Solvent-free allylic oxidation of alkenes with O2 mediated by Fe- and Cr-MIL-101

Article abstract of DOI:10.1016/j.jcat.2012.11.003

Catalytic properties of Fe-MIL-101 and Cr-MIL-101 metal-organic frameworks in the solvent-free oxidation of cyclohexene and α-pinene with molecular oxygen have been explored. Both catalysts allow alkene oxidation under mild conditions (1 bar O₂, 40-60°C) and afford allylic oxidation products. The nature of catalysis and the product distribution strongly depend on the nature of the transition metal. Cr-MIL-101 behaves as truly heterogeneous catalyst to give predominantly α,β-unsaturated ketones. Catalysis over Fe-MIL-101 has true heterogeneous nature only at 40°C, producing mainly 2-cyclohexene-1-ol. At 50-60°C, iron leaching into solution occurs, leading to cyclohexenyl hydroperoxide as the major product. Under optimal conditions, both catalysts can be reused several times without suffering a loss of the catalytic properties. Rate-retarding and rate-accelerating effects of inhibitors and initiators, respectively, indicate radical chain mechanism. Different pathways for transformation of hydroperoxide have been suggested to rationalize the observed differences in the reaction selectivities over Cr- and Fe-MIL-101.

Full text of DOI:10.1016/j.jcat.2012.11.003

Journal of Catalysis 298 (2013) 61–69
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Solvent-free allylic oxidation of alkenes with O2 mediated by Fe- and Cr-MIL-101
Igor Y. Skobelev a,b, Alexander B. Sorokin b, Konstantin A. Kovalenko c, Vladimir P. Fedin c,
Oxana A. Kholdeeva a,
⇑
a Boreskov Institute of Catalysis, pr. Lavrentieva 5, Novosibirsk 630090, Russia
b IRCELYON, CNRS, 2, Avenue A. Einstein, 69626 Villeurbanne Cedex, France
c Nikolaev Institute of Inorganic Chemistry, pr. Lavrentieva 3, Novosibirsk 630090, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 July 2012
Revised 9 October 2012
Accepted 3 November 2012
Available online 23 December 2012
Catalytic properties of Fe-MIL-101 and Cr-MIL-101 metal–organic frameworks in the solvent-free oxida-
tion of cyclohexene and a-pinene with molecular oxygen have been explored. Both catalysts allow alkene
oxidation under mild conditions (1 bar O2, 40–60 °C) and afford allylic oxidation products. The nature of
catalysis and the product distribution strongly depend on the nature of the transition metal. Cr-MIL-101
behaves as truly heterogeneous catalyst to give predominantly
a,b-unsaturated ketones. Catalysis over
Fe-MIL-101 has true heterogeneous nature only at 40 °C, producing mainly 2-cyclohexene-1-ol. At
50–60 °C, iron leaching into solution occurs, leading to cyclohexenyl hydroperoxide as the major product.
Under optimal conditions, both catalysts can be reused several times without suffering a loss of the cat-
alytic properties. Rate-retarding and rate-accelerating effects of inhibitors and initiators, respectively,
indicate radical chain mechanism. Different pathways for transformation of hydroperoxide have been
suggested to rationalize the observed differences in the reaction selectivities over Cr- and Fe-MIL-101.
Ó 2012 Elsevier Inc. All rights reserved.
Keywords:
Allylic oxidation
Alkenes
Molecular oxygen
Heterogeneous catalysis
Metal–organic framework
MIL-101
1. Introduction
Metal–organic frameworks (MOFs) have attracted considerable
attention due to a unique combination of properties, such as extre-
The allylic oxidation of olefins is an important synthetic process
since the oxidation products, ,b-unsaturated ketones and alcohols,
mely high surface areas, crystalline open structures, tunable pore
size, and functionality. These features allow considering them as
prospective materials for gas storage, molecular recognition, sepa-
ration, and catalysis [11–19]. From a catalytic point of view, these
materials are very attractive for applications in liquid-phase pro-
cesses since they potentially combine advantages of both homoge-
neous and heterogeneous catalysts. Normally, MOFs are composed
of spatially isolated metal atoms or clusters linked by polydentate
organic ligands, resulting in a rigid porous network. An important
feature of MOFs is a high content of uniform active metal sites reg-
ularly distributed over the material. These sites are readily accessi-
ble for reagents due to the rather large size of pore entrances. Until
recently, low thermal, chemical, and solvolytic stabilities (relative
to completely inorganic zeolites) limited the interest to the applica-
tion of MOFs in catalysis. However, in 2005, Férey and co-workers
reported the discovery of the mesoporous chromium terephthalate
Cr-MIL-101, which showed a good resistance to air, water, common
solvents, and thermal treatment up to 300 °C [20]. In 2009, Taylor-
Pashow and co-workers published the synthesis of the iron-con-
taining analog, Fe-MIL-101 [21]. The MIL-101 materials have a zeo-
type crystal structure comprising two kinds of cages with free
internal diameters of 29 and 34 Å, accessible through microporous
windows of ca. 12 and 16 Å. They have extremely large surface areas
and numerous transition metal sites, which can be activated by
heating in vacuum [22].
a
are valuable intermediates for the fine chemical industry [1–3].
Usually, homogeneous catalytic systems or stoichiometric pro-
cesses based on such oxidants as manganese dioxide, selenium
dioxide, or chromium trioxide are employed to accomplish allylic
oxidations [2–5]. However, the use of the hazardous oxidants and
difficulties related to separation and recycling of homogeneous cat-
alysts make such synthetic procedures environmentally unfriendly
and cost consumable. The development of catalytic systems based
on clean oxidants and easily recyclable heterogeneous catalysts is,
therefore, a challenging goal [2].
Chromium catalysts are well known for the selective oxidation of
cyclic alkenes to
a,b-unsaturated ketones using tert-butyl hydro-
peroxide (TBHP) as oxidant [2–4]. Cr(III)-containing pillared clays
[6], aluminophosphates [7], and mesoporous silicates CrMCM-41
and CrMCM-48 [8,9] were reported to catalyze this reaction. How-
ever, hot catalyst filtration tests [10] revealed that in most cases,
the observed catalysis was due to chromium species leached from
the solid catalyst into solution during the oxidation process, and
hence, had homogeneous nature.
⇑
Corresponding author. Fax: +7 383 330 95 73.
E-mail address: khold@catalysis.nsk.su (O.A. Kholdeeva).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.11.003

62
I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
Activated Cr-MIL-101 was found to catalyze cyanosilylation of
2.3. Catalytic runs and product analysis
benzaldehyde [23], sulfoxidation of thioethers with H2O2 [22], ben-
zylic oxidation of tetralin [24], allylic oxidation of alkenes with
TBHP [25], and oxidation of cyclohexane with O2 and/or TBHP
[26]. The allylic oxidation of alkenes was also possible using aque-
ous hydrogen peroxide as oxidant after modification of MIL-101
with active polyoxometalate complexes [27]. However, the MIL-
101-based catalytic systems reported so far for the allylic oxidation
required the use of toxic and/or flammable organic solvents such as
benzene, chlorobenzene [25], or acetonitrile [27] and a hydroper-
oxide oxidant (H2O2 or TBHP). At the same time, solvent-free oxi-
dations that employ molecular oxygen are significantly more
attractive from both economic and environmental viewpoints.
