Design of highly porous single-site catalysts through two-step postsynthetic modification of mixed-linker MIL-53(Al)

Article abstract of DOI:10.1002/cplu.201402123

Defined mixed-linker metal-organic frameworks based on MIL-53-NH₂(Al) were successfully synthesized under ambient pressure, thus, enabling easy scaling up of the synthesis. The two-step postsynthetic modification reaction of these materials with, first, maleic anhydride and, second, palladium(II) acetate resulted in immobilized palladium(II) complexes at the side chain of the linker molecules. The high porosity of the mixed-linker materials was retained throughout the modification process, which was in contrast to a previous study with pure MIL-53-NH₂(Al). The novel palladium-containing materials were applied in Heck-type C-C coupling reactions of bromo- or chlorobenzene with styrene, in which they exhibited high catalytic activity and selectivity.

Full text of DOI:10.1002/cplu.201402123

CHEMPLUSCHEM
FULL PAPERS
DOI: 10.1002/cplu.201402123
Design of Highly Porous Single-Site Catalysts through
Two-Step Postsynthetic Modification of Mixed-Linker
MIL-53(Al)
Meike A. Gotthardt,[a] Roland Schoch,[b] Tobias S. Brunner,[c] Matthias Bauer,*[b] and
Wolfgang Kleist*[a, d]
Defined mixed-linker metal–organic frameworks based on MIL-
53-NH2(Al) were successfully synthesized under ambient pres-
sure, thus, enabling easy scaling up of the synthesis. The two-
step postsynthetic modification reaction of these materials
with, first, maleic anhydride and, second, palladium(II) acetate
resulted in immobilized palladium(II) complexes at the side
chain of the linker molecules. The high porosity of the mixed-
linker materials was retained throughout the modification pro-
cess, which was in contrast to a previous study with pure MIL-
53-NH2(Al). The novel palladium-containing materials were ap-
plied in Heck-type CÀC coupling reactions of bromo- or chloro-
benzene with styrene, in which they exhibited high catalytic
activity and selectivity.
Introduction
Owing to their defined crystalline structure and high porosity,
metal–organic frameworks (MOFs) have emerged as interesting
materials for postsynthetic modifications (PSMs)[1] and catalytic
applications.[2] MOFs consist of metal ions or clusters that are
connected by organic linker molecules with two or more func-
tional groups which can bind to the metal atoms.[3] A multi-
tude of linker molecules can be utilized as building blocks in
MOFs (e.g., di- or tricarboxylic acids,[4] diphosphonates[5]), re-
sulting in a high variety of structures and properties of the ma-
terials. Furthermore, linker molecules with additional functional
groups that do not bind to the metal ions themselves can be
used. The most prominent example for this approach are the
isoreticular metal–organic framework (IRMOF) series based on
MOF-5[6] and isoreticular frameworks with various functional
groups based on the structure of MIL-53(Al),[4a,7] which have
been utilized in various applications, including gas adsorp-
tion,[8] separation,[9] and catalysis.[10] Such functional groups
(e.g., amine groups, formyl groups) can also be seen as poten-
tial binding sites for more complex organic functionalities,
which can be introduced using the PSM approach.[1b,c] Through
this approach, it is also possible to insert more complex func-
tionalities that cannot be introduced during MOF synthesis be-
cause they would hinder the framework formation or result in
different crystalline or amorphous structures. If chelating func-
tionalities are incorporated, the modified side chains can be
employed to bind catalytically active metal centers, thus, facili-
tating the immobilization of highly active and defined metal
complexes in the framework materials.[1d,11] The MOF-based
catalyst materials should combine the beneficial characteristics
of both homo- and heterogeneous catalyst systems. On one
hand, they contain defined metal complexes that should result
in high catalytic activity, whereas, on the other hand, the sepa-
ration and reuse of the solids should be feasible.
[a] Dipl.-Chem. M. A. Gotthardt, Dr. W. Kleist
Institute for Chemical Technology and Polymer Chemistry
Karlsruhe Institute of Technology (KIT)
Engesserstrasse 20, 76131 Karlsruhe (Germany)
[b] Dipl.-Chem. R. Schoch, Prof. Dr. M. Bauer
Department Chemie, Fakultꢀt fꢁr Naturwissenschaften
Universitꢀt Paderborn, Warburgerstrasse 100
33098 Paderborn (Germany)
We have previously shown that MIL-53-NH2(Al), which con-
sists of AlÀOHÀ chains connected by 2-aminoterephthalate
(ABDC) linkers,[4a] can be synthesized at ambient pressure,[12]
thus, facilitating easy scaling up of the synthesis. The amine
groups of the linker molecules were successfully modified in
a two-step postsynthetic reaction and the resulting catalyst
material exhibited high activity in the Heck-type CÀC coupling
reaction of bromo- (BrBz) or chlorobenzene (ClBz) and sty-
rene.[12] Although the structure was retained throughout the
modification process, a drastic decrease of the specific surface
area and pore volume was observed. This observation was ex-
plained by the fact that modification of the amine groups pref-
erentially took place on the outer surface, thus leading to
blocking of the pore entrances. Because of this finding, we de-
cided to tune the number and distribution of amine groups
E-mail: matthias.bauer@upb.de
[c] Dipl.-Chem. T. S. Brunner
Institute for Inorganic Chemistry
Karlsruhe Institute of Technology (KIT)
Engesserstrasse 15, 76131 Karlsruhe (Germany)
[d] Dr. W. Kleist
Institute of Catalysis Research and Technology
Karlsruhe Institute of Technology (KIT)
Hermann-von-Helmholtz-Platz 1
76344 Eggenstein-Leopoldshafen (Germany)
Fax: (+49)721-60844820
E-mail: wolfgang.kleist@kit.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201402123.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
1
&
ÞÞ
These are not the final page numbers!

