Epoxidation of styrene with H2O2 catalyzed by alanine-salicylaldehyde schiff base chromium (III) complexes immobilized on mesoporous materials

Article abstract of DOI:10.1007/s10562-010-0301-8

Alanine-salicylaldehyde Schiff base chromium (III) complex was immobilized on mesoporous silica gel (SiO₂), MCM-41 and SBA-15. The resulting immobilized complexes were promising catalysts for the epoxidation of styrene with 30% hydrogen peroxide, and they all showed much higher catalytic performance than their homogeneous analogue. Simultaneously, the catalytic performance of immobilized complexes was found to be closely related to the textual and surface properties of the supports used. The complex immobilized on methyl-containing MCM-41 exhibited the highest catalytic performance. Under optimal reaction conditions, the highest conversion of styrene reached 80.2% with 77.0% selectivity to epoxide. In addition, the catalytic performance remained stable after six times of recycling. Graphical Abstract: A series of immobilized alanine-salicylaldehyde Schiff base chromium (III) complexes were promising catalysts for the epoxidation of styrene with 30% hydrogen peroxide. Moreover, the different textural and surface properties of supports significantly influenced their catalytic performance.

Full text of DOI:10.1007/s10562-010-0301-8

Catal Lett (2010) 136:96–105
DOI 10.1007/s10562-010-0301-8
Epoxidation of Styrene with H2O2 Catalyzed by Alanine–
Salicylaldehyde Schiff Base Chromium (III) Complexes
Immobilized on Mesoporous Materials
•
•
•
Xiaoli Wang Gongde Wu Wei Wei
Yuhan Sun
Received: 9 August 2009 / Accepted: 5 February 2010 / Published online: 24 February 2010
Ó Springer Science+Business Media, LLC 2010
Abstract Alanine–salicylaldehyde Schiff base chromium
(III) complex was immobilized on mesoporous silica gel
(SiO2), MCM-41 and SBA-15. The resulting immobilized
complexes were promising catalysts for the epoxidation of
styrene with 30% hydrogen peroxide, and they all showed
much higher catalytic performance than their homogeneous
analogue. Simultaneously, the catalytic performance of
immobilized complexes was found to be closely related to
the textual and surface properties of the supports used. The
complex immobilized on methyl-containing MCM-41
exhibited the highest catalytic performance. Under optimal
reaction conditions, the highest conversion of styrene
reached 80.2% with 77.0% selectivity to epoxide. In
addition, the catalytic performance remained stable after
six times of recycling.
1 Introduction
Catalytic epoxidation of alkenes to epoxides is a versatile
reaction from a synthetic point of view. Epoxides can be
easily converted into polyethers, diols and aminoalcohols,
which are of great importance in bulk chemistry, fine
chemistry and pharmaceutical industry [1]. Though a large
number of methods have been developed for the conver-
sion of alkenes to epoxide, only a few useful catalytic
systems with H2O2 as oxidant have been reported [2–8].
Therefore, there is an urgent need to develop effective
catalysts for the epoxidation with H2O2.
Transition metal Schiff base complexes bear resem-
blance to enzymatic catalysts and are eye-catching since
they provide advantages due to their relatively easy syn-
thesis and versatile coordination structures. Among the
various transition metal complexes reported [9–17], the
amino acid Schiff base complexes have been proven to be
excellent catalysts for catalytic epoxidation [13, 14].
However, such homogeneous complexes suffer from the
drawbacks of poor catalyst recovery and product separation
[18]. Thus, in the past decade, much effort has been paid
to immobilize the homogeneous catalysts onto various
supports, especially the siliceous mesoporous materials
[19–21]. The method used frequently was to immobilize
the active homogeneous complexes on supports through
surface-bound linkers [22]. In order to tailor the immobi-
lized catalysts for the desired performance, several groups
investigated the role of linkers, supports and immobiliza-
tion methods to improve the catalytic performance. Lau
et al. reported that the dispersion effect of supports played
an important role in the catalytic performance of immo-
bilized complexes [23]. Corma and coworkers demon-
strated that the linker length and immobilized methods
influenced the catalytic performance of immobilized
Keywords Chromium complexes ꢀ Immobilized ꢀ
Mesoporous materials ꢀ Styrene ꢀ Epoxidation
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-010-0301-8) contains supplementary
material, which is available to authorized users.
X. Wang ꢀ G. Wu (&)
Department of Environment and Technology, Nanjing Institute
of Technology, 211167 Nanjing, People’s Republic of China
e-mail: wugongde@njit.edu.cn
X. Wang
e-mail: wangxiaoli212@126.com
W. Wei ꢀ Y. Sun
State Key Laboratory of Coal Conversion, Institute of Coal
Chemistry, Chinese Academy of Sciences, 030001 Taiyuan,
People’s Republic of China
123

Epoxidation of Styrene with H2O2
97
complexes [24, 25]. We have also reported on the impor-
tance of the linker flexibility and the coordination abilities
of terminal functional groups in linkers [26]. However, to
the best of our knowledge, there are few reports which
systemically deal with the effect of the textural and surface
properties of supports such as their channels, pore sizes,
surface area, surface polarity and hydrophobicity on the
catalytic performance of immobilized complexes.
Here we report the epoxidation of styrene with 30%
H2O2 catalyzed by a series of alanine–salicylaldehyde
Schiff base chromium (III) complexes immobilized on
three typical mesoporous supports. Moreover, the roles of
the textural and surface properties of supports in the cata-
lytic performance of immobilized complexes are discussed
in detail.
