Aerobic epoxidation of olefins over the composite catalysts of Co-ZSM-5(L) with bi-/tridentate Schiff-base ligands

Article abstract of DOI:10.1016/j.molcata.2010.10.021

Three tridentate or bidentate Schiff-base ligands inclusive of salicylaldehyde benzoylhydrazone (L₁), vanillic aldehyde benzoylhydrazone (L₂) and 4-methyl benzaldehyde benzoylhydrazone (L₃), have been designed, synthesized and employed to coordinate with Co²⁺ ions and ion-exchanged Co-ZSM-5 forming several Co-L complexes and Co-ZSM-5(L) composite catalysts. The catalytic epoxidation of several alkenes with dry air has been carried out at 90 °C under atmospheric pressure using Co-L complexes and Co-ZSM-5(L) catalysts (using TBHP in small amounts as the initiator). In contrast, the catalysts Co-ZSM-5(L) shows higher catalytic activity than Co-ZSM-5 itself and Co-L complexes. Among three Co-ZSM-5(L) catalysts, Co-ZSM-5(L₁) exhibits the highest activity for the selective epoxidation of alkenes. Recycling studies show the recyclability of Co-ZSM-5(L₁) as a heterogeneous catalyst, which does not lose the catalytic activity after even eight reuses appreciably. Based on the experiments, one possible reaction mechanism is proposed.

Full text of DOI:10.1016/j.molcata.2010.10.021

Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Aerobic epoxidation of olefins over the composite catalysts of Co-ZSM-5(L) with
bi-/tridentate Schiff-base ligands
B. Qi, X.-H. Lu, S.-Y. Fang, J. Lei, Y.-L. Dong, D. Zhou, Q.-H. Xia∗
Ministry-of-Education Key Laboratory for the Synthesis and Application of Organic Functional Molecules, & School of Chemistry and Chemical Engineering,
Hubei University, Wuhan 430062, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 July 2010
Received in revised form 12 October 2010
Accepted 25 October 2010
Available online 30 October 2010
Three tridentate or bidentate Schiff-base ligands inclusive of salicylaldehyde benzoylhydrazone (L1),
vanillic aldehyde benzoylhydrazone (L2) and 4-methyl benzaldehyde benzoylhydrazone (L3), have been
designed, synthesized and employed to coordinate with Co2+ ions and ion-exchanged Co-ZSM-5 forming
several Co-L complexes and Co-ZSM-5(L) composite catalysts. The catalytic epoxidation of several alkenes
with dry air has been carried out at 90 ◦C under atmospheric pressure using Co-L complexes and Co-ZSM-
5(L) catalysts (using TBHP in small amounts as the initiator). In contrast, the catalysts Co-ZSM-5(L) shows
higher catalytic activity than Co-ZSM-5 itself and Co-L complexes. Among three Co-ZSM-5(L) catalysts,
Co-ZSM-5(L1) exhibits the highest activity for the selective epoxidation of alkenes. Recycling studies
show the recyclability of Co-ZSM-5(L1) as a heterogeneous catalyst, which does not lose the catalytic
activity after even eight reuses appreciably. Based on the experiments, one possible reaction mechanism
is proposed.
Keywords:
Epoxidation
Alkene
Co-L
Co-ZSM-5(L)
Schiff-base ligands
Air
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
aerobic oxidation of ␣-pinene over Co(II)–complexes without any
co-oxidant was studied, where verbenone was the main product
The catalytic epoxidation of alkenes to produce epoxides is
an important industrial reaction, since epoxides are key building
blocks in organic synthesis [1], as well as commercially impor-
tant intermediates used in the synthesis of chiral pharmaceuticals,
pesticides, epoxy paints, agrochemicals, perfume materials and
sweeteners [2,3]. Traditionally, epoxides are produced by the
chlorohydrin process and the Halcon process [4]; however, a lot
of by-products resulted from both processes are environmen-
tally undesirable. Thus, the pursuit for environmentally-benign
methods with clean and cheap oxidant has been a challenge in
chemical industrial and academic fields. Epoxidation of alkenes
using molecular oxygen as the oxidant is in accordance with our
desires [5]. However, most epoxidations using O2 as the oxidant
need sacrificial reductants, such as O2 with H2 [6–10], O2 with
Zn powder [11] or O2 with organic alcohols or aldehydes [12–17].
