Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks MIL-100 and MIL-101-a comparative study

Article abstract of DOI:10.1016/j.cattod.2014.01.010

Catalytic properties of Fe- and Cr-based metal-organic frameworks (MOFs) MIL-100 and MIL-101 have been assessed in two liquid-phase reactions: solvent-free allylic oxidation of alkenes (cyclohexene, α- and β-pinenes) with molecular oxygen and oxidation of anthracene (AN) with tert-butyl hydroperoxide (TBHP). In the oxidation of alkenes, the product selectivity strongly depends on the nature of metal (Fe or Cr) but, for the same metal, only slightly differs for the MIL-100 and MIL-101 structures. The Fe-containing MOFs afford the formation of unsaturated alcohols while Cr-based MOFs give mainly unsaturated ketones. Both Cr-MIL-100 and Cr-MIL-101 favor decomposition of cyclohexenyl hydroperoxide to produce 2-cyclohexen-1-one with 67-69% selectivity. Stability toward destruction reduced in the order Cr-MIL-101, Cr-MIL-100 > Fe-MIL-100 > Fe-MIL-101. In the oxidation of anthracene over both Cr-MOFs and Fe-MIL-101, the selectivity toward 9,10-anthraquinone (AQ) attained 100% at 92-100% AN conversion. The turnover frequency (TOF) decreased in the order Cr-MIL-101 > Fe-MIL-101 > Cr-MIL-100 > Fe-MIL-100. Cr-MIL-101 revealed superior catalytic performance in terms of AN conversion, AQ selectivity and TOF. Nearly quantitative yield of AQ was obtained after 1.5 h at 100 °C in chlorobenzene as solvent. No leaching of active metal occurred under optimal reaction conditions and the MOFs could be recycled several times without deterioration of the catalytic properties.

Full text of DOI:10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
Catalysis Today xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Hydrocarbon oxidation over Fe- and Cr-containing metal-organic
frameworks MIL-100 and MIL-101–a comparative study
a,b,∗
a
a
Oxana A. Kholdeeva
, Igor Y. Skobelev , Irina D. Ivanchikova ,
b,c
b,c
d
Konstantin A. Kovalenko , Vladimir P. Fedin , Alexander B. Sorokin
Boreskov Institute of Catalysis, Lavrentiev ave. 5, Novosibirsk 630090, Russia
Novosibirsk State University, Pirogova st. 2, Novosibirsk 630090, Russia
Nikolaev Institute of Inorganic Chemistry, Lavrentiev ave. 3, Novosibirsk 630090, Russia
Institut de Recherches sur la Catalyse et l’Environnement de Lyon (IRCELYON), UMR 5256, CNRS-Université Lyon 1, 2, av. A. Einstein, 69626 Villeurbanne
Cedex, France
a
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Catalytic properties of Fe- and Cr-based metal-organic frameworks (MOFs) MIL-100 and MIL-101 have
been assessed in two liquid-phase reactions: solvent-free allylic oxidation of alkenes (cyclohexene, ␣-
and ␤-pinenes) with molecular oxygen and oxidation of anthracene (AN) with tert-butyl hydroperoxide
Received 1 December 2013
Accepted 14 January 2014
Available online xxx
(TBHP). In the oxidation of alkenes, the product selectivity strongly depends on the nature of metal (Fe or
Cr) but, for the same metal, only slightly differs for the MIL-100 and MIL-101 structures. The Fe-containing
MOFs afford the formation of unsaturated alcohols while Cr-based MOFs give mainly unsaturated ketones.
Both Cr-MIL-100 and Cr-MIL-101 favor decomposition of cyclohexenyl hydroperoxide to produce 2-
cyclohexen-1-one with 67–69% selectivity. Stability toward destruction reduced in the order Cr-MIL-
Keywords:
MIL-101
MIL-100
Heterogeneous catalysis
Alkenes
1
01, Cr-MIL-100 > Fe-MIL-100 > Fe-MIL-101. In the oxidation of anthracene over both Cr-MOFs and Fe-
Anthracene
Selective oxidation
MIL-101, the selectivity toward 9,10-anthraquinone (AQ) attained 100% at 92–100% AN conversion. The
turnover frequency (TOF) decreased in the order Cr-MIL-101 > Fe-MIL-101 > Cr-MIL-100 > Fe-MIL-100.
Cr-MIL-101 revealed superior catalytic performance in terms of AN conversion, AQ selectivity and TOF.
◦
Nearly quantitative yield of AQ was obtained after 1.5 h at 100 C in chlorobenzene as solvent. No leaching
of active metal occurred under optimal reaction conditions and the MOFs could be recycled several times
without deterioration of the catalytic properties.
©
2014 Elsevier B.V. All rights reserved.
1
. Introduction
Until recently, low thermal, chemical and solvolytic stabilities
in comparison with inorganic zeolites) limited the application of
(
Metal-organic frameworks (MOFs) have attracted consider-
MOFs in catalysis. In 2005, Férey and co-workers reported the
synthesis of a mesoporous chromium terephthalate, Cr-MIL-101,
which demonstrated a good resistance to air, water, common sol-
vents and thermal treatment (up to 300 C) [11]. Taylor-Pashow
et al. synthesized an iron-containing analog, Fe-MIL-101 [12]. The
MIL-101 material has a zeotype crystal structure consisting of two
types of cages with a diameter of 34 and 29 A˚ (the corresponding
windows are 16 and 12 A˚ ). Coordinatively unsaturated metal sites
(CUS) can be easily generated in the structure of MIL-101 by heat-
ing the material in vacuum [13]. Changing the linker molecule from
1,4-benzenedicarboxylic acid for 1,3,5-benzenetricarboxylic acid
able attention owing to a unique combination of properties, such
as crystalline open structures, extremely high surface areas and
pore volumes, tunable pore size and functionality. All these allow
consideration of MOFs as prospective materials for molecular
recognition, separation, gas storage and catalysis [1–10]. These
materials potentially combine advantages of both homogeneous
and heterogeneous catalysts because their structure consists of iso-
lated metal clusters or atoms connected by polydentate organic
ligands to form a rigid porous framework. The important feature of
such framework is a high number of regularly distributed uniform
metal centers that are accessible for reagents the size of which is
comparable or smaller than the MOF cage entrances.
