solution, diers dramatically from that of
MIL-88B(Fe), the product of heterogeneous nucleation. This
eect is attributed to symmetry transfer, that is, the dierent
symmetry relationships between the COOH-terminated SAM
and the two crystal systems. No crystal growth was observed
with hydroxy- and alkyl-terminated SAMs, or with untreated
gold slides. Presented with a surface exposing (approximately)
hexagonal symmetry, the growth species, e.g. the SBUs and
bdc linkers, clearly prefer to assemble in the form of a
hexagonal MIL-88B structure type instead of monoclinic
MIL-53. However, we suggest using terms homogeneous and
heterogeneous nucleation with care in these cases. Virtually
nothing is known about the composition of the growth
solutions in terms of the molecular species of relevance for
nucleation and further growth. It is possible that nuclei, and
even larger nanocrystals of both MIL species, are formed in
the solution in a homogeneous manner, but only the MIL-88B
types matches to the sticky SAM surface, which then leads
to the observed selectivity. There is no direct evidence for a
heterogeneous mechanism of the nucleation step of
MIL-88B growth on the particular SAM. Also, one should
keep in mind that evidence on the persistence of SBUs during
the formation of MOFs under the typical solvothermal, and
in fact heterogeneous, conditions is quite shallow.
38
The
adsorption properties of the deposited crystals of
MIL-88B(Fe) were tested by taking advantage of the strong
breathing eect of the structure, depending on the accommo-
dation of guest molecules. The samples were exposed to a
vapour phase saturated with DMF for 24 h and the expected
breathing eect upon exposure/drying cycles was observed by
X-ray diraction. Obviously the attachment to the SAM, i.e.
some elasticity of the organic interface to the substrate, can
accommodate the shrinkage and expansion cycles.
The obvious and common drawback of all of the above
mentioned pioneering reports on the deposition of MOFs on
SAMs from somehow pre-treated mother solutions, is that in
fact smooth and dense MOF thin lms were not obtained at
all. Rather scattered, more or less isolated crystals, or island of
crystals, or rough coatings with many cracks were deposited,
even after reaction times of several days.
4. Stepwise layer-by-layer liquid epitaxy of MOFs
In contrast to the established synthesis protocols of MOFs,
where the educts (primary building blocks, typically two) are
mixed and reacted under solvothermal conditions, the stepwise
layer-by-layer growth mode of MOFs as introduced by
Shekhah et al.
23,39
is based on the combination of the reaction
partners in a sequential, stepwise fashion. The individual steps
are separated by removing unreacted components via rinsing
with a solvent, see Fig. 9. The principles of such a layer-
by-layer growth mode of supramolecular architectures on
surfaces is well known,
40
but has only recently been
transferred to the fabrication of MOF thin lms. As an
intriguing example for the bottom-up synthesis of functional
inorganicorganic hybrid thin lm materials in a more general
sense, we would like to cite the more recent work of Altman
et al. and refer to further references given therein.
41
Obviously, the layer-by-layer and bottom-up concepts of
the assembly of supramolecular structures rely on surface
chemistry. From a more general point of view, it is also related
to the well-established solid-phase synthesis of complex
(bio)organic polymers, such as peptides, DNA, etc. The ideal
substrates to start with in such a layer-by-layer deposition of
organic ligands or metal-oxo coupling units, are organic
surfaces as exposed by SAMs.
16
Since a fairly large number
of MOF structures reported so far are based on carboxylate
coupling units, the choice of COOH-terminated SAMs
exposing a COOH-terminated surface is quite obvious. It
has to be noted, however, that the fabrication of such
COOH-terminated surface from the corresponding organo-
thiols is not trivial.
42,43
Indeed, by using the COOH-
functionalized organic surface of a mercaptohexadecanoic
acid SAM (MHDA) as a 2D anchoring and nucleation site,
Fig. 8 Left: in the Fe
3+
/bdc system, MIL-53(Fe) forms as the
product of homogeneous nucleation, while MIL-88B(Fe) deposits
in an oriented fashion on COOH-terminated SAMs as the product of
heterogeneous nucleation. Right: a schematic representation of the
observed oriented growth of the MIL-88B(Fe) phase on mercapto-
