Catalytic properties of MIL-101

Article abstract of DOI:10.1039/b718371b

A very high catalytic activity in the cyanosilylation reaction was observed for MIL-101, a chromium based metal-organic framework; moreover, MIL-101 is also a remarkably stable support for palladium in hydrogenation reactions, with significantly higher activity than e.g. palladium on activated carbon. The Royal Society of Chemistry.

Full text of DOI:10.1039/b718371b

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Catalytic properties of MIL-101w
Antje Henschel,a Kristina Gedrich,a Ralph Kraehnertb and Stefan Kaskel*a
Received (in Cambridge, UK) 28th November 2007, Accepted 13th June 2008
First published as an Advance Article on the web 1st August 2008
DOI: 10.1039/b718371b
A very high catalytic activity in the cyanosilylation reaction was
observed for MIL-101, a chromium based metal–organic frame-
work; moreover, MIL-101 is also a remarkably stable support
for palladium in hydrogenation reactions, with significantly
higher activity than e.g. palladium on activated carbon.
In the last decade, metal–organic frameworks (MOFs) have
attracted considerable interest due to their remarkable properties
such as high specific surface areas combined with a high crystal-
linity.1,2 Due to these properties MOFs are promising materials
Fig. 1 Trimeric chromium(III)-cluster in MIL-101, the octahedra
for applications in gas storage,3 separation processes,4 and cata-
lysis.5 The use of MOFs as solid catalysts is particularly interest-
ing, since the pore size and functionality of the framework can be
adjusted over a wide range for a variety of catalytic reactions.2
Even though a large number of different MOFs is known, until
today only a few of them have been tested in catalytic reactions.
represent the chromium ions, carbon and oxygen atoms are in light
grey and black, respectively; the medium grey atoms represent either
an oxygen atom derived from a water molecule or a hydroxyl group or
a fluorine atom.
can be easily removed in vacuum and at elevated temperature, and
leave behind the chromium accessible for potential reactants.
MIL-101 samples were prepared according to a synthesis route
reported by Fe
´
rey et al.8 The X-ray diffraction pattern of the
In pioneering work, Ferey et al. published in 2005 the synthesis
´
and characterisation of the metal–organic framework MIL-101
(Cr3X(H2O)2O(bdc)3; X = F, OH; bdc = benzene-1,4-dicarbox-
ylate) showing a very high specific surface area as compared to
other porous coordination polymers known until today (SLangmuir
= 5900 m2 gꢀ1) and a high pore volume (approx. 2.0 cm3 gꢀ1).6
MIL-101 consists of m3-oxo bridged chromium(III)-trimers cross-
linked by terephthalate groups. The chromium ions of the
MIL-101 structure have a pseudo-octahedral coordination. The
vertices of these pseudo-octahedra are occupied by the m3-O atom
in the middle of the chromium trimers and four oxygen atoms
derived from the carboxylate groups of the terephthalate linkers.
The coordination sphere around each chromium is finally filled up
by a water molecule, a fluorine atom or a hydroxyl group (Fig. 1).
MIL-101 is an ideal MOF for applications in heterogeneous
catalysis due to the following reason. The pores and pore windows
are big enough to give access for even large reactant molecules
diffusing into the pores, but also large product molecules can
move out of the network. For the catalytic activity of solid
catalysts the accessibility of the catalytic sites is even more
essential.7 As mentioned above, the pseudo-octahedral coordina-
tion sphere around each chromium atom is completed with either
water molecules, fluorine or a hydroxyl group. In contrast to the
fluorine and the hydroxyl group, the coordinated water molecules
green powder matches the calculated pattern and thus confirms
the formation of MIL-101 (see ESIw, Fig. S1). The specific surface
area of the obtained porous material amounts to 2367 m2 gꢀ1
(SBET, p/p0 = 0.1) and was determined using nitrogen physisorp-
tion measurements (see ESIw, Fig. S5). In contrast to Ferey et al.
´
we could not observe any increase of the specific surface area via
post-reaction hydrothermal treatment of MIL-101 in ethanol.
To evaluate the catalytic activity of MIL-101, we have used the
cyanosilylation of aldehydes as a test reaction catalyzed by Lewis
acids. Fujita et al. were the first to demonstrate the catalytic
activity for a two-dimensional square network material [Cd(4,40-
bpy)2](NO3)2.9 The catalytic activity of Cu3(BTC)2 (BTC =
benzene-1,3,5-tricarboxylate) in the cyanosilylation of benzalde-
hyde was studied in detail, because this material also has well
defined accessible Lewis acidic sites, exposed to the inner surface.
