Selective oxidation of olefins with aqueous hydrogen peroxide over phosphomolybdic acid functionalized knitting aryl network polymer

Article abstract of DOI:10.1016/j.molcata.2015.12.012

A phosphomolybdic acid (PMA)-based heterogeneous catalyst, denoted as PMA/KAP, was prepared by immobilizing PMA onto a knitting aryl network polymer (KAP) based on triphenylphosphine (PPh₃). The catalytic property of PMA/KAP was investigated for the selective oxidation of olefins with aqueous hydrogen peroxide (H₂O₂) as oxidant. When using ethyl acetate (EAC) as reaction medium, PMA/KAP performs higher activity and selectivity to epoxide for a variety of olefins, and it can be reused for several times without obvious loss of activity. When the reaction was carried out in acetonitrile (AN) medium, deactivation of PMA/KAP catalyst can be observed immediately. A variety of characterization results suggest that the degradation of PMA unit to (PO₄[MoO(O₂)₂]₄)^3- occurs easily when the PMA/KAP catalyst is operated in H₂O₂/AN system, while such degradation behavior could be significantly inhibited when the catalyst is used in the system of H₂O₂/EAC. We proposed that the neighbouring P-containing ligands dispersed in the framework of KAP can produce a steric pocket with low electron density, which can promote the formation of multi-weak coordination interaction between PMA unit and several P ligands. Such multi-weak interaction can inhibit the degradation of PMA to (PO₄[MoO(O₂)₂]₄)^3-, thus avoiding the leaching of active species from the KAP support, and resulting in the formation of relatively stable heterogeneous PMA supported catalyst for olefin epoxidation with H₂O₂ in the media of EAC.

Full text of DOI:10.1016/j.molcata.2015.12.012

Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Selective oxidation of olefins with aqueous hydrogen peroxide over
phosphomolybdic acid functionalized knitting aryl network polymer
a
a
b
a
a
Xiaojing Song , Wanchun Zhu , Yan Yan , Hongcheng Gao , Wenxiu Gao ,
a
a,∗
Wenxiang Zhang , Mingjun Jia
Key Laboratory of Surface and Interface Chemistry of Jilin Province, College of Chemistry, Jilin University, 130021 Changchun, China
State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, College of Chemistry, Jilin University, Changchun 130012, China
a
b
a r t i c l e i n f o
a b s t r a c t
Article history:
A phosphomolybdic acid (PMA)-based heterogeneous catalyst, denoted as PMA/KAP, was prepared by
immobilizing PMA onto a knitting aryl network polymer (KAP) based on triphenylphosphine (PPh3).
The catalytic property of PMA/KAP was investigated for the selective oxidation of olefins with aqueous
hydrogen peroxide (H2O2) as oxidant. When using ethyl acetate (EAC) as reaction medium, PMA/KAP
performs higher activity and selectivity to epoxide for a variety of olefins, and it can be reused for several
times without obvious loss of activity. When the reaction was carried out in acetonitrile (AN) medium,
deactivation of PMA/KAP catalyst can be observed immediately. A variety of characterization results
Received 8 September 2015
Received in revised form 5 December 2015
Accepted 17 December 2015
Available online 23 December 2015
Keywords:
Phosphomolybdic acid
Knitting aryl network polymer
Immobilization
Selective oxidation
Olefins
3
−
suggest that the degradation of PMA unit to (PO4[MoO(O2)2]4) occurs easily when the PMA/KAP catalyst
is operated in H2O2/AN system, while such degradation behavior could be significantly inhibited when
the catalyst is used in the system of H2O2/EAC. We proposed that the neighbouring P-containing ligands
dispersed in the framework of KAP can produce a steric pocket with low electron density, which can
promote the formation of multi-weak coordination interaction between PMA unit and several P ligands.
3−
Such multi-weak interaction can inhibit the degradation of PMA to (PO4[MoO(O2)2]4) , thus avoiding
the leaching of active species from the KAP support, and resulting in the formation of relatively stable
heterogeneous PMA supported catalyst for olefin epoxidation with H2O2 in the media of EAC.
©
2015 Elsevier B.V. All rights reserved.
1
. Introduction
benzene with an external cross-linker, could be used as suitable
support to fabricate efficient heterogeneous catalysts. For instance,
Porous organic frameworks (POFs) have received increasing
Pd nanoparticles supported PPh -functionalized KAP exhibited
3
attention due to their unique properties like large surface, low
skeletal density, and high chemical stability [1–10]. Compared with
traditional inorganic materials, one particular advantage of POFs is
its ability to introduce various organic chemical functionalities for
constructing novel advanced materials. Recently, a variety of POFs
containing functional units have been reported, which have shown
potential applications in adsorption, ion exchange, nanotechnol-
ogy, and catalysis [11–16].