hexadecanoic acid SAMs on gold substrates at 25 1C from aged
mother solutions. Figures adapted from ref. 27 with permission,
Copyright 2008, Wiley-VCH.
1424 | Chem. Soc. Rev., 2009, 38, 14181429 This journal is c The Royal Society of Chemistry 2009
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
the growth of homogenous, structurally well-dened MOF
structures in a step-by-step fashion was demonstrated by
Shekhah et al. for the case of HKUST-1 (Fig. 9).
23
The two components, copper(II)acetate (CuAc
2
) and 1,3,5-
benzenetricarboxylic acid (H
3
btc), were separately dissolved in
ethanol and the substrate was immersed into each solution in a
cyclic way, while each step was followed by rinsing with pure
ethanol. By starting with copper(II)acetate, a linear increase of
thickness of the deposited HKUST-1 layer with the number of
(alternating) immersion cycles in CuAc
2
and btc could be
demonstrated in situ by surface plasmon resonance (SPR)
spectroscopy (see Fig. 10). Most importantly, XRD data
recorded for these deposited MOF layers revealed that
only one orientation is grown, with the [100] direction
perpendicular to the surface. Interestingly, when replacing
copper(II)acetate with zinc(II)acetate, instead of the growth
of oriented MOFs, only the deposition of a non-porous,
amorphous Zn
2+
/btc polymer was observed, in contrast to
the known solvothermal chemistry, which leads to MOFs.
44,45
When the MHDA SAM (COOH-terminated surface)
was replaced by a mercaptoundecanol (MUD) SAM
(OH-terminated surface) again an oriented growth of
HKUST-1 takes place.
46
This time, however, with the [111]
direction of HKUST-1 orientated normal to the substrate
surface.
The observation that on a COOH-terminated surface,
HKUST-1 grows along the [100] direction and on an
OH-terminated surface along the [111] direction, is in full
accord with the observation reported by Biemmi et al.
(see discussion above), who observed that the deposition of
MOFs from a mother liquor at elevated temperature also
leads to polycrystalline, but well-ordered, MOF lms with the
same preferential orientations. A particular advantage of
the layer-by-layer method is the possibility to directly monitor
the deposition of both ligands and coupling units using SPR
spectroscopy. The data in Fig. 10 show that subsequently
adding CuAc
2
and btc at room temperature leads to step-
by-step deposition of layers on both MHDA and MUD
SAMs. A closer inspection of the data reveals that the height
of the deposition steps is dierent for the rst layers. In future
work it will be important to investigate the importance of
deposition parameters (concentrations and temperature) and
the lm deposition kinetics. The XRD data shown in Fig. 11
(recorded after 40 deposition cycles) provides unambiguous
evidence for the formation of highly ordered HKUST-1 on the
MUD-modied Au substrate. This works therefore goes
signicantly beyond earlier sequential deposition of organic
layers, e.g. in the case of polyelectrolytes or dithiol-Cu
multilayers.
47,48
In the work by Shekhah et al.,
23
the
Fig. 9 A schematic diagram for the step-by-step growth of MOFs on SAMs by repeated growth cycles separated by washing: rst immersion in a
solution of metal precursor and subsequently in a solution of the organic ligand. Here, for simplicity, the scheme simplies the assumed structural
complexity of the carboxylic acid coordination modes. Reproduced from ref. 23 with permission, Copyright 2007, American Chemical Society.
Fig. 10 The SPR signal as a function of time, recorded in situ during
sequential injections of (A) CuAc
2
, (B) ethanol, and (C) btc in the SPR
cell containing MHDA SAM (above)
23
and MUD SAM (below).
46
Fig. 10 above adapted from ref. 23 with permission, Copyright 2007,
American Chemical Society.
This journal is c The Royal Society of Chemistry 2009 Chem. Soc. Rev., 2009, 38, 14181429 | 1425
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
gas-loading properties of the deposited HKUST-1 layers were
also studied via NH
3
/water exchange experiments. IR and
near-edge X-ray absorption ne structure spectroscopy
(NEXAFS) data allowed them to conclude that the loading
with NH
3
was similar to that observed for bulk HKUST-1,
49
including a substantial irreversibility.
Another particular advantage of using the organic surfaces
exposed by Au/organothiol SAMs is the availability of
methods to laterally pattern the organic surface, e.g. via micro
contact printing (mCP),