[30, 31]. The ligand Sal–Ala was prepared firstly by con-
densation of salicylaldehyde and b-alanine with a salicyl-
aldehyde/alanine at molar ratio of 1. Typically, alanine
(0.02 mol) was suspended in 60 mL of deionized water.
Then salicylaldehyde (0.02 mol) dissolved in 60 mL
absolute ethanol was added dropwise while vigorous stir-
ring the suspension. After refluxing for 3 h, the resulting
mixture was cooled, and produced precipitates upon addi-
tion of ether. To obtain crystals of the desired Schiff base
ligand, the solid was further recrystallized from ethanol.
Then, the ligand Sal–Ala (1.93 g) was dissolved in 50 mL
absolute ethanol. The resultant solution was alkalized with
the appropriate amount of NaOH to adjust the pH close to
7. A solution of CrCl3ꢀ6H2O (2.66 g) in 50 mL ethanol was
added dropwise while stirring vigorously the above mixture
so that the bright yellow solution turned green. After
refluxing for 6 h, the solid was filtered, washed with cold
ethanol, dried under vacuum (at 60 °C and 10-2 Pa for 2 h)
and recrystallized in absolute ethanol.
2 Experimental
2.1 Catalyst Preparation
2.1.4 Preparation of the Heterogenised Chromium (III)
Complexes
2.1.1 Preparation of Aminopropyl Functionalized
Supports (APTES-Supports)
Typically, the solution of the as-prepared homogeneous
complex (1.5 g) in 50 mL of dry toluene was added to the
suspension of freshly dried APTES-support or APTES-
support(CH3) (1 g) in 50 mL of dry toluene. The mixture
was vigorously stirred under reflux for 10 h. Then the
resulting suspension was cooled and filtered through a
Buchner funnel supplied with a fine-porous filter paper.
The collected powder was washed overnight in a Soxhlet
extractor using equivalent alcohol and acetonitrile as sol-
vent to remove the homogeneous complexes adsorbed on
the surface of the support, and then the solid was dried
in air at 80 °C for 10 h. By choosing the different
supports: APTES–SiO2, APTES–MCM-41, APTES–SBA-15,
APTES–SiO2(CH3), APTES–MCM-41(CH3) and APTES–
SBA-15(CH3), the corresponding immobilized materials
were obtained and denoted as Cr(Sal–Ala)–SiO2, Cr(Sal–
Ala)–MCM-41, Cr(Sal–Ala)–SBA-15, Cr(Sal–Ala)–SiO2
(CH3), Cr(Sal–Ala)–MCM-41(CH3) and Cr(Sal–Ala)–
SBA-15(CH3), respectively.
SiO2 was purchased from Shanghai Chemical Reagent
Company. The pure siliceous MCM-41 and SBA-15 were
synthesized as previously described [27, 28]. Aminopropyl
functionalized supports (APTES–SiO2, APTES–MCM-41
and APTES–SBA-15) were prepared as in the literature
[25].
2.1.2 Preparation of Methyl-Containing APTES-Supports
(APTES-Support(CH3))
The methyl-containing APTES-supports were prepared by
following the procedures in our previous report [29]. Typ-
ically, the as-prepared APTES-support (3 g) was degassed
at 50 °C under 10-2 Pa for 2 h. Then, an excess of dim-
ethyldiethoxysilane (0.03 mol) in dry toluene (100 mL)
was added, and the suspension was stirred at reflux tem-
perature under nitrogen flow for 24 h. The resulting solid
was filtered, Soxhlet-extracted with dichloromethane for
24 h and dried in air (at 60 °C under 10-2 Pa for 3 h). Then,
the corresponding methyl-containing supports were
obtained and denoted as APTES–SiO2(CH3), APTES–
MCM-41(CH3) and APTES–SBA-15(CH3).
2.2 Characterization
The contents of carbon and nitrogen in the samples were
determined using a Vario EL analyzer. The chromium
contents were measured by inductively coupled plasma
(ICP) emission spectroscopy (PerkinElmer ICP OPTIMA-
3000). Powder X-ray diffraction (XRD) experiments were
performed at room temperature on a Rigaku D Max III
VC instrument with Ni filtered Cu Ka radiation
2.1.3 Preparation of the Homogeneous Alanine–
Salicylaldehyde Schiff base Chromium (III) Complex
(Cr(Sal–Ala))
The synthetic procedures for the homogeneous Cr(Sal–Ala)
complex were similar as those reported in references
˚
(k = 1.5404 A) at 40 kV and 30 mA, in the 2h range of
123

98
X. Wang et al.
1–9° at a scan rate of 1°/min. The specific surface area and
average pore diameter were measured by N2 adsorption–
desorption using a Micromeritics ASAP-2000 instrument
(Norcross, GA). The samples were outgassed at 80 °C and
10-4 Pa overnight and then the adsorption–desorption
isotherms were measured at liquid nitrogen temperature.
The surface area was calculated by the BET method and
the average pore diameter was calculated by the BJH
method from the desorption isotherm. Water adsorption
isotherms were measured gravimetrically using a BEL-
SORP 18 instrument (Bel Japan Inc.) at room temperature.