◦
resulted in the formation of another MOF structure, MIL-100, which
contains cages of 27 and 24 A˚ accessible through microporous
windows of 9 and 6 A˚ , respectively. Both Cr-MIL-100 [14] and
Fe-MIL-100 [15] have been reported by Férey’s group. Schematic
representation of the MIL-101 and MIL-100 structures is shown in
Fig. 1.
∗ Corresponding author. Tel.: +7 383 326 9433; fax: +7 383 330 9573.
E-mail addresses: khold@catalysis.nsk.su, khold@catalysis.ru (O.A. Kholdeeva).
0
920-5861/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2014.01.010
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
O.A. Kholdeeva et al. / Catalysis Today xxx (2014) xxx–xxx
2
Fig. 1. Schematic representation of the MIL-101 and MIL-100 structures: M3O-carboxylate trimer—primary building unit, A and D—supertetrahedra, secondary building
units, B and E—small cages, C and F—large cages of MIL-101 and MIL-100, respectively.
Fe-MIL-100 revealed catalytic activity in the oxidation of
diphenylmethane and triphenylmethane with tert-butyl hydroper-
oxide (TBHP) [16] and Friedel–Crafts benzylation of benzene
without additional purification. TBHP was used as a solution in
decane (4.7 M; Aldrich) or as a 70 wt% aqueous solution. Cyclo-
hexenyl hydroperoxide was prepared according to the literature
[27]. The exact concentration of hydroperoxides was determined
iodometrically prior to use.
[
15]. Iron coordinated to 1,3,5-benzenetricarboxylate (BTC), the
so-called [Fe(BTC)] metal-organic framework, has been commer-
cialized by BASF under the trade name of Basolite F 300. Although
the exact crystal structure of Fe(BTC) is still unknown, it is likely that
it has a structure similar to that of Fe-MIL-100. [Fe(BTC)] is an effec-
tive catalyst for aerobic oxidation of cyclooctane [17] and styrene
2.2. MOF synthesis and characterization
Cr-MIL-101 [11] and Fe-MIL-101 [12] were prepared by the
[
18]. In turn, Cr-MIL-100 operates as effective Lewis acid catalyst in
solvothermal method following protocols reported elsewhere [25].
Fe-MIL-100 and Cr-MIL-100 were synthesized according to the
literature [14,15] with some modifications. In a typical synthe-
sis of Fe-MIL-100, 140 mg metal iron (2.5 mmol), 350 mg trimesic
acid (1.65 mmol), 5 mL HF (5 mmol), 1.2 mL HNO3 (1.2 mmol) and
various organic transformations [19]. Cr-MIL-101 is able to catalyze
cyanosilylation of benzaldehyde [20], carboxylation of epoxides
[
21], benzilyc oxidation of tetralin [22], sulfoxidation of thioethers
with H O [13], oxidation of cyclohexane [23] and allylic oxida-
2
2
tion of alkenes with TBHP [24] or O [25,26]. Recently, we reported
2
1
3 mL H O were put into a Teflon-lined stainless steel bomb and
2
the use of Fe-MIL-101 in the allylic oxidation of alkenes with
molecular oxygen and found a product distribution significantly
different from Cr-MIL-101 [25,26]. So far, no direct comparison of
the catalytic behavior of MIL-100 and MIL-101 materials in selec-
tive oxidation reactions has been reported. In the present work,
we assessed and compared catalytic properties of Fe- and Cr-based
MIL-100 and MIL-101 in two liquid-phase reactions: solvent-free
allylic oxidation of alkenes with molecular oxygen and oxidation
of anthracene (AN) to 9,10-anthraquinone (AQ) using TBHP as oxi-
dant.
◦
heated up to 150 C and then kept at this temperature for 24 h. The
resulting orange solid was filtered off and washed with hot (ca.
◦
8
0 C) water for 3 h. Then, the material was dried at room tem-
perature overnight. In a typical synthesis of Cr-MIL-100, 100 mg
metal chromium (2 mmol), 300 mg trimesic acid (1.4 mmol), 1 mL
HF (4 mmol) and 9 mL H O were placed into a Teflon-lined stain-
2
◦
◦
less steel bomb and heated with the rate of 20 C/min up to 220 C
during 96 h. Cooling was performed with the rate of 10 C/min.
◦
The resulting green solid was filtered off, washed with water and
acetone and dried at room temperature.
The MOFs were characterized by X-ray diffraction (XRD) tech-
nique and low temperature N2 adsorption measurements. XRD
patterns of all the materials were in accordance with the literature
2
. Experimental
[
11,12,14,15]. The average size of crystallites was ca. 0.7 and 7 ␮m
2.1. Materials
for the MIL-101 and MIL-100 materials, respectively. The specific
BET surface areas were in the range of 3200–3400 (MIL-101) and
Cyclohexene (CyH), ␣- and ␤-pinene (Sigma-Aldrich or Alpha-
2
2
000–2200 (MIL-100) m /g. The content of Cr (Fe) was ca. 23 and
Aesar) were purified prior to use by passing through a column
filled with neutral alumina to remove traces of possible oxi-
dation products. Anthracene (99%) and anthraquinone (99%)
were purchased from Aldrich and used without additional
2
6 wt% in activated MIL-101 and MIL-100, respectively.
