Effect of iron content on selectivity in isomerization of α-pinene oxide to campholenic aldehyde over Fe-MMM-2 and Fe-VSB-5

Article abstract of DOI:10.1016/j.apcata.2013.10.016

Isomerization of α-pinene oxide to campholenic aldehyde (CA) was investigated over Fe-containing mesoporous mesophase materials (Fe-MMM-2) and microporous nickel phosphate molecular sieves (Fe-VSB-5). Activity and selectivity of reaction towards CA over Fe-containing materials was found to depend on iron content in materials that affects oligomeric state of Fe and amount of Lewis acid sites. In the presence of Fe-VSB-5 selectivity towards CA increased with increase in iron content in structure, while that decreased in the presence of Fe-MMM-2. This phenomenon is related to change in agglomeration of iron species in Fe-MMM-2 structure. The high selectivity towards CA in the presence of Fe-VSB-5 was suggested to arise from unique structure of these materials, which favours shape selectivity.

Full text of DOI:10.1016/j.apcata.2013.10.016

Applied Catalysis A: General 469 (2014) 427–433
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Effect of iron content on selectivity in isomerization of ␣-pinene oxide
to campholenic aldehyde over Fe-MMM-2 and Fe-VSB-5
a,b,∗
a
a
c
c
M.N. Timofeeva
, V.N. Panchenko , Zubair Hasan , Nazmul Abedin Khan ,
a
a
d
c,1
M.S. Mel’gunov , A.A. Abel , M.M. Matrosova , K.P.Volcho , Sung Hwa Jhung
a
Boreskov Institute of Catalysis SB RAS, Prospekt Akad. Lavrentieva 5, 630090 Novosibirsk, Russian Federation
Novosibirsk State Technichal 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:
Received 12 July 2013
Received in revised form 7 October 2013
Accepted 8 October 2013
Available online 17 October 2013
Isomerization of ␣-pinene oxide to campholenic aldehyde (CA) was investigated over Fe-containing
mesoporous mesophase materials (Fe-MMM-2) and microporous nickel phosphate molecular sieves (Fe-
VSB-5). Activity and selectivity of reaction towards CA over Fe-containing materials was found to depend
on iron content in materials that affects oligomeric state of Fe and amount of Lewis acid sites. In the
presence of Fe-VSB-5 selectivity towards CA increased with increase in iron content in structure, while
that decreased in the presence of Fe-MMM-2. This phenomenon is related to change in agglomeration
of iron species in Fe-MMM-2 structure. The high selectivity towards CA in the presence of Fe-VSB-5 was
suggested to arise from unique structure of these materials, which favours shape selectivity.
Keywords:
Isomerization of ␣-pinene oxide
Fe-VSB-5
Fe-containing mesoporous materials
Lewis acidity
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
of PO isomerization, because they are highly valuable ingredients
for the production of flavors. It is well-known that Lewis acid sites
(LAS) favour the formation of CA, while Brønsted acid sites (BAS)
lead to the formation of trans-carv, trans-sobrerol (trans-sobr) and
p-cymene [3,10,11]. Note that the main attention in literatures is
focused on synthesis of CA which is used in the fragrance industry
for its sandalwood scent. ZnCl2 and ZnBr2 in benzene are the most
active homogeneous systems with 85% selectivity towards CA [3].
However, the application of ZnCl2 or ZnBr2 has several disadvan-
tages, such as lack of regeneration, corrosion problems, toxicity and
wastewater pollution.
Many research groups try to find a truly heterogeneous catalyst
system. Unfortunately, offered systems have not been competitive
with the homogeneous system. However, several heterogeneous
systems based on zeolites with good activities have been reported
[10,12,13]. Thus, Hölderich et al. [10] found that the three-
dimensional large pore system (7.4 A˚ ) with supercages of 12 A˚ and
Nowadays, considerable attention is focused on the develop-
ment of new solid Lewis catalysts, which can be applied not only
for works with fundamental scientific interest, but also for many
applications at industrial level [1,2]. The main goal of the design of
solid Lewis acids is the substitution of the traditional homogeneous
acid catalysts, such as ZnCl , AlCl , FeCl etc. [2,3] for solving all the
well-known problems of the homogeneous acids. Zeolites and zeo-
type materials containing transition metal ions are the promising
systems for catalysis. These materials are widely used as catalysts
in alkylation and acylation of aromatic compounds [4–6], processes
of isomerization and cyclyzation of terpenes [7–9].
Isomerization of terpene oxides is one of the important reactions
for synthesis of intermediates for production of drugs, vitamins
and fragrances. ␣-Pinene oxide (PO) is one of the key examples,
which isomerizes rapidly in the presence of acids, thereby forming
many products (Scheme 1). Trans-carveol (trans-carv) and camp-
holenic aldehyde (CA) are industrially the most desired products
2
3
3
large amount of mesopores make US-Y zeolites very suitable for
isomerization of PO. Dealuminated of H-US-Y zeolite by HCl allows
to obtain CA with a good yield. The selectivity towards CA was
◦
about 70% at 25 C, but it could be improved to 80%, when reac-
◦
tion was carried out at -30 C. It was suggested that the active
∗ Corresponding author. Tel.: +7 383 330 7284; fax: +7 383 330 8056.