One of the main technological advantages of solvent-free processes
is a possibility to obtain high concentrations of products in the fi-
nal mixture (high volume yield) even at rather low substrate
conversions.
Catalytic experiments were carried out in thermostated glass
vessels connected with an oxygen balloon (1 bar O2). The oxidation
reactions were performed at 40–60 °C under vigorous stirring
(500 rpm). Typically, the reactions were initiated by the addition
of a small amount of TBHP (2–3 ꢂ 10ꢁ2 mmol) to a mixture con-
taining an organic substrate (1 mL of cyclohexene or 1.5 mL of
a-
pinene) and 13–26 mg of the catalyst preliminary activated in vac-
uum (see above). The reaction time was 16 h. Concentration of
cyclohexenyl hydroperoxide (CyH-OOH) was determined by iodo-
metric titration in 80% acetic acid in the presence of H3PO4 (the lat-
ter was added as a 10% v/v aqueous solution). Other reaction
products were identified by gas chromatography–mass spectrom-
etry (GC–MS) and quantified by GC techniques. Aliquots of the
reaction mixture (25 lL) were withdrawn periodically during the
reaction course by a syringe through a septum. Each aliquot was
diluted with 0.4 mL of acetonitrile containing 60 mg of PPh3 and
internal standard (biphenyl). The treatment of the reaction mixture
with PPh3 is required to reduce CyH-OOH, otherwise it decom-
poses in the GC injector, affecting the ratio of 2-cyclohexene-1-ol
(CyH-ol) and 2-cyclohexene-1-one (CyH-one) [28]. The true con-
centration of CyH-ol was calculated as a difference between the va-
lue obtained by GC and the concentration of CyH-OOH determined
iodometrically (reduction in CyH-OOH with PPh3 gives an addi-
tional amount of CyH-ol). Before recycling, catalysts were filtered
off, washed two times with 15 mL of ethanol at 60 °C for 2 h, acti-
vated in vacuum, and then reused. Turnover frequency values
(TOF) were determined from initial rates of CyH consumption.
TOF was calculated on the basis of all metal ions of MIL-101 frame-
work (23 wt.%) although the real number of the catalytically active
sites could be smaller.
In the present work, we explore the potential of both Cr- and
Fe-MIL-101 as heterogeneous catalysts for the solvent-free allylic
oxidation of two representative alkenes, cyclohexene (CyH) and
a-pinene, with molecular oxygen under mild reaction conditions
(1 bar O2, 40–60 °C). A particular attention is given to investigation
into the nature of catalysis and evaluation of the catalysts stability
and reusability. An attempt of rationalizing the differences in the
product selectivities achieved over Cr-MIL-101 and Fe-MIL-101
under various reaction conditions is made, considering oxidation
mechanisms that could be involved.
2. Experimental
2.1. Materials
CyH and a-pinene were purchased from Sigma-Aldrich or Alfa-
Aesar and purified prior to use by passing through a column filled
with neutral alumina to remove traces of possible oxidation prod-
ucts, in particular, hydroperoxides. All other reactants were ob-
tained commercially and used without further purification.
Aqueous solution of TBHP (70%) was used as initiator.
2.4. Adsorption measurements
The product adsorption on Fe-MIL-101 and Cr-MIL-101 was
studied following the methodology reported elsewhere [29]. A
sample of 28 mg of Fe-MIL-101 or Cr-MIL-101 activated in vacuum
at 70 or 100 °C, respectively, was stirred with cyclohexane during
2 h. Then, MIL-101 material was filtered off using a syringe-driven
2.2. Preparation and characterization of catalysts
filter. A solution of product (0.05 M of CyH-one, CyH-ol or a-pinene
Cr-MIL-101 was prepared following a procedure similar to that
described by Férey et al. [20]. In a typical synthesis, a mixture of
1.2 g (3 mmol) of Cr(NO3)3ꢀ9H2O, 500 mg of terephthalic acid
(H2bdc, 3 mmol), and 0.6 mL of 5 M HF (3 mmol) in 15 mL H2O
was heated at 220 °C for 8 h in a Teflon-lined stainless steel bomb.
The resulting green solid was passed through a coarse glass filter to
remove unreacted H2bdc and then filtered through a dense paper
filter. The green raw product was washed in hot dimethylformam-
ide (DMF) (100 °C, 8 h, 2 times) and then in boiling ethanol (80 °C,
8 h, 2 times), filtered off, and dried overnight in an oven at 75 °C.
Fe-MIL-101 was prepared as described by Taylor-Pashow et al.
[21] with some modifications. In a typical synthesis, a mixture of
0.675 g (2.45 mmol) of FeCl3ꢀ6H2O, 206 mg of H2bdc (1.24 mmol),
and 15 mL DMF was heated at 110 °C for 20 h in a Teflon-lined
stainless steel bomb. The resulting brown solid was filtered off,
and the raw product was purified by washing in hot ethanol
(60 °C, 3 h, 2 times), filtered off, and dried in an oven (70 °C,
30 min).
oxide in cyclohexane, 1.5 mL) was passed through the filter con-
taining MIL-101 saturated with cyclohexane (it was taken instead
of cyclohexene to avoid a possible product formation), collected,
and passed again several times until the constant concentration
of the product in the filtrate was reached. The product concentra-
tion in the filtrate was monitored by GC. One passage took 30 s. All
operations were carried out at room temperature. The equilibrium
constant was evaluated using Langmuir adsorption model
KC
1 þ KC
h ¼
ð1Þ
and the mass balance equation
C0V ¼ CV þ nh
ð2Þ
where K is the adsorption constant, Mꢁ1; V is the total volume of the
product solution, L; C0 is the initial product concentration, M; C is
the equilibrium product concentration, M; n is the number of sorp-
tion sites (metal centers), mol; and h is the degree of filling of the
metal centers. Combining Eqs. (1) and (2), we can express the
adsorption constant K by:
The resulting MIL-101 materials were characterized with X-ray
diffraction (XRD) and N2 adsorption measurements. XRD patterns
of the MIL-101 materials were in accordance with the literature
[20,21]. The BET (Brunauer–Emmett–Teller) specific surface areas
were in the range of 3200–3400 m2 gꢁ1. Both catalysts were acti-
vated in vacuum at 100 or 150 °C (Cr-MIL-101) or at 70 °C
(Fe-MIL-101) for 6 h prior to catalytic studies.
VðC0 ꢁ CÞ
K ¼
ð3Þ
Cðn ꢁ VðC0 ꢁ CÞÞ
Eq. (3) was used for evaluation of the adsorption constants for CyH-
ol, CyH-one and -pinene oxide.
a

I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
63
2.5. Instrumentation
2,3-epoxy-cyclohexanone, 2,3-epoxy-cyclohexanol, bis-cyclohexe-
nyl, and 1-chloro-2-cyclohexanol (the latter was found only in
the presence of Fe-MIL-101) were identified by GC–MS technique.
No derivatives of carboxylic acids were identified after methylation
of the reaction mixture with trimethylsilyldiazomethane. A blank
experiment showed that CyH oxidation does not occur in the ab-
sence of catalyst under the conditions employed (Table 1, run 1).