Transition metal complexes, especially cobalt complexes, are well-
known catalysts for the selective oxidation of alkenes with O2, e.g.
cobalt(III) acetylacetonate (acac) for styrene, tert-butylethylene,
norbornylene and 1,1-dineopentylethylene [18], cobalt(II) complex
for terminal olefins with 2-ketones and 2-alcohols [19,20], as well
as Co(salen) complexes for alkenes with isobutyradehyde [21]. Also,
[22].
Cobalt-based heterogeneous catalysts applied for the epoxida-
tion of alkenes using O2 without any co-reductant have attracted
much attention. Tang et al. first reported that ion-exchanged
Co-faujasite-type zeolites and Co-MCM-41 could catalyze the epox-
idation of styrene with O2 in the absence of co-reductant [23,24].
Co2+-exchanged faujasite-type zeolites modified by increasing the
cobalt content or introducing alkali and alkaline cations were used
to catalyze the epoxidation of styrene, ␣-pinene and cycloalkenes
[25–27]. Co-TUD-1 exhibited a high reactivity for the epoxidation of
trans-stilbene by O2, and the selectivity of epoxides was very high
[28]. Co(II)Saloph-Y catalyzed the selective oxidation of ␣-pinene
with air to epoxide and D-verbenone with small amounts of azobis-
isobutyronitrile as the initiator, where the main products were
epoxide and verbenone [29]. Additionally, the composite catalyst
consisting of cis-MoO2-Schiff-base complexes and zeolite-Y could
efficiently catalyze the epoxidation of alkene with O2 [30–32].
Our group previously reported that the composite catalyst
Co-ZSM-5 (L) coordinated with tridentate ligands (e.g. vanillic alde-
hyde salicylhydrazone) could efficiently catalyze highly selective
epoxidation of styrene with dry air in the presence of small amounts
of TBHP as the initiator [33]. In the present study, we further
extend previous ideas to other tridentate or bidentate Schiff-base
ligands, such as three newly designed Schiff-bases salicylaldehyde
benzoylhydrazone (L1), vanillic aldehyde benzoylhydrazone (L2)
∗
Corresponding author. Tel.: +86 27 5086 5370; fax: +86 27 5086 5370.
E-mail addresses: xia1965@yahoo.com, xiaqh518@yahoo.com.cn (Q.-H. Xia).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.10.021

B. Qi et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
45
Table 1
and 4-methyl benzaldehyde benzoylhydrazone (L3), and to the
Physicochemical characterization of Co-zeolite(L) catalysts.
epoxidation of various alkene molecules with air, inclusive of ␣-
pinene, styrene, ␣-methyl styrene and cyclooctene. Fortunately,
some encouraging results have been attained.
Catalysts
Co-contenta (wt%) SiO2/Al2O3 ratiosa
Surface areab (m2/g)
Na-ZSM-5
Co-ZSM-5
Co-ZSM-5(L1) 1.16
Co-ZSM-5(L2) 1.15
Co-ZSM-5(L3) 1.15
0
1.21
25
25
25
25
25
406.5
375.6
335.2
329.1
327.3
821.3
720.4
665.2
155.7
120.6
97.5
2. Experimental
2.1. Materials
Na-Y
0
4.7
Co-Y
1.30
1.24
0
1.12
1.07
4.7
4.7
2.0
2.0
2.0
The main reagents used in the synthesis of Schiff-base ligands
and catalysts were 4-methyl benzaldehyde (99%), salicylalde-
hyde (99.5%), vanillic aldehyde (99.5%), benzoylhydrazide (99%),
cobalt(II) acetate tetrahydrate (Co(Ac)2·4H2O, 99.5%), cobalt(II)
chloride hexahydrate (CoCl2·6H2O, 99.5%), Na-ZSM-5, abso-
lute ethanol (EtOH, 99.5%). The freshly distilled solvents were
dimethylacetamide (DMA), dimethylformamide (DMF), dioxane,
cyclohexanol and toluene. Other reagents included distilled water,
␣-pinene (>98%), styrene (>99%), cyclooctene (>99%), ␣-methyl
styrene (>99%), NaClO, NaIO4, aqueous H2O2 (30%), aqueous tert-
butyl hydroperoxide (TBHP, >65%) and dry air.
