The remarkable catalytic activity of the saturated metal organic framework V-MIL-47 in the cyclohexene oxidation

Article abstract of DOI:10.1039/c0cc01506g

The remarkable catalytic activity of the saturated metal organic framework MIL-47 in the epoxidation of cyclohexene is elucidated by means of both experimental results and theoretical calculations. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/c0cc01506g

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
The remarkable catalytic activity of the saturated metal organic
framework V-MIL-47 in the cyclohexene oxidationw
Karen Leus,a Ilke Muylaert,a Matthias Vandichel,b Guy B. Marin,c Michel Waroquier,b
Veronique Van Speybroeck*b and Pascal Van Der Voort*a
Received 20th May 2010, Accepted 4th June 2010
First published as an Advance Article on the web 18th June 2010
DOI: 10.1039/c0cc01506g
The remarkable catalytic activity of the saturated metal organic
framework MIL-47 in the epoxidation of cyclohexene is elucidated
by means of both experimental results and theoretical calculations.
Fig. 1 depicts the cyclohexene conversion and the detailed
product distributions for MIL-47 and the homogeneous
VO(acac)2 catalyst. In both reactions, we used an identical
amount of V-sites (0.4 mmol, 0.095 g MIL-47 and 0.106 g
VO(acac)2). The four main reaction products formed are
cyclohexene oxide 2, cyclohexane-1,2-diol 3, tert-butyl-2-
cyclohexenyl-1-peroxide 4 and 2-cyclohexen-1-one 5. From
these figures it can be seen that the homogeneous catalyst
VO(acac)2 and the MIL-47 show high reaction rates for the
conversion of cyclohexene. The initial turnover frequencies
considered at the first 20 min are 57 hÀ1 and 43 hÀ1 for the
VO(acac)2 and the MIL-47 respectively. Both catalysts showed
an almost linear conversion rate during the first 2 h of reaction
and finally topped out at 60% conversion. The turnover
numbers calculated at the end of the reaction were 62 and
68 for VO(acac)2 and MIL-47 respectively. Moreover,
considering the product distributions, truly interesting similarities
between both catalysts are observed.
Metal organic frameworks (MOFs) are crystalline porous solids
composed of a three-dimensional network of metal ions held
in place by multidentate organic molecules.1,2 Recently, the
catalytic applications of MOFs have been reviewed by Hupp
et al. and Farrusseng et al.3,4 However, catalytic reaction
mechanisms in metal organic frameworks and in particular at
saturated coordinated centres are still poorly understood.
In this study, we describe the high catalytic activity of a
coordinative saturated V-MOF in the liquid phase oxidation of
cyclohexene. The metal organic framework MIL-47 consists of a
porous terephthalate framework built from infinite chains of
V4+O6 octahedra and has a three-dimensional orthorhombic
structure that exhibits large pores.5 Every structure unit in
MIL-47 contains an O–V+IV–O bridge. The linkage of the
V+IV to dicarboxylate groups of the terephthalate linkers results
in an octahedral structure unit. Hence each vanadium(+IV)
centre is saturated.
Firstly, the cyclohexene oxide is formed during the first 2 h
and is topped out at 15%, but its yield decreases with
increasing reaction time. At the same time and at the same
reaction rate the oxide is decreasing, the cyclohexane-1,2-diol
formation is observed, indicating a consecutive reaction is
taking place. After 5 h, the diol formation reaches a plateau
at 15%. Additionally, both catalysts produce a significant
amount of tert-butyl-2-cyclohexenyl-1-peroxide 4 (30%) as
determined by GC-GC-TOF-MS and a minimal amount of
2-cyclohexen-1-one 5 (5%).
The selective oxidation of cyclohexene is particularly
interesting as it can yield, depending on the catalyst and the
reaction conditions, several products as shown in Scheme 1.