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Scheme 1. Schematic representation of the two-step PSM of mixed-linker MIL-53-NH2(Al); yellow: Al, red: O, gray: C, blue: N, white: H, green: Pd, dark red: li-
gands (acetate or H2O).
inside the framework using the mixed-linker metal–organic
framework (MIXMOF) concept,[13] in which both functionalized
and nonfunctionalized linker molecules are incorporated in
a MOF in defined ratios. Previously, MIXMOFs based on MIL-
53(Al) were synthesized under hydrothermal conditions by
using terephthalic acid (H2BDC) and aminoterephthalic acid
(H2ABDC) in different ratios.[13d] Herein, we synthesized defined
MIXMOFs based on MIL-53-NH2(Al) under ambient pressure.
We also report a strategy for the design of Pd/MIXMOF cata-
lysts through a two-step PSM by using, first, maleic anhydride
and, second, palladium(II) acetate (Scheme 1).
The aim of this study was the introduction of targeted
amounts of well-defined anchored palladium complexes into
MIXMOFs with a tuned number of amine groups. Using this
new approach, the undesired pore blocking that was observed
for MIL-53-NH2(Al) should be prevented. Therefore, not only
the active palladium(II) centers on the outer surface, but also
those inside the pores should be accessible during catalysis. To
examine the catalytic activity of the palladium(II)-containing
MIXMOFs, we have chosen a Heck-type CÀC coupling reaction
of bromobenzene or chlorobenzene and styrene as a suitable
test reaction. Furthermore, the properties and catalytic activity
of the new materials were compared with those of modified
MIL-53-NH2(Al), which was previously studied in our group.[12]
Figure 1. XRD patterns of MIL-53-NH2(Al) and MIXMOFs based on MIL-53-
NH2; the number given in parentheses represents the percentage of ABDC
in the MIXMOF material.
linker in the MIXMOFs was reduced (Figure S2, left, in the Sup-
porting Information). Furthermore, the aromatic CÀH bending
vibrations of ABDC and BDC could be distinguished by their
different substitution patterns at the aromatic ring (Figure S2,
right, in the Supporting Information). Therefore, the IR spectra
clearly prove the successful incorporation of both linker mole-
cules into the MIXMOFs. The IR spectra did not show a band at
n˜ ꢀ1680 cmÀ1 (C=O stretching vibration of protonated linker
molecules), which led to the conclusion that there were no re-
sidual acid molecules present in the pores (Figure 2); this
would be the case for hydrothermally synthesized MIXMOFs
without further activation.
Results and Discussion
Owing to the observed loss of pore volume throughout the
modification process of pure MIL-53-NH2(Al),[12] we decided to
use MIXMOFs based on MIL-53-NH2(Al), with
a reduced
number of amine groups, as the starting materials. The dilution
of the amine groups should promote the distribution of immo-
bilized complexes inside the framework, and therefore, prevent
pore blocking. MIXMOFs containing 2-aminoterephthalate
(ABDC) and terephthalate (BDC) in defined ratios (4:1, 3:2, 2:3;
see the Experimental Section) were successfully synthesized
under ambient pressure at elevated temperature. In contrast
to the hydrothermal conditions previously used for the synthe-
sis of those MIXMOFs,[13d] the new route facilitates easy scaling
up. To confirm the isoreticular structures for the mixed-linker
materials and MIL-53-NH2(Al), powder X-ray diffraction (PXRD)
measurements were taken. The patterns shown in Figure 1
prove the successful synthesis of crystalline MIXMOFs.
Attenuated total reflectance (ATR) IR spectra were recorded
to confirm the different linker ratios qualitatively. In the spec-
tra, a decrease in intensity of the NÀH stretching vibrations
was observed when the amount of the 2-aminoterephtalate
Figure 2. IR spectra of MIL-53-NH2(40) (black) and modified MIL-53-NH2(40)-
Mal (dark gray). For comparison the spectrum of MIL-53-NH2(Al)_as contain-
ing free acid molecules is shown (light gray).
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
2
&
ÞÞ
These are not the final page numbers!

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
NMR spectroscopy of digested samples was used to quantify
the ratio of ABDC and BDC incorporated in the mixed frame-
works. Because the IR spectra revealed that there were no re-
sidual free acid molecules in the pore system, the ABDC/BDC
ratio determined by NMR spectroscopy could be directly relat-
ed to the composition of the MIXMOFs. The linker ratios deter-
mined by NMR spectroscopy (Table 1 and Figures S3–S6 in the
We previously used this two-step reaction with MIL-53-
NH2(100) as a starting material. In that case, we observed a sig-
nificant decrease of the micropore volume throughout the
modification process from 0.39 to only 0.01 cm3 gÀ1 (Table 2
and Figure 3, top).[12] Owing to this discovery, we decided to
Table 1. Theoretical and measured linker ratios.
ABDC/BDC
Applied
Determined
MIL-53-NH2(80)
MIL-53-NH2(60)
MIL-53-NH2(40)
4:1
3:2
2:3
3.7:1.0
2.8:2.0
2.0:3.0
Supporting Information) matched very well with the ratios ap-
plied during synthesis, which proved that both linker mole-
cules were built into the framework with the same preference.
To exclude the formation of core–shell particles and to
prove a mostly homogeneous distribution of the two linkers,
three additional samples with an initial linker ratio of 1:1 were
synthesized for 8, 24, and 72 h. The crystallinity of the resulting
frameworks increased with the reaction time (for XRD patterns,
see Figure S14 in the Supporting Information), whereas NMR
spectra (Figures S11–S13 in the Supporting Information) of all
three samples showed that both linker molecules were present
in the desired ratio. Those findings support the formation of
frameworks with homogeneously distributed linkers, rather
than the formation of core–shell structures. A detailed study
on the distribution of the individual linker molecules in mixed-
linker MIL-53(Al) materials was previously reported by Lescouet
et al.[14]
Figure 3. Adsorption (filled symbols) and desorption (open symbols) iso-
therms of MIL-53-NH2(100) (top) and MIL-53-NH2(40) (bottom) before and
after modification.