2.3. Catalytic tests and product analysis
purification. Chlorobenzene, toluene, dichloroethane and acetoni-
trile (Fluka) were dried and stored over activated 4 A˚ molecular
Catalytic oxidations of alkenes with O , product analysis and
2
sieves. All other reactants were obtained commercially and used
product adsorption measurements were performed as described
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
O.A. Kholdeeva et al. / Catalysis Today xxx (2014) xxx–xxx
3
Table 1
CyH oxidation with molecular oxygen over MIL-100 and MIL-101.
◦
Entry
Catalyst (mg)
T ( C)
Selectivity (%)
CyH conversion (%)
Reference
1
2
3
1
2
3
4
5
6
7
8
Fe-MIL-100 (13)
Fe-MIL-100 (13)
Fe-MIL-100 (26)
Fe-MIL-101 (13)
Fe-MIL-101 (13)
Fe-MIL-101 (26)
Cr-MIL-100 (26)
Cr-MIL-101 (26)
50
40
40
50
40
40
60
60
21
11
14
22
15
18
52
56
36
62
60
11
58
57
31
38
43
27
26
63
27
25
17
6
10
7
8
27
8
8
This work
This work
This work
[25]
[25]
This work
This work
[25]
12
16
Reaction conditions: 1 mL CyH, 1 bar O2, 0.02–0.03 mmol TBHP (as initiator), 16 h.
elsewhere [25,26]. Catalytic oxidations of AN were carried out in
thermostated glass vessels under vigorous stirring (600 rpm). Reac-
tions were started by the addition of TBHP to a mixture containing
anthracene, MOF, internal standard (biphenyl) and solvent. Prior
experiments.
MIL-100 or MIL-101
o
◦
40-60 C, 1 bar О
2
to use, MOFs were activated in vacuum: Cr-MIL-101–at 120 C
◦
◦
for 2 h and then 180 C for 4 h, Fe-MIL-101—at 120 C for 6 h and
◦
Cr(Fe)-MIL-100—at 150 C for 6 h. Samples of the reaction mix-
1
2
3
ture were withdrawn periodically during the reaction course by
a syringe. The oxidation products were identified by GS–MS. Yields
of AQ and conversions of AN were quantified by GC. Each experi-
ment was reproduced 3–4 times. Turnover frequency (TOF) values
were determined from initial rates of AN consumption. Catalyst re-
usability was studied in 4–6 time scaled experiments. Before reuse,
catalysts were filtered off, washed with hot chlorobenzene (PhCl)
In our previous work, we revealed that the product distribution
in CyH oxidation over Fe-MIL-101 strongly depended on the reac-
tion temperature, which, in turn, determined the nature of catalysis
◦
[
25]. Thus at 50–60 C, Fe-MIL-101 underwent iron leaching and Fe
species in solution favor the formation of 3. We did not investi-
gate the structure and composition of the soluble iron species. It
could be mono or di(polynuclear) complexes of Fe(III) with car-
boxylates (linker) or some other ligands (e.g. reaction products).
◦
and acetone, and dried in air at 30–40 C for 1–2 h. The nature of
catalysis was verified by hot filtration test [28].
◦
On the contrary, the MOF was stable to leaching at 40 C and pro-
duced 2 as the main oxidation product [25]. The same regularities
2
.4. Instrumentation
turned out to operate with Fe-MIL-100. Indeed, when the reac-
◦
tion was performed at 50 C, hydroperoxide 3 predominated for
GC analyses were performed using a gas chromatograph Agi-
both Fe-MIL-100 (Table 1, entry 1) and Fe-MIL-101 (entry 4). While
lent 4890D equipped with a flame ionization detector and a
0 m × 0.25 mm VF-5 MS capillary column or gas chromatograph
◦
decreasing the reaction temperature down to 40 C, cyclohexenol
3
became the main product over both Fe–MOFs and formed with
the selectivity of 58–62% at 7–8% conversion (entries 2 and 5). No
further substrate conversion occurred upon increasing the reac-
tion time. Enlarging the catalyst amount produced no significant
changes in CyH conversion and product selectivity (Table 1, com-
Chromos GC-1000 equipped with a flame ionization detector and
a 30 m × 0.25 mm BPX-5 capillary column. GC–MS analyses were
carried out using a gas chromatograph HP 6890 (a 50 m × 0.25 mm
DB-5 MS capillary column) equipped with a mass-selective detec-
tor Agilent MSD 5973. XRD measurements were performed on
Shimadzu XRD 7000S or DRON-3M. Nitrogen adsorption at 77 K
was measured using an ASAP-2020 instrument (Micrometrics) or
Autosorb iQ (Quantachrome) within a partial pressure range of
pare entries 2 and 3 or entries 5 and 6). Hot catalyst filtration test
◦
(
Fig. 2a) proved that the catalysis over Fe-MIL-100 at 40 C is truly
heterogeneous. At such temperature, both Fe–MOFs can be used
repeatedly without deterioration of the catalytic properties. Table 2
presents the results on CyH conversion and selectivity toward the
main product 2 obtained in four consecutive reuses. Importantly,
studies by XRD revealed that Fe-MIL-100 preserved its structure
after four reuses (Fig. 3a). This differs significantly Fe-MIL-100
from Fe-MIL-101 that partially lost the regular structure under the
same conditions [25]. Therefore, we may conclude that Fe-MIL-100
and Fe-MIL-101 demonstrate similar catalytic properties in CyH
−
6
1
0
–1.0.
3
. Results and discussion
3.1. Cyclohexene oxidation with molecular oxygen
Previously, we found that the presence of both MOF catalyst and
initiator (TBHP) is required to accomplish CyH oxidation at mild
◦
reaction conditions, viz., 1 atm of O2 and 60 C [25]. Hence, in all
Table 2
catalytic experiments, we added small amounts of TBHP to initiate
the reaction.
Recycling of MIL-101 and MIL-100 in CyH oxidation.