E-mail addresses: timofeeva@catalysis.ru, mariya-timofeeva@yandex.ru
sites are highly dispersed Lewis acid species (extra-framework alu-
minium) located within the zeolitic framework. Moreover, it was
demonstrated that the performance of H-US-Y depends on the bulk
(
M.N. Timofeeva), sung@knu.ac.kr (S.H. Jhung).
1
Tel: +82 53 950 5341; fax: +82 53 950 6330.
0
926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.10.016

4
28
M.N. Timofeeva et al. / Applied Catalysis A: General 469 (2014) 427–433
existence of LAS formed by Ni ions in VSB-5 was proved by IR
OH
spectroscopy using benzonitrile (PhCN) as a probe molecule. Note
that the insertion of iron ions into the VSB-5 framework leads
to the appearance of new LAS formed by Fe ions. At the same
time Fe-MMM-2 has an ordered hexagonal arrangement of uni-
form mesopores with a diameter of 3.75–3.85 nm and silicate wall
thickness 1.3-1.4 nm [19]. According to the comparative analysis of
nitrogen adsorption, Fe-MMM-2 has no micropores and possesses
high internal and low external surface areas. Moreover, these mate-
rials possess acid sites [20] which can favour the high selectivity
towards CA. We can suggest that Fe-VSB-5 and Fe-MMM-2 materi-
als, like zeolites, possess high surface areas, unique well-defined
pore openings and surface acidity and therefore, have potential
in this reaction. Therefore, in this work, we examined the role of
textural and acid properties in the catalytic performance of these
materials in the rearrangement of PO to CA.
-
H O
2
H+
O
trans-carveol
p-cymene
2 H O
(
trans-carv)
+
H O/H
2
α-pinene oxide
-
2
OH
trans-sobrerol
trans-sobr)
(
OH
O
2. Experimental
campholenic aldehyde
CA)
(
2.1. Materials
Scheme 1. Products obtained in the course of ␣-pinene oxide rearrangement.
␣-Pinene oxide (98.0%) was purchased from Acros Organics.
Octane, FeCl ·4H O, NiCl ·6H O, FeCl ·6H O and Na SiO were
2
2
2
2
3
2
2
3
purchased from Merck.
SiO /Al O ratio. Yield of CA increases with decreasing aluminum
2
2
3
content.
2.2. Synthesis of catalysts
van Bekkum et al. [8] also demonstrated that Ti-BEA zeolite
allows to obtain CA with high selectivity in both the liquid and
vapour phases. The high activity was attributed to the presence
of isolated, well-dispersed titanium sites in zeolite. Furthermore, it
was suggested that structure of the zeolite with pore size compara-
ble with size of reagents and intermediates may improve selectivity
towards CA by shape-selectivity.
Ravindra et al. [7] investigated isomerization of PO over Al-
MSU-SFAU (Si/Al 70), which was synthesized from nanoclustered
zeolite Y seeds as framework precursors. According to this synthe-
sis, Al-MSU-SFAU possesses a mesoporous structure with the walls
having microporosity. The selectivity towards CA was 86% at 54%
conversion of PO. It was suggested that reaction can occur within
microporous channels. However, due to the short length of chan-
nels, reaction products can diffuse away from the active site before
further reaction to other isomers occurs.
Fellenz et al. [14] demonstrated that the selectivity towards
CA was 90% at 10–15% conversion in isomerization of PO over Fe-
MCM-41 in toluene at room temperature for 10 min. However, the
selectivity towards CA was 67–70% at 42–70% conversion. Amount
of Fe in Fe-MCM-41 was 8.4–8.9 wt%. Similar result was demon-
strated in isomerization of PO in dichloroethane over Fe-MCM-41
containing iron 3.5–4.1 wt% [15]. Unfortunately, in both cases effect
of iron content on activity and selectivity of the reaction was not
investigated. Note that iron-containing microporous and meso-
porous materials have been mainly investigated as catalysts of
oxidation, whereas study about their application in isomerization
of PO was rarely reported.
VSB-5 and Fe-VSB-5 materials were hydrothermally synthe-
sized at pH = 7.3 with microwave irradiation according to a reported
procedure in Refs. [16,17]. H PO , FeCl ·4H O and NiCl ·6H O were
3
4
2
2
2
2
used as the sources of phosphorous, iron and nickel, respectively.
The designation of the samples, the reaction conditions of synthe-
sis, chemical composition and textural data of VSB-5 and Fe-VSB-5
samples are shown in Table 1 and Supporting Information.