On the other hand, only traces of the oxidation products (ca. 1% to-
tally) were observed after 16 h when Fe-MIL-101 catalyst was used
without TBHP initiator (Table 1, run 2). Hence, the blend of the cat-
alyst and initiator is required to accomplish the oxidation reaction.
The product distribution and substrate conversion acquired over
Fe-MIL-101 strongly depended on the reaction temperature. CyH-
OOH predominated among the oxidation products at 50 and
60 °C, and the ratio of CyH-one and CyH-ol versus CyH-OOH aug-
mented with increasing the reaction temperature. The total yield
of the allylic oxidation products (CyH-OOH, CyH-one and CyH-ol)
as well as the CyH-one/CyH-ol molar ratio also increased (Table 1,
run 3 vs 4 and run 5 vs 6). However, the product distributions ap-
peared to be quite similar for both temperatures while compared
at close substrate conversions. Thus, selectivities to CyH-one,
CyH-ol, and CyH-OOH were 22, 11, and 63%, respectively, at 27%
conversion achieved at 50 °C (Table 1, run 3), while at 60 °C, they
were 25, 10, and 60%, respectively, at 22% conversion (Fig. 1a). Inter-
estingly, a similar product distribution was reported by Van Sickle
et al. [30] for CyH autoxidation in the presence of conventional rad-
ical initiators: 80% CyH-OOH, 7% CyH-OH, 5% CyH-one, and 1% CyH
oxide. This similarity strongly suggests that at 60 °C, over Fe-MIL-
101 homogeneous cyclohexene autoxidation process contributes
significantly to the product formation, and indeed, this was con-
firmed by hot filtration experiments (vide infra). The increase in me-
tal to substrate molar ratio, [M]:[CyH], from 1:180 to 1:90 had a
small (if any) effect on the substrate conversion and on the ratio be-
GC analyses were performed using a gas chromatograph Agilent
4890D equipped with
a flame ionization detector and a
30 m ꢂ 0.25 mm VF-5 MS capillary column or gas chromatograph
Chromos GC-1000 equipped with a flame ionization detector and
a 30 m ꢂ 0.25 mm BPX-5 capillary column. GC–MS analyses were
carried out using a gas chromatograph HP 6890 (a 50 m ꢂ 0.25 mm
DB-5 MS capillary column) equipped with a mass-selective detec-
tor Agilent MSD 5973. Fourier transform infrared spectroscopy (FT-
IR) measurements were performed using KBr pellets containing
0.3 wt.% of sample on a Bruker Vector 22 or Bruker Vertex 80 spec-
trometer in the 400–4000 cmꢁ1 range. Nitrogen adsorption at 77 K
was measured using an ASAP-2020 instrument (Micrometrics) or
Sorbtometr M within the partial pressure range of 10ꢁ6–1.0. Ele-
mental analysis was performed on HORIBA Jobin Yvon Activa
ICP-OES (IRCELYON). For XRD measurements, Shimadzu XRD
7000S or DRON-3 M instruments were used.
3. Results and discussion
3.1. Product study
Both Fe-MIL-101 and Cr-MIL-101 mediated allylic oxidation of
CyH and a-pinene with molecular oxygen under mild reaction con-
ditions (1 atm O2, 40–60 °C) without use of any organic solvent
other than substrate. The oxidation reactions were initiated by
addition of a small amount of TBHP (2–3 ꢂ 10ꢁ2 mmol). The main
results are presented in Table 1 along with the results of blank
experiments. With CyH as substrate, both catalysts gave the unsat-
urated alcohol and ketone, CyH-ol and CyH-one, along with hydro-
peroxide CyH-OOH as major products. The formation of
cyclohexene oxide was a minor reaction (less than 4%). Traces of
Table 1
Alkene oxidation with O2 over Fe-MIL-101 and Cr-MIL-101a.
Run
Catalyst
[M]:[CyH]
T, °C
Product selectivity, %
Substrate conversion, %
O
OH
OOH
O
1
2
3
4
5
6
7
8
–
–
60
60
50
60
50
60
40
60
50
50
60
–
–
11
14
6
7
58
9
25
20
38
–
–
–
–
–
–
4
4
4
4
–
9
–
–
–
Fe-MIL-101b
Fe-MIL-101
Fe-MIL-101
Fe-MIL-101
Fe-MIL-101
Fe-MIL-101
1:180
1:180
1:180
1:90
<1
27
44
32
44
8
44
12
10
16
22
36
31
43
15
23
58
60
56
63
46
59
46
27
59
17
20
6
1:90
1:180
1:2 ꢂ 105
1:180
1:180
1:90
c
Fe(acac)3
9
10
11
Cr-MIL-101
Cr-MIL-101d
Cr-MIL-101
O
O
OOH
OH
O
12
13
14
–
–
60
40
60
1
33
31
1
-
39
2
25
14
–
33
8
–
9
8
4
12
26
Fe-MIL-101
Cr-MIL-101
1:180
1:90
a
b
c
Reaction conditions: 1 mL of alkene, 1 bar O2, 13 or 26 mg of catalyst activated at 70 °C (Fe) or 100 °C (Cr), TBHP 0.02 mmol, 16 h.
No TBHP was added.
5 ppm Fe.
d
Catalyst activation temperature 150 °C.

64
I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
CyH-one/-ol selectivity at the same conversion. The dependences
(a)
of the product distribution on the CyH conversion at 60 °C differ
significantly for Fe-MIL-101 (Fig. 1a) and Cr-MIL-101 (Fig. 1c),
suggesting different reaction mechanisms. While for Fe-MIL-101,
the selectivity to CyH-OOH strongly increased at the beginning of
the reaction, the opposite trend was observed for Cr-MIL-101.
The increase in the catalyst activation temperature led to minor
changes in the product distribution and substrate conversion (Ta-
ble 1, run 9 vs 10).
60
40
20
0
CyH-OOH
CyH-one
Oxidation of a-pinene (Table 1, runs 13 and 14) was studied un-
CyH-ol
20
der the conditions which allow true heterogeneous catalytic pro-
cess (vide infra). The total alkene conversion achieved over both
Cr- and Fe-MIL-101 was ca. 1.5 times higher than CyH conversion
attained under the same conditions (Table 1, compare runs 7 and
13 or 11 and 14, respectively). Cr-MIL-101 afforded mostly verbe-
nol (31%) and verbenone (39%) (Table 1, run 14) while Fe-MIL-101
favored the formation of verbenol (33%) and campholenic aldehyde
0
5
10
15
CyH conversion, %
(b) 70
60
CyH-ol
(33%) (Table 1, run 13). With both catalysts,
the minor reaction product (8–9% selectivity). Only traces of verbe-
none were found for -pinene oxidation over Fe-MIL-101. Thus,
with both cyclohexene and -pinene, only Cr-MIL-101 produces
a significant amount of the corresponding unsaturated ketones.
Hermans et al. have demonstrated that -pinene easily forms
a-pinene oxide was
50
40
30
20
10
a
a
CyH-OOH
a
epoxide in autoxidation processes upon the addition of peroxyl rad-
icals to the C@C bond followed by a unimolecular rearrangement
[31,32]. The low epoxide yields observed in the presence of
Fe-MIL-101 and Cr-MIL-101 can be explained by several reasons.