In this study, the reaction was carried out with tert-butyl
hydroperoxide (70 wt% TBHP in water) as oxidant and
chloroform as solvent. The internal standard used was
1,2,4-trichlorobenzene. The molar ratios were chloroform :
In order to examine the heterogeneity of the catalytic
reaction, a hot filtration experiment was performed. The
catalyst was separated from the reaction mixture after 1 h
using a combined nylon-membrane filter. Fig. 2 presents the
conversion curve of cyclohexene in the presence of MIL-47
and the cyclohexene conversion after filtration. Also the sum
of products 2, 3 and 5 is shown with and without the hot
filtration. It is clear from this figure that both the conversion
of cyclohexene and the formation of oxidized products are
strongly reduced after filtration of the catalyst. This indicates
that the catalytic reaction is predominantly heterogeneous.
There is however some homogeneous catalytic activity left.
This activity is due to a combination of radical mechanisms
cyclohexene : TBHP : 1,2,4-trichlorobenzene
= 462 : 1 : 2 : 1.
The reaction temperature was kept constant at 323 K. Samples
were taken out from the reaction mixture at certain times and
analyzed by gas chromatography. Next, MIL-47 was
compared with a homogeneous vanadium catalyst vanadyl
acetylacetonate VO(acac)2. Prior to the reaction, all the catalysts
were dried under vacuum. Blank tests were performed without
catalyst and no cyclohexene conversion was observed.
a Department of Inorganic and Physical Chemistry,
Centre for Ordered Materials, Organometallics and Catalysis,
Ghent University, Krijgslaan 281-S3, 9000 Ghent, Belgium.
E-mail: pascal.vandervoort@ugent.be
b Centre for Molecular Modeling, Ghent University,
Technologiepark 903, 9052 Zwijnaarde, Belgium.
E-mail: Veronique.vanspeybroeck@ugent.be
c Laboratory for Chemical Technology, Ghent University,
Krijgslaan 281-S5, 9000 Ghent, Belgium
w Electronic supplementary information (ESI) available: XRF analysis,
nitrogen adsorption isotherms, XRD patterns, experimental and
computational details. See DOI: 10.1039/c0cc01506g
Scheme 1
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 5085–5087 | 5085

View Online
completely clogged by organic compounds. Subsequently,
MIL-47 was regenerated by a treatment in a tubular furnace
under a nitrogen flow at 523 K for four hours to remove the
organic compounds.
After this calcination treatment the porosity was completely
regenerated as can be seen in Fig. S.2.w Additionally, X-ray
diffraction (XRD) patterns show that the crystalline structure
of MIL-47 is preserved after regeneration (see Fig. S.3w).
Finally, the catalytic activity of MIL-47 is tested in a second
catalytic run, in which 27 mg catalyst was recovered and used.
In order to make a fair comparison, we also show first runs,
using respectively 20 mg and 40 mg catalyst, in Fig. 3.
Although the kinetics are initially different, the final conversion
in the second run does not show any significant loss of activity.
Fig. 1 Total cyclohexene conversion and detailed product distribution
in the cyclohexene oxidation with (a) MIL-47 and (b) VO(acac)2 in
chloroform with tert-butyl hydroperoxide (70 wt% solution in water)
at 323 K. The numbers of the curves correspond with the numbers in
Scheme 1.
Fig. 3 Cyclohexene conversion for MIL-47 in its first run with (&)
40 mg MIL-47, (J) 20 mg MIL-47 catalyst mass and (’) after
regeneration with 27 mg MIL-47.
(formation of product 4 occurs via radical pathways (vide infra))
and leached V. By XRF analysis we have determined that
12.8% of the V-sites have leached out after 1 h of reaction (see
Fig. S.1 in the supplementary informationw). Finally, we also
show the catalytic conversion caused by dissolving exactly the
amount of VO(acac)2 that corresponds to the 12.8% leached
V-sites (.). This leads however to only 5% conversion.
Furthermore, the stability and the regenerability in terms of
morphology and catalytic performance of MIL-47 were studied.