Fe-MMM-2 were synthesized hydrothermally according to the
procedure which generally included mixing of three solutions
(0.2 M aqueous solution of cetyltrimethylammonium bromide
(CTAB) (pH 1.0, HCl); 0.5 M FeCl3 and 0.12 M aqueous solution of
a silica sol (pH 1.0, HCl)) [16,17]. Na Si O and FeCl ·6H O were
2
2
5
3
2
used as the sources of silicon and iron, respectively. All three solu-
tions were clear and transparent. The solutions were mixed and the
final pH was adjusted to 2.0 using 4 M HCl. The synthetic mixtures
were kept overnight under ambient conditions and then hydrother-
◦
mally treated at 50 C for 24 h. The resulting precipitate was filtered
◦
off, washed with distilled water, dried in air at 120 C for 24 h,
and calcined at 600 C for 5 h. Chemical composition and textural
◦
data of Fe-MMM-2 samples are shown in Table 2 and Supporting
Information.
2.3. Instrumental measurements
The porous structure of the materials was determined from the
◦
adsorption isotherm of N2 at −196 C on ɑ Micromeritics ASAP
2400 equipment. The specific surface area (SBET) was calculated
from adsorption data over the relative pressure range between 0.05
and 0.20. The total pore volume (Vtotal) was calculated from the
amount of nitrogen adsorbed at a relative pressure of 0.99. The
In this work, we examined the role of iron content in the
catalytic performance of new Fe-containing materials, such as
Fe-containing mesoporous mesophase silica materials (Fe-MMM-
2
) and microporous Fe-containing nickel phosphate molecular
X-ray diffraction patterns were measured on ɑn X-ray diffractome-
sieves (Fe-VSB-5). According to Refs. [16,17], the structure of
VSB-5 and Fe-VSB-5 is based on octahedral NiO6 units linked
by a tetrahedral HPO4 and PO , forming one-dimensional 24-
ter (ThermoARL) with Cu-K␣ (ꢀ = 1.5418 A˚ ) radiation. The chemical
analyses were done by means of an inductively coupled plasma-
atomic emission spectrometry (ICP-AES).
4
membered ring channel structures with unit cell formula of
To study Brønsted and Lewis surface acidity by pyridine adsorp-
tion, samples were pretreated within the IR cell. Fe-VSB-5 samples
were pretreated by heating for 1 h in air and for 1 h under vac-
Ni20[(OH)12((H O) )[(HPO ) (PO4) ]·12H O. Gao et al. [18] ana-
2
6
4
8
4
2
lyzed acidic properties of VSB-5 by method of desorption
◦
temperature of ammonia (NH -TPD) and found that VSB-5 pos-
uum at 200 C, while Fe-MMM-2 samples were pretreated by
3
◦
◦
sessed 263 ␮mol/g of weak acid sites (a peak at 200 C). The
heating for 1 h in air and for 1 h under vacuum at 450 C. The

M.N. Timofeeva et al. / Applied Catalysis A: General 469 (2014) 427–433
429
Table 1
The reaction conditions of synthesis, chemical composition and textural data of VSB-5 and Fe-VSB-5 samples.
Preparationgel(atomic%) Chemical composition
Textural datab
x a
Fe
Fe+Ni+P
Weight content (wt%)
Fe
Fe+Ni+P
2
3
3
Atomic ratio (%)
SBET (m /g)
Vꢁ (cm /g)
V␮ (cm /g)
Fe
Ni
P
VSB-5
0
0.09
0.18
0
5
10
0
3.1
6.5
48.5
43.5
39.8
12.8
12.8
12.0
0
4.58
9.92
284
202
203
0.24
0.13
0.16
0.13
0.08
0.09
3
6
.1%Fe-VSB-5
.5%Fe-VSB-5
a
Molar composition of reaction mixture is xFeCl2:0.63H3PO4:1.0NiCl2:3.0NH3:100H20.
SBET—specific surface area; Vꢁ—total pore volume; V␮—micropore volume.
b
Table 2
Textural data and chemical composition of Fe-MMM-2.
2
3
2
Fe (wt%)
SBET (m /g)
Vmeso (cm /g)
Sext (m /g)
D (nm)
d10 (nm)
w (nm)
Si-MMM-2
0
981
962
970
995
942
0.45
0.44
0.45
0.45
0.43
20
25
36
34
35
3.0
3.0
3.0
3.0
3.0
3.75
3.85
3.84
3.84
3.85
1.3
1.4
1.4
1.4
1.4
1
1
3
5
.0% Fe-MMM-2
.7% Fe-MMM-2
.9% Fe-MMM-2
.9% Fe-MMM-2
1.0
1.7
3.9
5.9
SBET—specific surface area; Vmeso—mesopore volume; Sext—external surface area of mesopore; d10—mesopore interlayer spacing; D—diameter of pore; w—pore wall thickness.
ꢀ
ꢁ
1
2
√
√
D = 2/ 3d10 Vmeso/1/ꢂ + Vmeso
,
w = 2/ 3d10 − D.
samples were exposed to saturated pyridine vapours at room tem-
perature for 10 min and were heated at 150 C for 15 min. Then
pyridine was desorbed for 30 min under vacuum at 150 C. The
GsBP1-MS 30 m × 0.32 mm, thickness 0.25 ␮m) was used for iden-
tifying the reaction products. A gas chromatograph (Agilent 7820)
with a flame ionization detector on capillary column HP-5 was used
to analyze products quantitatively.