Co-Y(L1)
Na-4A
Co-A
Co-A(L1)
a
Results obtained from ICP-AES.
Values obtained from N2-adsorption results.
b
N, 8.43; found: C, 50.57; H, 3.05; N, 8.38); Co-L2 is bidentate com-
plex (elemental analysis calcd (%) for C15H13N2O3CoCl2: C, 45.14;
H, 3.28; N, 7.02; found: C, 45.10; H, 3.31; N, 6.98); Co-L3 is biden-
tate complex (elemental analysis calcd (%) for C15H13N2OCoCl2: C,
49.07; H, 3.57; N, 3.81; found: C, 49.10; H, 3.60; N, 3.79).
2.2. Synthesis of Schiff-base ligands
Benzoylhydrazide (0.03 mol) was added into a 100-ml round-
bottom flask filled with 40 ml of absolute ethanol under stirring,
and the resulting mixture was refluxed at 78 ◦C until the dissolution
of benzoylhydrazide. Subsequently, 0.03 mol of salicylaldehyde (or
vanillic aldehyde, or 4-methyl benzaldehyde) was slowly dripped
into the above solution while stirring, which was refluxed for
another 3 h until the completion of reaction. After the solution
was cooled down to room temperature, a coarse product could
be recovered by filtration, which then underwent further recrys-
tallization with mixed ethanol and distilled water to obtain pure
salicylaldehyde benzoylhydrazone crystal (L1) (or vanillic aldehyde
benzoylhydrazone (L2), or 4-methyl benzaldehyde benzoylhydra-
zone (L3)) (Scheme 1).
2.4. Preparation of zeolite composite catalysts
Initially, ion exchange was carried out at 90 ◦C for 9–10 h by stir-
ring 5 g of Na-ZSM-5 in an aqueous solution (250 ml) of cobalt(II)
acetate tetrahydrate (0.375 g, 1.5 mmol). The solid product was
recovered by filtration, washed thoroughly with hot water till all
soluble acetate ion species were removed, and then dried in an oven
at 120 ◦C for 5 h. Certain amount of ion-exchanged zeolites were
separately added into hot ethanol solution of L1, L2 or L3 (1.5 mmol),
and stirred at 78 ◦C for 10 h to ensure the formation of composites.
In order to remove the excessive ligands, the recovered solids that
had been washed several times with hot ethanol were again sub-
jected to the Soxhlet extraction in ethanol under refluxing until
no free ligand was detected from the extract by the UV–vis spec-
trum. The resulting zeolite composites coordinated with ligands
were in vacuo dried at 80 ◦C for 24 h and stored in the desiccator
for catalytic uses.
L1: yield, 90.2%, M.p.: 171 ◦C; IR (KBr) (cm−1): 3270, v (N–H);
1668, v (C O); 1617, v (C N); 3070, v (O–H); 1H-NMR (DMSO-d6)
(600 MHz): 12.111 (s, 1H, N–H), 11.298 (s, 1H, Ar–OH), 8.651 (s, 1H,
N
C–H), 7.952–6.932 (m, 9H, Ar–H); UV–vis (in ethanol): 288, 299
and 330 nm.
L2: yield, 88.1%, M.p.: 192 ◦C; IR (KBr) (cm−1): 3250, v (N–H);
1655, v (C O); 1610, v (C N); 3420, v (O–H); 1H-NMR (DMSO-d6)
2.5. Structural characterizations of ligands and catalysts
(600 MHz): 12.666 (s, 1H, N–H), 9.549 (s, 1H, Ar–OH), 8.346 (s, 1H,
N
C–H), 7.906-6.843(m, 8H, Ar–H), 3.840 (s, 3H, -OCH3); UV–vis
Powder X-ray diffraction (XRD) profiles were recorded on
(in ethanol): 240 and 328 nm.
L3: yield, 86.4%, M.p.: 156 ◦C; IR (KBr) (cm−1): 3198, v (N–H);
1650, v (C O); 1610, v (C N); 1H-NMR (DMSO-d6) (600 MHz):
11.798 (s, 1H, N–H), 8.425 (s, 1H, N C–H), 7.918–7.276 (m, 9H,
Ar–H), 2.352 (s, 3H, Ar–CH3); UV–vis (in ethanol): 236 and 310 nm.