Catalyst
CyH conversiona (%)
Selectivity to 2 (%)
Fe-MIL-101 [25]
Fe-MIL-100
9 (8, 9, 8)
7 (8, 7, 7)
57 (55, 58, 56)
60 (63, 61, 62)
Selectivity to 1 (%)
56 (58, 56, 57)
50 (53, 52, 51)
3
.1.1. Fe-MIL-100 and MIL-101
Table 1 presents the results on solvent-free CyH oxidation
Cr-MIL-101 [25]
Cr-MIL-100
16 (15, 17, 16)
11 (13, 12, 11)
with molecular oxygen over both Fe- and Cr-containing MIL-100
and MIL-101. The main oxidation products were 2-cyclohexen-1-
one (1), 2-cyclohexen-1-ol (2) and cyclohexenyl hydroperoxide
Reaction conditions: 1 mL CyH, 1 bar O2, 0.02–0.03 mmol TBHP (as initiator), 16 h,
◦
◦
40 C for Fe-MILs (13 mg) or 60 C for Cr-MILs (26 mg).
a
(
3). Minor amounts of cyclohexene oxide were found in few
In parentheses, 2nd, 3rd and 4th reuses.
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
4
O.A. Kholdeeva et al. / Catalysis Today xxx (2014) xxx–xxx
7
6
5
4
3
2
1
0
(
a)
12 (b)
1
0
8
6
4
2
0
Catalyst filtration
Catalyst filtration
0
1
2
3
4
5
6
0
1
2
3
4
5
6
Time, h
Time, h
◦
◦
Fig. 2. Hot catalyst filtration test for (a) Fe-MIL-100 (40 C) and (b) Cr-MIL-100 (60 C). Reaction conditions as in Table 1, entries 2 and 7.
oxidation but the former has a higher stability toward degradation
under the reaction conditions.
high selectivity toward 1 realized over Cr-MIL-101 was attributed
to the ability of Cr centers to promote dehydration of 3.
Cr-MIL-101(100)
3
.1.2. Cr-MIL-100 and MIL-101
In contrast to Fe-MIL-100, the main product of CyH oxidation
o
6
0 C
over Cr-MIL-100 is ␣,␤-unsaturated ketone 1, which forms with
the selectivity of 52% at 12% CyH conversion (Table 1, entry 7).
The selectivity is close to that acquired with Cr-MIL-101 (entry
Earlier, Sheldon et al. demonstrated the possibility of such
transformations for both cyclohexenyl hydroperoxide [27] and
cyclohexyl hydroperoxide [29] over Cr-containing catalysts.
Indeed, we found that both Cr-MIL-101 and Cr-MIL-100 catalyze
decomposition of 3 to yield 1. Fig. 4 shows plots of the accumula-
tion of 1 versus time. Cyclohexenone selectivity remains constant
during the decomposition process (67–69%). The only byproduct
is 2 which might form along with 1 in the decomposition of 3 via
8
) although the substrate conversion over MIL-101 is higher (16%
versus 12%). Hence, again the product distribution is determined
mostly by the nature of the transition metal in the MOF frame-
work. Hot catalyst filtration test confirmed no leaching of active
◦
chromium species from Cr-MIL-100 at 60 C (Fig. 2b), pointing out
that Cr-MIL-100 is more stable toward metal leaching than the
iron analog that can not survive at such temperature. Both Cr-
MOFs could be used at least four times without loss of the catalytic
properties (Table 2). XRD showed no degradation of the Cr-MIL-
>
Haber–Weiss mechanism [30]. Similar values of cyclohexenone
selectivity (65–74%) were reported by Sheldon et al. for various
chromium catalysts, including Cr(acac) , Cr-APO5 and CrS-1 [27].
3
1
00 structure after four consecutive cycles (Fig. 3b). Previously, the
structural integrity of Cr-MIL-101 was also corroborated by XRD
25]. Hence, in contrast to the Fe-MILs, no difference in stability of
Possible routes leading to the preferable formation of unsaturated
alcohols over Fe-MIL-101 have been suggested earlier [25]. Similar
product distributions found for Fe-MIL-100 and Fe-MIL-101 imply
that similar pathways might operate with Fe-MIL-100.
[
Cr-MIL-100 and Cr-MIL-101 has been found.
Thus the product selectivities found for CyH oxidation in the
presence of Fe-MIL-100 and Cr-MIL-100 are similar to those
obtained with Fe-MIL-101 and Cr-MIL-101 [25], thereby indicat-
ing that the nature of transition metal rather than the nature of
the MOF linker determinates the oxidation mechanism. Like in the
case of Fe-MIL-101 and Cr-MIL-101 [25], radical scavengers (ionol)
impede the oxidation processes over both Fe-MIL-100 and Cr-MIL-
3.2. Terpene oxidation with molecular oxygen
Oxidation of two representative terpenes, ␣- and ␤-pinenes,
was explored under the conditions at which all the MOF catalysts
◦
remain stable, i.e. 40 and 60 C for Fe- and Cr-MILs, respectively.
1
00, indicating that radical chain steps are involved. To explain
The results are presented in Table 3. In the oxidation of ␣-pinene,
both Fe-MIL-100 and Fe-MIL-101 produced a significant amount of
verbenol along with campholenic aldehyde, hydroperoxides and
epoxide, but no verbenone was found in the reaction mixture
the difference in the product distribution in CyH oxidation over
Fe-MIL-101 and Cr-MIL-101, we suggested different pathways for
the transformation of the primary oxidation product, 3 [25,26]. The
(
a)
(b)
A
A
B
В
5
10
15
2
20
25
30
35
5
10
15
20
25
30
θ, degrees
2θ, degrees
Fig. 3. XRD pattern of (a) Fe-MIL-100 and (b) Cr-MIL-100: initial sample (A) and after four catalytic cycles (B). Reaction conditions as in Table 2.