◦
◦
strength of BAS of materials was characterized by the proton affin-
ity values (PA) according to Ref. [21]. The amount of BAS and
LAS was estimated from the intensity of the band of the stretch-
−
1
3. Results and discussion
ing vibration of pyridinium ions with a maximum at 1540 cm
−1
(
BAS) and 1450 cm
(LAS) and using extinction coefficients of
Here, we investigated effect of Fe content on the nature of
acid sites and catalytic performance of Fe-VSB-5 and Fe-MMM-
εLAS = 1.67 ± 0.1 and ε
= 2.22 ± 0.1 cm/␮mol [22]. The strength of
BAS
BAS of samples was characterized by the proton affinity values (PA)
2
, because variation of iron state in Fe-containing materials can
[
23]. For the analysis of the Lewis surface acidity by PhCN adsorp-
change amount of LAS, the dispersion of the Fe-active phase, and
accessibility of Fe ions for adsorption of reactants. The main atten-
tion was focused on distribution of products in isomerization of
PO, which strongly depends on nature of acid sites. We also tried
to reveal effect of textural properties of Fe-containing materials on
activity and selectivity of this reaction.
tion the VSB-5 and Fe-VSB-5 samples were exposed to saturated
PhCN vapours at room temperature. FT-IR spectra of the adsorbed
PhCN were recorded every 10 min up to saturation by PhCN. FT-IR
spectra were recorded on a Shimadzu FTIR-8300S spectrometer in
−
1
−1
the range of 400–6000 cm with a resolution of 4 cm .
2.4. Catalytic test
3.1. Catalytic performance of Fe-VSB-5 materials
◦
The isomerization of PO was carried out at 30 C in a glass reac-
tor equipped with magnetic stirrer. Dichloroethane was used as a
solvent. Before reaction all catalysts were activated at 150 C for
VSB-5 and Fe-VSB-5 were prepared in weakly basic conditions
◦
at pH 7.5. The textural data of VSB-5 and Fe-VSB-5 materials are
shown in Table 1. These data point that Fe-VSB-5 are microporous
materials with high specific area. The crystallinity, morphology and
textural properties do not change substantially after the incor-
poration of iron into framework (Table 1 and Fig. S1 (Supporting
information)).
4
2
5
h in order to remove adsorbed water. Then 25 or 75 mmol of PO,
ml of dichloroethane, 10 mmol of octane (internal standard) and
mg of catalyst were added into reactor. At different time inter-
vals aliquots was taken from reaction mixture and analyzed. A
mass-spectrometer (Shimadzu GCMS QP-2010 Ultra with column
Table 3
Isomerization of PO over Fe-containing materials ɑ.
Run
PO conversion (%)
Selectivity (%mol)
(
CA)
(trans-Carv)
(trans-Sobr)
Other
1
2
3
4
5
6
7
8
VSB-5
89
92
55
62
67(64)
14
58
53(55)
51
46
11
14
16(15)
61
16
14(15)
5
5
5(7)
11
5
5(6)
29
19
3.1%Fe-VSB-5
6.5%Fe-VSB-5
Si-MMM-2
1.0%Fe-MMM-2
1.7%Fe-MMM-2
3.9%Fe-MMM-2
5.9%Fe-MMM-2
95(96)b
15
b
b
b
12(14)b
14
97
21
93(95)b
92
b
b
b
28(24)b
28
16
14
5
5
97
34
ɑ
◦
◦
Experimental condition: 0.25 mmol PO in 2 ml dichloroethane, 5 mg catalyst, 30 C, 30 min.
b
Catalyst was filtered off after 30 min of reaction and the filtrate was stirred at 30 C for 30 min.

4
30
M.N. Timofeeva et al. / Applied Catalysis A: General 469 (2014) 427–433
A
100
B
70
80
Selectivity
6.5%Fe-VSB-5
60
50
40
6
0
VSB-5
Conversion
40
20
0
0
2
4
6
8
0
60
120
180
Tiime, min
Fe content in Fe-VSB-5, wt.%
◦
Fig. 1. (A) Kinetic curves of isomerization of PO over VSB-5 and 6.5%Fe-VSB-5 (reaction conditions: 0.75 mmol PO in 2 ml dichloroethane, 5 mg catalyst, 30 C); (B) effect of
iron content in Fe-VSB-5 on conversion of PO and selectivity towards CA (reaction time was 30 min).
