Rearrangement of α-pinene oxide to campholenic aldehyde over the trimesate metal-organic frameworks MIL-100, MIL-110 and MIL-96

Article abstract of DOI:10.1016/j.jcat.2013.11.006

The catalytic performance of porous metal-benzenetricarboxylates, such as MIL-100(Al, Fe and Cr), MIL-110(Al) and MIL-96(Al), was investigated with a combination of physicochemical and catalytic approaches in the rearrangement of α-pinene oxide to campholenic aldehyde (CA). The investigation of Lewis acidity was done by EPR and IR spectroscopy using 2,2′,6,6′- tetramethyl-1-piperidinyoxyl radical and benzonitrile as the probe molecules, respectively. For Al-BTCs, both these methods showed a decrease in the amount of Lewis acid sites as follows: MIL-100 > MIL-110 > MIL-96. The reaction rate and selectivity toward CA also decreased in the same order. A relatively good correlation between selectivity toward CA and the electronegativity of the metal ion was found for isostructural MIL-100(Al, Fe and Cr). The selectivity toward CA decreased in the order: MIL-100(Al) > MIL-100(Fe) > MIL-100(Cr). The high selectivity toward CA in the presence of MIL-100 has been suggested to originate from the unique structure of this material, which favors shape selectivity.

Full text of DOI:10.1016/j.jcat.2013.11.006

Journal of Catalysis 311 (2014) 114–120
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Rearrangement of
a-pinene oxide to campholenic aldehyde over
the trimesate metal–organic frameworks MIL-100, MIL-110 and MIL-96
a,b,
⇑
a
a
c
c
Maria N. Timofeeva
, Valentina N. Panchenko , Anna A. Abel , Nazmul Abedin Khan , Imteaz Ahmed ,
a
d
c,
⇑
Artem B. Ayupov , Konstantin P. Volcho , Sung Hwa Jhung
a
Boreskov Institute of Catalysis SB RAS, Prospekt Akad. Lavrentieva 5, 630090 Novosibirsk, Russian Federation
Novosibirsk State Technical University, Prospekt K. Marksa 20, 630092 Novosibirsk, Russian Federation
Department of Chemistry and Green-Nano Materials Research Center, Kyungpook National University, Sankyuck-Dong, Buk-Ku, Daegu 702-701, Republic of Korea
Novosibirsk Institute of Organic Chemistry SB RAS, Prospekt Akad. Lavrentieva 9, 630090 Novosibirsk, Russian Federation
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
The catalytic performance of porous metal–benzenetricarboxylates, such as MIL-100(Al, Fe and Cr), MIL-
Received 22 September 2013
Revised 1 November 2013
Accepted 7 November 2013
Available online 18 December 2013
1
10(Al) and MIL-96(Al), was investigated with
a combination of physicochemical and catalytic
approaches in the rearrangement of -pinene oxide to campholenic aldehyde (CA). The investigation
a
0
0
of Lewis acidity was done by EPR and IR spectroscopy using 2,2 ,6,6 -tetramethyl-1-piperidinyoxyl radical
and benzonitrile as the probe molecules, respectively. For Al–BTCs, both these methods showed a
decrease in the amount of Lewis acid sites as follows: MIL-100 > MIL-110 > MIL-96. The reaction rate
and selectivity toward CA also decreased in the same order. A relatively good correlation between selec-
tivity toward CA and the electronegativity of the metal ion was found for isostructural MIL-100(Al, Fe and
Cr). The selectivity toward CA decreased in the order: MIL-100(Al) > MIL-100(Fe) > MIL-100(Cr). The high
selectivity toward CA in the presence of MIL-100 has been suggested to originate from the unique struc-
ture of this material, which favors shape selectivity.
Keywords:
Metal–organic framework
Lewis acidity
Isomerization of
a-pinene oxide
Campholenic aldehyde
Ó 2013 Elsevier Inc. All rights reserved.
1
. Introduction
cyclization of (+)-citronellal and Diels–Alder cyclization reactions
[
1,3–5].
Isomerization of PO is commonly used to produce trans-
During the past two decades, metal–organic frameworks
(
MOFs) have provoked great interest because of their considerable
carveol (trans-carv) and campholenic aldehyde (CA), which are
highly valuable ingredients for the production of flavors
(Scheme 1). It is well known that the use of mild Lewis acids fa-
vors the formation of CA, while Brönsted acid sites will result in
the formation of trans-carv and trans-sorbrerol (trans-sobr)
applications in adsorption and catalysis. Nowadays, catalysis by
MOFs is a rapidly expanding field due to the unique textural and
physicochemical properties of these materials, including a high
specific surface area, the ability to control porosity by the nature
of the organic ligands, the high amount and dispersity of metal ions
in the framework, and the ability to control acid–base and redox
properties by varying both the nature of metals and the organic li-
gands [1–4]. The presence of coordinatively unsaturated metal
sites (CUS) in some of these MOFs allows their application in catal-
ysis as Lewis acids. Recently, metal–benzenetricarboxylates (M–
BTCs, M = Fe, Cu and Zr) were shown to be suitable as a heteroge-
2 2
[3,6,7]. ZnCl and ZnBr are the most active and selective homo-
geneous systems in this reaction. Zeolites, zeotype materials, and
MOFs have also been applied as Lewis acids in this reaction
[1,3,5,6,8].