The original synthesized MOF exhibits a specific surface area
around 800 m2 gÀ1. After the first catalytic run, the pores are
Besides MIL-47, two other V catalysts are tested for the
oxidation of cyclohexene, namely the supported VOx/SiO2 and
the microporous crystalline VAPO-5. The latter is a crystalline
AlPO zeotype, containing vanadium(+IV) centres. We observed
that VAPO-5 is completely catalytically inactive for the oxidation
of cyclohexene. This observation is consistent with other
research, outlining that VAPO-5 (and more generally MeAPOs)
are not suitable as catalysts for liquid phase oxidation
reactions.6
The supported VOx/SiO2 catalyst shows a good conversion
in a first run (see Fig. S.4w). However, after the first catalytic
run, 40% of the vanadium was leached from the silica
supported catalyst and no cyclohexene conversion was observed
in its second run, indicating the reaction occurs basically
homogeneously.
To get more insight into the possible reaction mechanisms
and intermediates, theoretical calculations were performed on
the VO(acac)2 catalyst, which showed a remarkable resemblance
to the MIL-47 catalyst from the experimental results. A
comparison of the optimized geometries of the VO(acac)2
system with the XRD data of the MIL-47 indicates that
the surroundings of the vanadium atom are very similar
geometrically wise (Fig. 4). The initial coordination of TBHP
with vanadium is expected to occur in a similar fashion apart
from steric constraints. More information on the computa-
tional details can be found in the supporting information.w
The formation of the epoxide, the diol and the peroxide will
Fig. 2 Hot filtration experiment. Cyclohexene conversion (’) and
yield of oxidized products (K) as a function of time. Cyclohexene
conversion (&) and yield of oxidized products (J) after filtration of
the catalyst. The catalytic activity of VO(acac)2 corresponding to the
12.8% leached V-sites (.).
ꢀc
This journal is The Royal Society of Chemistry 2010
5086 | Chem. Commun., 2010, 46, 5085–5087

View Online
Fig. 4 Comparison of distances (in A) for (a) the homogeneous
VO(acac)2 system and (b) the metal organic framework MIL-47.
now be discussed. Because the ketone is not present in large
quantities, this reaction pathway will not be considered here.
Fig. 5 shows the epoxidation pathway with VO(acac)2 with
the final regeneration step of the catalyst included. The reaction
cycle starts with an activation step (A) in which TBHP
coordinates with VO(acac)2. Here, an alkylperoxo species,
namely VO(acac)(OOtBu), is formed. All attempts to model
an epoxidation cycle on a complex with two acac ligands
failed, which suggests that first vacant coordination sites need
to be generated, in accordance with earlier literature data on
the activity of peroxo and alkylperoxo species for oxidations.7
In a second step the actual epoxidation (E) towards cyclohexene
oxide can occur easily with an activation barrier of 55.9 kJ molÀ1
calculated at the B3LYP/6-311+g(3df,2p) level of theory. This
step is irreversible which can be deduced from the reverse
barrier which is very high (236.8 kJ molÀ1). After epoxidation,
Hacac can coordinate back to the vanadium centre and
cyclohexene oxide is split off (P). Notice that a similar ligand
exchange with cyclohexene oxide could also occur with other
species present in the reaction medium (tBuOH, H2O, TBHP).
The final regeneration step occurs fast via a proton transfer
reaction. Both activation (A) and production (P) are endothermic
from our calculations but the equilibrium could significantly
shift on inclusion of a larger molecular environment.
Fig. 5 VO(acac)2 epoxidation cycle. Gray numbers indicate reaction
energies (kJ molÀ1), while black numbers indicate reaction barriers
(bottom, kJ molÀ1) and reaction rates (top, sÀ1).
showing a similar catalytic activity and product distribution
to the homogeneous vanadium complex VO(acac)2. Hot
filtration experiments show that the oxidation catalysis occurs
predominantly heterogeneously. MIL-47 can be regenerated and
reused. Theoretical calculations on the VO(acac)2 system give
insights into the reaction pathways towards the reaction products
and reveal that the successful epoxidation is accompanied by
a linker exchange with TBHP. A similar mobility of the
terephthalic linkers is required to explain the high catalytic
activity of the saturated vanadium centre in MIL-47.