The results of the catalytic performance of VSB-5 and Fe-VSB-
samples are presented in Table 3. Kinetic curves are shown in
Early it has been shown [24] that PhCN can be used for anal-
−
1
5
ysis of LAS of VSB-5 and Fe-VSB-5. Band at 2282 cm observed
in a spectrum of PhCN adsorbed on VSB-5 can be assigned to
the interaction between the nonbonding electrons of PhCN and
electron-deficient Ni ions (LAS) (Fig. 2A). At the same time in a
Fig. 1A. Experimental data clearly show that VSB-5 proved to be
◦
very active in isomerization of PO at 30 C in dichloroethane. CA
was the main product with 55% selectivity at 89% conversion of
.25 mmol PO for 30 min. Moreover, trans-carv and trans-sobr were
also detected in amount around 11% and 5%, respectively. The
obtained results indicate that VSB-5 possesses both LAS and BAS
−
spectrum of 6.5%Fe-VSB-5 we can reveal two bands at 2282 cm
1
0
−
1
and 2268 cm
(Fig. 2A). A new band can be attributed to the
interaction between the nonbonding electrons of PhCN and the
(
Scheme 1). After the insertion of Fe into VSB-5 network, conver-
electron-deficient Fe ions. We estimated the integral intensities (Sq)
−
1
−1
sion of PO and the isomer selectivity towards CA increase to 95%
and 67%, respectively. Reaction did not proceed after separation of
catalyst that was confirmed in special experiments. A 6.5%Fe-VSB-5
sample was filtered off after 30 min of reaction in dichloroethane at
of bands at 2282 cm and 2268 cm . The results show that the
increasing iron contents in Fe-VSB-5 leads to an increase in inte-
−
1
gral intensity of band at 2268 cm , which can be considered as an
indicator of the increase in the amount of LAS formed by Fe ions
◦
◦
−1
3
0 C, and then the filtrate was stirred at 30 C for 30 min (Table 3,
(Fig. 2B). At the same time integral intensity of band at 2282 cm
run 3). No conversion of PO was observed after catalyst removing,
providing evidence of heterogeneous catalysis.
decreases with increasing of Fe content in Fe-VSB-5 that can point
to the decrease in the amount of LAS formed by Ni ions. Therefore,
according to experimental evidence isomorphous substitution of a
nickel ion by iron ions leads to the change in nature of LAS. The
larger amount of Fe ions, the lower amount of LAS formed by Ni
ions and larger amount of LAS formed by Fe ions.
It is well-known that LAS favour formation of CA [3]. It is reason-
able to suggest that catalytic properties of Fe-VSB-5 are determined
by Lewis acidity (Fig. 1B and 2B), because selectivity towards CA and
As one can see from the experimental evidence (Fig. 1B), the cat-
alytic activity of Fe-VSB-5 samples and the isomer selectivity with
regard to CA can be adjusted by the Fe content. The larger amount
of iron in Fe-VSB-5, the higher yield of CA. We can assume that this
result is related with the change of nature of acid sites. Noteworthy,
no iron oligomeric species were observed in Fe-VSB-5 samples with
1
–6.5 wt% iron content (Fig. S3 (Supporting information)) [24].
A15
(Ni)
B
160
2282 (Ni)
120
(Fe)
1
0
5
0
8
0
0
0
VSB-5
4
2268 (Fe)
6.5%Fe-VSB-5
2
250
2280
2310
0
2
4
6
Wavenumber, cm-1
Fe content in Fe-VSB-5, wt.%
−
1
Fig. 2. (A) IR spectra of PhCN adsorbed on VSB-5 and 6.5%Fe-VSB-5; and (B) correlation between iron content in Fe-VSB-5 and integral intensities of bands at 2268 cm and
−
1
2
282 cm
.

M.N. Timofeeva et al. / Applied Catalysis A: General 469 (2014) 427–433
431
60
40
20
0
1
1
5
0
5
0
0
2
4
6
Fig. 3. 6.5%Fe-VSB-5 recycling in isomerization of PO (amount of reactants was cor-
Fe content in Fe-MMM-2, wt. %
rected based on reaction conditions: 0.75 mmol PO in 2 ml dichloroethane, 5 mg
◦
catalyst, 30 C, 30 min).
Fig. 4. (A) Effect of iron content in Fe-MMM-2 on yield of CA and amount of LAS
(
yield of CA was determined by GLH); (reaction conditions: 0.25 mmol PO in 2 ml
◦
dichloroethane, 5 mg catalyst, 30 C, 30 min).
integral intensities (Sq) of the band at 2268 cm 1 rise with increas-
ing of Fe content in Fe-VSB-5. Note that we do not exclude the effect
of Ni active sites on activity and selectivity. However, the insertion
of Fe into VSB-5 framework stronger affects selectivity towards CA
in comparison with Ni ions.