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
O.A. Kholdeeva et al. / Catalysis Today xxx (2014) xxx–xxx
5
0
.20
.16
.12
.08
.04
.00
(
a)
(b)
0
0
0
0
.20
.16
.12
.08
0
0
0
0
0
0.04
0.00
0
50
100
Time, min
150 200 250
0
50
100 150 200 250
Time, min
Fig. 4. Formation of cyclohexenone (ꢀ) in the course of decomposition of cyclohexenyl hydroperoxide 3 (᭹) over Cr-MIL-100 (a) and Cr-MIL-101 (b). Reaction conditions:
◦
0
.2 mmol of 3, 26 mg MOF, 1 mL cyclohexane, 60 C.
Table 3
Solvent-free oxidation of ␣- and ␤-pinenes with O2 over MIL-101 and MIL-100.
Entry
Catalyst
Conversion (%)
Selectivity (%)
ROOHɑ
1
2
3
4
Fe-MIL-100
Cr-MIL-100
Fe-MIL-101 [26]
Cr-MIL-101 [26]
9
32
15
33
31
–
30
–
9
9
9
8
15
9
33
8
44
37
25
14
20
12
26
39
ROOHɑ
7
8
9
Fe-MIL-100
Cr-MIL-100
Fe-MIL-101 [26]
Cr-MIL-101 [26]
12
18
13
22
64
20
57
32
–
2
–
23
10
33
–
26
45
43
18
1
0
27
◦
◦
Reaction conditions: 1.5 mL of alkene, 1 bar O2, 0.02–0.03 mmol TBHP (initiator), 16 h, 40 C for Fe-MILs (13 mg) or 60 C for Cr-MILs (26 mg).
Sum of hydroperoxides.
a
(
Table 3, entries 1 and 3). The conversion of ␣-pinene was slightly
lower for Fe-MIL-100 relative to Fe-MIL-101 (9% versus 12%). In
-pinene oxidation, both Fe-MILs gave two unsaturated alcohols
pinocarveol and myrtenol) along with hydroperoxides, but prac-
organic linker nature produces less effect on the alkene conversion
and product selectivity. Interstingly, the situation is different if we
compare adsorption properties of the MOFs with regards to the CyH
oxidation products. We evaluated the diffusion time and product
(1 and 2) adsorption constants for Fe- and Cr-MIL-100 following
the methodology reported in our previous work [25]. The results
are presented in Table 4 in comparison with the results acquired
ealier for the MIL-101 analogues. While the diffusion time does not
exceed 1.5 min for all the MOFs, the adsorption constants are higher
for MIL-100 than for MIL-101 and, surprisingly, do not depend on
the nature of metal.
Previously we demonstrated that the addition of both 1 and 2
could stop CyH oxidation over Cr- and Fe-MIL-101 [25]. We found
similar rate-retarding effects upon addition of these products to
the reaction with Fe- and Cr-MIL-100 (Fig. 5a and b). Water, the
product which forms in a stoichiometric amount along with 1, can
also act as an inhibitor of the alkene oxidation catalyzed by MOFs,
␤
(
tically no ketones formed (Table 3, entries 7 and 9). The substrate
conversion was close for both MOFs: 12% and 13% for Fe-MIL-100
and Fe-MIL-101, respectively.
Cr-containing MIL-100 and MIL-101 transformed ␣- and ␤-
pinene mainly to hydroperoxides and unsaturated ketones and
alcohols, with ketones predominating over alcohols (Table 3).
Minor amounts of epoxide and campholenic aldehyde also formed
in the reaction with ␣-pinene. Higher substrate conversions were
attained for Cr-MIL-101 than for Cr-MIL-100 (26% versus 20% and
2
2% versus 18% for ␣- and ␤-pinene, respectively). Although the
composition of products is more complicated in the oxidation
of terpenes than in case of CyH, most likely, because of isomer-
ization/rearrangement reactions typical of terpenoids, we may
conclude that, in general, Fe-MILs favor the formation of unsatu-
rated alcohols while Cr-MILs mediate the formation of unsaturated
ketones.
as confirmed by the experiment with H O additives (Fig. 5b and
2
Table 4
Adsorption of cyclohexenone (1) and cyclohexenol (2) on MIL-100 and MIL-101.
3.3. Product adsorption studies
Material
K1 (M−1)
K2, (M
−1
)
Diffusion time (min)
From the data presented in Tables 1 and 3, one can judge that the
Fe-MIL-100
Fe-MIL-101 [25]
Cr-MIL-100
8 ± 2
3 ± 1
8 ± 2
3 ± 1
8 ± 2
1
nature of metal strongly affects not only the product distribution
but also the attainable substrate conversion that is considerably
higher for Cr-containing MOFs than for the iron analogues. The
1 ± 0.3
8 ± 2
1.5
1.5
1.5
Cr-MIL-101 [25]
1 ± 0.3
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
O.A. Kholdeeva et al. / Catalysis Today xxx (2014) xxx–xxx
6
7
6
5
4
3
2
1
0
12 (b)
18
(
a)
(c)
1
6
4
1
0
8
6
4
2
0
1
12
10
8
+ 0.2 mmol of 1
and 0.1 mmol of 2
+ 0.3 mmol of 2
and 0.1 mmol of 1
+
2
0.3 mmol of H O
2
6
+
0.4 mmol H O
4
2
2
0
0
1
2
3
4
5
6
0
1
2
3
4
5
6
0
1
3
4
5
6
Time, h
Time, h
Time, h
Fig. 5. Inhibition effects of oxidation products on CyH oxidation over (a) Fe-MIL-100, (b) Cr-MIL-100 and (c) Cr-MIL-101.
c). Hence, we believe that adsorption of the oxidation products is
responsible for the rather low conversions of alkenes observed in
oxidation over the MOF catalysts. Meanwhile, we should mention
that solvent-free conditions ensure high volume yields of products
even at low substrate conversions. Indeed, volume yields of the
oxygenated products in the oxidation of CyH and pinenes attained
AN conversion, %
AQ selectivity, %
-1
TOF, h
1
00
10
8
8
6
4
2
0
0
0
0
0
1
38 and 246 g/L for Fe- and Cr-MILs, respectively.