Before the adsorption experiments, samples were pre-
treated at 110 °C for 6 h under vacuum. FTIR spectra of
samples were recorded in KBr disks at room temperature
on a Shimadzu (model 8201 PC) spectrophotometer. UV–
Vis absorption spectra were recorded on a Shimadzu
(model 2501 PC) spectrophotometer (for solid sample,
optical grade BaSO4 was used as reference). Solid state
NMR spectra were recorded on a Bruker MSL 300 NMR
spectrometer with resonance frequencies of 75.5 and
59.6 MHz for 13C and 29Si, respectively. Chemical shifts
(ppm) are reported relative to the external standard of
tetramethylsilane. The hydrophobicity of methyl-contain-
ing immobilized complexes was typically tested by mea-
suring the contact angle (h) of a water droplet placed on the
sample surface, using the formula: h ¼ tanꢁ1ð2h=WÞ;
where h is the height of the water droplet and W is the
width of the droplet touching the samples surface [32].
3 Results and Discussion
3.1 NMR Spectra of Organo-Modified Supports
Typical NMR spectra of APTES–MCM-41 and APTES–
MCM-41(CH3) are shown in Figs. 1, 2. In the 29Si NMR
spectrum of APTES–MCM-41 (see Fig. 1), the two reso-
nances at about -109 and -99 ppm can be attributed to
29Si nuclei having four Si–O–Si linkages (Q4) and 29Si
nuclei having three Si–O–Si linkages and one –OH (Q3),
respectively. The two resonances at about -55 and
-65 ppm are assigned to RSi(OSi)(OH)2 and RSi(OSi)3,
respectively [33]. The presence of those characteristic
resonances indicates the successful preparation of APTES–
MCM-41. For APTES–MCM-41(CH3), the characteristic
resonances originating from APTES–MCM-41 are all
present and simultaneously, two new resonances at about
-6.5 and -13 ppm, corresponding to the attached Si(Me)2
moiety [34], are also observed. This suggests that the
methyl groups have been successfully introduced.
(b)
(a)
2.3 Catalytic Test
The oxidation of styrene was carried out in a 100 mL
Teflon-lined and magnetically stirred autoclave. In a typi-
cal experiment, a mixture of 0.25 g heterogenised catalyst
or 0.025 g homogeneous complex and 0.025 mol styrene
was stirred for about 10 min, and then the corresponding
amount of 30% H2O2 was introduced. The autoclave was
adjusted to the desired temperature within 5 min. After the
reaction run for 2–16 h, the products were filtered out from
the catalyst and then analyzed using a gas chromatograph
with a capillary 30 m HP-5 column and an FID detector.
50
0
-50
-100
-150
-200
ppm
Fig. 1 29Si CP MAS NMR spectra of (a) APTES–MCM-41, (b)
APTES–MCM-41(CH3)
2.4 Adsorption Test
(b)
(a)
Typically, 0.05 g immobilized material was mixed sepa-
rately with the solution A (8.0 9 10-8 mol styrene in
15 mL n-hexane) and B (8.0 9 10-8 mol epoxide in
15 mL n-hexane). After stirring at 0 °C for 4 h, the mix-
tures were filtered and the UV–Vis absorptions of the fil-
trate solutions were recorded. The concentrations were
determined on the basis of UV–Vis spectra and compared
with standard solutions.
100
50
0
-50
-100
ppm
Fig. 2 13C CP MAS NMR spectra of (a) APTES–MCM-41, (b)
APTES–MCM-41(CH3)
123

Epoxidation of Styrene with H2O2
99
In the 13C NMR spectrum of APTES–MCM-41 (see
Fig. 2), three peaks with almost equal intensity are
observed at about 9.5, 22 and 43 ppm, respectively, which
can be attributed to the three different C-atoms of the
aminopropyl group [33]. This also indicates the successful
anchoring of aminopropyl groups on the surface of the
support. In the spectrum of APTES–MCM-41(CH3),
besides the characteristic resonances of APTES–MCM-41,
four new resonances at about -1.7, 16.8, 35 and 57.3 ppm
originating from methylated species are also found [35],
further confirming the successful introduction of methyl
groups.
preparation of the homogeneous complex. In the FTIR
spectra of the immobilized complexes, apart from the
bands in the overlapping regions of the silica backbone, the
other bands of Cr(Sal–Ala) complex are all clearly
observed despite some marginal shifts in the position of
bands due to the immobilization. Thus the coordination
structure of the homogeneous complex has survived the
immobilization.
The UV–Vis spectrum of the homogeneous complex
displays a peak at about 382 nm typical of the metal–ligand
band and a broad peak at about 600 nm associated with a
d–d transition (see Fig. 4). This is similar to related metal
(Sal–Ala) compounds described in the literatures [37], also
implying the successful preparation of the Cr(Sal–Ala)
complex. The UV–Vis spectra of all immobilized materials
are quite similar to the spectrum of the homogeneous
complex, further confirming that the immobilized com-
plexes as depicted in Scheme 1 were all successfully
prepared.
3.2 Coordination Structures of Immobilized Materials
The results of the elemental analysis reveal that the
obtained values of Cr(Sal–Ala) complex are quite compa-
rable with the calculated ones (see Table 1), indicating that
the as-prepared homogeneous complex maintains the
expected elemental composition. The N/Cr molar ratios of
all immobilized complexes are well consistent with the
expected value of 2, which provides a supporting evidence
for the successful immobilization of the homogeneous
complex on the supports. Furthermore, a much higher
carbon content in Cr(Sal–Ala)-support(CH3) than in
Cr(Sal–Ala)-support is also indicative of the successful
introduction of methyl groups.