Thus, Fe-containing M–BTCs, such as the commercial iron
trimesate Basolite F-300 (Fe(BTC)), MIL-100(Fe), MIL-88(Fe) and
MIL-127(Fe), have been used as catalysts with Lewis acid sites
(LAS) for the rearrangement of PO to CA in the absence of solvent
at 70 °C [5]. About 10% conversion of PO with 50% selectivity to-
ward CA was obtained in the presence of these materials. Of note,
the TON value for MIL-100(Fe) was comparable with that for Fe–Y
neous catalyst for the isomerization of
a-pinene oxide (PO), the
2
zeolite and was higher than that for homogeneous ZnCl .
⇑
Corresponding authors. Address: Boreskov Institute of Catalysis SB RAS,
According to De Vos et al. [1], the catalytic performance of MIL-
00(Fe) can be significantly changed by acid pretreatment. Acid
Prospekt Akad. Lavrentieva 5, 630090 Novosibirsk, Russian Federation.
Fax: +7 383 330 8056 (M.N. Timofeeva), fax: +82 53 950 6330 (S.H. Jhung).
E-mail addresses: timofeeva@catalysis.ru (M.N. Timofeeva), sung@knu.ac.kr (S.H.
Jhung).
1
pretreatment leads to an increase in both Lewis and Brönsted acid-
ity. Prolonged acid pretreatment of the catalyst favors a decrease in
0
021-9517/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2013.11.006

M.N. Timofeeva et al. / Journal of Catalysis 311 (2014) 114–120
115
O
OH
OH
H+
Lewis acid sites
O
+
α-pinene oxide
campholenic aldehyde
pinocarveol
(
CA)
trans-carveol
trans-carv)
(
Scheme 1. Products obtained in the course of PO rearrangement.
the conversion of PO and in the selectivity toward CA. The low
selectivity toward CA has been attributed to the presence of the
Brönsted acid groups as catalytic sites. Conversely, high selectivity
structure/Lewis acidity and catalytic activity of these M–BTCs
in the rearrangement of PO to CA.
3 2
toward CA has been shown in the presence of Cu (BTC) [9]. The
2
. Experimental
nature of the solvent affects the activity and selectivity of the reac-
tion. Dichloroethane and toluene are the best solvents and have
also been found to be the best reaction media for zeolites, Brön-
sted, and Lewis acids [6,7,10,11].
2.1. Materials
a
-Pinene oxide (98.0%) was purchased from Acros Organics. Oc-
tane and FeCl O were purchased from Merck. Commercial
ꢀ6H
dichloroethane, 2,2 ,6,6 -tetramethyl-1-piperidinyoxyl radical
TEMPO) (Aldrich), Al(NO O (98 wt.%, Junsei), 1,3,5-benzene-
ꢀ9H
tricarboxylic acid (H BTC) and 1,3,5-trimethyl-benzenetricarboxy-
late (Me –BTC, 98%, Aldrich), ortho-phosphoric acid (H
We can assume that M–BTCs, such as MIL-100(Cr, Fe and Al),
MIL-96(Al) and MIL-110(Al), possess high selectivity toward CA,
because these materials possess isolated and well-defined Lewis
acid sites (LAS). Furthermore, these materials, like zeolites, pos-
sess high surface areas and unique well-defined pore openings
and, therefore, have potential in heterogeneous catalysis with
possible shape selectivity. Noteworthy, MIL-100(Al), MIL-96(Al),
and MIL-110(Al) synthesized from identical precursor reactants,
and only the pH of the starting mixture and/or the reaction time
allow to control structures [12,13]. According to a significant
number of studies [12,14], the structure of mesoporous MIL-
3
2
0
0
(
3
)
3
2
3
3
3
PO
4
,
8
5 wt.%, Merck), sodium hydroxide (NaOH, 4 M), nitric acid
0
(
(
HNO
3
, 60 wt.%), iron powder (Fe , 99%, DC Chemical Co.), CrO
3
98%, Junsei), and hydrofluoric acid (HF, 48%, OCI Company Ltd.)
were used without any further purification.