6
3.4. Anthracene oxidation with TBHP
4
The development of a truly heterogeneous catalyst for the selec-
2
tive oxidation of AN to AQ is a challenging goal, and various types
of solids have been tested in this reaction [31–34]. Since the frame-
work of both MIL-100 and MIL-101 has a strong affinity to aromatic
hydrocarbons [35], we attempted to use these MOFs as catalytic
0
o
80
100 C
materials for AN oxidation with nonaqueous TBHP as oxidant.
Fig. 7. Effect of temperature on AN oxidation with TBHP over Cr-MIL-101. Reaction
O
conditions: AN 0.02 M, TBHP 0.09 M, Cr-MIL-101 1.5 mg, ClPh 2.5 mL.
Cr(Fe)-MIL-101(100)
solvent, TBHP
and a higher concentration (0.1 M) of the aromatic substrate can be
employed to produce AQ with excellent selectivity (Table 5). While
using 10 mol% excess of TBHP relative to AN (the stoichiometry of
AN oxidation with TBHP to produce AQ is 1:3) the conversion of
AN attained 95% (Table 5, entry 2). By increasing the amount of
the oxidant, quantitative conversion of AN to AQ can be accom-
plished (Table 5, entry 4). The efficiency of the oxidant utilization
was, therefore, in the range of 67–86%. The use of aqueous TBHP
reactant instead of its nonaqueous solution in decane did not lead
to deterioration of the product yield but decreased the reaction rate
(Table 5, entry 5). This is not surprising if we remember that water
can adsorb on the MOF active centers and compete with reactants
for the adsorption sites. It is precedented in the literature that cat-
alyst activation in vacuum enables improvement of the catalytic
performance [13,23,24]. Indeed, nonactivated Cr-MIL-101 showed
some reduction of the reaction rate in the oxidation of AN.
O
AN
AQ
Because of the poor solubility of AN in most organic solvents,
benzene is typically employed to perform AN oxidation under
liquid phase conditions [31–33]. We studied AN oxidation over Cr-
MIL-101 in a few organic solvents and found that chlorobenzene is
the best choice in terms of AN conversion, AQ selectivity and TOF
Fig. 6). Therefore, this solvent was used in all other experiments.
Selvaraj at al. [31] reported that temperature around 80 C was
optimal for AN oxidation with TBHP in the presence of Cr-SBA-15
as catalyst. For Cr-MIL-101, we found that the increase of the reac-
tion temperature from 80 to 100 C allowed the reaction rate to be
(
◦
◦
increased ca. 4 times without loss of the AQ selectivity (Fig. 7). Fur-
thermore, the solubility of AN in chlorobenzene at 100 C increases
◦
AN conversion, %
AQ selectivity, %
Fig. 8 shows a comparison of different parameters in oxidation of
AN over Cr- and Fe-containing MIL-101 and MIL-100. With Fe-MIL-
-1
TOF, h
1
01, the reaction was significantly slower than with the Cr analogue
1
00
10
8
and 92% conversion of AN was reached after 6 h (compare with
8
6
4
2
0
0
0
0
0
Table 5
Oxidation of anthracene with TBHP over Cr-MIL-101.
6
Entry Catalyst
[TBHP]
M)
Time Conversion AN
Select. AQ
(%)
TOFɑ
−
(
(h)
(%)
(h 1)
4
1
2
3
4
5
no catalyst
0.45
Cr-MIL-101 0.33
Cr-MIL-101 0.38
Cr-MIL-101 0.45
Cr-MIL-101 0.45
3
5
100
100
100
100
100
–
2.5
1.5
1
95
98
100
100
12
22
24
15
2
b
2.5
0
◦
ClPh
MePh DCE MeCN
Reaction conditions: AN 0.1 M, MOF 3.2 mg, chlorobenzene 2 mL, 100 C.
a
TOF = (moles of AN consumed)/[(moles of Cr)·time]; determined from the initial
rates.
Fig. 6. Effect of solvent nature on AN oxidation with TBHP over Cr-MIL-101. Reaction
conditions: AN 0.02 M, TBHP 0.09 M, Cr-MIL-101 2.5 mg, solvent 5 mL, 80 C, 4–6 h.
b
◦
Aqueous TBHP (70 wt%) was used.
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
O.A. Kholdeeva et al. / Catalysis Today xxx (2014) xxx–xxx
7
AN conversion, %
AQ selectivity, %
AN conversion, %
AQ selectivity, %
-1
-1
TOF, h
TOF, h
100
100
80
60
40
20
0
80
60
40
20
0
I
II
III
IV
V
VI
Catalytic cycles
Fig. 9. Recycling of Cr-MIL-101 in oxidation of AN with TBHP. Reaction conditions
Fig. 8. AN oxidation with TBHP over Cr- and Fe-MIL-101 and MIL-100. Reaction
conditions: AN 0.1 M, TBHP 0.45 M, MOF 1.6 mg, ClPh 5 mL, 100 C, 1.5 (Cr-MIL-101)
as in Table 5 (entry 4).
◦
or 6 h (other MOFs).