50
and thus to achieve a selective
growth of MOF lms on pre-dened areas of the surface.
In the work by Munuera et al., such patterned SAMs have
been used in connection with quantitative SFM to follow the
growth of oriented MOFs on the COOH-terminated surface
and to study their morphological characteristics (Fig. 12).
51
The results verify the selective growth of the HKUST-1 on the
COOH-terminated surface (Fig. 12A) and the homogeneity of
the deposited layers. It also demonstrates the strength of
the step-by-step preparation procedure employed and its
capability to fabricate high quality MOFs lms on surfaces.
The observed monotonous thickness increase indicates that,
within experimental error, the number of layers grown is
proportional to the number of immersion cycles, implying
that once the rst [Cu
3
btc
2
(H
2
O)
n
] layer nucleates on the
COOH-terminated regions, all of the subsequent material
is deposited on top of the previously nucleated layers.
Interestingly, in that work it was observed that the increase
in thickness per immersion cycle amounted to half a unit cell in
the [100] direction, see Fig. 12B.
With regard to applications, it is particularly relevant that
the topmost lm surface roughness is fairly low and does not
increase substantially with SURMOF thickness. In contrast to
the immersion of substrates into preconditioned solvothermal
mother liquors, the step-by-step synthesis yields extraordinary
homogeneous lms of 4100 nm thickness and a roughness in
the order of only one elementary cell in regions in the range of
several mm
2
. As a result, the layer-by-layer approach may also
be suited for the fabrication of thicker layers with a very
homogenous, at surface, e.g. in sensor applications and for
fabricating membranes.
Using the step-by-step approach, it was also possible
to graft a monocarboxylic-substituted polychlorotriphenyl-
methyl radical (PTMCOOH) onto a COOH-functionalized
MHDA SAM using Cu
2+
ions as linkers between the carboxyl
groups of the SAM and the PTMCOOH ligand (Fig. 13).
52
In
this case, the organic ligand only has one coupling unit, so that
only the deposition of a monolayer can be achieved. In
this study, the rather well-dened metal radical adlayer
was characterized thoroughly using dierent surface analysis
Fig. 11 Out-of-plane XRD data for HKUST-1 lms: (a) bulk,
(b) growth on a MHDA SAM (simulation), (c) growth on MHDA
SAM (experimental), (d) growth on MUD SAM (simulation),
(e) growth on MUD SAM (experimental).
46
Adapted from ref. 23
with permission, Copyright 2007, American Chemical Society.
Fig. 12 (A) Two dierent ways of measuring thickness (averaged
proles and histograms). Left: (a) a topographic image (6.5 6.5 mm)
and (b) a selected area for accurate thickness estimation; right: the
corresponding height histogram (top) and averaged prole (bottom)
calculated over the whole area in (b). The red lines in the histogram
represent the corresponding Gaussian ts. (B) (a) A series of topo-
graphic SFM images for dierent samples corresponding to n = 10,
20, 23, 30 and 45 immersion cycles from left to right, respectively. The
total colour scale (total height range) is 110 nm for all the images.
Because of the low topography of the 10 cycles sample, the inset shows
the same image with the scale magnied by a factor of two. (b) Film
thickness as a function of the number of immersion cycles. The red
dashed line corresponds to the proposed half-layer growth, whereas
the grey one corresponds to a single unit cell or complete layer growth
(see text). (c) The root mean square (rms) surface roughness as a
function of the number of immersion cycles calculated for dierent
scan sizes (see inset). The black horizontal line corresponds to the rms
of the starting substrate while the blue dashed line has been drawn as a
visual aid. Error bars represent the standard deviation values. Repro-
duced from ref. 51 with permission, Copyright 2008, RSC Publishing.
1426 | Chem. Soc. Rev., 2009, 38, 14181429 This journal is c The Royal Society of Chemistry 2009
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
techniques, such as contact angle, IRRAS, XPS, SPR,
ToF-SIMS, SFM and NEXAFS. The magnetic character of
the grafted radical ions was conrmed by EPR. The density of
unoccupied electronic states was investigated using X-ray
absorption spectroscopy and a low energy peak in the
NEXAFS spectrum directly revealed the presence of partially
occupied electronic levels, thus proving the open-shell