Knitting aryl network polymers (KAPs) are a new family of POFs,
which can be prepared though a simple one-step Friedel–Crafts
reaction using aromatic compounds as the monomer and formalde-
hyde dimethyl acetal as the external cross-linker [17–19]. In the
field of catalysis, a few recent works showed that a functionalized
excellent activity and selectivity for the Suzuki–Miyaura crossing
coupling reaction of aryl chlorides [18]. Rh supported PPh -based
3
KAP showed higher activity and stability than Rh supported sil-
ica catalyst for the hydroformylation of higher olefins [20]. It was
believed that the special coordinating ability of PPh ligands as well
3
as the porous characteristics of KAP play critical role in stabilizing
the nanoparticles of noble metal.
Recently, numerous contributions have been made to develop
highly active and stable polyoxometalates (POM)-based hetero-
geneous catalysts for the application in liquid-phase oxidation
(or epoxidation) of olefins. By selecting different supports and
preparation strategies, a variety of relatively active and stable POM-
supported heterogeneous catalysts have been obtained [21–23].
5−
KAP material, obtained by knitting triphenylphosphine (PPh ) and
For instance, Kholdeeva et al. immobilized [PW11CoO39
]
and
−
into MIL-101 through electrostatical interaction,
3
5
[
PW11TiO40]
and found that the resulting materials were active heterogeneous
catalysts for the oxidation of olefins with H2O2 or O2 as oxidant [21].
Bordoloi et al. reported that molybdovanadophosphoric acids sup-
∗ Corresponding author. Fax: +86 431 85168420.
E-mail address: jiamj@jlu.edu.cn (M. Jia).
http://dx.doi.org/10.1016/j.molcata.2015.12.012
1
381-1169/© 2015 Elsevier B.V. All rights reserved.

X. Song et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
33
ported amine-modified mesoporous silicas, were highly active and
stable catalysts for selective oxidation of anthracene with t-BuOOH
preparation of KAP and PMA/KAP are shown in Scheme 1. Element
analysis (%): KAP, C 83.8, H 6.2, Fe 5.2. PMA/KAP: C 76.9, H 5.6, Fe
2.4.
as oxidant [22]. Kasai et al. immobilized [r-1,2-H SiV W10O40
]
2
2
on N-octyldihydroimidazolium cation-modified SiO , to obtain an
For comparison, PMA-(PPh3)3 complex was synthesized
through adding some PMA into a methanol solution containing
2
efficient catalyst for the oxidation of olefins and sulfides with
H O as oxidant [23]. Our recent work showed that phospho-
PPh3 ligands (the ratio of PPh /PMA is about 3.5). The mixed solu-
2
2
3
molybdic acid supported COF-300 (a covalent organic framework
material) exhibits high activity and stability for the epoxidation of
cyclooctene and 1-octene with t-BuOOH as oxidant [24]. In spite of
these considerable progresses, most of the POM supported catalysts
commonly suffer from active species leaching from the supports,
tion presents yellow at first, and the solid complex with light green
color can be obtained finally after filtering and washing by CH OH.
3
The resultant complex is denoted as PMA-(PPh ) , in which the
3
3
mole ratio of PMA/PPh3 is determined by elemental analysis. The
process for preparation of PMA-(PPh3)3 is shown in Scheme 2.
which is particularly serious when H O is used as oxidant [25]. This
2
2
can be mainly attributed to the strong complexing and solvolytic
properties of H O and solvents, which can usually transfer PMA
2.3. Catalytic reaction
2
2
cluster into dissolvable compound, or turn it to smaller species
by oxidative degradation [26,27]. Besides, it is quite hard to get
truly heterogeneous supported POM catalysts, since the relatively
large size of POM clusters brings serious difficulty to build stable
linkages between supports and POM. Therefore, it is still a very
significant subject to develop novel efficient POM-based heteroge-
neous catalysts for the oxidation of olefins with H O , and to reveal
The catalytic properties of PMA/KAP and PMA-(PPh3)3 were
tested by the oxidation of olefins. The reactions were initiated by
adding oxidant H O2 into a 10 ml flask containing catalyst, solvent
2
and corresponding reactants under beforehand designed tempera-
ture. The course of the reactions were monitored and quantified by
Shimadzu GC-8A gas chromatograph with HP-5 capillary column.
2
2
the key factors favoring the improvement of the stability of the POM
supported catalysts.
2.4. Characterization techniques
In this work, we tried to use PPh -based KAP as support to
prepare phosphomolybdic acid (PMA) supported heterogeneous
catalyst, and the catalytic performance of the resulting material
3
FT-IR spectra were recorded on a Nicolet AVATAR 370 DTGS
−1
spectrometer in the range 4000–500 cm . Powder X-ray diffrac-
tion (XRD) patterns were recorded on a Shimadzu XRD-6000
diffractometer (40 kV, 30 mA), using Ni-filtered CuK␣ radiation.