However, only moderate yields of 57% (88.5% selectivity) were
achieved after prolonged duration (72 h) with a high catalyst
loading of 5 mol% referred to the amount of benzaldehyde.10
According to our results presented here, MIL-101 has a much
higher activity towards the addition of trimethylsilylcyanide
(TMSCN) to benzaldehyde (Scheme 1) as compared to other
MOFs such as Cu3(BTC)2, even if a lower catalyst loading of only
0.5 mol% is used. In a typical reaction, 15 mg MIL-101
(B0.02 mmol) were used for the conversion of 4 mmol benzalde-
hyde in heptane at 313 K. An excess of 100% TMSCN was
employed with respect to the amount of benzaldehyde. The
composition of the reaction mixture was determined by GC-MS
analysis (Fig. 2). All benzaldehyde is converted after 3 h leading to
a product yield of 98.5%. In addition to this remarkable high
catalytic activity, MIL-101 has the advantage that chromium(III)
cannot be as easily reduced by the benzaldehyde, whereas
a Department of Inorganic Chemistry, Dresden University of
Technology, Mommsenstr. 6, D-01069 Dresden, Germany.
E-mail: stefan.kaskel@chemie.tu-dresden.de;
Fax: +49-351-46337287; Tel: +49-351-46334885
b Leibniz-Institute for Catalysis e.V. at the University of Rostock,
Branch Berlin, Richard-Willstaetter-Str. 12, D-12489 Berlin,
Germany
w Electronic supplementary information (ESI) available: Synthetic
details, powder X-ray diffraction patterns, nitrogen physisorption
isotherms, additional catalytic data. See DOI: 10.1039/b718371b
ꢁc
This journal is The Royal Society of Chemistry 2008
4192 | Chem. Commun., 2008, 4192–4194

View Online
MIL-101 can be reused several times as a catalyst in the cyano-
silylation of benzaldehyde with a somewhat reduced activity.
Besides the demonstrated catalytic activity of pure MIL-101,
MIL-101 is also an attractive candidate for a catalyst support
due to its high porosity and a relatively high thermal stability.
Recently, we have shown that MOF-5 is a suitable catalyst
support for palladium.11 The supported catalyst, denoted as
Pd/MOF-5, showed a higher activity towards the hydrogena-
tion of styrene than palladium on activated carbon with similar
palladium loadings. Unfortunately, one serious disadvantage of
the Pd/MOF-5 catalyst is its instability in contact with water or
humid air. Therefore, the aim of our continuous research was
the preparation of other palladium supported MOFs combining
both, high activity towards hydrogenation reactions and stabi-
lity towards air and moisture. MIL-101 seemed to be the ideal
palladium support for this aim, since it is long-term stable in air
atmosphere and has a very high specific pore volume. Thus,
a catalyst with palladium supported on MIL-101 (denoted
Pd/MIL-101) was prepared and tested for catalytic activity
and stability in liquid-phase and gas-phase hydrogenation,
respectively, namely the hydrogenation of styrene, and the
hydrogenation of acetylene in the presence of ethylene.
Scheme 1 (a) Addition of trimethylsilylcyanide to benzaldehyde, and
(b) hydrogenation of styrene.
copper(II) in Cu3(BTC)2 is easily reduced leading to a decomposi-
tion of the framework at higher reaction temperatures. However,
the framework of MIL-101 is intact after usage as a catalyst
judging from the powder X-ray diffraction pattern of the catalyst
after cyanosilylation (see ESIw, Fig. S2). Compared to the XRD
pattern of the as made material a redistribution of the peak
intensities but no change of the peak positions is observed. This
finding is in accordance with the results of Ferey et al. describing
´
the same observation for a MIL-101 sample with incorporated
Keggin polyanions.6 The change in intensities can therefore be
assigned to high-boiling residues of the catalytic reaction that are
not completely removed in vacuo.
To prove the heterogeneous mechanism of the cyanosilyl-
ation catalysed by MIL-101, the catalyst was filtered off after
15 min reaction time and stirring of the filtrate continued at
the same reaction temperature. The reaction was again moni-
tored by GC-MS analysis and the concentrations of benzalde-
hyde and the product are shown in Fig. 2 (broken line). These
results confirm the assumption of a heterogeneous mechanism
since neither additional benzaldehyde is consumed nor is
product formed after filtration.