100
Hot catalyst filtration
1
00% conversion after 1.5 h for Cr-MIL-101). The selectivity toward
8
0
0
AQ remained 100%. If we compare Cr-MIL-101 with Cr-MIL-100,
we can see that the reaction rate expressed in TOFs decreases ca. 4
times for the latter, pointing out that diffusion limitation may take
place for the bulky AN and AQ molecules within MIL-100 that pos-
sesses smaller cages and larger crystallites than MIL-101. Although
6
40
both aromatic substrate and product molecules are flat, the size of
AN molecule in two other dimensions is 9.2 × 5.0 A˚ while the size
20
of AQ molecule is 9.3 × 5.4 A˚ . The size of the TBHP molecule is esti-
mated as 4.4 × 3.7 × 4.0 A˚ . At the same time, the average diameter of
cage entrances in MIL-100 is 9 and 6 A˚ for the large and small win-
0
dows, respectively [14,15]. So, we may suppose that both AN and
AQ molecules can diffuse only through the large windows. Given
in mind that the cage entrances for MIL-101 are 12 and 16 A˚ , it is
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Time, h
´
not surprising that this MOF is considerably more active than MIL-
Fig. 10. Hot catalyst filtration test for AN oxidation with TBHP over Cr-MIL-101.
Reaction conditions: AN 0.02 M, TBHP 0.09 M, Cr-MIL-101 1 mg, ClPh 2 mL, 100 C.
◦
1
00 in the oxidation of AN. Among the MOFs studied, Fe-MIL-100
showed inferior catalytic properties (only 18% AN conversion after
h and 90% AQ selectivity), indicating the influence of both the MOF
6
completely, indicating that no active chromium species had leached
into solution during the oxidation process. Additionally, elemen-
tal analysis confirmed that no chromium species (≤0.1 ppm) were
present in the filtrate. XRD patterns of Cr-MIL-101 before the cat-
alytic reaction and after reuse are practically identical (Fig. 11),
structure (determined by the nature of the organic linker) and the
nature of transition metal (Fe versus Cr) on the oxidation process.
The lower selectivity of Fe-MIL-100 relative to Fe-MIL-101 could be
due to the lower substrate conversion (and, therefore, the presence
of some intermediate products) over the former. Finally, we may
conclude that Cr-MIL-101 revealed superior catalytic performance
in terms of AN conversion, AQ yield and TOF in comparison with
other catalysts. For example, mesostructured chromium-silicate
Cr-SBA-15 catalyzed AN oxidation with TBHP to produce AQ with
1
00% selectivity at 90% AN conversion [31] while the attainable
selectivity and conversion were 98% and 79%, respectively, with Cr-
MCM-41 as catalyst [32]. Furthermore, the total reaction time was
by order of magnitude greater with chromium-silicates than with
Cr-MIL-101 (20 h versus 1.5 h), indicating much higher activity of
the MOF catalyst.
after reuse
Importantly, Cr-MIL-101 can be used repeatedly without loss
of the catalytic activity and selectivity. Fig. 9 demonstrates the
recycling behavior of this MOF in six consecutive runs. The cata-
lyst was simply separated by filtration, washed, dried in air and
used in the next run. No MOF reactivation was required between
the catalytic runs, which makes the recycling a convenient and fast
procedure. The hot catalyst filtration test (Fig. 10) proved heteroge-
neous nature of catalysis over Cr-MIL-101: after fast removal of the
catalyst at ca. 45% conversion, the reaction in the filtrate stopped
initial material
2
3
4
5
6
7
8
9
10
2
Θ, degrees
Fig. 11. XRD patterns for Cr-MIL-101: initial material and after reuse.
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010

G Model
CATTOD-8857; No. of Pages8
ARTICLE IN PRESS
O.A. Kholdeeva et al. / Catalysis Today xxx (2014) xxx–xxx
8
which implies that no structural damage of the MOF occurred under
the turnover conditions of AN oxidation with TBHP.
(grant N 13-03-00413) and the Ministry of Education and Science
of the Russian Federation. I.Y.S. acknowledges French Embassy in
Moscow for the doctoral fellowship.
4
. Conclusions
References
The iron- and chromium-containing MOFs of the MIL-100 and
MIL-101 structure are efficient catalysts for both solvent-free
alkene oxidation with molecular oxygen and AN oxidation to AQ
with TBHP in chlorobenzene as solvent. Studies on cyclohexene
oxidation revealed that the Cr-MILs are more stable than their
iron analogs. While Cr-MIL-100 and Cr-MIL-101 were stable to
[1] J.L.C. Rowsell, O.M. Yaghi, Microporous Mesoporous Mater. 73 (2004) 3.
[
[
[
2] A.K. Cheetham, C.N.R. Rao, R.K. Feller, Chem. Commun. (2006) 4780.
3] G. Férey, Chem. Soc. Rev. 37 (2008) 191.
4] A.U. Czaja, N. Trukhan, U. Muller, Chem. Soc. Rev. 38 (2009) 1284.
[5] D. Farrusseng, S. Aguado, C. Pinel, Angew. Chem. Int. Ed. 48 (2009) 7502.
6] A. Corma, H. García, F.X. Llabrés i Xamena, Chem. Rev. 110 (2010) 4606.
[7] M. Yoon, R. Srirambalaji, K. Kim, Chem. Rev. 112 (2011) 1196.
8] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Catal. Sci. Technol. 1 (2011) 856.
[
◦
metal leaching at 60 C, both Fe-MIL-101 and Fe-MIL-100 could
[
◦
resist only 40 C. According to XRD, Fe-MIL-100 preserved its
[9] Thematic issue on metal-organic frameworks, Chem. Rev. 112 (2012) 673.
[10] Y.K. Hwang, G. Férey, U.H. Lee, J.S. Chang, in: M.G. Clerici, O.A. Kholdeeva (Eds.),
Liquid Phase Oxidation via Heterogeneous Catalysis: Organic Synthesis and
Industrial Applications, Wiley, Hoboken, New Jersey, 2013, p. 371.
11] G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I. Mar-
giolaki, Science 309 (2005) 2040.