(
PMA/KAP) was studied for the oxidation of olefins with H O2 as
2
oxidant in different reaction media. Besides, a homogeneous cata-
lyst named PMA-(PPh3)3 was also prepared by coordinating PMA
with free PPh3 ligands for the purpose of comparison. By com-
bining a variety of characterization results, it can be revealed that
multi-interactions are present between one PMA unit and several
P-containing ligands dispersed in the framework of KAP support,
which can result in the formation of relatively active and stable PMA
Thermogravimetry (TG) were carried out using A Netzsch Thermo-
◦
analyser STA 449F3 with a N flow and heating rate of 10 /min, from
2
◦
room temperature to 800 C. Transmission electron microscopy
(
TEM) images were taken with a H8100-IV electron microscope
operating at 200 kV. The samples were suspended in ethanol by
sonication and then picked up on a Cu grid covered with a car-
bon film. Element analyzes of C and H were implemented with
a PerkinElmer 2400CHN elemental analyzer. For determining the
Fe amount remained in KAP support or PMA/KAP catalyst, the
sample was firstly treated by 3 M HNO3 aqueous solution, and
then the Fe concentration in the filtration was detected by induc-
tively coupled plasma-optical emission spectroscopy (ICP-AES).
N2 adsorption/desorption isotherms were measured at 77 K using
a Micromeritics ASAP 2010N analyzer. Samples were degassed
supported KAP catalyst for the epoxidation of olefins with H O in
2
2
the presence of EAC.
2
. Experimental section
2
.1. Materials
◦
Benzene (PhH), triphenylphosphine (PPh ), FeCl3 (anhydrous),
at 110 C for 8 h before measurements. Specific surface areas
3
methanol, 1,2-dichloroethane (DCE), dimethylformamide (DMF),
were calculated using BET model. XPS measurements were made
on a VGESCA LAB MK-II X-ray electron spectrometer using Al
K␣ radiation. Solid 3 P MAS NMR spectra were recorded on a
400 MHZ Bruker spectrometer. The 31P MAS NMR chemical shifts
are referenced to the resonances of monoammonium phosphate
acetonitrile (AN), ethylacetate (EAC) and 30 wt% H O2 aqueous
2
1
solution were purchased from China National Medicines Corpora-
tion Ltd., all of which were of analytical grade and were used as
received. Cyclooctene, cyclohexene, styrene, 1-hexene, 1-octene,
a-piene and formaldehyde dimethylacetal (FDA) were purchased
from Aldrich.
3
1
(NH H PO ) standard. Liquid P NMR spectra were recorded on a
4
2
4
31
500 MHZ AVANCEIII500. P chemical shifts are referenced to 85%
H PO as an external standard.
3
4
2.2. The preparation of catalysts
3. Results and discussion
KAP was synthesized according to literature procedure [18].
Typically, 1.56 g PhH, 5.25 g PPh3 and 4.56 g FDA were dissolved in
3.1. Catalyst characterization
2
0 ml DCE containing 9.75 g FeCl3 (anhydrous). The resulting mix-
◦
ture was stirred at 45 C for 5 h to form original network, and then
heated at 80 C for 67 h to complete the reaction. The as-synthesized
Fig. 1 shows the FT-IR spectra of KAP, PMA, PMA-(PPh3)3 and
PMA/KAP. The characteristic bands of KAP are in agreement with
the related literature results [18]. The benzene skeleton vibration
◦
KAP was washed with methanol in a Soxhlet to remove residual
−
1
FeCl . Catalyst PMA/KAP was prepared by adding 0.30 g KAP sup-
peaks appear in the range of 1600–1450 cm , while the peaks
3
−
1
port into a methanol solution containing 0.053 g PMA, the mixture
was then stirred for 24 h at room temperature. The resulting solid
was washed three times with methanol, and then washed with
at around 1250–950 and 900–650 cm
result from C H bend-
−
1
ing vibrations of benzene ring. 1435 cm could be attributed to
the P-CH2 which shows that the phosphine ligands are embedded
◦
methanol in a Soxhlet’s extractor at 80 C for 24 h, then dry in
into the skeleton of KAP [28]. For PMA-(PPh3)3 and PMA/KAP, the
◦
−1
vacuum at 60 C for 8 h. The PMA loading determined by atomic
characteristic bands of PMA appear in the range of 1100–700 cm
confirming the presence of PMA units in the resulting catalysts.
,
adsorption spectroscopy (AAS) was 0.05 mmol/g. The processes for

3
4
X. Song et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
Scheme 1. Preparation process of KAP support and PMA/KAP catalyst.
Scheme 2. Preparation process of PMA-(PPh3)3.
Fig. 1. FT-IR spectra of KAP, PMA, and PMA/KAP.
Fig. 2. The wide angle XRD patterns of PMA/KAP and pure PMA (insert).