This research is co-funded by Ghent University, GOA grant
no. 01G00710 and Methusalem grant no. 01M00409,
BELSPO in the frame of IAP 6/27 and the European Research
Council (FP7(2007–2013) ERC grant no. 240483). I.M. acknowl-
edges the IWT grant no. IWT/SB/71325. We are grateful to
Steven Pijl for his experimental help with the GC-GC-TOF-MS
measurement and to Jan Musschoot for the XRF analysis.
In the case of the VO(acac)2 catalysts substantial amounts
of acetic acids are formed upon decomposition of Hacac with
TBHP.8 The subsequent formation of the cyclohexane-1,2-diol
can be explained by opening of the epoxide ring by an acid
catalyzed hydrolysis mechanism.9 In the MIL-47, a small
fraction of terephthalic acid is anticipated to catalyze the
opening of the epoxide ring.
Notes and references
1 N. L. Rosi, J. Eckert, M. Eddaoudi, D. T. Vodak, J. Kim,
M. O’Keeffe and O. M. Yaghi, Science, 2003, 300, 1127.
2 G. Ferey, Chem. Soc. Rev., 2008, 37, 191.
´
3 (a) J. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen
and J. T. Hupp, Chem. Soc. Rev., 2009, 38, 1450; (b) J. Perles,
N. Snejko, M. Iglesias and M. A. Monge, J. Mater. Chem., 2009,
19, 6504; (c) F. Gandara, E. G. Puebla, M. Iglesias,
D. M. Proserpio, N. Snejko and M. A. Monge, Chem. Mater.,
2009, 21, 655; (d) M. C. Bernini, F. Gandara, M. Iglesias,
N. Snejko, E. Gutierrez-Puebla, E. V. Brusau, G. E. Narda and
M. A. Monge, Chem.–Eur. J., 2009, 15, 4896.
4 D. Farrusseng, S. Aguado and C. Pinel, Angew. Chem., Int. Ed.,
2009, 48, 7502.
5 K. Barthelet, J. Marrot, D. Riou and G. Ferey, Angew. Chem., Int.
´
Ed., 2002, 41, 281.
6 M. J. Haanepen, A. M. ElemansMehring and J. H. C. Van Hooff,
Appl. Catal., A, 1997, 152, 203.
7 E. P. Talsi, V. D. Chinakov, V. P. Babenko and K. I. Zamaraev,
J. Mol. Catal., 1993, 81, 235; H. Mimoun, M. Mignard, P. Brechot
and L. Saussine, J. Am. Chem. Soc., 1986, 108, 3711.
8 L. Stepovik and M. Gulenova, Russ. J. Gen. Chem., 2009, 79, 1663.
9 A. Lundin, I. Panas and E. Ahlberg, J. Phys. Chem. A, 2007, 111,
9087.
During the first catalytic cycle, a majority of the reaction
product is tert-butyl-2-cyclohexenyl-1-peroxide. This product
is formed due to a radical reaction pathway between the
substrate and the oxidant TBHP and has been recently
observed for unsaturated Co-MOFs.10 Freely diffusing peroxy
radicals can be generated by a variety of mechanisms in which
the oxidation state of vanadium changes.
Similar reaction pathways in the MIL-47 system can be
considered, provided one terephthalic acid ligand folds away
from the vanadium centre to create a vacancy for the
coordination of a TBHP molecule. Subsequently, the actual
epoxidation can occur. Further theoretical modelling is in
progress to fully investigate the proposed reactions in the
MIL-47.
10 M. Tonigold, Y. Lu, B. Bredenkotter, B. Rieger, S. Bahnmuller,
¨
¨
In conclusion, the saturated metal organic framework
MIL-47 is catalytically active in the oxidation of cyclohexene,
J. Hitzbleck, G. Langstein and D. Volkmer, Angew. Chem., Int.
Ed., 2009, 48, 7546.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 5085–5087 | 5087