character of the grafted ligands. SEM measurements on a
laterally patterned sample prepared by mCP of MHDA in a
matrix of hexadecane thiolate (a CH
3
-terminated SAM) was
performed to demonstrate that the metal-assisted anchoring of
the open-shell ligand occurs selectively on the COOH-
terminated SAM. These results represent an easy and new
approach to anchor organic radicals on surfaces, and consti-
tute the rst step towards the growth of magnetic metal
organic radical-based frameworks on solid substrates.
5. Conclusions and perspectives
Although comparably few groups have studied the deposition
of porous coordination polymers (PCPs) and/or MOFs at
surfaces, some signicant advances have been made, as
discussed in this article above. The rst examples of phase
pure MOF thin lms have been described, showing promising
morphology. However, more detailed characterisation of the
adsorption properties of the lms or coatings in comparison to
the known bulk materials needs to be done. Liu et al. most
recently reported on a rather convincing, and in fact rst
example, of a true continuous and well-intergrown MOF-5
membrane, which was successfully prepared on porous
a-alumina substrate by in situ solvothermal synthesis.
53
The
BET measurements on crystals taken from the same mother
liquor that was used for membrane synthesis yielded a
Langmuir surface area of 2259 m
2
g
1
, and a narrow pore
size distribution centered at 1.56 nm. The permeation data for
H
2
, CH
4
, N
2
, CO
2
and SF
6
of the grown MOF-5 membrane
show that the diusion of simple gases follows the Knudsen
diusion behaviour.
Despite the fact that self-assembled organic monolayers
(SAMs) have been shown to be very useful to direct the
growth and even allow the control of the orientation, as well
as the deposited MOF phase, the morphologies of the
obtained MOF lms from the solvothermal mother solutions
are poor. The demonstration of the oriented layer-by-layer
growth of smooth, very homogeneous and quasi epitaxial
MOFs on SAMs/Au points in a novel direction, which holds
much promise and is quite similar to related oriented layer-by-
layer growth and bottom-up assemblies of other hybrid
inorganicorganic materials at surfaces.
40,41,5457
However,
the particular combination of porosity with chemical and
physical functionalities of the coordination framework
8
will
be most interesting, and will possibly transgress the limitation
of less ordered and non-porous thin lms of coordination
polymers also grown at surfaces, which, for example, are
interesting as redox or photo-functional molecular systems
on electrodes.
56
In this context, we would like to quote another
recent and very nice example of unique perspectives on MOF
thin lms as pars pro toto in order to stimulate further
research. Allendorf et al. communicated the concept of
stress-induced chemical detection using MOFs by integrating
a thin lm of the HKUST-1 with a microcantilever surface.
58
Their results showed that the energy of molecular adsorption,
which causes slight distortions in the MOF crystal structure, is
converted to mechanical energy to create a highly responsive,
reversible and selective sensor. This sensor responds to water,
methanol and ethanol vapours, but yields no response to either
N
2
or O
2
. The magnitude of the signal, which is measured by a
built-in piezoresistor, is correlated with the concentration and
can be tted to a Langmuir isotherm.
Aside from the applications for MOF thin lm device
fabrication indicated above, the layer-by-layer preparation
method oers new prospects to study the kinetics and mechanism
of MOF formation itself in more detail from a new
perspective.
37,59
The systematic in situ SPR monitoring of
MOF lm growth is expected to provide a new insight on
the assembly process of the frameworks being not only
dependent on solvents, temperature, pH etc., but also depen-
dent on the oered building blocks, in particular on the
precursors for the metal-containing SBUs. Recently, we found
evidence for the preferred layer-by-layer growth of HKUST-1
using Cu(Ac)
2
as a Cu
2+
source (see Figs. 1012) in contrast to
Cu(NO
3
)
2
3H
2
O.
60
Since solutions of Cu(Ac)
2
in fact contain
the SBU-like dimeric species, [Cu
2
(CH
3
COO)
4
], this nding
directly supports the mechanistic implications of the
Fig. 13 A schematic representation of (a) mononuclear and (b) dinuclear copper(II) complexes obtained by reacting PTMCOOH or its
carboxylate PTMCOO