Nitrogen physisorption measurements revealed that the mi-
cropore volumes of the mixed-linker materials were lower than
that of pure MIL-53-NH2(Al) (Table 2). Whereas the micropore
volume of MIL-53-NH2(Al) was 0.39 cm3 gÀ1, the micropore vol-
umes of the MIXMOFs were between 0.17 and 0.30 cm3 gÀ1.
After thorough characterization of the starting materials, cat-
alytically active palladium(II) centers were immobilized in the
frameworks through a two-step PSM reaction. First, the reac-
tion of maleic anhydride with the amine groups in the mixed
materials led to the formation of functionalized side chains,
which could then in the second step be used as chelating li-
gands for the immobilization of palladium(II) ions (Scheme 1).
use MIXMOFs with a reduced number of amine groups to pre-
vent blocking of the pores during the immobilization of the
palladium(II) complexes.
To test our hypothesis, we performed nitrogen physisorption
measurements of the modified MIXMOFs (Table 2). Whereas
MIL-53-NH2(100) lost most of its micropore volume throughout
the modification process (Figure 3, top), the MIXMOFs re-
mained highly porous with a micropore volume of up to
0.11 cm3 gÀ1 (Figure 3, bottom and Figure S1 in the Supporting
Information). The adsorption and desorption isotherms for
MIL-53-NH2(40) before and after modification are also shown in
Figure 3, bottom. Clearly, the dilution and distribution of the
amine functionalities inside the MIXMOFs minimized blocking
of the pores during modification, thus, proving the benefit of
the MIXMOF approach.
Table 2. Micropore volumes of MIXMOFs and modified materials ob-
tained from nitrogen physisorption measurements.
The XRD patterns affirmed that the structure and crystallinity
of the MIXMOFs were retained throughout the modification
process. The reflections representing distances perpendicular
to the pore system ((hkl)=(200) and (400)) remained at the
same position (2q=9.28 and 18.628, respectively), whereas
a small shift was observed for all other reflections (Figure 4).
This shift can be explained by the so-called “breathing effec-
t”[4a,15] of MIL-53(Al), which describes the ability of the frame-
MIL-53-
NH2(100)
[cm3 gÀ1
MIL-53-
NH2(80)
[cm3 gÀ1
MIL-53-
NH2(60)
[cm3 gÀ1
MIL-53-
NH2(40)
[cm3 gÀ1
]
]
]
]
unmodified
maleic anhydride
palladium(II)
acetate
0.39
0.03
0.01
0.17
0.08
0.08
0.22
0.20
0.10
0.30
0.12
0.11
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
3
&
ÞÞ
These are not the final page numbers!

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
large micropore volume was retained for the MIXMOFs, thus
proving that pore blocking was successfully minimized, despite
the higher degree of modification. This observation again illus-
trates the beneficial effects of the diluted amine groups in the
MIXMOFs. Notably, in the literature, a higher degree of modifi-
cation of approximately 40% was reported for MIL-53-
NH2(100), which could be explained by the fact that, in this
study, a 15 times higher amount of maleic anhydride was used
relative to that used herein.[16] In addition, no pore volume or
specific surface area was reported for the materials after PSM.
To determine the palladium loading of the modified frame-
works, atomic absorption spectroscopy (AAS) was used. The
measurements revealed that the palladium loading of the
modified MIXMOFs increased with a decreasing number of
amine groups inside the framework. Whereas MIL-53-NH2(100)-
Mal-Pd contained only 2.1 wt% of Pd, the metal loading could
be increased to 3.1 wt% for MIL-53-NH2(40)-Mal-Pd (Table 5).
Figure 4. XRD patterns of MIL-53-NH2(80), MIL-53-NH2(80)-Mal, and MIL-53-
NH2(80)-Mal-Pd. The dotted lines illustrate the shift of the reflections that
are not perpendicular to the pores.
work to change its geometry, depending on the substrates
inside the pores. Thus, the shift also confirmed the successful
modification of the amine groups. The position of the second
reflection ((hkl)=(110); Table 3) might also be a qualitative in-
dication of the degree of modification.
Table 5. Results of the AAS measurements of the modified MIXMOFs.
MIL-53-NH2(100)- MIL-53-NH2(80)- MIL-53-NH2(60)- MIL-53-NH2(40)-
Mal-Pd
Mal-Pd
Mal-Pd
Mal-Pd
2.1 wt% Pd
2.8 wt% Pd
2.9 wt% Pd
3.1 wt% Pd
Table 3. Position of the reflection (hkl)=(110) in the XRD pattern
throughout the modification process.
Despite their higher Pd content, the micropore volume of all
MIXMOFs was significantly larger after the second modification
step than that of MIL-53-NH2(100)-Mal-Pd.
X
MIL-53-NH2(X) [8] MIL-53-NH2(X)-Mal [8] MIL-53-NH2(X)-Mal-Pd [8]
100
80
60
12.32
12.33
12.33
12.31
12.13
12.22
12.27
12.28
12.12
12.10
12.21
12.23
Although the degree of modification per amine group could
be increased in the mixed materials compared with that of
MIL-53-NH2(100), the overall degree of modification was identi-
cal. Therefore, the increasing palladium loading of the modi-
fied MIXMOFs may be explained by the increased accessibility
of the chelating ligands inside the pores owing to the higher
micropore volume of the modified mixed-linker frameworks. It
is also possible that undesired palladium nanoparticles were
formed, thus increasing the palladium content.