Fig. 2 reveals the XRD patterns of the PMA/KAP catalyst. There
TEM images of the KAP support and the PMA/KAP catalyst are
shown in Fig. 3. Amorphous nanosize particles are observed in the
image of the KAP support. Compared with the KAP support, the
corresponding region of PMA/KAP turns to darker, which might be
caused by the introduction of PMA units.
is only one broad peak, suggesting the amorphous characteristic of
the resulting material. The absence of diffraction peaks associated
with the crystalline phase of PMA suggests that PMA cluster should
be highly dispersed on the surface or the channel of KAP support.

X. Song et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
35
Fig. 3. The TEM images of (A) KAP and (B) PMA/KAP.
ence of abundant micropores [18]. The appearance of hysteresis
loop at relative pressure between 0.4 and 0.7 indicates the existence
of a certain amount of mesopores in KAP and PMA/KAP. Moreover,
a sharp rise at the high pressure regions (above 0.8) reveals the
presence of macropores in these materials. Compared with KAP,
the N2 adsorption capacities for PMA/KAP decreases somewhat,
which can be assigned to the introduction of PMA units to the
surface/channel of KAP support. According to the traditional BET
2
method, BET surface area of KAP is found to be 922 m /g, and it
2
decreases to 761 m /g after introducing PMA.
The P 2p and Mo 3d XPS spectra of KAP, PMA, PMA-(PPh3)3 and
PMA/KAP are shown in Fig. 6. The P 2p spectrum of KAP support
shows a peak at 133.3 eV, which is higher than that of pure PPh3
(132.2 eV), but quite consistent to the binding energy (BE) value for
the reported triphenylphosphine oxide (OPPh ) [29]. These results
suggest that the P atoms in PPh3 ligands have been oxidized to P O
3
◦
Fig. 4. TG analysis of PMA/KAP and KAP at a heating rate of 10 C/min under nitrogen.
species possibly by oxygen/FeCl3 during the process for synthesiz-
ing KAP support. The spectrum of PMA shows a signal at 134.5,
which is attributed to the BE of P in PMA. After supporting PMA on
KAP, a broad peak centered at 133.8 can be observed, which should
be due to the partial overlap of the P signals derived from KAP
support and PMA. The Mo 3d XPS spectrum of PMA displays char-
acteristic peaks at 236.3 and 233.2 eV, which are attributed to Mo
(
VI) 3d3/2 and Mo (VI) 3d5/2, respectively [30]. For PMA-(PPh ) , the
3 3
binding energy of Mo shifts to more negatively (235.9 and 232.8 eV)
than that of pure PMA, which can be attributed to the nucleophilic
attack of the phosphine lone pair (in KAP) on a ␲* Mo O orbital
[
31]. It is found that the Mo binding energies of PMA/KAP are simi-
lar to that of PMA-(PPh ) , implying that the redox property of Mo
3
3
species in both compounds are quite similar.
The 31P MAS-NMR spectra of PPh , PMA, KAP, PMA-(PPh )
3
3 3
and PMA/KAP are shown in Fig. 7. Free PPh3 ligand and PMA dis-
play phosphorous resonance signals at −10.88 ppm and −4.84 ppm,
respectively. For the P signal of PPh3 ligand in PMA-(PPh3)3, it
changes from −10.88 ppm (free PPh3 ligand) to 1.33 ppm, suggest-
ing that relatively strong coordination interaction should be built
between the PPh3 ligands and the PMA units. In previous litera-
Fig. 5. Nitrogen adsorption and desorption isotherms of KAP and PMA/KAP.
The TG profiles of KAP and PMA/KAP are shown in Fig. 4. A slight
◦
weight loss at around 100 C can be attributed to desorption of a
tures, it has been proposed that the phosphines (PR , R Me, Et, Ph)
3
trace amount of water and methanol. For KAP, significant weight
◦
might interact with the molybdenum (VI) oxides to form molybde-
num complexes O Mo O · · · PR3 linkages [31–34]. The 31P NMR
spectrum of KAP shows a broad signal centered at about 25 ppm,
indicating that the chemical environment of P species in KAP is
quite different from the pure PPh3 ligand. These results suggest
that the electron density of P species in KAP support decreases
significantly after forming the knitting aryl network polymers, con-
loss starts at about 300 C due to the thermal decomposition of this
polymer support. The decomposition rate of PMA/KAP decreases
somewhat in comparison with that of KAP support, which might
be an indicative of the existence of some interactions between KAP
support and PMA units.
The N2 adsorption–desorption isotherms of KAP and PMA/KAP
are shown in Fig. 5. For both samples, a steep nitrogen gas uptake at
low relative pressure (0–0.1) can be observed, indicating the pres-
firming further that the PPh ligands have already been oxidized to
3

3
6
X. Song et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
Fig. 6. XPS spectra for the BE of (A) P in KAP, PMA/KAP and PMA and (B) Mo in PMA, PMA/KAP and PMA-(PPh3)3.
Table 1
The oxidation results of cyclooctene with H2O2 as oxidant, PMA/KAP and PMA-
PPh3)3 as catalysts in different solvents.