with copper(II)acetate, and (c) idealized representation for grafting the PTMCOO

ligand on top of a COOH-terminated

SAM. Reproduced from ref. 52 with permission, Copyright 2008, American Chemical Society.
This journal is c The Royal Society of Chemistry 2009 Chem. Soc. Rev., 2009, 38, 14181429 | 1427
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
controlled SBU concept of MOF synthesis.
1,38
The overall
kinetics of layer-by-layer MOF growth were found to be
strictly linear.
23,60
However, non-liner growth modes for the
related formation of nanoarchitectures at surfaces were
observed in the case of self replicating amphiphilic mono-
layers, for example in case of polyelectrolytes.
61
A similar
non-liner mechanism was recently found for inorganic
organic hybrid systems, similar to MOFs or surface coordina-
tion polymers (SCPs).
62
It will be interesting to compare these
dierent chemical systems, including MOFs, in terms of the
underlying growth mechanisms, possibly even aimed at an
accelerated self-propagating growth of oriented MOFs at
surfaces. Furthermore, the step-by-step method obviously
oers the unique opportunity to grow MOF-like ordered
structures, which cannot be obtained by established solvo-
thermal routes. For example, the deposition of MOFs with
alternating layers (heterostructures), possibly with non-
periodic combinations of dierent metal ions and/or dierent
linkers should be feasible. As an example for this direction of
MOF thin lm research, we nally like to highlight the recent
work of Kanaizuka et al. on the construction of highly
oriented crystalline, but non-porous, SCPs, which are
composed of copper dithiooxamide complexes. The authors
suggest that such homo- and also heterostructured SCPs,
quite similar to MOF thin lms, may be useful for many
applications, including Josephson junctions of super-
conductors, magnetic spin valves, capacitance, screen displays,
fuel cells and catalytic devices.
63
Acknowledgements
The authors acknowledge nancial support by the EU STREP
SURMOF (NMP4-CT-2006-032109) and the Priority
Program 1362 Metalorganic Frameworks of the German
Research Foundation. D. Z. is grateful for additional support
by the Ruhr University Research School (http://www.
research-school.rub.de).
References
1 G. Ferey, Chem. Soc. Rev., 2008, 37, 191.
2 O. M. Yaghi, M. OKeee, N. W. Ockwig, H. K. Chae,
M. Eddaoudi and J. Kim, Nature, 2003, 423, 705.
3 O. Delgado-Friedrichs, M. D. Foster, M. OKeee,
D. M. Proserpio, M. J. Treacy and O. M. Yaghi, J. Solid State
Chem., 2005, 178, 2533.
4 O. Delgado-Friedrichs, M. OKeee and O. M. Yaghi, Phys.
Chem. Chem. Phys., 2007, 9, 1035.
5 N. W. Ockwig, O. Delgado-Friedrichs, M. OKeee and
O. M. Yaghi, Acc. Chem. Res., 2005, 38, 176.
6 K. Uemura, R. Matsuda and S. Kitagawa, J. Solid State Chem.,
2005, 178, 2420.
7 S. Bureekaew, S. Shimomura and S. Kitagawa, Sci. Technol. Adv.
Mater., 2008, 9, 014108.
8 S. Kitagawa and R. Matsuda, Coord. Chem. Rev., 2007, 251, 2490.
9 F. Di Renzo and F. O. Fajula, in Zeolites and Ordered Mesoporous
Materials: Progress and Prospects, ed. J. C

ejka and H. van

Bekkum, Elsevier, Amsterdam, 2005, vol. 157, pp. 1.
10 T. Bein and S. Mintova, Zeolites and Ordered Mesoporous
Materials: Progress and Prospects, ed. J. C