40
The decreasing intensity of the NÀH stretching vibrations
(n˜ =3498 and 3387 cmÀ1) in the IR spectra further indicated
the formation of the amide. The additional band at n˜
ꢀ1700 cmÀ1, which appeared for the samples modified with
maleic anhydride (Figure 2) could be ascribed to the additional
C=O vibration of the inserted carbonyl functionalities.
X-ray absorption spectroscopy (XAS) is a useful method to
gain information about the short-range coordination structure
and oxidation state of the metal centers.[17] Therefore, XAS
measurements were performed to confirm the successful im-
mobilization of palladium(II) complexes and to exclude the
possible formation of nanoparticles. For MIL-53-NH2(100)-Mal-
Pd, the nature of the palladium species immobilized in the
framework strongly depended on the amount of the palladium
precursor used during the modification process. The desired
immobilization of palladium(II) complexes was achieved when
small amounts of palladium(II) acetate were used, whereas
higher concentrations of the precursor led to the additional
formation of undesired palladium nanoparticles.[12] Thus, we
again used small amounts of the palladium(II) precursor for the
modification of the MIXMOFs to prevent particle formation.
The X-ray absorption near-edge structure (XANES) spectra
(Figure 5) of all modified MIXMOFs showed a double-peak
structure, which is characteristic of palladium(II).[18] The extend-
ed X-ray absorption fine structure (EXAFS) spectra of the three
To quantify the degree of modification with maleic anhy-
dride, we again used NMR spectroscopy of digested samples
(Table 4 and Figures S7–S10 in the Supporting Information). Al-
though the overall degree of modification, that is, the total
number of functional groups, was identical for all materials,
the degree of modification per amine group increased for the
mixed-linker frameworks. At the same time, a comparatively
Table 4. Degree of modification of the frameworks, as determined by
NMR spectroscopy.
Degree of modification Degree of modification
of amine groups [%]
of the framework [%]
MIL-53-NH2(100)-Mal
MIL-53-NH2(80)-Mal
MIL-53-NH2(60)-Mal
MIL-53-NH2(40)-Mal
5
5
6
5
4
4
5
11
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
4
&
ÞÞ
These are not the final page numbers!

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Figure 5. Normalized and self-absorption-corrected X-ray absorption spectra
of MIL-53-NH2(80)-Mal-Pd (bottom), MIL-53-NH2(60)-Mal-Pd (middle), and
MIL-53-NH2(40)-Mal-Pd (top).
Figure 6. Fourier-filtered EXAFS spectra (left) and corresponding Fourier-
transformed functions (right) of the samples MIL-53-NH2(80)-Mal-Pd
(bottom), MIL-53-NH2(60)-Mal-Pd (middle), and MIL-53-NH2(40)-Mal-Pd (top).
samples (Figure 6) were similar
and the structural parameters
Table 6. Number, type, and distances of backscatterers of palladium in MIL-53-NH2(80)-Mal-Pd, MIL-53-NH2(60)-
Mal-Pd, and MIL-53-NH2(40)-Mal-Pd obtained by EXAFS analysis.
obtained from EXAFS analysis
were identical, within experi-
mental error. In addition to the
type and number of fitted shells,
the distances of adjusted back-
scatterers were also very similar
(Table 6). Fitting of the spectra
resulted in four oxygen atoms
between 1.9 and 2.0 ꢁ, which
could be assigned to the carbox-
yl groups of the modified linker
and acetate or H2O ligands. Two
carbon shells at distances of 2.9
and 3.9 ꢁ could be attributed to
the carbon scaffold of the modi-
fied linker. Therefore, XAS meas-
urements back up the desired
immobilization of palladium(II)
complexes in the MIXMOFs.
However, palladium contribu-
tions at distances of 2.6 and
3.6 ꢁ indicated the additional
presence of a low amount of
palladium clusters. Regarding
the coordination numbers, ap-
proximately 0.3 atoms in the
shell at 2.6 ꢁ and 0.6 atoms at
Abs-Bs[a]
N(Bs)[b]
R(Abs-Bs) [ꢁ][c]
s [ꢁ]À1[d]
R [%][e]
Ef [eV][f]
Afac[g]
PdÀO
PdÀO
PdÀPd
PdÀC
PdÀPd
PdÀC
1.7Æ0.08
2.0Æ0.2
0.3Æ0.03
2.0Æ0.2
0.7Æ0.07
2.0Æ0.2
1.908Æ0.0190
2.072Æ0.0207
2.640Æ0.0264
2.914Æ0.0291
3.605Æ0.0360
3.963Æ0.0396
0.050Æ0.005
0.032Æ0.003
0.032Æ0.003
0.112Æ0.011
0.032Æ0.003
0.039Æ0.003
MIL-53-NH2(80)-
Mal-Pd
9.168
À6.113
0.4346
PdÀO
PdÀO
PdÀPd
PdÀC
PdÀPd
PdÀC
1.7Æ0.08
2.7Æ0.2
0.7Æ0.07
1.5Æ0.1
1.0Æ0.1
3.1Æ0.3
1.891Æ0.0189
2.048Æ0.0204
2.658Æ0.0265
2.852Æ0.0285
3.587Æ0.0358
3.956Æ0.0395
0.032Æ0.003
0.032Æ0.003
0.087Æ0.008
0.112Æ0.011
0.045Æ0.004
0.045Æ0.004
MIL-53-NH2(60)-
Mal-Pd
12.77
À8.871
0.3633
PdÀO
PdÀO
PdÀPd
PdÀC
PdÀPd
PdÀC
2.0Æ0.1
1.8Æ0.1
0.2Æ0.02
2.0Æ0.2
0.5Æ0.05
2.0Æ0.2
1.929Æ0.0192
2.077Æ0.0207
2.670Æ0.0267
2.894Æ0.0289
3.592Æ0.0359
3.941Æ0.0394
0.067Æ0.006
0.032Æ0.003
0.055Æ0.005
0.110Æ0.011
0.032Æ0.003
0.050Æ0.005
MIL-53-NH2(40)-
Mal-Pd
9.957
À5.034
0.4222
[a] Abs: X-ray absorbing atom, Bs: backscattering atom. [b] Number of backscattering atoms. [c] Distance be-
tween the absorbing and backscattering atom. [d] Debye–Waller-like factor. [e] Fit index. [f] Fermi energy that
accounts for the shift between theory and experiment. [g] Amplitude reducing factor.