−
The larger amount of LAS in 6.5%Fe-VSB-5 in comparison with
VSB-5 was also confirmed by IR spectroscopy using pyridine as
probe molecules (Table 4). BAS were also identified in samples. The
amount of BAS is about 1 ␮mol/g that is nearly twice less than LAS
in VSB-5. At the same time 6.5%Fe-VSB-5 possesses about 6 ␮mol/g
of BAS that is nearly two times less than that of LAS. We can assume
that the acid sites arise from the equilibrium between bridging
OH-groups, paired centres of Lewis sites and P-OH groups. Prob-
ably, the increasing of LAS amount arises from the different atomic
radius of Ni(II) and Fe(II–III) that can lead to the appearance of
defect electron-deficient sites (LAS). The difference in amount of
BAS between VSB-5 and Fe-VSB-5 materials affects difference in
selectivities towards trans-carv (Tables 3 and 4). The low selectiv-
ity towards trans-carv (11%) in the presence of VSB-5 is related to
the low amount of BAS.
Importantly, 6.5%Fe-VSB-5 can be used repeatedly without sig-
nificant loss of catalytic activity during at least four catalytic cycles
(
Fig. 3). The used 6.5%Fe-VSB-5 can be separated from the reaction
Fig. 5. Isomerization of PO over Fe-containing materials (reaction condition:
0.25 mmol PO in 2 ml dichloroethane, 5 mg catalyst, 30 C, 30 min).
mixture by simple filtration, washed by dichloroethane, dried in air
and activated at 150 C for 4 h.
◦
◦
3
.2. Catalytic performance of Fe-MMM-2 materials
in Table 2, the wall thickness of Fe-MMM-2 synthesized at pH 2.0
is slightly higher (1.4 nm) compared to the pure siliceous MMM-2
(1.3 nm). This can be explained by the incorporation of larger Fe
ions into the silica framework of MMM-2 to replace smaller ions
Fe-MMM-2 samples were prepared in acidic conditions at pH
.0. The textural data of samples are shown in Table 2, Figs. S3 and
2
4+
S5 (Supporting information). As one can see from the data shown
Si . The synthesized Fe-MMM-2 materials have relatively thick
Table 4
Brønsted and Lewis acidity of Fe-VSB-5 and Fe-MMM-2 materials determined by FTIR spectroscopy using pyridine as a probe molecule.
Run
Brønsted acid sites
NLAS/NBAS (mol/mol)
NBASa (␮mol/g)
PAb (kJ/mol)
1
2
3
4
5
6
7
8
VSB-5
1.0
3.2
6.0
n.d.
0.5
2.8
6.2
5.9
1213
1180
1152
1390 c
1185
1175
1180
1188
1.8
1.9
2.1
-
11.4
3.7
2.8
2.5
3.1%Fe-VSB-5
6.5%Fe-VSB-5
Si-MMM-2
1% Fe-MMM-2
1.7% Fe-MMM-2
3.9% Fe-MMM-2
5.9% Fe-MMM-2
a
NBAS and NLAS—amount of BAS and LAS, correspondingly.
PA—the proton affinity.
b
c
1
390 kJ/mol corresponds to the PA of surface OH-groups of Si-MMM-2 (IR spectroscopy using CO as probe molecule) [23].

4
32
M.N. Timofeeva et al. / Applied Catalysis A: General 469 (2014) 427–433
Fe-VSB-5
7
A100 3
B
3
6
7
60
5
1
5
2
2
1
80
6
8
Fe-MMM-2
4
0
6
0
1 - VSB-5
1
2
3
4
- VSB-5
2
3
- 3.1%Fe-VSB-5
- 6.5%Fe-VSB-5
- 3.1%Fe-VSB-5
- 6.5%Fe-VSB-5
- Si-MMM-2
40
4 - Si-MMM-2
5
6
7
8
- 1.0%Fe-MMM-2
- 1.7%Fe-MMM-2
- 3.9%Fe-MMM-2
- 5.9%Fe-MMM-2
20
5 - 1.0%Fe-MMM-2
6 - 1.7%Fe-MMM-2
7 - 3.9%Fe-MMM-2
20
4
8
- 5.9%Fe-MMM-2
0
0
1
150 1200 1250 1300 1350 1400
0
5
10
15
20
Amount of LAS, μmol/g
PA, kJ/mol
Fig. 6. (A) Correlation between conversion of PO and strength of BAS for Fe-MMM-2 and Fe-VSB-5, (B) Correlation between yield of CA and amount of LAS for Fe-MMM-2
◦
and Fe-VSB-5 (reaction condition: 0.25 mmol PO in 2 ml dichloroethane, 5 mg catalyst, 30 C, 30 min).
Table 5
by the difference in surface acid properties. We can assume that
because of the basic conditions of synthesis, 4.1%Fe-MCM-41 has
lower amount of BAS in comparison with 3.9%Fe-MMM-2. The low
amount of BAS leads to the increase in selectivity towards CA.
Isomerization of PO over Fe-containing mesoporous silica materials.