ejka and H. van

Bekkum, Elsevier, Amsterdam, 2005, vol. 157, pp. 263.
11 H. Y. Jiang, B. Q. Zhang, Y. S. Lin and Y. D. Li, Chin. Sci. Bull.,
2004, 49, 2547.
12 E. E. McLeary, J. C. Jansen and F. Kapteijn, Microporous
Mesoporous Mater., 2006, 90, 198.
13 A. S. T. Chiang and K. J. Chao, J. Phys. Chem. Solids, 2001, 62,
1899.
14 L. Tosheva and V. P. Valtchev, Chem. Mater., 2005, 17, 2494.
15 A. M. Spokoyny, D. Kim, A. Sumrein and C. Mirkin, Chem. Soc.
Rev., 2009, DOI: 10.1039/b807085g.
16 (a) D. L. Allara and R. G. Nuzzo, Langmuir, 1985, 1, 52;
(b) J. C. Love, L. A. Estro, J. K. Kriebel, R. G. Nzzo and
G. M. Whitesides, Chem. Rev., 2005, 105, 1103.
17 S. Mann, Nature, 1993, 365, 499.
18 S. Feng and T. Bein, Nature, 1994, 368, 834.
19 S. Hermes, F. Schro der, R. Chelmowski, C. Wo ll and
R. A. Fischer, J. Am. Chem. Soc., 2005, 127, 13744.
20 S. Hermes, D. Zacher, A. Baunemann, C. Wo ll and R. A. Fischer,
Chem. Mater., 2007, 19, 2168.
21 Y. Yoo and H. K. Jeong, Chem. Commun., 2008, 2441.
22 D. Zacher, A. Baunemann, S. Hermes and R. A. Fischer, J. Mater.
Chem., 2007, 17, 2785.
23 O. Shekhah, H. Wang, S. Kowarik, F. Schreiber, M. Paulus,
M. Tolan, C. Sternemann, F. Evers, D. Zacher, R. A. Fischer
and C. Wo ll, J. Am. Chem. Soc., 2007, 129, 15118.
24 (a) E. Biemmi, C. Scherb and T. Bein, J. Am. Chem. Soc., 2007,
129, 8054; (b) E. Biemmi, A. Darga, N. Stock and T. Bein,
Microporous Mesoporous Mater., 2008, 113, 380.
25 J. Gascon, S. Aguado and F. Kapteijn, Microporous Mesoporous
Mater., 2008, 113, 132.
26 M. Arnold, P. Kortunov, D. J. Jones, Y. Nedellec, J. Karger and
J. Caro, Eur. J. Inorg. Chem., 2007, 60.
27 C. Scherb, A. Schodel and T. Bein, Angew. Chem., Int. Ed., 2008,
47, 5777.
28 M. Kubo, W. Chaikittisilp and T. Okubo, Chem. Mater., 2008, 20,
2887.
29 J. Hazovic, M. Bjorgen, U. Olsbye, P. D. C. Dietzel, S. Bordiga,
C. Prestipino, C. Lamberti and K. P. Lillerud, J. Am. Chem. Soc.,
2007, 129, 3612.
30 D. N. Dybtsev, H. Chun, S. H. Yoon, D. Kim and K. Kim, J. Am.
Chem. Soc., 2004, 126, 32.
31 M. Arnold, P. Kortunov, D. J. Jones, Y. Nedellec, J. Karger and
J. Caro, Eur. J. Inorg. Chem., 2007, 6064.
32 D. J. Tranchemontagne, J. R. Hunt and O. M. Yaghi, Tetrahedron,
2008, 64, 8553.
33 S. Hausdorf, F. Baitalow, J. Seidel and F. Mertens, J. Phys. Chem.
A, 2007, 111, 4259.
34 S. Hermes, T. Witte, T. Hikov, D. Zacher, S. Bahnmuller,
G. Langstein, K. Huber and R. A. Fischer, J. Am. Chem. Soc.,
2007, 129, 5324.
35 T. Loiseau, C. Serre, C. Huguenard, G. Fink, F. Taulelle,
M. Henry, T. Bataille and G. Ferey, Chem.Eur. J., 2004, 10, 1373.
36 C. Mellot-Draznieks, C. Serre, S. Surble, N. Audebrand and
G. Ferey, J. Am. Chem. Soc., 2005, 127, 16273.
37 S. Surble, C. Serre, C. Mellot-Draznieks, F. Millange and
G. Ferey, Chem. Commun., 2006, 284.
38 S. Surble, F. Millange, C. Serre, G. Ferey and R. I. Walton, Chem.
Commun., 2006, 1518.
39 O. Shekhah, H. Wang, T. Strunskus, P. Cyganik, D. Zacher,
R. Fischer and C. Wo ll, Langmuir, 2006, 23, 7740.