2.2. Synthesis of M–BTCs (M = Al, Fe and Cr)
1
00(Al), [Al
3
O(OH)(H
2
O)
2
[BTC]ꢀ24H
2
O], is assembled from tri-
3
+
mers of -O-bridged Al octahedra, which are connected by
l
3
Al–BTCs were synthesized from Al(NO
tricarboxylic acid (H BTC) or trimethyl 1,3,5-benzenetricarboxy-
late (Me –BTC), sodium hydroxide (NaOH, 4 M), nitric acid
HNO , 60%), and deionized water similar to the reported methods
3
)
3 2
ꢀ9H O, 1,3,5-benzene-
BTC linkers into a large-pore framework structure. The corre-
sponding three-dimensional framework exhibits two types of
cavities. The first type of cavity is delimited by 12 pentagonal
windows with a size of 5.5 Å (dodecahedral cage); the second
cavity is delimited by 12 pentagonal windows and 4 hexagonal
windows with a size of 8.6 Å (hexadodecahedral cage). Therefore,
the porosity of MIL-100 originates from both 25 Å and 29 Å mes-
opores, which are accessible via 5.5 Å and 8.6 Å windows,
respectively. MIL-96(Al) and MIL-110(Al) also exhibit a three-
dimensional framework. MIL-110(Al), with a chemical formula
3
3
(
3
for MIL-100 [12,14], MIL-110 [12,15], and MIL-96 [17] under
autogenous pressure at 210 °C. The reactant compositions for the
desired phases are shown in Table 1. For conventional electric crys-
tallization, the reaction mixture of 20 g was loaded in a Teflon-
lined autoclave and put in a preheated electric oven. The crystalli-
zation times for MIL-100(Al) and MIL-110(Al) were 2 h, while for
MIL-96(Al) this was 2 days. MIL-100(Fe) was synthesized accord-
ing to a reported procedure [22]. In this case, iron powder,
of Al
8
(OH)12{(OH)
3
(H
2
O)
3
}[BTC]
3
ꢀ42H
2
O, is also a porous Al–BTC
with a honeycomb topology. Its structure is built up from the
connection of a new type of octahedrally coordinated aluminum
octameric unit through trimesate ligands delimiting one-dimen-
sional large hexagonal channels (16 Å) [13,15,16]. The structure
3 3
H BTC, hydrofluoric acid (HF, 48%), HNO and water were mixed,
then the mixture was transferred to a Teflon-lined autoclave and
heated in an electric oven at 160 °C for 12 h. MIL-100(Cr) was syn-
thesized similar to a reported method [23]. For the preparation of
MIL-100(Cr), chromium (VI) oxide, HF, H BTC, and H O were mixed
3 2
and then the reactant mixture was transferred to a Teflon-lined
autoclave and heated in a conventional electric oven for 4 days at
of MIL-96(Al), with a chemical formula of [Al12O(OH)18(H
2
O)
-O-
3
(
Al (OH) )[BTC] O], contains isolated trinuclear l
2
4
6
ꢀ24H
2
3
bridged aluminum clusters and infinite chains of aluminum octa-
hedra forming a honeycomb lattice with 18-membered rings
[
17]. The structure of MIL-96(Al) has three types of cages. The
pore-opening diameters of these cavities are rather small and
were estimated to be in the range of 2.5–3.5 Å. Previously
Table 1
The reaction conditions of synthesis of M–BTC materials.
[
6,10,11,18], it was suggested that microporosity and structure
of zeolites and zeotype materials are important for the reaction
selectivity. We can assume that structure of Al–BTCs also may
allow to control reaction selectivity toward CA. For analysis of
main factors which can affect the reaction rate and selectivity
for the dominant isomer, we used combination of spectroscopic
and catalytic methods. Note that investigations of Lewis acidity
of MOFs [19–21] and demonstration of correlations between
amount of LAS and catalytic activity of MOFs [1–5,9] are rather
limited or nonsystematic. The aims of this study were to devel-
oped new spectroscopic approaches for investigation of Lewis
acidity of MOFs and to determine correlations between
Reaction conditions
Reaction mixture (mol.%)
MIL-100(Cr)
MIL-100(Fe)
MIL-100(Al)
MIL-110(Al)
MIL-96(Al)
CrO
3
:HF:H
3
BTC:H
2
O
220 °C, 4 days
160 °C, 12 h
210 °C, 2.5 h
210 °C, 2.5 h
210 °C, 2 days
1
.0:1.0:0.67:265
0
3 3 2
Fe :H BTC:HF:HNO :H O
1.0:0.67:2.0:0.6:277
Al(NO
1.0:0.67:255:1.26
3
)
3
ꢀ9H
2
O/Me
3
–BTC/H
2
O/HNO
3
Al(NO
3
)
3
ꢀ9H
2
O/Me
3
–BTC/NaOH/H
2
O
1.0:0.5:2.3:310
Al(NO
3
)
3
ꢀ9H
2
3 2
O/H BTC/H O
1.0:0.67:400

1
16
M.N. Timofeeva et al. / Journal of Catalysis 311 (2014) 114–120
a temperature of 220 °C. After crystallization for a fixed period of
time, the autoclaves were cooled to room temperature and the so-
lid products were recovered by filtration. The solid products were
washed with a water/ethanol mixture (1:1 v/v) and stirred in eth-
3. Results and discussion
3.1. Investigation of Lewis acidity
anol at 60 °C for 5 h to remove the unreacted H
3 3
BTC or Me –BTC.
3.1.1. Analysis of Lewis acidity by IR spectroscopy
After filtration, the purified solids were dried overnight at 100 °C.
The designation of the samples, chemical composition, and textural
data of the M–BTC samples are presented in Table 2.
The Lewis acidity of Al–BTCs with different structures, such as
MIL-110(Al), MIL-110(Al) and MIL-96(Al), was studied by the
adsorption of PhCN as a probe molecule in combination with FT-
IR spectroscopy. PhCN as a soft base is usually used as a probe mol-
ecule for the identification of the LAS of oxides and zeolites [24,25].