3.3 Textural Properties of Immobilized Complexes
The powder XRD patterns of the complexes immobilized
on MCM-41 and on SBA-15 show one single peak around
2h angles of 2–3° and 1–2°, respectively (see Fig. 5). This
peak can be attributed to the (100) plane of the hexagonal
unit cell, indicating that the structural ordering of parent
MCM-41 and SBA-15 channels remains intact after the
immobilization of the Cr(Sal–Ala) complex. However, the
higher angle peaks associated with (110) and (200)
reflections present in their parent samples no longer appear,
and an overall decrease in the intensity of (100) reflection
is also observed. These changes are due to the decrease in
The FTIR spectrum of the homogeneous Cr(Sal–Ala)
complex shows the characteristic bands at 1621 (m(C=N)),
À
Á
À
Á
ꢁÞ
ꢁÞ
1570 masðCOO ; 1400 msðCOO
(m(Cr–O) (phenolic)), 612 (m(Cr–O) (carboxylic)) and
463 cm-1 (m(Cr–N)) (see Fig. 3). This agrees well with the
published literature data [31, 36], indicating the successful
; 1332 (m(ph–O)), 661
Table 1 Composition and textural parameters of samples
Samples
Composition
C (wt%)
Textural parameters
N (wt%)
Cr (wt%)
N/Cra
SBET (m2/g)
dp (nm)
Cr(Sal–Ala)
43.05 (43.09)b
5.00 (5.03)b
18.6 (18.67)b
0.99 (1.00)b
–
–
327
1,050
712
235
650
586
210
517
505
485
–
SiO2
–
–
–
9.73
2.76
7.61
8.35
2.27
5.60
7.52
2.12
2.10
5.36
MCM-41
–
–
–
–
SBA-15
–
–
–
–
Cr(Sal–Ala)–SiO2
Cr(Sal–Ala)–MCM-41
Cr(Sal–Ala)–SBA-15
Cr(Sal–Ala)–SiO2(CH3)
Cr(Sal–Ala)–MCM-41(CH3)
Cr(Sal–Ala)–MCM-41(CH3)c
Cr(Sal–Ala)–SBA-15(CH3)
5.08
5.38
5.10
6.20
7.26
6.81
6.05
0.91
0.97
0.92
0.86
0.87
0.82
0.81
1.7
1.8
1.7
1.6
1.6
1.5
1.5
1.99 (2.00)b
2.00 (2.00)b
2.01 (2.00)b
2.00 (2.00)b
2.02 (2.00)b
2.03 (2.00)b
2.01 (2.00)b
a
Molar ratio
b
Value in parenthesis corresponds to the calculated results
The recovered catalyst after reusing six times
c
123

100
X. Wang et al.
CH2
Cr
CH2
C O
CH2
CH2
C O
N
O
N
O
Cl
O
Cl
O
Cr
N
(c)
N
(b)
CH3
CH3
O
Si
Si
Si
O O O
O O
O
OH O
(a)
Cr(Sal-Ala)-SiO2(CH3)
Cr(Sal-Ala)-MCM-41(CH3)
Cr(Sal-Ala)-SBA-15(CH3)
Cr(Sal-Ala)-SiO2
Cr(Sal-Ala)-MCM-41
Cr(Sal-Ala)-SBA-15
2000
1500
1000
500
Scheme 1 Structural representation of the as-prepared immobilized
chromium complexes
Wavenumber (cm-1)
Fig. 3 FTIR spectra of (a) Cr(Sal–Ala), (b) Cr(Sal–Ala)–MCM-41,
(c) Cr(Sal–Ala)–MCM-41(CH3)
MCM-41
Cr(Sal-Ala)-MCM-41
Cr(Sal-Ala)-MCM-41(CH3)
SBA-15
Cr(Sal-Ala)-SBA-15
(c)
Cr(Sal-Ala)-SBA-15(CH3)
(b)
0
1
2
3
4
5
6
7
8
2 Theta (deg.)
Fig. 5 XRD patterns of samples
(a)
of the IUPAC classification, suggesting that the mesosized
structures of the supports remain intact after the immobi-
lization of Cr(Sal–Ala) complex [41]. However, the
immobilized materials all exhibit a decrease in surface area
and pore size in comparison with their parent samples (see
Table 1), which may be due to the attaching of organic
moieties on the walls of the supports [42].
200
300
400
500
600
700
800
Wavelength (nm)
Fig. 4 UV–vis spectra of (a) Cr(Sal–Ala), (b) Cr(Sal–Ala)–MCM-
41, (c) Cr(Sal–Ala)–MCM-41(CH3)
local order as previously mentioned by Lim et al. [38].
Moreover, the (100) reflections of the immobilized com-
plexes show clear shifts to higher 2h values compared to
those of the corresponding parent samples, which can be
attributed to the contraction of the unit cells of the grafted
samples originating from the immobilization of the bulky
organometallic groups inside the channels of MCM-41 and
SBA-15 [39, 40].
3.4 Surface Properties of Immobilized Complexes
The density of Cr(Sal–Ala) complexes per surface area in
the two complexes immobilized on SiO2 is two times
higher than that on MCM-41 and on SBA-15 (see Table 2).