(
Catalyst
Reaction media Time (h) Conversion (%) Selectivity (%)
Epoxide Diols
PMA/KAP
PMA-(PPh3)3
PMA/KAP
PMA-(PPh3)3
PMA/KAP
PMA-(PPh3)3
DMF
AN
9
9
36.9
56.0
26.4
89.0
78.6
95.9
>99
>99
>99
>99
>99
93.6
–
–
–
–
–
6.4
EAC
9
3
Reaction conditions: catalyst 10 mg, solvent 2 ml, cyclooctene 0.12 ml, H2O2 0.30 ml,
◦
temperature 70 C.
PMA/KAP, 1-hexene can be converted to epoxide rapidly, and 70.2%
of conversion with >99% selectivity of epoxide is obtained. The
conversion of 1-octene is rather low (10.6%) under the test con-
ditions, possible due to the fact that the molecular of 1-octene has
larger steric hindrance (caused by the increase of chain length).
Notably, PMA/KAP exhibits much higher selectivity to 1-hexene
epoxide than that of PMA-(PPh ) , although the catalytic activity
Fig. 7. 31P MAS-NMR spectrum of PPh3, PMA, PMA-(PPh3)3, KAP and PMA/KAP.
OPPh3 during the polymerizing process as revealed by the above
XPS results. For both PMA/KAP and PMA-(PPh ) , the P signals
3
3
belonged to PMA are quite similar to the pure PMA, indicating that
the basic structure of the introduced PMA units is remained. For the
3
3
of PMA/KAP is slightly lower than the corresponding homogenous
catalyst. Similar situation can also be observed for the epoxidation
of styrene, in which PMA/KAP establishes lower activity but higher
epoxide selectivity than that of PMA-(PPh ) . Besides, PMA/KAP is
31
P NMR signal in KAP support, it shifts slightly toward high field
after introducing the PMA units into the KAP support, suggesting
the existence of weak coordination interaction between the Mo
species in PMA units and the P atoms in KAP support. These results
suggest that the OPPh3 ligands existed in KAP support still have
the coordination ability with the PMA units, thus may bring con-
siderable effect on the structure stability (redox capacity) and the
catalytic activity of the introduced PMA compounds.
3
3
also active for the oxidation of ␣-piene and cyclohexene, which
can convert the olefins to the corresponding epoxide and enol (or
enone).
For studying the stability of PMA/KAP catalyst, leaching test
was conducted in the system of H O /EAC or H O /AN. In dupli-
2
2
2
2
cate reaction, the solid catalyst was separated by hot filtration,
and the filtrate was then further stirred. As shown in Fig. 8A, no
further conversion of cyclooctene could be detected in the filtrate
for catalyst PMA/KAP in the system of H O /EAC, which demon-
3
.2. Catalytic performance of PMA/KAP and PMA-(PPh3)3
The catalytic properties of PMA/KAP and PMA-(PPh3)3 were
2
2
investigated for the oxidation of cyclooctene with H O2 as oxidant
strates that the oxidation reaction is a truly heterogeneous process.
As for the H O /AN system, cyclooctene can still be converted to
2
in the presence of different solvent media (Table 1). Both PMA/KAP
and PMA-(PPh3)3 are catalytically active for the epoxidation of
cyclooctene. Compared with the homogeneous PMA-(PPh ) , the
2
2
epoxide rapidly after removing the PMA/KAP catalyst by filtrating,
suggesting that leaching of active sites from the solid catalyst occurs
during the reaction course. These results suggest that different sol-
vent/oxidant systems can bring serious difference in influencing
the stability of PMA/KAP catalyst.
3
3
solid PMA/KAP catalyst gives lower activity under test conditions.
Moreover, the catalytic properties of both catalysts are solvent
dependent, the conversion of cyclooctene follows the order of
EAC > DMF > AN. When EAC is used as medium, PMA/KAP catalyst
gives a 78.6% conversion of cyclooctene with nearly 100% selectivity
of epoxide after 9 h reaction.
In addition, PMA/KAP and PMA-(PPh3)3 are also catalytically
active for the oxidation/epoxidation of other olefins as shown in
Table 2. They can even oxidize the relative inert terminal olefins
of 1-hexene and 1-octene to their corresponding epoxides. For
The recycling experiments of PMA/KAP for cyclooctene epoxi-
dation were carried out in the systems of H O /EAC and H O /AN.
2
2
2
2
Before reusing, the recovered catalysts were washed with corre-
◦
sponding solvent and dried at 60 C. In the system of H O /EAC,
2
2
the activity of the recycled PMA/KAP catalyst decreases somewhat
at the second run, it then keeps nearly consistent with further run.
These results suggest that catalyst PMA/KAP is recyclable in the

X. Song et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
37
Table 2
The oxidation of olefins catalyzed by PMA/KAP and PMA-(PPh3)3 with H2O2 as oxidant.