First, some amounts of the epoxide could be adsorbed by the
CyH-one
3
2
4
5
6
7
8
CyH conversion, %
catalyst. Indeed, the adsorption constant for
a-pinene oxide on
(c)
Fe-MIL-101 was evaluated to be ca. 30 Mꢁ1 (see Experimental for
details). Consequently, under the concentration conditions of run
CyH-one
CyH-ol
60
13 (Table 1), the maximum amount of
on Fe-MIL-101 could be estimated as 38
of the epoxide in the solution is 150 mol. Note that the real amount
of the adsorbed epoxide could be lower due to the concurrent
adsorption of -pinene and other oxidation products and higher
reaction temperature. Second, -pinene oxide could be transformed
a-pinene oxide adsorbed
l
mol, while the amount
40
20
0
l
a
a
during the oxidation process into campholenic aldehyde and/or
other products. To verify this assumption, we performed two addi-
tional experiments. In the first one, we added
CyH-OOH
15
CyH conversion, %
a-pinene oxide
5
10
20
(0.03 M) to the reaction mixture containing -pinene and catalyst
a
before starting the reaction. No significant increase in the rate of
campholenic aldehyde formation was observed during, at least,
Fig. 1. Product selectivity versus substrate conversion for Fe-MIL-101 at 60 °C (a)
and 40 °C (b) as well as Cr-MIL-101 at 60 °C (c).
5 h. In the second experiment, where 0.5 M of
added to an inert solvent, cyclohexane, containing the catalyst, we
observed a pronounced decrease in -pinene oxide concentration
a-pinene oxide was
tween the sum of CyH-one and CyH-ol and CyH-OOH. However, the
relative amount of CyH-one with respect to CyH-ol was increased at
higher catalyst loadings (Table 1, run 3 vs 5 and run 4 vs 6).
Interestingly, decreasing the reaction temperature from 50 to
40 °C resulted in a dramatic change in the product distribution (Ta-
ble 1, run 7). Indeed, CyH conversion achieved only 8% after 16 h
and CyH-ol became the main product formed with 58% selectivity,
while selectivity to the other products was rather low (15% CyH-
one and 27% CyH-OOH). Importantly, at 60 °C, Fe-MIL-101 provided
62% of CyH-OOH and only 20% of CyH-ol at the same substrate con-
version (Fig. 1a). Therefore, the observed difference in the product
distribution (Fig. 1a and b) cannot be associated with the effect of
conversion. Most probably, different profiles of product distribution
at 40 and 60 °C might be caused by changes in the rate-limiting step
and/or entire reaction mechanism.
a
(ca. 40% conversion after 2 h at 40 °C), and GC–MS identified cam-
pholenic aldehyde as the main reaction product. Therefore, trans-
formation of
a-pinene oxide to the aldehyde cannot be ruled out.
Finally, the catalyst could affect the peroxyl over alkoxyl radical
concentration ratio. Indeed, the increase in the alkoxyl radical con-
centration would favor allylic oxidation versus epoxidation which
occurs via addition of peroxyl radical to the alkene double bond.
However, further studies are needed to verify this hypothesis.
3.2. Catalyst stability and recycling
Hot catalyst filtration tests [10] were performed to verify the
nature of catalysis in the catalytic systems studied. The test carried
out at 60 °C with Fe-MIL-101 revealed a significant contribution of
homogeneous catalysis (Fig. 2a). Elemental analysis of the filtrate
showed minor iron leaching (7 ppm). It is well known that transi-
tion metals are able to catalyze oxidation reactions, even when
they are present in trace amounts [7,10]. Indeed, CyH oxidation
performed in the presence of 5 ppm of Fe(acac)3 (Table 1, run 8) re-
sulted in 44% CyH conversion. The product selectivities were close
Cr-MIL-101 showed a lower CyH conversion under the same
reaction conditions than Fe-MIL-101 (Table 1, run 3 vs 9 and run
6 vs 11). The selectivity toward CyH-one/-ol was much higher for
Cr-MIL-101 at the same reaction temperature and conversion
(compare Fig. 1a and c). Thus, the CyH-one/-ol selectivity attained
94% at 16% CyH conversion, while Fe-MIL-101 provided only 32%

I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
65
Table 2
40
35
30
25
20
15
10
5
Recycling of Fe-MIL-101 and Cr-MIL-101 in CyH autoxidation.
(a)
Catalyst
Substrate conversion,%
Main product selectivity,%
Fe-MIL-101a
Cr-MIL-101d
9 (8, 9, 8)b
57 (55, 58, 56)b,c
56 (58, 56, 57)b,g
16 (15, 17, 16)b
a
b
c
1 mL CyH, [Fe]:[CyH] = 1:180, 40 °C, 16 h.
In parentheses, 2nd, 3rd and 4th reuses.
CyH-ol selectivity.
1 mL CyH, [Cr]:[CyH] = 1:90, 60 °C, 16 h.
CyH-one selectivity.
d
g
filtration of the catalyst
0
0
2
4
6
8
10
toward degradation during the catalytic process correlates with
its higher thermal stability (300 °C [20] versus 150 °C for Fe-
MIL-101 [33]). According to the N2 adsorption data, the surface
area decreased for both catalysts after their reuse. Thus, the surface
area of Cr-MIL-101 diminished from 3400 to 1582 m2 gꢁ1, most
probably, due to pore blockage by reaction products. In turn, the
surface area of Fe-MIL-101 dropped from 3200 to 188 m2ꢀgꢁ1 in
accordance with the partial destruction of the MOF framework
established by XRD (Fig. 4b).
Time, h
18
(b)
16
14
12
10
8
6
filtration of the catalyst
4
3.3. Comparison with other catalytic systems
2
A comparison of the catalytic properties of Fe- and Cr-MIL-101
with some representative catalysts previously reported for CyH oxi-
dation with molecular oxygen is given in Table 3. For all Cr-contain-
ing catalysts, including Cr-MIL-101, the main oxidation product
was CyH-one, which indicates that the nature of active metal deter-
mines the oxidation mechanism. Chromium-containing molecular
sieves allow achieving higher conversions of CyH than Cr-MIL-
101 or Fe-MIL-101. Mesoporous Cr-MCM-41 showed the highest
substrate conversion and enone selectivity but suffered from signif-
icant chromium leaching [34]. For Cr-APO5, the issue of metal
leaching was not addressed, and the nature of catalysis remained
unclear [34]. It is worth of noting that Cu-MOF [35] and Co-MOF
[36] produced CyH-OOH as the main oxidation product. Co-MOF
provided a high substrate conversion but in the presence of the or-
ganic solvent (0.6 M CyH in MeCN). The Fe/SiO2 catalyst demon-
strated a high CyH conversion at 78 °C in acetonitrile solution
(1 M CyH) [37]. Unfortunately, the problem of the iron leaching
was not addressed for this system, and no analysis for hydroperox-
ide formation was performed. Fe(BTC) metal–organic framework
showed CyH conversion close to Fe-MIL-101 but at higher temper-
ature and oxygen pressure [38]. Over both Fe(BTC) and Fe/SiO2 cat-
alysts, CyH-one predominated among the oxidation products. In
contrast, Fe-MIL-101 afforded CyH-ol as the major product. This dif-
ference might be rationalized if we take into account more severe
conditions applied for Fe(BTC) (100 °C, 5 bar O2) or Fe/SiO2 (78 °C,
1 bar O2) compared to Fe-MIL-101 (40 °C and 1 bar O2). Yet, differ-
ent reaction mechanisms cannot be ruled out.