This indicates a low dispersion of the active moieties in the
complexes immobilized on SiO2. Moreover, with the
introduction of non-polar methyl groups, the amounts of
polar residual silanol groups on the support surface of
The N2 adsorption–desorption isotherms of the three
immobilized complexes all correspond to the type IV class
123

Epoxidation of Styrene with H2O2
101
Table 2 Surface group density of samples
Samples
Density of Cr(Sal–Ala)
(910-6 mol/m2)
Residual silanol groupsa
per gram sample (910-3 mol/g)
Water adsorbedb
per gram sample (g/g)
Contact
angle (°)
SiO2
–
2.2
1.0
0.6
5.0
2.1
0.7
4.8
2.0
0.6
0.10
0.09
0.07
0.12
0.11
0.07
0.13
0.11
0.08
–
–
Cr(Sal–Ala)–SiO2
Cr(Sal–Ala)–SiO2(CH3)
MCM-41
1.39
1.47
–
92
–
Cr(Sal–Ala)–MCM-41
Cr(Sal–Ala)–MCM-41(CH3)
SBA-15
0.53
0.60
–
–
95
–
Cr(Sal–Ala)–SBA-15
Cr(Sal–Ala)–SBA-15(CH3)
a
0.56
0.59
–
97
Calculated from FTIR spectroscopy
b
The saturation adsorption capacities of water for samples were obtained from the adsorption isotherms of water
immobilized complexes decrease. So the surface polarity of
immobilized complexes most likely decreases as well.
Simultaneously, the amount of water adsorbed in the
immobilized complexes decreases as the samples become
richer in methyl groups, indicating that the hydrophobic
character of the surface increases with the loading of
methyl groups. However, a further study on the hydro-
phobicity of the three methyl-containing samples reveals
that their contact angels for water are all less than 100°
suggesting that they are only moderately hydrophobic.
Thus, the introduction of methyl groups mainly leads to a
modification in the surface polarity of the immobilized
complexes.
as their homogeneous analogue, and then exhibit signifi-
cantly enhanced catalytic performance. On the other hand,
the introduction of the support changes the local environ-
ment of active sites, and then influences the local con-
centration of the reactant and product around active sites,
which in turn may contribute to the enhanced catalytic
performance of immobilized complexes [26].
The differences in the catalytic performance of immo-
bilized complexes with different supports were also
investigated in detail, and the results are listed in Table 3.
It is found that the two complexes immobilized on SiO2
exhibit the worst catalytic performance. Simultaneously,
the two complexes immobilized on SBA-15 exhibit rela-
tively higher styrene conversion and much lower epoxide
selectivity than the two complexes immobilized on MCM-
41. Based on the yield of epoxide, the catalytic perfor-
mance of the complexes immobilized on different supports
decreases in the following order: on MCM-41 [ on SBA-
15 ꢂ on SiO2.
3.5 Catalytic Performance
The so-obtained immobilized complexes and their homo-
geneous analogue were tested in the epoxidation of styrene
with 30% H2O2. Six products, epoxide, benzaldehyde,
hyacintin, hypnone, benzoic acid and benzyl benzoate were
detected under the conditions used. The results in Table 3
reveal that only a small amount of epoxide (\5%) is pro-
duced in the absence of catalyst. With Cr(Sal–Ala) as
catalyst, the styrene conversion and epoxide selectivity are
slightly increased to 30.2 and 16.9%, respectively. How-
ever, when the reaction is performed over immobilized
Schiff base complexes, the catalytic performance is found
to be significantly improved, and the optimal epoxide yield
reaches 61.8% over Cr(Sal–Ala)–MCM-41(CH3). Such
obviously enhanced catalytic performance, on the one
hand, can be attributed to the dispersion effect of supports
[23, 43–47]. Compared to the active sites in homogeneous
systems, the dispersed catalytic sites in immobilized
complexes exhibit more difficulty in forming the inactive
l-oxo dimers or other polymeric species. Thus, the
immobilized complexes do not undergo rapid degradation
A comparison of the immobilized complexes in terms of
dispersion of their active complexes indicates that the poor
catalytic performance of the complexes immobilized on
SiO2 can be attributed to the low dispersion of their active
moieties as depicted in Table 2. The excessive active
complexes per surface area in the complexes immobilized
on SiO2 also have probably formed the inactive polymers
[23]. Considering that the complexes immobilized on
MCM-41 and SBA-15 exhibit similar dispersion of active
complexes, surface polarity and hydrophobicity, the sig-
nificant differences in their catalytic performance can be
reasonably related to the different channels and pore sizes
of their supports. The different textural properties of the
two supports might show different degrees of influence on
the catalytic performance of the immobilized complexes
through modifying the local amounts of reactant (styrene)
and objective product (epoxide) around active sites [48].