◦
structure after several reuses at 40 C, which contrasts with Fe-
MIL-101 that lost partially the structural integrity under the same
conditions. Hence, the MOF stability increases in the sequence
Fe-MIL-101 < Fe-MIL-100 < Cr-MIL-100, Cr-MIL-101. Whereas the
MOF structure (MIL-101 versus MIL-100) just slightly affects the
catalytic performance in the alkene oxidation, the nature of tran-
sition metal determines both the attainable substrate conversion
and the reaction selectivity. In general, Fe-MILs favor the forma-
tion of unsaturated alcohols, while Cr-MILs afford the formation
of unsaturated ketones. The high selectivity of the Cr-containing
MOFs toward ketones is, most likely, due to the high activity of these
materials in dehydration of corresponding hydroperoxides to pro-
duce unsaturated ketones. Adsorption of the oxidation products,
including water, inhibits the oxidation process over the MOFs. The
product (cyclohexenol and cyclohexenone) adsorption constants
are higher for MIL-100 than for MIL-101 and do not depend on the
nature of the metal.
[
[
[
[
[
[
[
12] K.M.L. Taylor-Pashow, J. Della Rocca, Z. Xie, S. Tran, W. Lin, J. Am. Chem. Soc.
131 (2009) 14261.
13] 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.
14] G. Ferey, C. Serre, C. Mellot-Draznieks, F. Millange, S. Surble, J.D.I. Margiolaki,
Angew. Chem. 116 (2004) 6456.
15] P. Horcajada, S. Surblé, C. Serre, D.-Y. Hong, Y.-K. Seo, J.-S. Chang, J.-M. Grenèche,
I. Margiolaki, G. Férey, Chem. Commun. (2007) 2820.
16] A. Dhakshinamoorthy, M. Alvaro, Y.K. Hwang, Y.-K. Seo, A. Corma, H. Garcia,
Dalton Trans. 40 (2011) 10719.
17] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Chem. Eur. J. 17 (2011) 6256.
[18] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, ACS Catal. 1 (2011) 836.
19] L. Mitchell, B. Gonzalez-Santiago, J.P.S. Mowat, M.E. Gunn, P. Williamson, N.
Acerbi, M.L. Clarke, P.A. Wright, Catal. Sci. Technol. 3 (2013) 606.
20] A. Henschel, K. Gedrich, R. Kraehnert, S. Kaskel, Chem. Commun. (2008) 4192.
[
[
[21] O.V. Zalomaeva, A.M. Chibiryaev, K.A. Kovalenko, O.A. Kholdeeva, B.S. Balzhin-
imaev, V.P. Fedin, J. Catal. 298 (2013) 179.
[
[
[
[
[
[
22] J. Kim, S. Bhattacharjee, K.-E. Jeong, S.-Y. Jeong, W.-S. Ahn, Chem. Commun.
2009) 3904.
In AN oxidation with TBHP, Cr- and Fe-MIL-101 are more
active than their MIL-100 analogues, which indicates that diffu-
sion of bulky AN molecules meets limitation within MIL-100 that
has smaller cage entrances and larger crystallites than MIL-101.
The nature of the metal also affects the catalytic performance.
The catalytic activity decreases in the order Cr-MIL-101 > Fe-MIL-
(
23] N.V. Maksimchuk, K.A. Kovalenko, V.P. Fedin, O.A. Kholdeeva, Chem. Commun.
48 (2012) 6812.
24] N.V. Maksimchuk, K.A. Kovalenko, V.P. Fedin, O.A. Kholdeeva, Adv. Synth. Catal.
352 (2010) 2943.
25] I.Y. Skobelev, A.B. Sorokin, K.A. Kovalenko, V.P. Fedin, O.A. Kholdeeva, J. Catal.
298 (2013) 61.
26] I.Y. Skobelev, K.A. Kovalenko, V.P. Fedin, A.B. Sorokin, O.A. Kholdeeva, Kinet.
Catal. 54 (2013) 607.
1
01 > Cr-MIL-100 > Fe-MIL-100. Cr-containing MIL-101 revealed
superior catalytic performance in terms of AN conversion, AQ selec-
tivity and TOF. No leaching of chromium into solution occurred and
the MOF could be reused several times without loss of the catalytic
properties.
27] H.E.B. Lempers, J.D. Chen, R.A. Sheldon, Stud. Surf. Sci. Catal. 94 (1995) 705.
[28] R.A. Sheldon, M. Wallau, I.W.C.E. Arends, U. Schuchardt, Acc. Chem. Res. 31
1998) 485.
(
[
[
[
[
29] J.D. Chen, J. Dakka, R.A. Sheldon, Appl. Catal. 108 (1994) LI–L6.
30] F. Haber, J. Weiss, Proc. R. Soc. London, Ser. A 147 (1934) 332.
31] M. Selvaraj, S. Kawi, Microporous Mesoporous Mater. 101 (2007) 240.
32] N. Srinivas, V. Radha Rani, S.J. Kulkarni, K.V. Raghavan, J. Mol. Catal. A: Chem.
Acknowledgment
179 (2002) 221.
[
33] A. Bordoloi, F. Lefebvreb, S.B. Halligudi, J. Catal. 247 (2007) 166.
The authors thank Dr. A. N. Shmakov and V. A. Utkin for some
XRD and GC–MS measurements, respectively. The research was
partially supported by the Russian Foundation for Basic Research
[34] C. Pérollier, C. Pergrale-Mejean, A.B. Sorokin, New J. Chem. 29 (2005) 1400.
35] D.-Y. Hong, Y.K. Hwang, C. Serre, G. Férey, J.-S. Chang, Adv. Funct. Mater. 19
2009) 1537.
[
(
Please cite this article in press as: O.A. Kholdeeva, et al., Hydrocarbon oxidation over Fe- and Cr-containing metal-organic frameworks
MIL-100 and MIL-101–a comparative study, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.01.010