0
0
1
2
3
4
5
6
Time, h
8
7
6
5
4
3
2
1
0
(c)
filtration of the catalyst
0
2
4
6
8
10
12
Time, h
Fig. 2. Hot catalyst filtration test for Fe-MIL-101 (a) and Cr-MIL-101 (b) at 60 °C as
well as Fe-MIL-101 at 40 °C (c).
to those observed for the Fe-MIL-101-catalyzed reaction at 60 °C
(Table 1, run 4). Thus, we may conclude that catalysis observed
with Fe-MIL-101 at 60 °C is due to active iron species leached into
solution during the oxidation process. On the contrary, Cr-MIL-101
demonstrated much better stability than Fe-MIL-101 and behaved
as truly heterogeneous catalyst at 60 °C, as proved by the hot filtra-
tion test (Fig. 2b). It should be noted that at a lower temperature
(e.g., 40 °C), Fe-MIL-101 can also act as truly heterogeneous cata-
lyst (Fig. 2c). According to the elemental analysis data, iron leach-
ing in this case was below 1 ppm.
Finally, the recycling behavior of Fe-MIL-101 and Cr-MIL-101
was studied in CyH oxidation under the conditions where these
catalysts certainly work as true heterogeneous ones. Both MIL-
101 materials retained their catalytic properties during at least
four consecutive runs (Table 2).
3.4. Mechanistic study
To assess the role of external diffusion in the oxidation process,
we investigated the influence of the rate of stirring of the reaction
mixture on the oxidation process. The increase in the rate from 500
to 1000 rpm did not affect the rate of CyH oxidation over Cr-MIL-
101 at 60 °C (TOF 7 hꢁ1). In contrast, the reaction rate over Fe-
MIL-101 at 40 °C increased significantly (TOF 7 versus 2 hꢁ1), indi-
cating more efficient oxygen diffusion into the reaction mixture.
Importantly, the product distribution remained unchanged. No fur-
ther increase in TOF was observed upon increasing the stirring rate
from 1000 to 1500 rpm.
FT-IR spectroscopic studies revealed the presence of the charac-
teristic bands of the MIL-101 metal–organic framework after four
catalyst reuses (Fig. 3). The XRD measurements confirmed that
Cr-MIL-101 preserved its structure after several reuses (Fig. 4a).
However, a partial destruction of Fe-MIL-101 during the oxidation
reaction occurred (Fig. 4b). The higher stability of Cr-MIL-101
The question of internal diffusion limitation is very important in
the investigation into liquid-phase heterogeneous reactions over

66
I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
(b)
(a)
A
B
A
B
2000
1600
1200
800
400
2000
1600
1200
800
400
Wavenumber, cm-1
Wavenumber, cm-1
Fig. 3. FT-IR spectra of initial MIL-101 sample (curves A) and after 4 catalyst reuses (curves B) for Fe-MIL-101 (a) and Cr-MIL-101 (b).
(b)
(a)
A
B
A
B
5
10
15
20
25
30
35
5
10
15
20
25
30
2
θ, degrees
2θ, degrees
Fig. 4. XRD patterns of initials sample (curves A) and after 4 catalytic cycles (curves B) for Cr-MIL-101 (a) and Fe-MIL-101 (b).
Table 3
Catalytic performances of MIL-101 in solvent-free CyH oxidation with O2 (1 bar) in comparison with other catalysts.
Catalyst
T, °C (time, h)
Product selectivity, %
Substrate conversion, %
Metal leaching
Ref.
O
OOH
O
OH
Cr-MIL-101
Fe-MIL-101
Cr-MCM-41
Cr-APO5
60 (16)
40 (16)
70 (24)
70 (24)
45 (15)
35 (24)
78 (10)
100 (2)
38
58
11.2
28
4
8
17
38
56
6
–
–
3.4
5
2
–
5
3
16
8
52
25
8
35
98
12
<1 ppm
<1 ppm
56 ppm
n.d.a
No leaching
<1.1 ppm
n.d.a
This work
This work
[34]
[34]
[35]
[36]
[37]
[38]
15
71.2
51
7
16
76
59
27
14.2
16
87
76
b
Cu-MOF
c,d
Co-MOF
FeIII/SiO2
–
d,e
h
f,g
Fe-MOF
–
No leaching
a
b
c
Not determined.
[Co4O(bdpb)3], bdpb = 1,4-bis[(3,5-dimethyl)-pyrazol-4-yl]benzene.
Cu(bpy)2(H2O)2(BF4)2, bpy = 4,4-bipyridine.
Acetonitrile was used as solvent.
Oxygen flow 20 mL/min.
Basolite F300 [Fe(BTC)] MOF, BTC = benzene-1,3,5-tricarboxylate.
5 bar O2.
d
e
f
g
h
No peroxide analysis was performed.
solid porous catalysts. Recently, it was found that oxidation of
diphenylmethane (DPM) with TBHP catalyzed by Fe-MIL-100 re-
vealed similar reaction rates regardless of the average particle size
of the catalyst, while the activity of the Fe-MIL-100 samples in the
oxidation of more bulky triphenylmethane (TPM) largely depended
on this parameter, indicating that diffusion limitation within the
MOF pores takes place for the bulky TPM substrate [39]. Molecules
of CyH and its oxidation products are smaller than DPM molecule,
while the cage entrances of MIL-101 (12 and 16 Å) are larger than
windows of MIL-100 (8.6 Å). Since no internal diffusion limitation
was observed for the DPM oxidation over MIL-100, one can suggest
that the same should be true in the case of CyH oxidation over MIL-
101.
To verify this assumption, the adsorption kinetics of CyH-one
and CyH-ol onto Cr- and Fe-MIL-101 was studied (see Experimen-
tal for details). The adsorption curves show that the adsorption
equilibrium was reached after 1.5 and 1 min for Cr-MIL-101 and
Fe-MIL-101, respectively (Fig. 5). Given in mind that the reaction

I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
67
1,04
1,00
0,96
0,92
0,88
0,84
0,80
1,04
CyH-one
CyH-ol
CyH-one
(a)
(b)
CyH-ol
1,00
K=1 M-1
K=1 M-1
0,96
0,92
K=3 M-1
K=3 M-1
0,88
0,84
0
1
2
3
4
5
0
1
2
3
4
5
Cycle
Cycle
Fig. 5. Sorption of CyH-one and CyH-ol on Fe-MIL-101 (a) and Cr-MIL-101 (b) at room temperature. One cycle corresponds to 30 s.
time was typically 16 h, and the internal diffusion is hardly ex-
pected to be the rate-limiting step of the overall oxidation process.