123

102
X. Wang et al.
Table 3 Catalytic performance of catalysts in the oxidation of styrene
Samples
Con.a (%) Sel.b (%)
H2O2
efficiencyc (%)
Epoxide Benzaldehyde Hyacinthin Hypnone Benzoic acid Benzyl benzoate
Blank
12.2
30.2
43.6
72.6
73.0
51.5
2.7
16.9
35.1
64.3
61.5
44.3
77.0
67.2
92.5
75.7
50.3
22.6
23.3
42.8
16.5
19.1
1.2
2.4
5.6
4.3
4.0
4.5
2.2
3.7
0
3.6
2.8
4.7
4.5
5.5
5.0
3.5
5.6
0
7.8
20.1
32.8
55.2
58.5
44.5
74.6
75.0
Cr(Sal–Ala)
1.5
2.1
2.0
2.5
2.1
0.6
2.0
0.7
2.2
2.3
3.2
1.3
0.2
2.4
Cr(Sal–Ala)–SiO2
Cr(Sal–Ala)–MCM-41
Cr(Sal–Ala)–SBA-15
Cr(Sal–Ala)–SiO2(CH3)
Cr(Sal–Ala)–MCM-41(CH3) 80.2
Cr(Sal–Ala)–SBA-15(CH3) 85.8
Reaction conditions: Styrene 25 mmol, H2O2 100 mmol, heterogenised catalyst 0.25 g or homogeneous catalyst 0.025 g, CH3CN 10 mL, 0 °C,
4 h
a
Conversion = (moles of styrene reacted/moles of styrene in the feed) 9 100
b
Selectivity = (moles of styrene converted to the products/moles of styrene reacted) 9 100
c
Efficiency of H2O2 for styrene oxidation = (moles of H2O2 converted to the products/moles of H2O2 consumed) 9 100
the modification in the surface properties of immobilized
complexes originating from the methyl groups which fur-
ther influence the access of styrene and epoxide molecules
to the active sites [49]. Thus, we investigated in detail the
differences in the adsorption capacities of immobilized
complexes with or without methyl groups for styrene and
epoxide. Upon the introduction of methyl groups, the
adsorption capacities of immobilized complexes for sty-
rene only show a moderate decrease, while a sharp
decrease in their adsorption capacities for epoxide is
observed (see Table 4). Such differences in the adsorption
capacities may be due to the different polarity of styrene
and epoxide. As mentioned above, with the introduction of
methyl groups, the surface polarity of immobilized com-
plexes significantly decreases, so their adsorption capaci-
ties for epoxide with higher polarity tend to sharply
decrease as well. This also indicates that the amount of
epoxide accessible to active sites of methyl-containing
immobilized complexes significantly decreases. Thus, the
probability of the freshly produced epoxide being overox-
idized decreases, which could afford the significantly
improved epoxide selectivity in the presence of methyl-
containing catalysts. Simultaneously, the moderate
decrease in the adsorption capacities of methyl-containing
immobilized complexes for styrene indicates that the sty-
rene molecules had more difficulty to access the active sites
of methyl-containing catalysts. In this respect, the methyl-
containing catalysts should exhibit relatively lower styrene
conversion, which is contrary to the factual results. Thus,
the enhanced styrene conversion is mainly attributed to the
improved H2O2 efficiency, and consequently to the
decreased amount of residual silanol groups in methyl-
containing immobilized complexes. We have discussed the
correlation between the H2O2 efficiency, the conversion of
Table 4 Adsorption of styrene and epoxide with different samples
Samples
Adsorption
of styrene
(910-6 mol/g)
Adsorption
of epoxide
(910-6 mol/g)
Cr(Sal–Ala)–SiO2
1.34
1.28
1.36
1.22
1.20
1.25
1.55
1.51
1.60
1.26
1.25
1.31
Cr(Sal–Ala)–MCM-41
Cr(Sal–Ala)–SBA-15
Cr(Sal–Ala)–SiO2(CH3)
Cr(Sal–Ala)–MCM-41(CH3)
Cr(Sal–Ala)–SBA-15(CH3)
Here, a series of adsorption tests were designed to inves-
tigate the local amounts of styrene and epoxide molecules
around active sites by UV–Vis spectroscopy. The results
reveal that the two complexes immobilized on SBA-15
both display higher adsorption capacities for styrene and
epoxide than their corresponding complexes immobilized
on MCM-41 (see Table 4). This may be related to the
crossed channels and big pore sizes of SBA-15, which
facilitate the adsorption of styrene and epoxide. The results
of the adsorption tests indicate that more styrene molecules
can access the active sites in the complexes immobilized
on SBA-15, so the immobilized complexes exhibit much
higher styrene conversion. However, the much higher
adsorption capacities for epoxide of the complexes
immobilized on SBA-15 simultaneously increase the
probability of this object product being overoxidized, thus
a relatively lower selectivity to epoxide is obtained.
Furthermore, the results in Table 3 also display that the
introduction of hydrophobic methyl groups lead to an
improvement in the H2O2 efficiency and the catalytic
performance of immobilized complexes independent of the
supports used. Such an improvement might be attributed to
123

Epoxidation of Styrene with H2O2
103
reactant and the amounts of residual silanol groups in our
previous report [29].
present investigation, epoxide is found to be the major
product, indicating that the hydroperoxide O–O bond is
mainly cleaved heterolytically. Accordingly, the coordin-
atively unsaturated chromium (III) mainly seems to acti-
vate H2O2 via heterolytic cleavage of the peroxide bond
resulting in high-valent oxochromium (V) intermediate,
and then reacts with styrene to yield epoxide.
In order to obtain the best catalytic results, the effect of
condition parameters on the catalytic performance of
Cr(Sal–Ala)–MCM-41(CH3), which is the most active
catalyst in the epoxidation of styrene among all immobi-
lized materials, was investigated. It is found that the dosage
of oxidant, the reaction time and temperature all signifi-
cantly influence the catalytic performance of this immo-
bilized complex (see Fig. 6). When the reaction ran for 4 h
at 0 °C with a H2O2/styrene molar ratio of 4, the best
styrene conversion is of 80.2% with 77.0% selectivity to
epoxide, and the main byproduct is benzaldehyde with
16.5% selectivity.