Fig. 1A shows the spectra of PhCN adsorbed on Al–BTCs. The band
at 2178 cm was observed in the IR spectra of all samples. This
band can be assigned to the interaction between the non-bonding
electrons of PhCN and electron-deficient Al ions (LAS), which exist
in the structure of Al–BTCs dehydrated at 200 °C. The results show
2.3. Instrumental measurements
ꢁ1
The porous structure of the materials was determined from the
adsorption isotherm of N
400 equipment. The specific surface area (SBET) was calculated
from the adsorption data over the relative pressure range between
.05 and 0.20. The total pore volume (V ) was calculated from the
2
at ꢁ196 °C using Micromeritics ASAP
2
that the intensity (absorbance/
q
) and integral intensity (S
band at 2178 cm depend on the structure of the Al–BTC. Integral
intensity (S ) decreased as follows:
q
) of the
0
R
ꢁ1
amount of nitrogen adsorbed at a relative pressure of 0.99. The
X-ray diffraction patterns were measured on a X-ray diffractome-
q
ter (ThermoARL) with Cu Ka (k = 1.5418 Å) radiation.
MIL ꢁ 100ðAlÞ > MIL ꢁ 110ðAlÞ > MIL ꢁ 96ðAlÞ
To study Lewis surface acidity by benzonitrile (PhCN), Al–BTC
samples were pretreated within the IR cell at 200 °C for 2 h, then
samples were exposed to saturated PhCN vapors at room temper-
ature. The FT-IR spectra of PhCN adsorbed on the catalysts were re-
corded every 10 min up to saturation by PhCN. FT-IR spectra were
recorded on a Shimadzu FTIR-8300S spectrometer in the range of
Therefore, the amount of LAS decreased in this order.
3.1.2. Analysis of Lewis acidity by EPR spectroscopy
Information on the nature and concentration of LAS (for exam-
3þ
ꢁ
1
ꢁ1
ple, AlCUS) on the surface of solids can be obtained with the spin
probe method, which is a powerful tool to investigate coordin-
atively unsaturated sites in zeolites and oxides [26,27]. Using this
method, we tried to determine the concentration of surface LAS
4
00–6000 cm with a resolution of 4 cm
The ESR spectra were measured with an ERS-221 EPR spectrom-
eter working in the X-band ( = 9.3 GHz). The EPR spectra were re-
corded at 20 dB attenuation with a typical microwave power of
mW. All samples for EPR measurements were prepared using
‘break seal’’ techniques. For this, 0.015–0.025 g of Al–BTC was
.
m
ꢀ
ꢁ
3þ
Al
by sample titration with TEMPO radicals. It is well known
CUS
3
‘
[
26–28] that the TEMPO molecule reacts with only one acid site;
therefore, the concentration of acid sites can be calculated by the
maximum amount of radicals adsorbed (until the EPR spectrum
of the TEMPO radical appears in the solution).
The EPR spectra of TEMPO complex adsorbed on MIL-100(Al)
are shown in Fig. 2. An anisotropic EPR signal (A) appeared when
TEMPO adsorbed and interacted with the LAS (Fig. 2). On MIL-
added into a quartz cell and evacuated at 200 °C until an absolute
pressure of 20 mTorr was reached and maintained for 3 h. After
cooling to room temperature under vacuum conditions, 1 ml of
ꢁ
5
ꢁ4
3
.2 ꢂ 10 –3.2 ꢂ 10 M TEMPO solution in toluene was added
to the sample. The EPR spectra were recorded at room tempera-
ture. The concentration of TEMPO was calculated after the calibra-
tion procedure (error ± 10%).
1
00(Al) sample, an EPR spectrum with g = 2.006 was observed.
For TEMPO complexes with Lewis acids, the superfine splitting
constant was equal to 35 G. These values and the shape of the
EPR signal show that the distribution of TEMPO molecules was ran-
dom in MIL-100(Al). The same values of g and the superfine split-
ting constant were obtained for MIL-110(Al) and MIL-96(Al).
The experimental data on the estimation of the amount of LAS
in MIL-100(Al), determined by the EPR method, are given in Table 3.
2.4. Catalytic tests
The isomerization of PO was carried out at 30 °C in a glass reac-
tor equipped with a magnetic stirrer. Dichloroethane was used as
the solvent. Before the reaction, all catalysts were activated at
ꢀ
ꢁ
3
CUS
þ
2
7
00 °C for 4 h in order to remove the adsorbed water. Then, 25 or
5 mmol of PO, 2 ml of dichloroethane, 10 mmol of octane (inter-
According to the experimental data, the amount of LAS Al
0.98 mmol/g for MIL-100(Al) calcined at 200 °C. This value is lower
was
nal standard), and 5 mg of the catalyst were added to the reactor.
At different time intervals, aliquots were taken from the reaction
mixture and analyzed. A mass spectrometer (Shimadzu GCMS
QP-2010 Ultra with a GsBP1-MS column, 30 m ꢂ 0.32 mm, thick-
than that determined by IR spectroscopy using CO (2180–
ꢁ1
2220 cm ) as the probe molecule (about 1.7 ± 0.2 mmol/g) [21].
This difference is probably related to the difference in the size of
the probe molecule, which affects the accessibility of LAS. Of note,
to explain the correlation between the acid–base properties and
catalytic activity of the catalyst, it is preferable if the size of the
probe molecule is comparable with the size of the reagents and
reaction products.
ness 0.25 lm) was used to identify the reaction products. A gas
chromatograph (Agilent 7820) with a flame ionization detector
on an HP-5 capillary column was used to analyze the products
quantitatively.
The results of the EPR investigation of the Lewis acidity of
Al–BTCs with different structures are presented in Table 3. The
amount of Lewis acid sites decreased in the order: MIL-100(Al) >
MIL-110(Al) > MIL-96(Al). These data are in agreement with those
obtained by IR spectroscopy using PhCN as the probe molecule.