3.6 ꢁ (Table 6) were found, which indicate a minor contribution
of very small particles that cannot be detected by XRD. The
presence of small palladium nanoparticles was also confirmed
by the observed PdÀPd distances, which were contracted com-
pared with those in bulk palladium.
coupling reactions. Moreover, even negligible concentrations
of active palladium species (10À5 mol%) can be sufficient to
catalyze Heck-type reactions of aryl iodides.[21] Previously, we
optimized the reaction parameters (reaction temperature,
base, additives) for a Heck-type reaction of less-activated bro-
mobenzene or chlorobenzene with styrene with MIL-53-
NH2(100)-Mal-Pd as a heterogeneous catalyst.[12] Therefore, the
novel catalyst materials were investigated in the same reaction
(Scheme 2) under optimized conditions to evaluate and com-
pare the catalytic activity of the palladium-containing MIX-
MOFs to that of MIL-53-NH2(100)-Mal-Pd.
Palladium-based catalysts exhibit high activity in CÀC cou-
pling reactions, for example, Heck reactions of aryl halides and
olefins. MOFs containing palladium nanoparticles[19] or immobi-
lized palladium species[20] have previously been used in Heck
reactions. The material with immobilized species was tested
with iodobenzene, which is a very reactive substrate for CÀC
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
5
&
ÞÞ
These are not the final page numbers!

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
species or truly heterogeneous,
we conducted hot filtration ex-
periments. To enable fast separa-
tion of the catalyst, it was placed
in a paper filter that was re-
moved after 30 min. The reac-
tion was then resumed under
the same reaction conditions for
Scheme 2. Heck-type CÀC coupling reaction of aryl halides and styrene with palladium-containing MIXMOFs as
the catalyst materials.
We discovered that all catalyst materials exhibited similar ac-
tivities (conversion and turnover numbers (TONs)) in the inves-
tigated CÀC coupling reactions (Table 7). In the reaction of
BrBz with styrene, the selectivity towards trans-stilbene was
the same for all investigated catalyst materials (92–93%). In
another 2.5 h. To exclude external diffusion limitations, a control
experiment was performed, for which the catalyst was also put
in a filter, but remained in the reaction mixture for 3 h. For
MIL-53-NH2(100)-Mal-Pd, we found that the reaction was partly
heterogeneous,[12] which was in contrast to the quasi-homoge-
neous mechanism widely accepted for conventional supported
catalyst systems.[21b,22] During the hot filtration tests with the
palladium-containing MIXMOFs, the conversion and yield fur-
ther increased, although the catalyst had been removed. How-
ever, after 3 h, the conversion of the hot filtration test was sig-
nificantly lower than that of the experiment without catalyst
removal (Figure 7). Hence, contributions from a truly heteroge-
neous pathway were found. The observed activity after remov-
al of the catalyst might originate from leached species or very
small catalyst particles, which could not be retained by the
filter.
Table 7. Results of the catalytic tests.
Substrate Conversion Yield Selectivity TON
[%]
[%]
[%]
MIL-53-NH2(100)-Mal-Pd
MIL-53-NH2(80)-Mal-Pd
MIL-53-NH2(60)-Mal-Pd
MIL-53-NH2(40)-Mal-Pd
MIL-53-NH2(100)-Mal-Pd
MIL-53-NH2(80)-Mal-Pd
MIL-53-NH2(60)-Mal-Pd
MIL-53-NH2(40)-Mal-Pd
BrBz[a]
BrBz[a]
BrBz[a]
BrBz[a]
ClBz[b]
ClBz[b]
ClBz[b]
ClBz[b]
96
94
93
91
22
21
19
18
88
86
86
84
17
18
16
15
92
92
93
93
78
84
84
83
8236
8710
8369
8625
1740
1975
1730
1685
[a] Reaction conditions: BrBz (10 mmol), styrene (15 mmol), sodium ace-
tate (12 mmol), Pd (0.01 mol%), N-methyl-2-pyrrolidone (NMP; 10 mL),
1408C, 3 h. [b] Reaction conditions: ClBz (10 mmol), styrene (15 mmol),
calcium hydroxide (12 mmol), tetrabutylammonium bromide (TBAB;
6 mmol), Pd (0.01 mol%), NMP (10 mL), 1608C, 6 h.
the coupling of ClBz and styrene, the palladium-containing
MIXMOFs exhibited a slightly higher selectivity of 84% com-
pared with 78% for MIL-53-NH2(100)-Mal-Pd. The low micro-
pore volume of MIL-53-NH2(100)-Mal-Pd suggests that the pal-
ladium(II) complexes are located mainly on the outer surface
of the material. The modified MIXMOFs exhibited a significantly
higher micropore volume, which indicates a good distribution
of the immobilized palladium complexes inside the porous
framework structure. However, the same TONs were observed
for the palladium-containing mixed-linker frameworks and MIL-
53-NH2(100)-Mal-Pd, which led to the conclusion that, for the
MIXMOFs, not only the immobilized palladium complexes on
the outer surface, but also those inside the pores could be
reached and were catalytically active. Additional catalytic stud-
ies involving substrates of different sizes might be suitable to
further corroborate this claim in future studies.