˚
a Rigaku D/MAX − IIIC diffractometer with CuK␣ (ꢀ = 1.54184 A)
operating at 30 kV and 25 mA. Infrared (IR) spectra were recorded
on a Shimadzu IR Prestige-21 Fourier Transform Infrared spec-
trophotometer. UV–vis spectra were collected on a Shimadzu
UV–vis UV-2550 spectrometer. 1H-NMR spectra of samples dis-
solved in DMSO-d6 were measured on
a Varian Inova-600
2.3. Synthesis of Co-L complexes
(600 MHz) NMR instrument using 0.03% TMS ((CH3)4Si) as an
internal standard. Thermogravimetric analysis (TGA) of Co-L and
Co-ZSM-5(L) were conducted on a TGA-7 analyzer. Elemental anal-
yses (C, H, N) of Co-complexes were conducted on an Elementar
VarioEl-III instrument. The content of Co in the sample was deter-
mined by the inductively coupled plasma (ICP) technique. The
BET specific surface area was calculated using the BET equation
with relative pressure (p/p0) at 0.05–0.25. Metal-containing com-
plexes and zeolites were determined to contain 16.4 wt% Co for
Co-L1, 15.2 wt% Co for Co-L2, 15.7 wt% Co for Co-L3, 1.21 wt% Co
for Co-ZSM-5. Co contents, BET surface areas and SiO2/Al2O3 ratios
of Co-ZSM-5(L) were also shown in Table 1. The electrochemi-
cal measurements were performed on a CHI 660A electrochemical
workstation.
Thus-synthesized ligand (0.03 mol) (L1, L2 or L3) was added into
a 100-ml round-bottom flask filled with 30 ml of absolute ethanol
under stirring, and the mixture was refluxed at 78 ◦C until the
dissolution of the ligand. Then, a solution of 0.03 mol cobalt(II) chlo-
ride hexahydrate dissolved in 30 ml absolute ethanol was slowly
dripped into the above solution, which was again refluxed for
another 2 h until the completion of reaction. After the solution was
cooled down to room temperature, a coarse product was recov-
ered by filtration, which was further recrystallized with ethanol to
obtain pure Co-L crystals (Co-L1, Co-L2 or Co-L3) (in Scheme 1). The
C, H, N contents of Co-L complexes were determined close to the
theoretical values, in which Co-L1 is tridentate complex [34] (ele-
mental analysis calcd (%) for C14H10N2O2CoCl: C, 50.62; H, 3.03;

46
B. Qi et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
R2
R1
O
R1
R3
R2
R3
CHO
NHNH2
EtOH
+
reflux
H
N
N
C
H
O
L1: R1=OH, R2=H,R3=H
L2: R1=H, R2=OCH3, R3=OH
L3: R1=H, R2=H, R3=CH3
OH
O
O
Cl
Co
Cl
Cl
Cl
O
Co
Cl Co
O
O
N
N
N
N
N
N
Co-L
Co-L
Co-L
3
2
1
Scheme 1. Synthetic routes of Schiff-base ligands and the structure of Co-L complexes.
2.6. Epoxidation of alkenes
Co-L3. Obviously, the absorption at about 315 nm for the Co-L com-
plexes is slightly shifted compared with that for the L complexes,
accompanied with the appearance of one band at about 235 nm.
Co-ZSM-5 produces two absorption bands with maxima at 246 and
520 nm (Fig. 1), in which the peak at 520 nm may represent Co2+
species coordinated octahedrally as [Co(H2O)6]2+ [35]. However,
there are recognizable three bands at 252, 310, 394 nm for Co-ZSM-
5(L1), at 245, 327, 526 nm for Co-ZSM-5(L2), and at 246, 307, 527 nm
for Co-ZSM-5(L3). For Co-ZSM-5(L), the absorption band at about
520 nm disappears or becomes very weak, suggesting the coordi-
nation of the majority of Co2+ species with Schiff-base ligands.
TG curves of Co-L and Co-ZSM-5(L) samples contain three-stage
weight losses in air (Supplementary data, Fig. 3 in the supplement).