The values of the sorption equilibrium constants K estimated from
eq. 3 are 1 and 3 Mꢁ1 for CyH-ol and CyH-one, respectively.
Although these values were determined for room temperature
and cannot be directly used for computation of the product sorp-
tion at the reaction temperatures, we may assume that adsorption
of the oxidation products on the active metal sites could, in princi-
pal, hinder the reaction. To check this possibility, we added a solu-
tion of CyH-one and CyH-ol (ca. 0.5 M each, the concentration close
to the value that was normally observed at the end of the reaction)
into the reaction mixture at half of the maximal CyH conversion.
The reaction stopped completely (Fig. 6). When only CyH-one
(1.0 M) was added, the effect was similar. Hence, inhibition by
the reaction products can be responsible for the termination of
the CyH oxidation process after achieving a certain level of sub-
strate conversion and product concentration. The situation
changes dramatically when the active metal leaches into solution.
Indeed, in the case of Fe-MIL-101 and the reaction temperature
above 40 °C, much higher CyH conversions could be achieved,
and the product distribution changed drastically (Table 1).
total amount of the metal sites (50–100 lmol), catalyst poisoning
by the inhibitor can hardly be expected. Hence, the reaction stopped,
most likely, because of the interaction of the inhibitor with radicals
responsible for the chain propagation. The observed effects of the
initiator and inhibitor on the reaction rate point out radical chain
nature of the CyH oxidation process over both MIL-101 catalysts.
Since the control reaction without catalyst (but with TBHP initiator)
exhibited a negligible conversion (Table 1, run 1), it is reasonable to
suppose that thermal homolytic dissociation of hydroperoxide can-
not be the principal initiation step and that MIL-101 participate, at
least, in the chain initiation.
It is well known that the primary oxidation product in radical
chain oxidation of hydrocarbons is a hydroperoxide, ROOH [40–
42]. Further transformations of ROOH may include its homolytic
decomposition over a catalyst (this process is responsible for the
chain initiation/branching) and/or heterolytic decomposition lead-
ing to molecular products. The homolytic decomposition is
described by the widely accepted Haber–Weiss mechanism that
normally operates in the presence of transition metal ions [43,44],
while the heterolytic decomposition may involve dehydration
and/or disproportionation reactions.
As was mentioned before, the presence of both catalyst and initi-
ator (TBHP) is necessary to accomplish alkene oxidation with molec-
ular oxygen. In turn, the addition of a conventional inhibitor of
radical chainprocesses (ionol)at half of themaximalCyH conversion
immediately stopped the reaction over both Fe- and Cr-MIL-101
(Fig. 7). This could be, in principle, explained either by suppression
of the growth of radical chains in autoxidation process or by adsorp-
tion of the inhibitor on the active metal sites, which may lead to the
catalyst poisoning. Since the total amount of ionol introduced into
In principle, unsaturated ketones can favor chain branching [45]
through interaction with hydroperoxide ROOH:
ROOH þ R@O ! ROÅ þ RꢁÅ @O þ H2O
ð4Þ
aH
If this occurred, one might expect that addition of CyH-one to
the reaction mixture would increase the oxidation rate. However,
an opposite effect (Fig. 6) was observed, and hence, we may ex-
clude a significant contribution of reaction (4) to the overall oxida-
tion process.
the reaction mixture (5 lmol) was at least 10 times less than the
Since the product distribution in CyH and
a-pinene oxidation
differs markedly for Cr-MIL-10 and Fe-MIL-101, one can assume
that further transformation of initially formed hydroperoxides pro-
ceeds via different pathways over these catalysts. Based on the lit-
erature [2,40–42,44] and the results obtained in this work, we can
envisage a tentative mechanism for CyH oxidation over the MIL-
101 catalysts as a sequence of steps shown in Scheme 1.
Indeed, Cr-containing catalysts are known to produce a high
yield of cyclohexanone in the oxidation of cyclohexane due to their
ability to favor dehydration of cyclohexyl hydroperoxide [46,47]
and oxidation of alcohol to ketone [48,49]. Sheldon and co-workers
studied CyH-OOH decomposition over chromium-containing por-
ous solids, CrAPO-5 and CrS-1, and soluble complex Cr(acac)3 and
found that all of them produced ca. 70% of CyH-one and 30% of
CyH-ol [50], which is close to the results that we acquired in CyH
oxidation using Cr-MIL-101. Consequently, the product distribu-
tion obtained over Cr-MIL-101 is likely due to CyH-OOH heterolytic
decomposition and/or dehydration on the chromium active sites.
Additionally, we found that Cr-MIL-101 is able to catalyze the oxi-
dation of CyH-ol to CyH-one with molecular oxygen, providing 18%
alcohol conversion after 4 h at 60 °C in cyclohexane solution.
Reaction plot
Reaction plot with product
addition after 2 hours
18
16
14
12
10
8
Products added
6
4
2
0
0
1
2
3
4
5
Time, h
Fig. 6. Effect of the addition of CyH-one/CyH-ol mixture on the reaction over
Cr-MIL-101.

68
I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
8
7
6
5
4
3
2
1
0
18
(a)
(b)
16
14
12
10
8
6
4
2
0
Inhibitor added
Inhibitor added
0
2
4
6
8
10 12 14 16 18
0
2
4
6
8
10 12 14 16 18
Time, h
Time, h
Fig. 7. Effect of inhibitor (ionol, 5 lmol) on the CyH oxidation rate over Fe-MIL-101 (a) and Cr-MIL-101 (b).
•
OO
OH
O
cat.*
cat.*
O2
OOH
•
O2
or
ROOH
Fe**
Cr**
TBHP
cat.*
Cr
•
O
OH
Scheme 1. Possible CyH oxidation pathways over MIL-101 catalysts: ꢄ radical chain pathway, ꢄꢄ non-radical chain pathway. Chain termination steps are not shown.
It is well known that conclusions about reaction mechanisms
The mechanism of the OAO bond cleavage strongly depends on
the nature of the hydroperoxide and ligand environment L. Hetero-
lytic pathway frequently occurs in biological and bio-inspired sys-
tems when porphyrin-like macrocyclic ligand stabilizes the high
oxidation state via participation of the ligand in the charge delocal-
ization [57]. The LFeV@O or L+ꢀFeIV@O species are known to be strong
epoxidizing agents [58–60]. However, the amount of epoxide was
low in the oxidation of CyH over Fe-MIL-101, especially, at 40 °C.