Supported palladium was prepared via incipient wetness
impregnation of activated MIL-101 using Pd(acac)2 (acac =
acetylacetonate) as precursor diluted in chloroform.11 The
required volume of chloroform was determined by measuring
the specific total pore volume (VP = 1.46 cm3 gꢀ1, p/p0 = 0.9)
of the synthesized MIL-101. The green powder was dried under
reduced pressure at ambient temperature. For the conversion
of the precursor into metallic palladium, Pd(acac)2/MIL-101
was heated in a hydrogen flow at 473 K for 1 h. The specific
surface area of palladium supported MIL-101 decreased to
2046 m2 gꢀ1 (SBET, p/p0 = 0.1) (see ESIw, Fig. S5). Thus, the
loss of the specific surface area amounts to only 14% compared
to 67% previously found for Pd/MOF-5. The powder X-ray
diffraction patterns of MIL-101 and Pd/MIL-101 do not differ
(see ESIw, Fig. S3) indicating the stability of the MIL-101
network under the employed reduction conditions.
For a more comprehensive study of the catalytic activity of
MIL-101 in the cyanosilylation of benzaldehyde, a recycling test
with three consecutive runs was performed (see ESIw, experimen-
tal section and Fig. S8). As mentioned before, a product yield of
98.5% is achieved in the first run after 3 h. In the second and the
third run the product yields, determined after the same reaction
time (3 h), decrease to 91.5% and 91.3%, respectively. In addi-
tion, a decrease of the reaction rate is observed from the first to
the third run as can be seen from the declining slopes. Thus,
The dispersion of palladium in the fresh Pd/MIL-101 material
was analysed via hydrogen chemisorption. In the procedure,
Pd/MIL-101 was flushed extensively in argon at 443 K to desorb
all hydrogen and decompose palladium hydride species. There-
after, the sample was cooled in argon, exposed to a flow of 5% H2
in argon, and heated at 10 K minꢀ1 to 443 K. The recorded total
H2 consumption amounted to 2.16 mmol H2 for 63.5 mg of Pd/
MIL-101. Assuming that only palladium interacts with hydrogen
and that each surface palladium atom accommodates one hydro-
gen atom, this translates into a dispersion of about 72%, suggest-
ing the diameter of the palladium particles is around 1.5 nm.12
In contrast to previously reported Pd/MOF-5, all Pd/MIL-101
materials could be synthesized and handled in an air atmosphere
due to the superior stability of MIL-101. A nitrogen physisorption
measurement repeated on a Pd/MIL-101 sample stored in air
for two months confirmed the intended higher stability also for
Pd/MIL-101, since storage-induced changes of the specific surface
area were found to be negligible (see ESIw, Fig. S7).
Fig. 2 Reactant and product concentration during the MIL-101
catalysed cyanosilylation of benzaldehyde, Solid line: K concentration
of benzaldehyde, ’ concentration of product. Broken line: filtration
test, J concentration of benzaldehyde, & concentration of product.
The new air-stable catalyst was tested in the hydrogenation of
styrene to evaluate the catalytic activity of Pd/MIL-101 and to
compare the results with those obtained with other palladium
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4192–4194 | 4193

View Online
was tested in the gas phase hydrogenation of an acetylene–
ethylene mixture in a continuous fixed-bed reactor operated in
repeated temperature cycles between 313 and 373 K. For the test,
20 mg of Pd/MIL-101 were diluted 1 : 10 with SiC. A gas mixture
containing acetylene (1.7%), excess ethylene (42.9%), hydrogen
(condition A: 3.4%, condition B: 2.1%) and nitrogen (balance)
was continuously dosed to the reactor operated at 2 bar with a
flow rate of 13.7 ml minꢀ1 (STP). The temperature was increased
from 313 to 373 K in 10 K steps, holding the temperature
constant for about 3.3 h at each temperature, and repeating this
program alternating with higher and lower hydrogen feed (i.e.
conditions A, B).
Compared to the same total amount of palladium, Pd/MIL-
101 showed an initial activity superior to that of all other tested
catalysts such as 0.05 wt% Pd/ZnO or 0.05 wt% Pd/Al2O3,
corresponding to the presumed high palladium dispersion (H2
chemisorption) facilitated by the framework of MIL-101. For
the feed condition ‘‘A’’ with higher hydrogen content, between
80 and 100% of acetylene was converted along with smaller
amounts of ethylene (see ESIw, Fig. S10), ethane being the main
product. Besides the high initial activity, Pd/MIL-101 showed
also significant activity after about 145 h time-on-stream as
shown in Fig. 4, indicating a remarkably slow deactivation of
the catalyst also under harsh reaction conditions. The observed
slow and continuous degradation in performance with time-
on-stream could possibly be explained by blocking of the pores
with residual hydrocarbon, but deserves further investigations.