Substrate
Catalyst
Time
Conversion (%)
Selectivity (%)
Epoxide
Others
PMA/KAP
PMA-
6
3
6
70.2
60.2
88.7
>99
34.0
16.4
–
a
66.0
(
PPh3)3
83.6a
PMA/KAP
6
10.6
>99
–
PMA/KAP
PMA/KAP
PMA-(PPh3)3
6
9
9
76.2
43.0
68.1
>99
27.3
7.3
–
b
72.7
92.7b
PMA/KAP
9
27.1
2.6
97.4c
◦
Reaction conditions:catalyst 10 mg, substrate 1 mmol, H2O2 3 mmol, temperature 70 C, EAC 2 ml.
a
1
,2-Hexanediol.
b
c
Benzaldehyde.
cyclohexane-1,2-diol (5.1%) + 2-cyclohexen-1-ol (23.9%) + 2-cyclohexene-1-one (68.4%).
Fig. 8. (A) Leaching experiments of catalyst PMA/KAP in H2O2/EAC and H2O2/AN. Dashed lines indicate the conversions after the removal of the catalysts. (B) Recycling
experiments of cyclooctene with catalyst PMA/KAP in H2O2/EAC and H2O2/AN respectively. Reaction conditions are analogous to those in Table 1.
H O /EAC system under the given reaction conditions. The slightly
ing that PMA can be retained well on the KAP support when the
solid catalyst is operated in the system of H O /EAC. It is worth to
2
2
decrease in catalytic activity might be caused by the adsorption of
trace amount of reactants or products on the surface of PMA/KAP. As
for the H O /AN system, deactivation of PMA/KAP can be observed
2
2
31
mention that there is almost no obvious change for the P NMR
signals belonged to phosphine ligands between the used PMA/KAP
catalysts (recovered either from H O /EAC or from H O /AN) and
2
2
when the catalyst is reused for the second run. These results suggest
that the types of solvent media have significant effect on the sta-
bility of PMA/KAP catalyst during the oxidation process of olefins
with H O .
2
2
2
2
the fresh catalyst, revealing that the state of P species for the KAP
support is well kept during the catalytic oxidation process.
In order to further understand the effect of solvent/oxidant on
2
2
3
Solid 1P MAS-NMR was carried out to characterize the used
the catalytic properties of PMA/KAP catalyst, liquid P NMR was
31
PMA/KAP catalyst, which was recovered from the H O /EAC or
used to characterize the homogeneous PPh , PMA and PMA-(PPh )
in CD CN (or in CD CN/H O ). As shown in Fig. 10, the P signal
3 3 2 2
2
2
3
3 3
31
the H O /AN system. As shown in Fig. 9, the P NMR signals
2
2
belonged to PMA units have disappeared for the catalyst recovered
from H O /AN, while the signals of the catalyst recovered from
of PPh3 ligand (dissolved in CD CN) appears at −5.58 ppm, and
3
it shifts to 35.81 ppm after treating with H O due to the forma-
2
2
2
2
H O /EAC are quite consistent with the fresh catalyst, establish-
tion of OPPh . Similarly, the P signal of PMA shifts from −3.53 ppm
2
2
3

3
8
X. Song et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
Fig. 9. Solid 31P MAS-NMR of (a) fresh PMA/KAP, (b) PMA/KAP recovered from
H2O2/AN and (c) PMA/KAP recovered from H2O2/EAC.
3 3 2 2
Fig. 11. The FT-IR spectra of PMA-(PPh ) , dissolved in H O /EAC. (a) Precipita-
tion obtained by adding Bu4NBr in EAC layer, and (b) solid obtained by drying the
H2O2/H2O layer.
lates, thiolates) dissolves easily in the presence of coordinating
ipr
solvents like AN and DMF, which can form [Tp MoOX(solv)] and
OPR3 species rapidly. In the case of noncoordinating solvents, such
as CHCl , EAC et al., the structure of the compound can be kept well.
3
Concerning the relatively high stability/recyclability of
PMA/KAP in the system of H O /EAC, it can be imagined that the
2
2
3−
degradation of PMA to (PO [MoO(O ) ] ) has been completely
4
2 2 4
inhibited when PMA units are introduced into the surface/pores
of KAP support, as already confirmed by the above solid 31
P
NMR results. Obviously, the physical/chemical environment of
PMA units are different between PMA/KAP and PMA-(PPh3)3
complex. During the catalytic oxidation progress, the PPh3 ligands
in PMA-(PPh3)3 could be easily converted to OPPh , while the
3
decomposition of PMA-(PPh3)3 complexes occurred rapidly. These
results suggest that the interaction between the OPPh3 ligands
and PMA units should be not very strong, which could not inhibit
the decomposition of PMA during the catalytic reaction course.