Therefore, the formation of FeV@O seems unlikely, while the forma-
tion of LFeIV@O and ROÅ via homolytic cleavage of OAO bond is in
agreement with free radical oxidation mechanism deduced from
the experimental observations. As was discussed above, radicals
ROÅ abstract hydrogen atom from the allylic position of alkene, thus
favoring allylic oxidation. In general, LFeIV@O species show sluggish
oxidizing properties. However, the increase in the oxidizing ability
was previously observed for binuclear iron structures due to the
presence of second metal ion in the active site [57]. A similar effect
might be expected for FeIV@O species generated within tri-nuclear
iron clusters which constitute the MIL-101 framework.
based on the product analysis solely should be drawn with precau-
tions [51]. Nevertheless, the sharp difference in the product selec-
tivity observed at 60 °C in the presence of homogeneous iron
species and at 40 °C for the truly heterogeneous catalyst Fe-MIL-
101 certainly indicates different oxidation pathways. In the former
case, the product selectivity is classic for radical chain autoxidation
[30,40–42]. At the same time, the product distribution observed for
Fe-MIL-101 at 40 °C (58% of CyH-ol and just 15% of CyH-one) is
quite unusual for radical-driven autoxidation processes where
the alcohol to ketone (A/K) molar ratio normally does not exceed
2 [49,52–54]. Thus, the A/K molar ratio of ca. 1.4 (at least, at the
beginning of the reaction) was found for cyclohexane autoxidation
in the presence of Co(acac)2 [49]. On the other hand, some exam-
ples of well-disguised radical reactions showing unusual selectiv-
ity, for example, the oxidation of alkanes with TBHP as oxidant
catalyzed by FeIII(TPA) (TPA = tris(2-pyridylmethyl)amine) com-
plexes where alcohol formed as a major product, illustrate a com-
plex nature of this chemistry [55,56]. Although we cannot rule out
completely a contribution of the Haber–Weiss mechanism [43] to
the overall alkene oxidation process over Fe-MIL-101, the unusu-
ally high A/K molar ratio (ca. 3.9) encouraged us to assume the
existence of an alternative pathway.
We suppose that the following reaction sequence may lead to
the predominant formation of the enol product in the case of the
heterogeneous alkene oxidation over Fe-MIL-101 at 40 °C:
LFeIV@O þ R ꢁ H ! RÅ þ LFeIVAOH ! RAOH þ LFeIII
ð8Þ
Once formed, ROOH may react with active metal sites to form
an alkylhydroperoxo complex:
Radical RÅ formed in reaction (8) could either interact with O2 to
form ROÅ2 radical and initiate a new chain or react with the LFeIV-
AOH species to give an alcohol product via rebound reaction.
When the oxidation process takes place within the MIL-101 cages,
where high concentration of the iron centers is available, the inter-
action of RÅ with FeIVAOH may predominate. The situation changes
dramatically when the active metal leaches from the Fe-MIL-101
matrix to the solution. The reaction of radical RÅ with O2 to produce
ROOÅ followed by H atom abstraction from alkene and formation of
ROOH þ LFeIII ! LFeIIIAOOR
ð5Þ
This complex can undergo either homolytic or heterolytic cleav-
age of the OAO bond to form LFeIVAO or formally LFeVAO adapting
L+ꢀFeIV@O state according to reactions (6) and (7), respectively:
LFeIIIAOOR ! LFeIV@O þ ROÅ
ð6Þ
ð7Þ
LFeIIIAOOR ! ½LFeV@O $ LþÅFeIV@Oꢃ þ ROꢁ

I.Y. Skobelev et al. / Journal of Catalysis 298 (2013) 61–69
[3] E.F. Murphy, T. Mallat, A. Baiker, Catal. Today 57 (2000) 115.
69
ROOH becomes the main pathway because the concentration of
iron species in solution is very low (<7 ppm), while the concentra-
tion of dioxygen is about 0.01 M [61]. Indeed, the ROOH product
predominates at high temperature when Fe-MIL-101 is destroyed,
and catalysis becomes homogeneous. However, further mechanis-
tic studies are needed to verify the possibility of the formation of
the iron-oxo species in the Fe-MIL-101 framework.
[4] J. Muzart, Chem. Rev. 92 (1992) 113.
[5] N. Chidambaram, S. Chandrasekaran, J. Org. Chem. 52 (1987) 5048.
[6] B.M. Choudary, A.D. Prasad, V. Swapna, V.L.K. Valli, V. Bhuma, Tetrahedron 48
(1992) 953.
[7] H.E.B. Lempers, R.A. Sheldon, Appl. Catal. A 143 (1996) 137.
[8] A. Sakthivel, S.E. Dapurkar, P. Selvam, Appl. Catal. A 246 (2003) 283.
[9] S.E. Dapurkar, A. Sakthivel, P. Selvam, New J. Chem. 27 (2003) 1184.
[10] R.A. Sheldon, M. Wallau, I.W.C.E. Arends, U. Schuchardt, Acc. Chem. Res. 31
(1998) 485.
[11] J.L.C. Rowsell, O.M. Yaghi, Micropor. Mesopor. Mater. 73 (2004) 3.
[12] A.K. Cheetham, C.N.R. Rao, R.K. Feller, Chem. Commun. (2006) 4780.
[13] G. Férey, Chem. Soc. Rev. 37 (2008) 191.
4. Conclusions
[14] A.U. Czaja, N. Trukhan, U. Muller, Chem. Soc. Rev. 38 (2009) 1284.
[15] D. Farrusseng, S. Aguado, C. Pinel, Angew. Chem. Int. Ed. 48 (2009) 7502.
[16] A. Corma, H. García, F.X. Llabrés i Xamena, Chem. Rev 110 (2010) 4606.
[17] M. Yoon, R. Srirambalaji, K. Kim, Chem. Rev. 112 (2011) 1196.
[18] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Catal. Sci. Technol. 1 (2011) 856.
[19] Thematic issue on metal-organic frameworks, Chem. Rev. 112 (2012) 673.
[20] G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I.
Margiolaki, Science 309 (2005) 2040.
[21] K.M.L. Taylor-Pashow, J. Della Rocca, Z. Xie, S. Tran, W. Lin, J. Am. Chem. Soc.
131 (2009) 14261.
[22] Y.K. Hwang, D.-Y. Hong, J.-S. Chang, H. Seo, M. Yoon, J. Kim, S.H. Jhung, C. Serre,
G. Férey, Appl. Catal. A 358 (2009) 249.
[23] A. Henschel, K. Gedrich, R. Kraehnert, S. Kaskel, Chem. Commun. (2008) 4192.
[24] J. Kim, S. Bhattacharjee, K.-E. Jeong, S.-Y. Jeong, W.-S. Ahn, Chem. Commun.
(2009) 3904.
[25] N.V. Maksimchuk, K.A. Kovalenko, V.P. Fedin, O.A. Kholdeeva, Adv. Synth.
Catal. 352 (2010) 2943.
[26] N.V. Maksimchuk, K.A. Kovalenko, V.P. Fedin, O.A. Kholdeeva, Chem. Commun.
48 (2012) 6812.
[27] N.V. Maksimchuk, M.N. Timofeeva, M.S. Melgunov, A.N. Shmakov, Yu.A.
Chesalov, D.N. Dybtsev, V.P. Fedin, O.A. Kholdeeva, J. Catal. 257 (2008) 315.
[28] G.B. Shul’pin, J. Mol. Catal. A: Chem. 189 (2002) 39.
[29] H.L. Wang, J.L. Duda, C.J. Radke, J. Colloid Interface Sci. 66 (1978) 153.
[30] D.E. Van Sickle, F.R. Mayo, R.M. Arluck, J. Am. Chem. Soc. 87 (1965) 4832.
[31] U. Neuenschwander, F. Guignard, I. Hermans, ChemSusChem 3 (2010) 75.
[32] U. Neuenschwander, I. Hermans, Phys. Chem. Chem. Phys. 12 (2010) 10542.