TEMPO adsorption resulted in changes in the EPR spectra as well
as in the IR spectra of coordinatively bonded PhCN (band at
2178 cm ). The quite good linear correlation between the EPR
and IR spectroscopy data shown in Fig. 1B indicates that EPR spec-
troscopy can be used to analyze the Lewis acidity on the surfaces of
M–BTCs.
Table 2
Chemical composition and textural data of M–BTC materials.
Al content (wt.%)
Textural data
2
3
3
S
BET (m /g)
V
R
(cm /g)
l
V (cm /g)
MIL-100(Cr)
MIL-100(Fe)
MIL-100(Al)
MIL-110(Al)
MIL-96(Al)
–
–
15.9
17.3
17.6
1462
1296
1486
679
0.77
0.81
0.70
0.58
0.14
0.31
0.34
0.38
0.27
0.13
ꢁ
1
315

M.N. Timofeeva et al. / Journal of Catalysis 311 (2014) 114–120
117
0
0
0
.4
.2
1.0
B
MIL-100(Al)
MIL-110(Al)
A
0
.8
.6
0
MIL-100(Al)
0.4
MIL-110(Al)
MIL-96(Al)
0.2
MIL-96(Al)
.0
2
0.0
0.00
140
2160
2180
2200
0.05
0.10
0.15
0.20
-1
Wavenumber, cm-1
Integral intensity of band at 2178 cm
Fig. 1. (A) IR spectra of PhCN adsorbed on MIL(Al) samples and (B) correlation between the amount of LAS (integral intensity of band at 2178 cmꢁ1) and the amount of TEMPO
adsorbed on MILs.
adsorption values of the probe molecules (TEMPO and PhCN).
The effect of the Al–BTC structure on the adsorption values of the
ge
probe molecules can be estimated from the effective density of
3þ
O.
Al , i.e. the amount of LAS based on the surface area and the
CUS
micropore volume. These values for MIL-100(Al) and MIL-110(Al)
are larger compared to MIL-96(Al), which is in good agreement
with the differences in their structures. Therefore, the experimen-
tal data clearly show that the amount of LAS in these three Al–BTCs
depends on their structure and textural properties.
TEMPO
A
MIL-100(Al)
3.2. Catalytic performance of metal–benzenetricarboxylates
0
.98 mmol/g
Particular interest was given to a study on the catalytic proper-
ties of isostructural M–BTCs [31], such as MIL-100(Al), MIL-
x8 0.05 mmol/g
1
3
00(Fe), and MIL-100(Cr). The empirical formula of MIL-100 is
D-{M O(X)(H O) [BTC] O, where M = Cr, Fe and Al; and
3
2
2
2
}ꢀnH
2
X = OH, F; BTC = benzene-1,3,5-tricarboxylate (BTC). The textural
properties of the MIL-100 samples are shown in Table 2. Note that
the textural properties of the MIL-100(Al, Cr, Fe) materials are not
much different from one another and similar to previously re-
ported results [32].
3
330
3360
3390
3420
3450
Magnetic field, Gauss
ꢁ
5
Fig. 2. EPR spectra of TEMP in toluene (3.2 ꢂ 10 mol/l) and TEMPO complex
According to the experimental data in Fig. 3, the conversion of
PO was 96–98% after 30 min of reaction in the presence of the
MIL-100 samples. The heterogeneous character of the reaction
was confirmed in special experiments for the most active sample,
MIL-100(Al). The MIL-100(Al) sample was filtered off after
adsorbed on MIL-100(Al) (0.98 mmol/g and 0.05 mmol/g).
Table 3
Amount of LAS in Al–BTC materials determined by EPR spectroscopy using TEMPO as
probe molecule.
1
5 min of reaction in dichloroethane at 30 °C, where the conver-
Altotal (mmol/g)
3þ
3þ
Al
(mmol/g)
Al =Altotal (mol/mol)
sion of PO was about 75%. Then, the filtrate without the catalyst
was stirred at 30 °C for 30 min. After removing the catalyst, further
conversion of PO (76%) was negligible, providing evidence of heter-
ogeneous catalysis.