For PMA/KAP catalyst, it was known that the P-containing ligands
are also present in the form of OPPh3 as revealed by the XPS and
NMR results. The main feature KAP support is that the OPPh3
Fig. 10. 31P NMR of (a) PPh3 (b) PMA (c) PPh3 + 0.3 ml H2O2 aqueous (d) PMA + 0.3 ml
H2O2 aqueous (e) PMA-(PPh3)3 + 0.3 ml H2O2 aqueous in CD3CN.
to 9.21 ppm after treating with H O , which can be attributed to
2
2
3−
the formation of degraded product of (PO [MoO(O ) ] ) [36]. For
4
2 2 4
complex PMA-(PPh ) , the P signals for PMA and PPh ligand shift to
3
3
3
9
.21 ppm (from −3.53) and to 35.81 ppm (from −5.58), respectively,
ligands have been linked through Ph-(CH )x- bridges, resulting
2
after treating them with excess H O . That is to say, in the system
2
2
in the formation of three-dimensional network. In this case, the
porous framework of KAP support, which contains numerous
bridge-linked phosphine ligands, can be regarded as multi-dentate
phosphine ligands with delocalized electrons, just like a steric
pocket with low-electron density [31,35]. Associated with the
relatively high stability of PMA/KAP catalyst, it can be deduced that
the introduced PMA unit might be stabilized by such steric pocket
by building multiple weak coordination interactions between the
PMA unit and the framework of KAP containing numerous OPPh3
ligands.
of H O /AN, PMA-(PPh ) can be rapidly dissolved/degraded into
2
2
3 3
3−
smaller species of (PO [MoO(O ) ] ) and OPPh3 species. On the
4
2 2 4
basis of these results, it can be deduced that the deactivation of
PMA/KAP catalyst in the system of H O /AN should be also caused
2
2
3−
by the formation of degraded (PO [MoO(O ) ] ) species, which
4
2 2 4
are easily leached from the solid KAP support.
It should be pointed out here that PMA-(PPh3)3 complexes are
nearly insoluble when EAC is used as solvent medium, although
PPh3 ligands are easily dissolved in EAC. Upon adding a certain
amount of H O aqueous solution, PMA-(PPh3)3 complexes dis-
2
2
solve partially. After standing for a while, two separating phase,
organic phase (EAC layer) and aqueous phase (H O /H O layer)
4. Conclusions
2
2
2
can be observed. On the basis of the FT-IR results (Fig. 11) and
the related literatures [37], it can be concluded that PMA units
PMA functionalized KAP catalyst was synthesized successfully
3−
and applied into the oxidation/epoxidation of olefins with H O2 as
and (PO [MoO(O ) ] ) species are present in the EAC layer and
2
4
2 2 4
oxidant. It is found that the resultant PMA/KAP catalyst can perform
relatively high activity and stability for the epoxidation of olefins
when using EAC as the reaction media. The relative good stabil-
ity of PMA/KAP catalyst can be mainly assigned to the presence of
multiple bridge-linked phosphine ligands dispersed in the bulky
KAP support, which can produce a steric pocket with low-electron
density for stabilizing the introduced PMA units through multiple
weak interaction between the PMA unit and the KAP support.
the H O /H O layer, respectively. These results suggest that using
2
2
2
EAC as solvent medium (rather than AN) can decrease the con-
3−
verting rate of PMA-(PPh3)3 complexes to (PO [MoO(O ) ] )
4
2 2 4
species. This might be mainly due to the fact that EAC has weaker
polarity and lower ability of solvent-coordination in comparison
with AN. Similar phenomena have already been reported in litera-
ipr
ture [31], and it was proposed that [Tp MoO(OPR )X] compound
3
ipr
−
(
Tp
hydrotris(3-isopropylpyrazol-1-yl) borate; X Cl , pheno-

X. Song et al. / Journal of Molecular Catalysis A: Chemical 413 (2016) 32–39
[17] B. Li, R. Gong, W. Wang, X. Huang, W. Zhang, H. Li, C. Hu, B. Tan,
39
Acknowledgment
Macromolecules 44 (2011) 2410–2414.
[
18] B. Li, Z. Guan, W. Wang, X. Yang, J. Hu, B. Tan, T. Li, Adv. Mater. 24 (2012)
Financial support from the National Natural Science Foundation
of China (No. 21173100 and 21320102001) is gratefully acknowl-
edged.
3390–3395.
[19] R. Dawson, L.A. Stevens, T.C. Drage, C.E. Snape, M.W. Smith, D.J. Adams, A.I.
Cooper, J. Am. Chem. Soc. 134 (2012) 10741–10744.
[
20] M. Jiang, Y. Ding, L. Yan, X. Song, R. Lin, Chin. J. Catal. 35 (2014) 1456–1464.
21] N. Maksimchuk, M. Timofeeva, M. Melgunov, A. Shmakov, Y. Chesalov, D.
Dybtsev, V. Fedin, O. Kholdeeva, J. Catal. 257 (2008) 315–323.
22] A. Bordoloi, F. Lefebvre, S. Halligudi, J. Catal. 247 (2007) 166–175.