[33] S. Bauer, C. Serre, T. Devic, P. Horcajada, J. Marrot, G. Férey, N. Stock, Inorg.
Chem. 47 (2008) 7568.
[34] S.E. Dapurkar, H. Kawanami, K. Komura, T. Yokoyama, Y. Ikushima, Appl. Catal.
A 346 (2008) 112.
[35] D. Jiang, T. Mallat, D.M. Meier, A. Urakawa, A. Baiker, J. Catal. 270 (2010) 26.
[36] M. Tonigold, Y. Lu, A. Mavrandonakis, A. Puls, R. Staudt, J. Möllmer, J. Sauer, D.
Volkmer, Chem. Eur. J. 17 (2011) 8671.
[37] J. Mao, X. Hu, H. Li, Y. Sun, C. Wang, Z. Chen, Green Chem. 10 (2008) 827.
[38] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, J. Catal. 289 (2012) 259.
[39] A. Dhakshinamoorthy, M. Alvaro, Y.K. Hwang, Y.-K. Seo, A. Corma, H. Garcia,
Dalton Trans. 40 (2011) 10719.
[40] E.T. Denisov, N.I. Mitskevich, V.E. Agabekov, Liquid-Phase Oxidation of
Oxygen-Containing Compounds, Consultants Press, New York, 1977.
[41] R.A. Sheldon, J.K. Kochi, Metal-Catalyzed Oxidations of Organic Compounds,
Academic Press, New York, 1981.
This study demonstrates that Fe and Cr-containing metal–or-
ganic frameworks of the MIL-101 family catalyze efficiently oxida-
tion of alkenes with molecular oxygen under mild solvent-free
conditions. By varying the nature of transition metal in the MIL-
101 framework, it is possible to govern the oxidation selectivity.
Cr-MIL-101 favors the formation of a,b-unsaturated ketones, while
Fe-MIL-101 can produce higher amounts of allylic alcohols. The
nature of catalysis over Cr-MIL-101 was proved to be true hetero-
geneous under all range of the conditions studied. In contrast, Fe-
MIL-101 revealed a lower stability under turnover conditions and
behaved as true heterogeneous only at 40 °C. The XRD and N2
adsorption measurements clearly showed that the structure of
Fe-MIL-101 partially collapsed after several reuses although FT-IR
spectroscopy revealed the presence of the main characteristic
bands of the MOF. Meanwhile, under optimal conditions both Cr-
and Fe-catalysts could be recycled at least four times without loss
of the catalytic properties. Since the reaction products inhibit the
oxidation process through adsorption on the MOF active centers,
the feasible alkene conversions are not high for one catalytic run.
However, the employment of solvent-free conditions ensures high
volume yield of the oxidation products even at rather low conver-
sions. At the temperature higher than 40 °C, Fe-MIL-101 starts suf-
fering from iron leaching, and catalysis becomes homogeneous,
which results in much higher substrate conversion and change of
the product selectivity (hydroperoxide turns out to be the main
oxidation product).
With both Fe- and Cr-catalysts, the oxidation of alkenes pro-
ceeds through radical chain mechanism, as pointed out by the
rate-retarding effect of the inhibitor (ionol) and rate-accelerating
effect of the initiator (TBHP); however, different pathways (both
radical chain and non-radical chain) may contribute to the product
formation, resulting in different reaction selectivities. The metal-
centered oxidation steps leading to unsaturated alcohol product
are probably involved in the case of Fe-MIL-101, while dehydration
of hydroperoxide and/or oxidation of alcohol to ketone over Cr-
MIL-101 are responsible for the predominate formation of unsatu-
rated ketones.
[42] E.T. Denisov, I.B. Afanas’ev, Oxidation and Antioxidants in Organic Chemistry
and Biology, Taylor & Francis, Boca Raton, 2005.
[43] F. Haber, J. Weiss, Proc. R. Soc. London, A 147 (1934) 332.
[44] H. Weiner, A. Trovarelli, R.G. Finke, J. Mol. Catal. A: Chem. 191 (2003) 217.
[45] U. Neuenschwander, I. Hermans, J. Catal. 287 (2012) 1.
[46] J.D. Chen, J. Dakka, R.A. Sheldon, Appl. Catal. A 108 (1994) L1.
[47] W. Buijs, R. Raja, J.M. Thoma, H. Wolters, Catal. Lett. 91 (2003) 253.
[48] J.D. Chen, J. Dakka, E. Neeleman, R.A. Sheldon, J. Chem. Soc., Chem. Commun
(1993) 1379.
[49] I. Hermans, J. Peeters, P.A. Jacobs, Top. Catal. 48 (2008) 41.
[50] H.E.B. Lempers, J.D. Chen, R.A. Sheldon, Stud. Surf. Sci. Catal. 94 (1995) 705.
[51] K.U. Ingold, P.A. MacFaul, in: B. Meunier (Ed.), Biomimetic Oxidation Catalyzed
by Transition Metal Complexes, Imperial College Press, London, 2000, p. 45.
[52] A. Ramanathan, M.S. Hamdy, R. Parton, T. Maschmeyer, J.C. Jansen, U. Hanefeld,
Appl. Catal. A 355 (2009) 78.
[53] A.P. Singh, N. Torita, S. Shylesh, N. Iwasa, M. Arai, Catal. Lett. 132 (2009) 492.
[54] N. Turrà, A.B. Acuña, B. Schimmöller, B. Mayr-Schmölzer, P. Mania, I. Hermans,
Top. Catal. 54 (2011) 737.
Acknowledgments
Dr. M.S. Melgunov is gratefully acknowledged for some N2
adsorption measurements. The research was partially supported
by the Russian Foundation for Basic Research (Grant 09-03-
93109). I.Y.S. acknowledges French Embassy in Moscow for the
doctoral fellowship. Work at the Nikolaev Institute of Inorganic
Chemistry was supported by the State Contracts (1729.2012.3
and 02.740.11.0628).
[55] P.A. MacFaul, K.U. Ingold, D.D.M. Wayner, L. Que Jr., J. Am. Chem. Soc. 119
(1997) 10594.
[56] J. Kim, R.G. Harrison, C. Kim, L. Que Jr., J. Am. Chem. Soc. 118 (1996) 4373.
[57] A.B. Sorokin, E.V. Kudrik, Catal. Today 159 (2011) 37.
[58] S. Tanase, E. Bouwman, Adv. Inorg. Chem. 58 (2006) 29.
[59] P.D. Oldenburg, L. Que Jr., Catal. Today 117 (2006) 15.
[60] O.Y. Lyakin, K.P. Bryliakov, G.J.P. Britovsek, E.P. Talsi, J. Am. Chem. Soc. 131
(2009) 10798.
References
[1] M. Hudlicky, Oxidations in Organic Chemistry, ACS Monograph Series,
American Chemical Society, Washington, DC, 1990. p. 84.
[2] R.A. Sheldon, in: R.A. Sheldon, H. van Bekkum (Eds.), Fine Chemicals through
Heterogeneous Catalysis, Wiley-VCH, Weinheim, 2001, p. 519.
[61] R. Battino, T.R. Rettich, T. Tominaga, J. Phys. Chem. Ref. Data 12 (1983) 163.