CUS
CUS
MIL-100(Al)
MIL-110(Al)
MIL-96(Al)
5.89
6.40
6.52
0.98
0.62
0.14
0.17
0.10
0.02
One can see from Table 4 that selectivity toward CA also de-
pends on the type of metals used to form the MIL-100 structure;
the selectivity decreased in the following order:
The correlation between the Al–BTC type and the amount of Le-
wis sites could be explained by the difference in the textural prop-
MIL ꢁ 100ðAlÞ > MIL ꢁ 100ðFeÞ > MIL ꢁ 100ðCrÞ
erties, i.e. the specific surface area (SBET) and porosity (V
R
)
At the same time, the yield of trans-carv changed as follows:
(
Table 2). Another explanation may be related to differences in
MIL ꢁ 100ðCrÞ > MIL ꢁ 100ðAlÞ ꢃ MIL ꢁ 100ðFeÞ
the structures of Al–BTCs. Thus, these structural features of the
Al–BTCs are known to affect their adsorption properties; thus,
Rustad et al. [33] estimated the intrinsic acidity of Al(III), Cr(III),
the amount of CO
MIL-96(Al) was 20 mmol/g, 16 mmol/g and 6.5–9.3 mmol/g
29,30], respectively. Based on these data, we also can expect that
the structure, i.e. the size of the cavities and channels, affects the
2
adsorbed on MIL-100(Al), MIL-110(Al) and
and Fe(III)
structure calculations. It was found that @Fe
groups have nearly the same intrinsic acidity, while ꢃCr
3
l -hydroxo functional groups from ab initio electronic
3 3
OH and @Al OH
[
3
OH
groups are significantly more acidic. The order of selectivity

1
18
M.N. Timofeeva et al. / Journal of Catalysis 311 (2014) 114–120
1
00
MIL-100(Al)
MIL-110(Al)
6
4
2
0
0
0
0
7
5
2
5
0
5
0
MIL-96(Al)
0
60
120
180
1,6
2,0
2,4
Xi
2,8
3,2
Time, min
Fig. 3. Kinetic curves of isomerization of PO over MIL-100(Al), MIL-110(Al), and
MIL-96(Al) (reaction conditions: 25 mmol PO in 2 ml dichloroethane, 5 mg catalyst,
i
Fig. 4. The relationship between electronegativity of the metal ions (X ) and yield of
CA in the isomerization of PO over isostructural M–BTCs (the yield of CA was
determined by GLH; reaction conditions: 25 mmol PO in 2 ml dichloroethane, 5 mg
catalyst, 30 °C, 30 min).
3
0 °C).
toward trans-carv is in agreement with the order of intrinsic acid-
ity of Al(III), Fe(III), and Cr(III) (Table 4). It is well known that the
Strong differences were observed in PO isomerization in the
presence of Al–BTCs with different structures (MIL-110(Al), MIL-
110(Al), and MIL-96(Al)). Conversion of PO and selectivity toward
CA decreased in the order MIL-100(Al) > MIL-110(Al) > MIL-96(Al)
(Fig. 3 and Table 4). It is well known that LAS favor the formation
of CA [3,7]; therefore, it is reasonable to suggest that the catalytic
properties of Al–BTCs are determined by Lewis acidity. Fig. 5 shows
the effect of the concentration of LAS determined by EPR and IR
spectroscopy (see Section 3.1) on the reaction rate and selectivity
toward CA. As can be seen from the experimental evidence
(Fig. 5), both the reaction rate and selectivity toward CA are pro-
portional to the total amount of LAS.
electronegativity of ions (X
is a parameter of the electron-withdrawing ability of the metal ion
34,35]. The value of X was calculated from the oxidation state of
i
) is proportional to Lewis acidity, which
[
i
the ion (Z) and the Pauling electronegativity (X
o
) using the follow-
ing equation [36]:
X
i
¼ X
o
ð1 þ 2ZÞ
ð1Þ
The relatively good correlation between the parameter X
i
and the
yield of CA is shown in Fig. 4. The yield of CA decreased steadily
with the parameter X . This interdependency can be explained by
the inductive effect of the metal ions on Lewis acidity in the solids.
Note that X has been used very often to characterize the acidity and
i
We can suggest that distribution of LAS in structure of Al–BTCs
can be also important for activity and selectivity of the reaction.
This effect was revealed from the change of TON calculated using
the following equation:
i
catalytic activity of solids [37,38].
The high yield of CA shows that the MIL-100 samples possess
LAS accessible to reactant molecules. The effect of LAS on the selec-
tivity toward CA was demonstrated by De Vos et al. [1] in the isom-
erization of PO over MIL-100(Fe). Both the conversion and
selectivity toward CA decreased after acid pretreatment of MIL-
TON ¼ moles of PO consumed=moles of LASCUS
ð2Þ
TON decreased in the order: MIL-96(Al) > MIL-110(Al) > MIL-
100(Al). This order may point that surface density of LASCUS in
structure of MIL-96(Al) is higher than that in MIL-110(Al) and
MIL-100(Al). Interestingly, high surface density of LASCUS does not
favor the high selectivity toward CA.
Another explanation for the correlation between the Al–BTC
type and the catalytic properties may be the difference in the
structure of these three Al–BTCs, which can affect the reagents
accessibility to active sites (see Section 1). For optimal catalytic
activity, the active sites should be freely accessible to reagent
1
00(Fe), which was explained by the formation of Brönsted acid
sites in the structure of MIL-100(Fe). Dhakshinamoorthy et al. [5]
investigated the isomerization of PO in the absence of a solvent
over a series of Fe–BTCs, including MIL-100(Fe). About 22% conver-
sion of PO with 45% selectivity toward CA was obtained in the pres-
ence of MIL-100(Fe) at 70 °C for 6 h. MIL-100(Fe) was found to
exhibit about two times greater conversion compared to Fe(BTC).
Note that the catalytic properties of Fe(BTC) are similar to those
of Al(OH)(BTC).
Table 4
Isomerization of PO over M–BTC materials.a
Run
PO conversion (%)
Selectivity (mol.%)
(
CA)
(trans-carv)
(trans-sobr)
Other
1
2
3
4
5
MIL-100(Cr)
MIL-100(Fe)
MIL-100(Al)
MIL-110(Al)
MIL-96(Al)
96
96
98
74
71
51
56
61
59
55
20
16
16
15
14
6
7
7
5
5
23
21
16
21
26
a
Reaction conditions: 25 mmol PO in 2 ml dichloroethane, 5 mg catalyst, 30 °C, 30 min.