40 K. Ariga, J. P. Hill and Q. M. Ji, Phys. Chem. Chem. Phys., 2007,
9, 2319.
41 M. Altman, A. D. Shukla, T. Zubkov, G. Evmenenko, P. Dutta
and M. E. van der Boom, J. Am. Chem. Soc., 2006, 128, 7374.
42 H. J. Himmel, A. Terfort and C. Wo ll, J. Am. Chem. Soc., 1998,
120, 12069.
43 R. Arnold, W. Azzam, A. Terfort and C. Wo ll, Langmuir, 2002, 18,
3980.
44 O. Yaghi, C. E. Davis and G. Li amd H. Li, J. Am. Chem. Soc.,
1997, 119, 2861.
45 S. S. Y. Chui, S. M. F. Lo, J. P. H. Charmant, A. G. Orpen and
I. D. Williams, Science, 2004, 238, 1148.
46 O. Shekhah and C. Wo ll, unpublished results 2008/2009; informa-
tion can be retrieved directly by contacting the authors.
47 A. Ulman and S. D. Evans, J. Am. Chem. Soc., 1991, 113, 5866.
48 G. Decher, Science, 1997, 277, 1232.
49 S. S. Y. Chui, S. M. F. Lo, J. P. H. Charmant, A. G. Orpen and
I. D. Williams, Science, 2004, 238, 1148.
1428 | Chem. Soc. Rev., 2009, 38, 14181429 This journal is c The Royal Society of Chemistry 2009
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online
50 Y. Xia and G. M. Whitesides, Annu. Rev. Mater. Sci., 1998, 28, 84.
51 C. Munuera, O. Shekhah, H. Wang, C. Wo ll and C. Ocal, Phys.
Chem. Chem. Phys., 2008, 10, 7257.
52 O. Shekhah, N. Roques, V. Mugnaini, C. Munuera, C. Ocal,
J. Veciana and C. Wo ll, Langmuir, 2008, 24, 6640.
53 Y. Liu, Z. Ng, E. A. Khan, H.-K. Jeong, Chi-bun Ching and
Z. Lai, Microporous Mesoporous Mater., 2009, 118, 296.
54 R. Yerushalmi, A. Scherz and M. E. van der Boom, J. Am. Chem.
Soc., 2004, 126, 27002701.
55 P. W. Zhu, M. E. van der Boom, H. Kang, G. Evmenenko,
P. Dutta and T. J. Marks, Chem. Mater., 2002, 14, 4982.
56 H. Nishihara, K. Kanaizuka, Y. Nishimori and Y. Yamanoi,
Coord. Chem. Rev., 2007, 251, 2674.
57 D. Maspoch, D. Ruiz-Molina and J. Veciana, Chem. Soc. Rev.,
2007, 36, 770.
58 M. D. Allendorf, R. J. T. Houk, L. Andruszkiewicz, A. A. Talin,
J. Pikarsky, A. Choudhury, K. A. Gall and P. J. Hesketh, J. Am.
Chem. Soc., 2008, 130, 14404.
59 J. A. Rood, W. C. Boggess, B. C. Noll and K. W. Henderson,
J. Am. Chem. Soc., 2007, 129, 13675.
60 O. Shekhah, H. Wang, D. Zacher, R. A. Fischer and C. Wo ll,
unpublished results 2008/2009; manuscript in preparation;
information can be retrieved directly by contacting the authors.
61 R. Maoz, S. Matlis, E. DiMasi, B. M. Ocko and J. Sagiv, Nature,
1996, 384, 150.
62 L. Motiei, M. Altman, T. Gupta, F. Lupo, A. Gulino,
G. Evmenenko, P. Dutta and M. E. van der Boom, J. Am. Chem.
Soc., 2008, 130, 8913.
63 K. Kanaizuka, R. Haruki, O. Sakata, M. Yoshimoto, Y. Akita and
H. Kitagawa, J. Am. Chem. Soc., 2008, 130, 15778.
This journal is c The Royal Society of Chemistry 2009 Chem. Soc. Rev., 2009, 38, 14181429 | 1429
D
o
w
n
l
o
a
d
e
d

b
y

N
A
T
I
O
N
A
L

T
E
C
H
N
I
C
A
L

U
N
I
V
E
R
S
I
T
Y

O
F

A
T
H
E
N
S

o
n

2
4

F
e
b
r
u
a
r
y

2
0
1
3
P
u
b
l
i
s
h
e
d

o
n

0
9

M
a
r
c
h

2
0
0
9

o
n

h
t
t
p
:
/
/
p
u
b
s
.
r
s
c
.
o
r
g

|

d
o
i
:
1
0
.
1
0
3
9
/
B
8
0
5
0
3
8
B
View Article Online