In addition, a further study on the stability of Cr(Sal–
Ala)–MCM-41(CH3) was also carried out. After the first
catalytic run, this immobilized catalyst was separated from
the reaction solution, washed several times with solvent to
remove any physisorbed molecules, dried and then reused
in another five catalytic runs. The results in Fig. 7 reveal
that the catalyst is still active up to six times of reusage,
however, with the further increase in the reusage, the cat-
alytic performance decreases significantly. The elemental
composition and textural properties of the recovered cata-
lyst after six times were also investigated, and the results
are listed in Table 1. The recovered catalyst reveals almost
the same elemental composition as the fresh catalyst,
indicating that the anchoring between chromium
The mechanistic probe for the epoxidation of olefins
with peroxide by oxometal complexes is attracting con-
tinuous interest, and the prevailing accepted views are the
enzyme mimetic mechanism (the heterolytic bond cleavage
of peroxide) and the free-radical mechanism (the homolytic
bond cleavage of peroxide). Generally, the heterolytic bond
cleavage of peroxide tends to form epoxide [50, 51]. In the
90
80
70
60
50
40
30
90
80
70
60
50
40
30
90
80
70
60
50
90
80
70
60
50
0
4
8
12
16
1
2
3
4
5
n(H2O2)/n(styrene)
Reaction time (h)
90
80
70
60
50
90
80
70
60
50
-20
0
20
40
60
Reaction temperature (oC)
Fig. 6 Effect of condition parameters on the catalytic performance of Cr(Sal–Ala)–MCM-41(CH3). Reaction conditions: Styrene 25 mmol,
catalyst 0.25 g, CH3CN 10 mL
123

104
X. Wang et al.
90
85
80
75
70
65
60
properties. The best styrene conversion reached 80.2% with
77.0% selectivity to epoxide over Cr(Sal–Ala)–MCM-
41(CH3), and the main byproduct was benzaldehyde
(Selectivity 16.5%). In addition, this immobilized material
could be reused for several runs.
Conversion of styrene
Selectivity to epoxide
Acknowledgements The authors acknowledge the financial sup-
ports from National Science Technology Foundation of China (No.
2006BAC2A08) and NJIT Scientific Research Foundation for Intro-
duce Talents (KXJ 08033 and KXJ 08034).
References
1
2
3
4
5
6
Reused times
1. Mandelli D, van Vliet MCA, Arnold U, Sheldon RA, Schuchardt
U (2001) J Mol Catal A 168:165
Fig. 7 Reusability of Cr(Sal-Ala)-MCM-41(CH3). Reaction condi-
tions: Styrene 25 mmol, H2O2 100 mmol, catalyst 0.25 g, CH3CN
10 mL, 0 °C, 4 h
2. Ishii Y, Yamawaki K, Ura T, Yamada H, Yoshida T, Ogawa M
(1988) J Org Chem 53:3587
3. De Vos DE, Sels BF, Reynaers M, Subba Rao YV, Jacobs PA
(1998) Tetrahedron Lett 39:3221
´
4. Coperet C, Adolfsson H, Sharpless KB (1997) Chem Commun
1565
complexes and support is strong enough to afford six cat-
alytic runs, and the chromium leaching is negligible.
Therefore, the decreasing catalytic performance over the
recovered catalyst with increasing catalytic runs may be
reasonably attributed to the increasing residual adsorbed
molecules which prevent the reactants from accessing the
active sites. This is confirmed by the decreased surface
areas and pore sizes of the recovered catalyst in compari-
son with the fresh catalyst due to the attaching of residual
adsorbed molecules to the walls of support. The obtained
immobilized complexes here also encounter the problem of
recycling in more catalytic runs for immobilized catalysts
prepared by coordinatively bonding active sites on the
support, and similar phenomena have been reported pre-
viously [25]. Thus, further increasing the stability of such
promising oxidative catalyst will be the focus of our fol-
lowing work, which we think is of significance for highly
effective heterogeneous catalyst design.