Summarizing, pure MIL-101 has a significantly higher catalytic
activity in the cyanosilylation of benzaldehyde than reported for
Cu3(BTC)2 and other MOFs. Moreover, it is a suitable support
for palladium which can be incorporated via incipient wetness
impregnation. Compared to Pd/MOF-5 and commercial sup-
ported palladium catalysts, Pd/MIL-101 shows a better catalytic
performance in the hydrogenation of styrene and cyclooctene,
and a high and sustained activity in the gas phase hydrogenation
of acetylene–ethylene mixtures, the essential advantage of
Pd/MIL-101 over Pd/MOF-5 being its stability towards air, mois-
ture and also (reducing) reaction conditions up to at least 373 K.
Fig. 3 Hydrogenation of styrene with different palladium supported
catalysts (1 wt%): E Pd/MIL-101, ’ Pd/MOF-5,
Pd/Norit A,
Pd/C; the broken line represents the filtration test.
supported catalysts. The reaction was carried out in a static
hydrogen atmosphere (1 atm) at 308 K (Scheme 1) and was
monitored by GC-MS analysis for 24 h. Fig. 3 depicts that the use
of Pd/MIL-101 as a hydrogenation catalyst leads to a complete
conversion of styrene to ethylbenzene after 7 h. Under the same
reaction conditions, Pd/MOF-5 yields B80 wt% hydrogenation
product after the same reaction time whereas Pd/Norit A affords
B65 wt% and Pd/C less than 50 wt% ethylbenzene. To assure
comparability, only supported catalysts with a palladium content
of 1 wt% were used. The results prove the higher activity of
Pd/MIL-101 towards the hydrogenation of styrene. Moreover,
Pd/MIL-101 clearly shows the highest activity in the hydrogena-
tion of cis-cyclooctene compared to all other palladium supported
catalysts tested (see ESIw, Fig. S9). The higher activity as com-
pared to Pd/MOF-5 may be caused by the larger pore size of
Pd/MIL-101 and is an argument for the efficient integration of
active sites inside the pore system and not outside.
As observed for MIL-101 used in cyanosilylation, the X-ray
diffraction pattern of the used Pd/MIL-101 catalyst shows a
similar redistribution of peak intensities compared to the one ob-
tained for Pd/MIL-101 before hydrogenation (see ESIw, Fig. S4)
caused by non-volatile residues of the hydrogenation reaction.
To eventually elucidate the long-term activity and stability of
Pd/MIL-101 also at more severe reaction conditions, the catalyst
Notes and references
1 F. Schuth, K. S. W. Sing and J. Weitkamp, Handbook of Porous
¨
Solids, Wiley-VCH, Weinheim, 2002.
2 J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous
Mater., 2004, 73, 3–14.
3 M. Hirscher and B. Panella, Scr. Mater., 2007, 56, 809–812.
4 P. S. Barcia, F. Zapata, J. A. C. Silva, A. E. Rodrigues and B.
Chen, J. Phys. Chem. B, 2007, 111, 6101–6103.
5 S. Kitagawa, R. Kitaura and S.-I. Noro, Angew. Chem., Int. Ed.,
2004, 43, 2334–2375.
6 G. Fe
´
Surble
rey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S.
and I. Margiolaki, Science, 2005, 309, 2040–2042.
7 L. Alaerts, E. Seguin, H. Poelman, F. Thibault-Starzyk, P. A.
´
Jacobs and D. E. De Vos, Chem.–Eur. J., 2006, 12, 7353–7363.
´
8 P. Horcajada, C. Serre, M. Vallet-Regı, M. Sebban, F. Taulelle and
G. Ferey, Angew. Chem., Int. Ed., 2006, 45, 5974–5978.
´
9 M. Fujita, Y. J. Kwon, S. Washizu and K. Ogura, J. Am. Chem.
Soc., 1994, 116, 1151–1152.
10 K. Schlichte, T. Kratzke and S. Kaskel, Microporous Mesoporous
Mater., 2004, 73, 81–88.
Fig. 4 Hydrogen conversion vs. time-on-stream in hydrogenation of
a mixture of acetylene–ethylene using 1 wt% Pd/MIL-101 in a
continuous fixed-bed reactor operated in repeated temperature cycles
between 313 and 373 K feeding a gas mixture with a hydrogen content
of 3.4% (condition A) or 2.1% (condition B).
11 M. Sabo, A. Henschel, H. Frode, E. Klemm and S. Kaskel, J.
¨
Mater. Chem., 2007, 17, 3827–3832.
12 G. Ertl, H. Knozinger and J. Weitkamp, Handbook of Hetero-
¨
geneous Catalysis, Wiley-VCH, Weinheim, 2008.
ꢁc
This journal is The Royal Society of Chemistry 2008
4194 | Chem. Commun., 2008, 4192–4194