Evidently, the first weight loss below 270 ◦C is assigned to the evap-
oration of physically adsorbed and intra-zeolite water molecules
[36], the second between 270 and 420 ◦C to the combustion of
organic ligand molecules coordinated onto the outer surface of
zeolite, and the third ranging from 420 to 750 ◦C to the oxida-
tive decomposition of organic ligand molecules adsorbed onto
the metal sites in the pores. Quantitatively, the total weight-loss
In a typical run, the catalyst (10 mg of Co-L or 100 mg of Co-
zeolite (L)), 3 mmol of alkene, 10 g of solvent (DMF) and 0.5 mmol
of TBHP were mixed in a 50-ml three-necked flask equipped with a
cryogenic-liquid condenser and with an air pump under magnetic
stirring and heated to desired temperatures. Then dry air with a sta-
ble flowrate of 30 ml/min controlled by a flowmeter was introduced
by bubbling into the bottom of reactor at atmospheric pressure.
Five hours later, the reaction was terminated and the solid cata-
lyst was filtered off. The liquid filtrate was quantitatively analyzed
by GC-900A equipped with an FID detector and an SE-30 column
(30 m × 0.25 mm × 0.25 ␮m), where hydrogen was used as carrier
gas. GC error for the determination was within 3%. Chloroben-
zene was used as an internal standard to quantify the components.
In order to investigate the stability and recyclability of Co-ZSM-
5(L1), the solid catalyst was recovered from the reaction mixture
by filtration, washed successively with distilled water and Soxhlet
extraction using ethanol under refluxing, dried at 80 ◦C for 6 h, and
reused in the next run.
3. Results and discussion
3.1. Characteristics of catalysts
a
As revealed by IR spectra of
L and Co-L complexes
(Supplementary data, Fig. 1(a) and (b) in the supplement), once
coordination with cobalt metal, the C O band of the ligand L at
about 1650 cm−1 disappears totally. Instead, characteristic C
N
bands emerges at about 1610 cm−1 and 635 cm−1 for Co-L. IR
spectra of Co-ZSM-5(L) (Supplementary data, Fig. 1(c) in the
supplement) show additional weak infrared signals in the range
of 1400–1650 and 2900–3550 cm−1, which can be associated with
the ligand molecules coordinated onto the surface of Co-ZSM-5
[33]. This coordination also leads to the emergence of one weak
infrared band at 630 cm−1, attributed to the Co-N vibration in Co-
ZSM-5(L). XRD patterns of Co-ZSM-5(L) exhibit the maintenance of
integrated ZSM-5 frameworks after ion-exchange and coordination
treatments.
b
c
d
220
260
300
340
380
420
460
500
540
580
λ/nm
UV–vis spectra of L and Co-L (Supplementary data, Fig. 2 in
the supplement) show several absorbance bands at 225, 235, 312,
418 nm for Co-L1, at 230, 331 nm for Co-L2 and at 240, 312 nm for
Fig. 1. UV–vis spectra of samples: (a) Co-ZSM-5(L1), (b) Co-ZSM-5(L2), (c) Co-ZSM-
5(L3), and (d) Co-ZSM-5.

B. Qi et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
47
percentage of Co-ZSM-5(L1), Co-ZSM-5(L2) and Co-ZSM-5(L3) are
measured to be about 5.2, 4.4 and 4.0 wt% in the temperature
range of 270–750 ◦C, respectively. It is known that the oxida-
tion of Co2+ ions exchanged into Na-ZSM-5 is difficult, because
of the strong interaction between Co2+ and anionic zeolite ZSM-
5 framework [24]. Thus, the Co species introduced into Na-ZSM-5
by ion-exchange method exist mainly in the form of Co(II) state.
with the rising order of the epoxide selectivity: cyclohexanol
(27.8%) < toluene (45.0%) < DMA (64.6%) < dioxane (71.2%) < DMF
(88.4%). As shown in Table 4, the solvent plays an important role
in the epoxidation reactions, similar to the reports in the litera-
ture [24,33]. Additionally, when adding 9 mmol of DMF into the
catalytic system using dioxane or toluene as the solvent, the con-
version of ␣-pinene and the selectivity of epoxide merely increased
to 20.1–34.3 mol% and 62.4–75.0%, respectively. However, when
adding 9 mmol of isobutylaldehyde as the reductant [25], the
conversion of ␣-pinene and the selectivity of epoxide can reach
97.4 mol% and 83.7%. Note that after the completion of reaction
the amount of oxo-DMF in the solution was detected by GC to be
ca. 0.18–0.30 mmol in optimal experiments, merely corresponding
to 1/17–1/10 of ␣-pinene quantity, i.e. the molar ratios of oxo-
DMF/epoxide lie in the range of 0.06–0.11. Thus, the DMF cannot
be recognized as a real reductant under the present experimental
conditions.