Conversion
of PO, (%)
Selectivity of
CA (%mol)
TONc
(mol/mol)
1
3
4
.7%Fe-MMM-2a
78
84
70
56
52
70
385
190
29
.9%Fe-MMM-2a
3.3. Comparison of catalytic properties
.1%Fe-MCM-41b [15]
a
Experimental condition: 0.75 mmol PO in 2 ml dichloroethane, 5 mg catalyst,
A comparison of the catalytic properties of the Fe-containing
◦
3
0
C, 30 min.
b
◦
zeotype materials is presented in Fig. 5. All tested materials showed
high activity and selectivity towards CA. In the presence of all sam-
ples conversion of PO is about 89–96% (Table 3). As one can see from
Fig. 6A strength of BAS affects the conversion of PO, i.e. activity of
samples. The higher the strength of BAS, the higher the reaction
rate. Fig. 6B demonstrates correlation between amount of LAS and
0
.75 mmol PO in 5 ml dichloroethane, 25 mg catalyst, 40 C, 3 h.
c
TON = ꢃPO/Fe.
walls, which is one of the factors responsible for their hydrothermal
stability.
Catalytic properties of iron-containing mesoporous silica mate-
rials (Fe-MMM-2) are somewhat different from that of Fe-VSB-5.
The main results of catalytic tests of Fe-MMM-2 samples are pre-
sented in Table 3. Similar to the reactions with Fe-VSB-5 catalyst,
reaction did not proceed either after removing Fe-MMM-2 catalyst
yield of CA. The yield of CA increases with increasing iron content
of microporous Fe-VSB-5 catalysts with the pore diameter of 11 A˚
[
25]. At the same time in the presence of mesoporous Fe-MMM-2
catalysts with the pore diameter of 30 A˚ yield of CA decreases with
increasing of iron content in Fe-MMM-2. Note that yield of CA in the
presence of Fe-VSB-5 is larger than that in the presence of Fe-MMM-
(Table 3, run 6). The main product, with Fe-MMM-2, was CA with
selectivity around 46–58%.
2
. The difference in yields of CA between microporous Fe-VSB-5 and
The iron content in Fe-MMM-2 also affects the reaction rate
and distribution of product (Table 3 and Fig. 4). Note that reaction
proceeds in the presence of Si-MMM-2. Conversion of PO is 15%
for 30 min (Table 3, run 4). However trans-carv is the main product
with selectivity of 61% that is likely related with the acidic medium
of synthesis (pH 2.0). The increase in iron content in Fe-MMM-2
leads to the decrease in selectivity towards CA. This phenomenon
cannot be explained by the change in amount of LAS, because
according to IR spectroscopy using pyridine as a probe molecule
mesoporous Fe-MMM-2 materials is difficult to explain by only dif-
ference in amount of LAS and BAS (Table 4). Thus, amount of LAS
and BAS is the same in 6.5%Fe-VSB-5 and 5.9%Fe-MMM-2, while
selectivity towards CA in the presence of 5.9%Fe-MMM-2 is lower
(46%) in comparison with 6.5%Fe-VSB-5 (67%). We can assume that
this difference is related to the difference of pore diameters. The
unique structure of Fe-VSB-5 with the pore diameter of 11 A˚ [25]
comparable with size of reactants and intermediates may affect the
improved selectivity towards CA by a shape-selectivity.
(Table 4), amount of LAS increases with increasing of iron content
in Fe-MMM-2. We can assume that this phenomenon is related to
change in agglomeration of iron species in an Fe-MMM-2 structure.
In our earlier work [19], we demonstrated that the agglomeration
of iron and formation of oligomerized iron species was observed
in Fe-MMM-2 when Fe content was more than 3.7 wt% (Fig. S4
4. Conclusions
Catalytic properties of Fe-containing mesoporous Fe-MMM-2
materials and microporous nickel phosphate molecular sieves Fe-
VSB-5 were studied in isomerization of PO. It was demonstrated
that these materials favour the rearrangement of PO to CA with
50–67% selectivity in dichloroethane under mild conditions.
Reaction rate and selectivity towards CA over Fe-containing
microporous Fe-VSB-5 and mesoporous Fe-MMM-2 materials
depend on iron content, which affects the oligomeric state of Fe
and amount of LAS. The high yield of CA is attributed to the pres-
ence of isolated iron sites in matrix of Fe-VSB-5 and Fe-MMM-2.
The yield of CA rises with increasing of iron content in Fe-VSB-5
due to the increase in amount of LAS. However, on the contrary, in
spite of increasing of amount of LAS the yield of CA decreases with
(
Supporting nformation)).
Of particular interest was a comparison of the catalytic activities
of Fe-containing mesoporous silica materials, i.e. activities of Fe-
MMM-2 prepared under acidic conditions (pH 2.0) and Fe-MCM-41
prepared under basic conditions (pH ∼ 11) [15]. As one can see from
Table 5, activity of 4.1%Fe-MCM-41 is lower in comparison with
3
4
.9%Fe-MMM-2. Conversion of PO over 4.1%Fe-MCM-41 was 70% at
◦
0 C for 3 h, while conversion of PO over 3.9%Fe-MMM-2 was 84%
◦
at 30 C for 30 min. At the same time, selectivity towards CA in the
presence of 4.1%Fe-MCM-41 (70%) is higher than that in the pres-
ence of 3.9%Fe-MMM-2 (56%). This phenomenon can be explained

M.N. Timofeeva et al. / Applied Catalysis A: General 469 (2014) 427–433
433
increasing of iron content in Fe-MMM-2 that is related to the for-
mation of oligomeric iron oxides species. The higher yield of CA in
the presence of Fe-VSB-5, in comparison to Fe-MMM-2, is proposed
to be due to transition-state shape selectivity induced by structure
of VSB-5.