23] J. Kasai, Y. Nakagawa, S. Uchida, K. Yamaguchi, N. Mizuno, Chem. Eur. J. 12
[
References
[
[
[
[
[
1] B. Li, F. Su, H. Luo, L. Liang, B. Tan, Micropor. Mesopor. Mat. 138 (2011)
07–214.
2] X. Du, Y. Sun, B. Tan, Q. Teng, X. Yao, C. Su, W. Wang, Chem. Commun. 46
2010) 970–972.
3] D. Dang, P. Wu, C. He, Z. Xie, C. Duan, J. Am. Chem. Soc. 132 (2010)
4321–14323.
(
2006) 4176–4184.
2
[
24] W. Gao, X. Sun, H. Niu, X. Song, K. Li, H. Gao, W. Zhang, J. Yu, M. Jia, Micropor.
Mesopor. Mat. 213 (2015) 59–67.
25] K. Yamaguchi, C. Yoshida, N. Mizuno, J. Am. Chem. Soc. 127 (2005) 530–531.
26] Dean C. Duncan, C.R. Chambers, Eric Hecht, Craig L. Hill, J. Am. Chem. Soc. 117
(
[
[
1
(
1995) 681–691.
[
[
[
4] B. James, R. Holst, A.I. Cooper, Adv. Mater. 22 (2010) 5212–5216.
5] H. Furukawa, O.M. Yaghi, J. Am. Chem. Soc. 131 (2009) 8875–8883.
6] R.W. Tilford, W.R. Gemmill, H.z. Loye, J.J. Lavigne, Chem. Mater. 18 (2006)
[
27] N. Melanie Gresley, W.P. Griffith, A.C. Laemmel, H.I.S. Nogueira, B.C. Parkin, J.
Mol. Catal. A: Chem. 117 (1997) 185–198.
28] W. He, F. Zhang, H. Li, Chem. Sci. 2 (2011) 961–966.
[
[
5
296–5301.
29] D.G. Larrude, M.E.H. Maia Costa, F.H. Monterio, F.L. Freire Jr., J. Appl. Phys. 111
[
[
7] C. Zhao, L. Zhong, J. Phys. Chem. C 113 (2009) 16860–16862.
8] J. Jiang, F. Su, A. Trewin, C.D. Wood, H. Niu, Y.Z. Khimayk, J.T.A. Jones, A.I.
Cooper, J. Am. Chem. Soc. 130 (2008) 7710–7720.
(
2012) 064315.
[
[
30] M. Nakayama, T. Ii, K. Ogura, J. Mater. Res. 18 (2003) 2509–2514.
31] A.J. Millar, C.J. Doonan, P.D. Smith, V.N. Nemykin, P. Basu, C.G. Young, Chem.
Eur. J. 11 (2005) 3255–3267.
32] J. Leppin, C. Forster, K. Heinze, Inorg. Chem. 53 (2014) 12416–12427.
33] P.D. Smith, A.J. Millar, Ch.G. Young, A. Ghosh, Partha Basu, J. Am. Chem. Soc.
[9] J. Weber, N. Du, M.D. Guiver, Macromolecules 44 (2011) 1763–1767.
[10] J. Lee, C.D. Wood, D. Bradshaw, M.J. Rosseinsky, A.I. Cooper, Chem. Commun.
[
[
(
2006) 2670–2672.
[
[
11] L. Chen, Y. Honsho, S. Seki, D. Jiang, J. Am. Chem. Soc. 132 (2010) 6742–6748.
12] J. Jiang, C. Wang, A. Laybourn, T. Hasell, R. Clowes, Y.Z. Khimyak, J. Xiao, S.J.
Higgins, D.J. Adams, A.I. Cooper, Angew. Chem. Int. Ed. 50 (2011) 1072–1075.
13] L. Chen, Y. Yang, D. Jiang, J. Am. Chem. Soc. 132 (2010) 9138–9143.
14] H.J. Mackintosh, P.M. Budd, N.B. McKeown, J. Mater. Chem. 18 (2008)
122 (2000) 9298–9299.
[
[
[
[
34] C.G. Laughlin, L.J. Young, Inorg. Chem. 35 (1996) 1050–1058.
35] V.N. Nemykin, J. Laskin, P. Basu, J. Am. Chem. Soc. 126 (2004) 8604–8605.
36] Ji. Hu, R.C. Burns, J. Mol. Catal. A: Chem. 184 (2002) 451–464.
37] Y. Ding, B. Ma, Q. Gao, G. Li, L. Yan, J. Suo, J. Mol. Catal. A: Chem. 230 (2005)
[
[
5
73–578.
121–128.
[
[
15] A.M. Shultz, O.K. Farha, J.T. Hupp, S.T. Nguyen, Chem. Sci. 2 (2011) 686.
16] Z. Xie, C. Wang, K.E. deKrafft, W. Lin, J. Am. Chem. Soc. 133 (2011) 2056–2059.