M.N. Timofeeva et al. / Journal of Catalysis 311 (2014) 114–120
119
Amount of TEMPO, mmol/g
Amount of TEMPO, mmol/g
0
.0
0.4
0.8
1.2
0,0
0,4
0,8
1,2
2
1
1
0
0
.0
65
MIL-100(Al)
MIL-100(Al)
.5
.0
60
55
50
MIL-110(Al)
MIL-96(Al)
MIL-110(Al)
.5 MIL-96(Al)
.0
0
0
5
10
15
20
5
10
15
20
-1
-1
Integral intensity of band at 2178 cm , u.e.
Integral intensity of band at 2178 cm , u.e.
Fig. 5. Dependence of the reaction rate and selectivity toward CA on the integral intensity of the band at 2178 cmꢁ1 (s) and the amount of TEMPO adsorbed on MILs (j)
reaction conditions: 25 mmol PO in 2 ml dichloroethane, 5 mg catalyst, 30 °C).
(
molecules, preferably via large channels or cavities. The effect of
the structure on selectivity toward CA was observed in the pres-
ence of Al-MSU-SFAU (Si/Al 70) [10] and Ti-BEA zeolites [11]. It
was suggested that the rearrangement of PO to CA occurs within
the micropores of the zeolites. The high selectivity toward CA
was a result of (i) LAS, (ii) the transition state shape selectivity in-
duced by the pore structure of the zeolite and (iii) the impact of the
solvent, which should favor low intraporous concentrations of PO.
Thus, dichloroethane as a non-polar solvent yielded a low intrapor-
ous PO concentration due to the hydrophobic nature of Ti-Beta and
led to high selectivity toward CA. Based on this approach, the high
selectivity, in the presence of mesoporous MIL-100, very likely ar-
ose from the unique structure, i.e. the 29 Å supercages that are tet-
rahedrally connected via ꢃ9 Å windows, which are large enough to
allow the free passage of PO and the reaction products. Three-
dimensional connectivity of the supercages will greatly increase
the ease of molecular transport through the crystals.
Acknowledgments
This work was supported by SB RAS Project V.44.2.12, the Fed-
eral Program ‘‘Scientific and Educational Cadres of Russia’’ (No.
2012-1.5-12-000-1013-002) and the Basic Science Research Pro-
gram through the National Research Foundation of Korea (NRF)
funded by the Korea government (MSIP) (Grant Number:
2013R1A2A2A01007176).
References
[
1] F. Vermoortele, R. Ameloot, L. Alaerts, R. Matthessen, B. Carlier, E.V.R.
Fernandez, J. Gascon, F. Kapteijn, D.E. De Vos, J. Mater. Chem. 22 (2012) 10313.
[2] P. Valvekens, F. Vermoortele, D. De Vos, Catal. Sci. Technol. 3 (2013) 1435.
[
[
3] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Chem. Commun. 48 (2012) 11275.
4] A. Dhakshinamoorthy, M. Alvaro, A. Corma, H. Garcia, Dalton Trans. 40 (2011)
6344.
[5] A. Dhakshinamoorthy, M. Alvaro, H. Chevreau, P. Horcajada, T. Devic, C. Serre,
H. Garcia, Catal. Sci. Technol. 2 (2012) 324.
[
6] W.F. Holderich, J. Roseler, G. Heitmann, A.T. Liebens, Catal. Today 37 (1997)
53.
3
[
[
[
7] J.B. Lewis, G.W. Hedrick, J. Org. Chem. 30 (1965) 4271.
8] A.T. Liebens, C. Mahaim, W.F. Holderich, Stud. Surf. Sci. Catal. 108 (1997) 587.
9] L. Alaerts, E. Seguin, H. Poelman, F. Thibault-Starzyk, P.A. Jacobs, D.E. De Vos,
Chem. Eur. J. 12 (2006) 7353.
4
. Summary
The effect of Lewis acid sites on the catalytic properties of M–
[
[
10] B. Ravindra, Y.T. Nie, S. Jaenicke, G.K. Chuah, Catal. Today 96 (2004) 147.
11] P.J. Kunkeler, J.C. van der Waal, J. Bremmer, B.J. Zuurdeeg, R.S. Downing, H. van
Bekkum, Catal. Lett. 53 (1998) 135.
BTCs, such as MIL-100(Al, Fe, and Cr), MIL-110(Al) and MIL-
6(Al), were studied in the isomerization of PO. The investigation
9
[
[
[
[
[
[
[
12] N.A. Khan, J.S. Lee, J. Jeon, C.-H. Jun, S.H. Jhung, Micropor. Mesopor. Mater. 152
of Lewis acidity was done by EPR and IR spectroscopy using TEMPO
and PhCN as the probe molecules, respectively. Both these meth-
ods showed a decrease in the amount of LAS in the order MIL-
(
2012) 235.
13] M. Haouas, C. Volkringer, T. Loiseau, G. Férey, F. Taulelle, Chem. Mater. 24
(2012) 2462.
14] C. Volkringer, D. Popov, T. Loiseau, G. Férey, M. Burghammer, C. Riekel, M.
Haouas, F. Taulelle, Chem. Mater. 21 (2009) 5695.
15] M. Haouas, C. Volkringer, T. Loiseau, G. Férey, F. Taulelle, Chem. Eur. J. 15
(2009) 3139.