Figure 7. Conversion of the hot filtration test after 0.5 and (0.5+2.5) h and
of the control experiment after 3 h. Reaction conditions: BrBz (10 mmol), sty-
rene (15 mmol), sodium acetate (12 mmol), Pd (0.01 mol%), NMP (10 mL),
1408C.
Conclusion
Isoreticular MIXMOFs based on MIL-53-NH2 were successfully
synthesized with defined ratios of 2-aminoterephthalate
(ABDC) and terephthalate (BDC) under ambient pressure. The
linker ratios found by NMR spectroscopy matched the applied
ratios very well; this confirmed that both linker molecules
were built into the framework with the same preference under
the applied reaction conditions. Furthermore, we could prove
that using MIXMOFs for the immobilization of defined palladiu-
m(II) complexes led to catalyst materials with high micropore
volumes of up to 0.11 cm3 gÀ1. Clearly, blocking of the pore
structure observed for MIL-53-NH2(100)-Mal-Pd could success-
fully be minimized by dilution of the amine groups in the
mixed frameworks. All new materials showed similar high ac-
For conventional supported palladium catalysts, it is well
known in the literature that the Heck reaction is exclusively
catalyzed by leached species in the form of defined palladium
complexes and not by the bulk material. The reaction mecha-
nism is “quasi-homogeneous”.[21b,22] In our case, defined palla-
dium(II) complexes are already immobilized in the framework
materials and, consequently, leaching from the solid catalyst
might not be necessary. To answer the question of whether
the reaction pathway was homogeneous owing to leached
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
6
&
ÞÞ
These are not the final page numbers!

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
tivity and selectivity in the Heck-type CÀC coupling reaction of
bromobenzene or chlorobenzene and styrene. Owing to the
high micropore volume of the modified MIXMOFs, the palladi-
um complexes immobilized inside the pores could be reached,
whereas for MIL-53-NH2(100)-Mal-Pd only the complexes on
the outer surface were catalytically active. Therefore, MIXMOFs
turned out to be interesting materials for further immobiliza-
tion of different metal complexes and might also be promising
for size- or shape-selective catalytic reactions.
Nitrogen physisorption
The samples were activated for 20 h at 1308C under vacuum and
analyzed using a Belsorp mini II instrument from BEL Japan.
FTIR spectroscopy
IR data were acquired using a Vertex 70 FTIR spectrometer from
Bruker Optics equipped with a Golden Gate Single Reflection ATR
sample cell from Specac. The data were collected from n˜ =4500 to
600 cmÀ1 and for each spectrum the arithmetic average of 400
measurements was taken.
Experimental Section
NMR spectroscopy
Synthesis of MIXMOFs
Samples (10 mg) were digested in NaOH/D2O (0.5 mL). Spectra
were recorded on a Bruker Ascend 400 MHz NMR spectrometer.
Chemical shifts were referenced to internal solvent resonances and
reported relative to tetramethylsilane (TMS).
Based on the synthesis of MIL-53-NH2(Al) under ambient pres-
sure,[12] MIXMOFs containing 80, 60, and 40% ABDC were synthe-
sized. The linker molecules (5.33 mmol; H2ABDC and H2BDC, ratios
are listed in Table 8) were dissolved in H2O (25 mL) and DMF
(50 mL) at 90 8C. A solution of Al(NO3)3·9H2O (2.0 g, 5.33 mmol) in
H2O (5 mL) was added and the reaction mixtures were stirred at
90 8C for 24 h. After filtration, the materials were washed with DMF
(3ꢂ25 mL) and H2O (1ꢂ25 mL). The samples were dried overnight
at room temperature and then for 3 days at 1308C in air.
Atomic absorption spectroscopy (AAS)
For AAS measurements, a Z-6100 Polarized Zeeman atomic absorp-
tion spectrometer from Hitachi was used. The palladium-containing
solid frameworks were digested in aqua regia (7 mL) and diluted
with distilled water to 100 mL.
Table 8. Ratios of H2ABDC and H2BDC that were applied in the synthesis
of MIXMOFs.
H2ABDC [%]
H2ABDC [g]
H2BDC [%]
H2BDC [g]
X-ray absorption spectroscopy (XAS)
MIL-53-NH2(80)
MIL-53-NH2(60)
MIL-53-NH2(40)
80
60
40
0.7730
0.5800
0.3860
20
40
60
0.1900
0.3800
0.5690
XAS experiments were performed at beamline BM25A at the Euro-
pean Synchrotron Radiation Facility (ESRF) in Grenoble, France. The
measurements were performed at the palladium K-edge at
24350 keV with a Si(311) double-crystal monochromator and a max-
imum synchrotron beam current of 200 mA. Spectra of undiluted
powder between two Kapton windows were recorded in fluores-
cence mode at ambient temperature.
PSM of MIXMOFs
The MIXMOFs were modified in a two-step PSM reaction. First
step: modification with maleic anhydride: maleic anhydride
(4 equiv, based on the number of amine groups) was dissolved in
acetonitrile (50 mL). MIXMOFs (2 mmol) were suspended in the so-
lution and the reaction mixtures were heated to 808C for 24 h.
After filtration, the resulting materials were washed with acetoni-
trile (5ꢂ20 mL), DMF (1ꢂ20 mL), and H2O (1ꢂ20 mL). The samples
were dried overnight at room temperature and then for 3 days at
1308C in air.