[9] B. Jarry, F. Launay, J.P. Nogier, J.L. Bonardet, Stud. Surf. Sci. Catal. 165 (2007)
91–794.
10] W.F. Holderich, J. Roseler, G. Heitmann, A.T. Liebens, Catal. Today 37 (1997)
53–366.
[11] J.B. Lewis, G.W. Hedrick, J. Org. Chem. 30 (1965) 4271–4275.
7
[
3
[
[
[
12] A. Dhakshinamoorthy, M. Opanasenko, J. Cejka, H. Garcia, Adv. Synth. Catal. 355
2013) 247–268.
13] A.T. Liebens, C. Mahaim, W.F. Holderich, Stud. Surf. Sci. Catal. 108 (1997)
587–594.
(
Acknowledgements
14] N.A. Fellenz, J.F. Bengoa, S.G. Marchetti, A. Gervasini, Appl. Catal. A 435–436
(
2012) 187–196.
This work was supported by SB RAS project V.44.2.12, the Fed-
eral Program “Scientific and Educational Cadres of Russia” (No.
[15] J.V. Coelho, A.L.P. de Meireles, K.A. da S. Rocha, M.C. Pereirac, L.C.A. Oliveira, E.V.
Gusevskay, Appl. Catal. A 443–444 (2012) 125–132.
[
16] S.H. Jhung, J.W. Yoon, J.-S. Hwang, A.K. Cheetham, J.-S. Chang, Chem. Mater. 17
2005) 4455–4460.
2
012-1.5-12-000-1013-002) and Basic Science Research Program
(
through the National Research Foundation of Korea (NRF) funded
by the Ministry of Education (Grant No. 2012004528).
[
17] S.H. Jhung, J.-S. Chang, Y.K. Hwang, J.-M. Greneche, G. Ferey, A.K. Cheetham, J.
Phys. Chem. B 109 (2005) 845–850.
18] D. Gao, G. Gao, Catal. Commun. 8 (2007) 681–685.
19] M.N. Timofeeva, M.S. Mel’gunov, O.A. Kholdeeva, M.E. Malyshev, A.N. Shmakov,
V.B. Fenelonov, Appl. Catal. B 75 (3–4) (2007) 290–297.
20] M.N. Timofeeva, M.E. Malyshev, V.N. Panchenko, A.N. Shmakov, A.G. Potapov,
M.S. Mel’gunov, Appl. Catal. B 95 (2010) 110–119.
21] A.A. Davydov, Molecular spectroscopy of oxide catalyst surfaces, John Wiley,
England, 2003, pp. 1–668.
[
[
References
[
[
[
[
[
1] C.B. Dartt, M.E. Davis, Catal. Today 19 (1994) 151–186.
2] A. Corma, H. Garcia, Chem. Rev. 103 (2003) 4307–4365.
3] J. Kaminska, M.A. Schwegler, A.J. Hoefnagel, H. van Bekkum, Recl. Trav. Chim.
Pays-Bas. 111 (1992) 432–437.
[
[
22] G. Kosova, Stud. Surf. Sci. Catal. 158 (1998) 59–66.
23] E.A.
Paukshtis,
Infrakrasnaya
Spektroskopiya
v
Geterogennom
[
[
[
4] H. van Bekkum, H.W. Kuowenhoven, Stud. Surf. Sci. Catal. 157 (2005) 311–336.
5] P.-S. E. Dai, Catal. Today 26 (1995) 3–11.
6] L. Kurfirtova, Y.-K. Seo, Y.K. Hwang, J.-S. Chang, J. Cejka, Catal. Today 179 (2012)
Kislotno–Osnovnom Katalize (Infrared Spectroscopy in Heterogeneous
Acid–Base Catalysis), Nauka, Novosibirsk, 1992.
24] M.N. Timofeeva, A.Yu. Z. Hasan, V.N. Orlov, Yu. Panchenko, A. Chesalov, I.P.
Soshnikov, S.H. Jhung, Appl. Catal. B 107 (1–2) (2011) 197–204.
25] N. Guillou, Q. Gao, P. Forester, J.-S. Chang, M. Nogues, S.-E. Park, G. Ferey, A.K.
Cheetham, Angew. Chem. Int. Ed. 40 (2001) 2831–2834.
[
[
8
5–90.
[
[
7] D.B. Ravindra, Y.T. Nie, S. Jaenicke, G.K. Chuah, Catal. Today 96 (2004) 147–153.
8] P.J. Kunkeler, J.C. van der Waal, J. Bremmer, B.J. Zuurdeeg, R.S. Downing, H. van
Bekkum, Catal. Lett. 53 (1998) 135–138.