16] C. Volkringer, D. Popov, T. Loiseau, G. Guillou, G. Ferey, M. Haouas, F. Taulelle,
C. Mellot-Draznieks, M. Burghammer, C. Riekel, Nat. Mater. 6 (2007) 760.
17] T. Loiseau, L. Lecroq, C. Volkringer, J. Marrot, G. Ferey, M. Haouas, F. Taulelle, S.
Bourrelly, P.L. Llewellyn, M. Latroche, J. Am. Chem. Soc. 128 (2006) 10223.
18] M.N. Timofeeva, V.N. Panchenko, Z. Hasan, N.A. Khan, M.S. Mel’gunov, A.A.
Abel, M.M. Matrosova, K.P. Volcho, S.H. Jhung, Appl. Catal. A: Gen. 469 (2014)
1
00(Al) > MIL-110(Al) > MIL-96(Al).
It was shown that the rearrangement of PO to CA in the pres-
ence of all materials proceeded in dichloroethane at 30 °C. The cat-
alytic activity and selectivity of the Al–BTCs significantly depend
on the structure of the Al–BTCs and the amount of LAS, and de-
creased as follows: MIL-100(Al) > MIL-110(Al) > MIL-96(Al). The
investigation into the catalytic properties of isostructural M–BTCs
based on the structure of MIL-100 showed that the selectivity
toward CA depends on the type of metal that forms the
MIL-100 structure and decreased in the order: MIL-100(Al) >
MIL-100(Fe) > MIL-100(Cr). This order is in agreement with the
electronegativity of the metal ion. It was suggested that the high
selectivity toward CA in the presence of MIL-100s arises from the
unique structures that allow free passage of PO and the reaction
products into 29 Å supercages via 9 Å windows, which favors
shape selectivity.
427.
[19] H. Leclerc, A. Vimont, J.-C. Lavalley, M. Daturi, A.D. Wiersum, P.L. Llwellyn, P.
Horcajada, G. Ferey, C. Serre, Phys. Chem. Chem. Phys. 13 (2011) 11748.
20] A. Vimont, J.-M.l. Goupil, J.-C. Lavalley, M. Daturi, S. Surble, C. Serre, F.
Millange, G. Ferey, N. Audebrand, J. Am. Chem. Soc. 128 (2006) 3218.
21] C. Volkringer, H. Leclerc, J.-C. Lavalley, T. Loiseau, G. Ferey, M. Daturi, A.
Vimont, J. Phys. Chem. C 116 (2012) 5710.
22] Y.-K. Seo, J.W. Yoon, J.S. Lee, U.-H. Lee, Y.K. Hwang, C.-H. Jun, P. Horcajada, C.
Serre, J.-S. Chang, Micropor. Mesopor. Mater. 157 (2012) 137.
23] T.K. Trung, N.A. Ramsahye, P. Trens, N. Tanchoux, C. Serre, F. Fajula, G. Férey,
Micropor. Mesopor. Mater. 134 (2010) 134.
[
[
[
[

1
20
M.N. Timofeeva et al. / Journal of Catalysis 311 (2014) 114–120
[
24] T. Montanari, M. Bevilacqua, C. Resini, G. Busca, J. Phys. Chem. B 108 (2004)
120.
25] T.S. Glazneva, N.S. Kotsarenko, E.A. Paukshtis, Kinet. Catal. 49 (2008) 859.
26] L.N. Ikryannikova, G.L. Markaryan, A.N. Kharlanov, E.V. Lunina, Appl. Surf. Sci.
[31] S.H. Jhung, N.A. Khan, Z. Hasan, CrystEngComm 14 (2012) 7099.
[32] F. Jeremias, A. Khutia, S.K. Henninger, C. Janiak, J. Mater. Chem. 22 (2012)
10148.
[33] J.R. Rustad, D.A. Dixon, A. Felmy, Geochim. Cosmochim. Acta 64 (10) (2000)
1675.
2
[
[
2
07 (2003) 100.
[
27] E.V. Lunina, G.L. Markaryan, O.O. Parenago, A.V. Fionov, Colloids Surf. A:
Physicochem. Eng. Aspects 72 (1993) 333.
[34] J. Penzien, C. Haesner, A. Jentys, K. Kohler, T.E. Muller, J.A. Lercher, J. Catal. 221
(2004) 302.
[
[
28] D.C. Doetschman, G.D. Thomas, Chem. Phys. 228 (1998) 103.
29] P.L. Llewellyn, S. Bourrelly, C. Serre, A. Vimont, M. Daturi, L. Hamon, G. De
Weireld, J.-S. Chang, D.-Y. Hong, Y.K. Hwang, S.H. Jhung, G. Férey, Langmuir 24
[35] P.C. Stair, J. Am. Chem. Soc. 104 (1982) 4044.
[36] K. Li, D. Xue, J. Phys. Chem. A 110 (2006) 11332.
[37] K.I. Tanaka, A. Ozaki, J. Catal. 8 (1967) 1.
(
14) (2008) 7245.
[38] K. Shimizu, T. Higuchi, E. Takasugi, T. Hatamachi, T. Kodama, A. Satsuma, J.
Mol. Catal. A: Chem. 284 (2008) 89.
[
30] Y. Lin, C. Kong, L. Chen, Chem. Asian J. 8 (2013) 1873.