Gas chromatography (GC)
For GC measurements, a GC-2010 Plus instrument from Shimadzu
with a nonpolar column (Rxi-5Sil MS, length: 30 m, diameter:
0.25 mm, film thickness: 0.25 mm) and a flame-ionization detector
(FID) was used. The sample (1 mL) was injected and vaporized at
2508C. The column was heated from 50 to 2808C at a rate of
Second step: modification with palladium acetate: palladium(II)
acetate (0.15 mmol) was dissolved in DMF (20 mL). The modified
MIXMOFs (1.5 mmol) were suspended in the solution and the reac-
tion mixtures were heated to 608C for 4 h. After filtration, the re-
sulting materials were washed with DMF (3ꢂ20 mL) and H2O (1ꢂ
20 mL). The samples were dried overnight at room temperature
and then for 3 days at 1308C in air.
10 KminÀ1
.
Heck reaction of bromobenzene and styrene
In a typical experiment, bromobenzene (BrBz; 10 mmol), styrene
(15 mmol), sodium acetate (12 mmol), and diethylene glycol dibu-
tyl ether (2.3 mmol as an internal GC standard) were dissolved in
NMP (10 mL) in a sealed glass tube. The catalyst (0.01 mol%, based
on Pd in regard to BrBz) was suspended in the reaction mixture
and the mixture was then heated to 1408C for 3 h. For the hot fil-
tration test, the catalyst was put into a paper filter bag, which was
removed from the reaction mixture after 0.5 h. The reaction was
then resumed for another 2.5 h. Samples for GC analysis were
taken after 0.5 and (0.5+2.5) h.
Powder X-ray diffraction (PXRD)
The measurements were performed using a Bruker D8 Advance dif-
fractometer. The samples were analyzed in the range of 2q=6–508
using CuKa radiation. The step width was 2q=0.00828 with a dwell
time of 2 s.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
7
&
ÞÞ
These are not the final page numbers!

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
[6] a) H. Li, M. Eddaoudi, M. O’Keeffe, O. M. Yaghi, Nature 1999, 402, 276–
279; b) M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe,
O. M. Yaghi, Science 2002, 295, 469–472.
[7] S. Biswas, T. Ahnfeldt, N. Stock, Inorg. Chem. 2011, 50, 9518–9526.
[8] a) S. Couck, T. Remy, G. V. Baron, J. Gascon, F. Kapteijn, J. F. M. Denayer,
Phys. Chem. Chem. Phys. 2010, 12, 9413–9418; b) E. Stavitski, E. A.
Pidko, S. Couck, T. Remy, E. J. M. Hensen, B. M. Weckhuysen, J. Denayer,
J. Gascon, F. Kapteijn, Langmuir 2011, 27, 3970–3976.
[9] a) L. Alaerts, M. Maes, L. Giebeler, P. A. Jacobs, J. A. Martens, J. F. M. De-
nayer, C. E. A. Kirschhock, D. E. De Vos, J. Am. Chem. Soc. 2008, 130,
14170–14178; b) S. Couck, E. Gobechiya, C. E. A. Kirschhock, P. Serra-
Crespo, J. Juan-AlcaÇiz, A. Martinez Joaristi, E. Stavitski, J. Gascon, F.
Kapteijn, G. V. Baron, J. F. M. Denayer, ChemSusChem 2012, 5, 740–750.
[10] M. G. Goesten, J. Juan-AlcaÇiz, E. V. Ramos-Fernandez, K. B. Sai Sankar
Gupta, E. Stavitski, H. van Bekkum, J. Gascon, F. Kapteijn, J. Catal. 2011,
281, 177–187.
Heck reaction of chlorobenzene and styrene
In a typical experiment, chlorobenzene (ClBz) (10 mmol), styrene
(15 mmol), calcium hydroxide (12 mmol), TBAB (6 mmol), and di-
ethylene glycol dibutyl ether (2.3 mmol as an internal GC standard)
were dissolved in NMP (10 mL) in a sealed glass tube. The catalyst
(0.01 mol%, based on Pd in regard to ClBz) was suspended in the
reaction mixture and the mixture was heated to 1608C for 6 h.
Acknowledgements
We thank the ESRF (European Synchrotron Radiation Facility) in
Grenoble, France, for providing beamtime at beamline BM25A.
Prof. P. W. Roesky (KIT) is acknowledged for providing the NMR
spectrometer. T.S.B. thanks the Helmholtz Research School
Energy-Related Catalysis and the FCI for financial support. This
study was supported by the KIT competence portfolio “Matter
and Materials–Applied and New Materials”.
[11] a) S. Bhattacharjee, D.-A. Yang, W.-S. Ahn, Chem. Commun. 2011, 47,
3637–3639; b) K. Oisaki, Q. Li, H. Furukawa, A. U. Czaja, O. M. Yaghi, J.
Am. Chem. Soc. 2010, 132, 9262–9264.
[12] M. A. Gotthardt, A. Beilmann, R. Schoch, J. Engelke, W. Kleist, RSC Adv.
2013, 3, 10676–10679.
[13] a) D. N. Bunck, W. R. Dichtel, Chem. Eur. J. 2013, 19, 818–827; b) S. E.
Wenzel, M. Fischer, F. Hoffmann, M. Frçba, J. Mater. Chem. 2012, 22,
10294–10302; c) A. D. Burrows, CrystEngComm 2011, 13, 3623–3642;
d) S. Marx, W. Kleist, J. Huang, M. Maciejewski, A. Baiker, Dalton Trans.
2010, 39, 3795–3798; e) Y. Jiang, J. Huang, S. Marx, W. Kleist, M.
Hunger, A. Baiker, J. Phys. Chem. Lett. 2010, 1, 2886–2890; f) H. Deng,
C. J. Doonan, H. Furukawa, R. B. Ferreira, J. Towne, C. B. Knobler, B.
Wang, O. M. Yaghi, Science 2010, 327, 846–850; g) W. Kleist, F. Jutz, M.
Maciejewski, A. Baiker, Eur. J. Inorg. Chem. 2009, 3552–3561.