´
5. Rudler H, Gregorio JR, Denise B, Bregeault J-M, Deloffre A
(1998) J Mol Catal A 133:255
6. van Vliet MCA, Arends IWCE, Sheldon RA (1999) Chem
Commun 821
7. Herrmann WA, Fischer RW, Marz DW (1991) Angew Chem Int
Ed 30:1638
8. Herrmann WA, Eder SJ, Kiprof P (1991) J Organomet Chem
412:407
9. Tsuji Y, Ohta T, Ido T, Minbu H, Watanabe Y (1984) J Orga-
nomet Chem 270:333
10. Barak G, Dakka J, Sasson Y (1988) J Org Chem 53:3553
11. Marko IE, Gautier A, Chelle-Regnaut I, Giles PR, Tsukazaki M,
Urch CJ, Brown SM (1998) J Org Chem 63:7576
´
´
12. Fernandez I, Pedro JR, Rosello AL, Ruiz R, Castro I, Ottenwa-
elder X, Journaux Y (2001) Eur J Org Chem 7:1235
13. Wang RM, Hao CJ, He YF, Wang YP, Xia CG (2002) Polym
Adv Technol 13:6
14. Zhao J, Han J, Zhang YC (2005) J Mol Catal A 231:129
15. Abdi SHR, Kureshy RI, Khan NH, Jasra RV (2004) Catal Surv
Jpn 8:187
16. Halligudi SB, Devassy BM, Kala Raj NK, Degaonkar MP,
Gopinathan S (2000) React Kinet Catal Lett 71:289
17. Yang DH, Gao L, Zhao WJ (2008) Catal Lett 126:84
18. Bahramian B, Mirkhani V, Moghadam M, Tangestaninejad S
(2006) Catal Commun 7:289
4 Conclusions
19. Abrantes M, Sakthivel A, Roma˜o CC, Ku¨hn FE (2006) J Orga-
nomet Chem 691:3137
20. Baleiza˜o C, Gigante B, Garcia H, Corma A (2003) J Catal
215:199
21. Luts T, Suprun W, Hofmannb D, Klepel O, Papp H (2006) J Mol
Catal A 261:16
22. Zhou XG, Yu XQ, Huang JS, Li SG, Li LS, Che CM (1999)
Chem Commun 1789
23. Lau S-H, Caps V, Yeung K-W, Wong K-Y, Tsang SC (1999)
Microporous Mesoporous Mater 32:279
24. Corma A, Garcia H, Moussaif A, Sabater MJ, Zhiber R,
Redouane A (2002) Chem Commun 1058
25. Baleiza˜o C, Gigante B, Sabater MJ, Garcia H, Corma A (2002)
Appl Catal A 228:279
A series of alanine–salicylaldehyde Schiff base chromium
(III) complexes immobilized on three typical mesoporous
supports (SiO2, MCM-41 and SBA-15) were prepared, and
they exhibited promising catalytic performance towards the
epoxidation of styrene with 30% H2O2. The textural
properties of the supports such as the surface area, channels
and pore sizes were found to play important roles in the
catalytic performance of the resulting immobilized com-
plexes. Simultaneously, the further introduction of methyl
groups significantly improved the catalytic performance of
immobilized complexes by modifying their surface
123

Epoxidation of Styrene with H2O2
105
26. Wang XL, Wu GD, Li JP, Zhao N, Wei W, Sun YH (2007) J Mol
Catal A 276:86
40. Jia MJ, Seifert A, Berger M, Giegengack H, Schulze S, Thiel WR
(2004) Chem Mater 16:877
27. Kresge CT, Leonowicz ME, Roth WJ, Vartuli JC, Beck JS (1992)
Nature 359:710
41. Sing KSW, Everett DH, Haul RAW, Moscow L, Pierotti RA,
Rouquerol T, Siemienewska T (1985) Pure Appl Chem 57:603
´
28. Zhao DY, Feng JL, Huo QS, Melosh N, Fredrickson GH, Chm-
elka BF, Stucky GD (1998) Science 279:548
29. Wang XL, Wu GD, Li JP, Zhao N, Wei W, Sun YH (2007) Catal
Lett 119:89
30. Heinert D, Martell AE (1962) J Am Chem Soc 84:3257
31. Shanthi R, Nagaraja KS, Udupa MR (1987) Inorg Chem Acta
133:211
32. Bikerman JJ (1958) Surface chemistry: theory and applications,
2nd edn. Academic Press, New York, p 343
33. Carvalho WA, Wallau M, Schuchardt U (1999) J Mol Catal A
144:91
42. Brunel D, Bellocq N, Sutra P, Cauvel A, Lasperas M, Moreau P,
Renzo FD, Galarneau A, Fajula F (1998) Coord Chem Rev
180:1085
43. Karandikar P, Dhanya KC, Deshpande S, Chandwadkar AJ,
Sivasanker S, Agashe M (2004) Catal Commun 5:69
44. Belal R, Meunier B, Aurangzeb N (1988) J Mol Catal A 44:187
45. Aurangzeb N (1994) J Chem Soc Chem Commun 1153
46. Silva AR, Figueiredo JL, Freire C, de Castro B (2004) Micro-
porous Mesoporous Mater 68:83
47. Joseph T, Hartmann M, Ernst S, Halligudi SB (2004) J Mol Catal
A 207:131
34. Sindorf DW, Maciel GE (1981) J Am Chem Soc 103:4263
´
(2000) J Catal 193:283
36. Percy GC (1975) J Inorg Nucl Chem 37:2071
37. Ren T, Yan L, Zhang HP, Suo JS (2003) J Mol Catal (China)
19:310
38. Lim MH, Stein A (1999) Chem Mater 11:3285
39. Sakthivel A, Zhao J, Hanzlik M, Chiang AST, Herrmann WA,
K}uhn FE (2005) Adv Synth Catal 347:473
48. Maldotti A, Molinari A, Varani G, Lenarda M, Storara L, Bigi F,
Maggi R, Mazzacani A, Sartori G (2002) J Catal 209:210
´
´
35. Dıaz I, Marquez-Alvarez C, Mohino F, Perez-Pariente J, Sastre E
´
49. Corma A, Domine M, Gaona JA, Jorda JL, Navarro MT, Rey F,
Perez-Pariente J, Tsuji J, McCulloch B, Nemeth LT (1998) Chem
´
Commun 2211
50. Nam W, Choi HJ, Han HJ, Cho SH, Lee HJ, Han SY (1999)
Chem Commun 387
51. Liu CJ, Yu WY, Li SG, Che CM (1998) J Org Chem 63:7364
123