[14] T. Lescouet, E. Kockrick, G. Bergeret, M. Pera-Titus, S. Aguado, D. Farrus-
seng, J. Mater. Chem. 2012, 22, 10287–10293.
[15] G. Fꢃrey, C. Serre, Chem. Soc. Rev. 2009, 38, 1380–1399.
[16] S. J. Garibay, Z. Wang, S. M. Cohen, Inorg. Chem. 2010, 49, 8086–8091.
[17] a) S. Gross, M. Bauer, Adv. Funct. Mater. 2010, 20, 4026–4047; b) M.
Bauer, C. Gastl, Phys. Chem. Chem. Phys. 2010, 12, 5575–5584.
[18] S. Schuster, E. Klemm, M. Bauer, Chem. Eur. J. 2012, 18, 15831–15837.
[19] Y. Huang, Z. Zheng, T. Liu, J. Lꢄ, Z. Lin, H. Li, R. Cao, Catal. Commun.
2011, 14, 27–31.
Keywords: CÀC coupling · heterogeneous catalysis · metal–
organic frameworks · palladium · synthesis design
[1] a) A. D. Burrows, L. L. Keenan, CrystEngComm 2012, 14, 4112–4114;
b) K. K. Tanabe, S. M. Cohen, Chem. Soc. Rev. 2011, 40, 498–519; c) Z.
Wang, S. M. Cohen, Chem. Soc. Rev. 2009, 38, 1315–1329; d) M. J. Ingle-
son, J. P. Barrio, J.-B. Guilbaud, Y. Z. Khimyak, M. J. Rosseinsky, Chem.
Commun. 2008, 2680–2682; e) A. D. Burrows, C. G. Frost, M. F. Mahon,
C. Richardson, Angew. Chem. Int. Ed. 2008, 47, 8482–8486; Angew.
Chem. 2008, 120, 8610–8614.
[2] a) H. Furukawa, K. E. Cordova, M. O’Keeffe, O. M. Yaghi, Science 2013,
341, 1230444; b) A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Chem.
Commun. 2012, 48, 11275–11288; c) S. T. Meek, J. A. Greathouse, M. D.
Allendorf, Adv. Mater. 2011, 23, 249–267; d) A. Corma, H. Garcia, F. X.
Llabres i Xamena, Chem. Rev. 2010, 110, 4606–4655; e) J. Lee, O. K.
Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen, J. T. Hupp, Chem. Soc. Rev.
2009, 38, 1450–1459.
[20] S.-N. Kim, S.-T. Yang, J. Kim, J.-E. Park, W.-S. Ahn, CrystEngComm 2012,
14, 4142–4147.
[21] a) W. Kleist, S. S. Prçckl, K. Kçhler, Catal. Lett. 2008, 125, 197–200; b) K.
Kçhler, W. Kleist, S. S. Prçckl, Inorg. Chem. 2007, 46, 1876–1883.
[22] a) A. F. Schmidt, A. A. Kurokhtina, Kinet. Catal. 2012, 53, 714–730; b) S.
Reimann, J. Stçtzel, R. Frahm, W. Kleist, J.-D. Grunwaldt, A. Baiker, J. Am.
Chem. Soc. 2011, 133, 3921–3930; c) N. T. S. Phan, M. Van Der Sluys,
C. W. Jones, Adv. Synth. Catal. 2006, 348, 609–679; d) S. S. Prçckl, W.
Kleist, M. A. Gruber, K. Kçhler, Angew. Chem. Int. Ed. 2004, 43, 1881–
1882; Angew. Chem. 2004, 116, 1917–1918; e) F. Zhao, M. Shirai, Y.
Ikushima, M. Arai, J. Mol. Catal. A: Chem. 2002, 180, 211–219.
[3] a) M. O’Keeffe, Chem. Soc. Rev. 2009, 38, 1215–1217; b) G. Fꢃrey, Chem.
Soc. Rev. 2008, 37, 191–214; c) J. L. C. Rowsell, O. M. Yaghi, Microporous
Mesoporous Mater. 2004, 73, 3–14.
[4] a) T. Ahnfeldt, D. Gunzelmann, T. Loiseau, D. Hirsemann, J. Senker, G.
Ferey, N. Stock, Inorg. Chem. 2009, 48, 3057–3064; b) T. Loiseau, C.
Serre, C. Huguenard, G. Fink, F. Taulelle, M. Henry, T. Bataille, G. Ferey,
Chem. Eur. J. 2004, 10, 1373–1382; c) S. S. Y. Chui, S. M. F. Lo, J. P. H.
Charmant, A. G. Orpen, I. D. Williams, Science 1999, 283, 1148–1150.
[5] a) M. T. Wharmby, J. P. S. Mowat, S. P. Thompson, P. A. Wright, J. Am.
Chem. Soc. 2011, 133, 1266–1269; b) J. A. Groves, S. R. Miller, S. J. War-
render, C. Mellot-Draznieks, P. Lightfoot, P. A. Wright, Chem. Commun.
2006, 3305–3307.
Received: April 25, 2014
Revised: September 9, 2014
Published online on && &&, 0000
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
8
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
M. A. Gotthardt, R. Schoch, T. S. Brunner,
M. Bauer,* W. Kleist*
&& – &&
Design of Highly Porous Single-Site
Catalysts through Two-Step
Postsynthetic Modification of Mixed-
Linker MIL-53(Al)
Holey MOFs! Defined single-site palladi-
um catalysts have been synthesized
through postsynthetic modification
(PSM) of mixed-linker metal–organic
frameworks (MIXMOFs). The high poros-
ity of the MIXMOF catalysts was re-
tained throughout the modification pro-
cess. This resulted in accessible and
highly active catalytic sites for Heck-
type CÀC coupling reactions.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 0000, 00, 1 – 9
&
9
&
ÞÞ
These are not the final page numbers!