The impact of reaction temperature on the epoxidation of ␣-
pinene over Co-ZSM-5(L1) in the temperature range 50–100 ◦C
is shown in Supplementary data, Fig. 4 in the supplement. With
increasing the reaction temperature from 50 to 90 ◦C, the conver-
sion of ␣-pinene and the selectivity of epoxide are increased from
14.2 mol% to 95.3 mol% and from 87.2% to 90.1%, respectively; how-
ever, both decrease to 80.0 mol% and 86.0% at 100 ◦C.
3.2. Catalytic epoxidation of ˛-pinene with dry air
3.2.1. Epoxidation of ˛-pinene with TBHP + air over different
catalysts
The catalytic activity of several Co-L complexes and Co-
zeolite(L) catalysts for the epoxidation of ␣-pinene has been tested
in DMF at 90 ◦C, with air as the oxidant and TBHP as the initiator.
Under our experimental conditions, different catalysts with simi-
lar Co content show a great difference in the catalytic conversion of
␣-pinene (Table 2). When no catalyst is added, only 10.9 mol% of ␣-
pinene is converted with 67.4% the epoxide selectivity. Very clearly,
Co-ZSM-5 (L) catalysts are more active but less selective for the
epoxide than the corresponding Co-L complexes. For homogeneous
catalysts, the catalytic activity and the epoxide selectivity grad-
ually decrease in the sequence of Co-L1 > Co-L2 > Co-L3 > Co(Ac)2.
For heterogeneous catalysts, the catalytic activity gradually
decreases in the order of Co-ZSM-5(L1) > Co-ZSM-5(L2) > Co-
ZSM-5(L3) > Co-Y(L1) > Co-ZSM-5 > Co-A(L1), slightly different from
the descending order of the epoxide selectivity: Co-ZSM-5(L1)
≈ Co-ZSM-5(L3) > Co-A(L1) > Co-ZSM-5(L2) > Co-Y(L1) > Co-ZSM-5.
The catalyst Co-ZSM-5(L1) show higher catalytic performance than
Co-ZSM-5(L2) and Co-ZSM-5(L3), since L1 is tridentate Schiff-base
ligand while L2 and L3 are bidentate Schiff-base ligands. Fukuda
and Katsuki have reported that the Co(salen) complex bearing
electron-donating methoxy group at 3-carbons showed better
activity than one bearing methyl (hydrogen) group [37]. A sim-
ilar phenomenon can be observed from our results. Meanwhile,
the TON value characterizing the activity of metal ions decreases
gradually in the sequence of Co-ZSM-5(L1) (145.4) > Co-ZSM-
5(L2) (141.1) > Co-ZSM-5(L3) (137.6) > Co-Y (L1) (107.8) > Co-A
(L1) (103.0) > Co-ZSM-5 (100.1) > Co-L1 (74.2) > Co-L2 (62.8) > Co-L3
(59.0) > Co(Ac)2 (19.5).
3.3. Catalytic epoxidation of various alkenes with dry air over
Co-ZSM-5(L)
Aside from the catalytic epoxidation of ␣-pinene with dry air
as stated above, that of several other alkenes over Co-ZSM-5(L)
composites is as well described in Table 5. For the epoxida-
tion of styrene, the catalytic activity of those catalysts gradually
decreases in the sequence of Co-ZSM-5(L1) (93.9 mol%) > Co-
ZSM-5(L2) (92.6 mol%) > Co-ZSM-5(L3) (83.7 mol%), different from
the descending order of the epoxide selectivity: Co-ZSM-5(L3)
(91.7%) ≈ Co-ZSM-5(L2) (91.1%) > Co-ZSM-5(L1) (89.7%). For ␣-
methyl styrene, almost 100% conversions are attained over
Co-ZSM-5(L) (99.7–100 mol%); while the epoxide selectivity
shows a descending order of Co-ZSM-5(L1) (74.1%) > Co-ZSM-5(L3)
(67.9%) > Co-ZSM-5(L2) (66.1%). For cyclooctene, the epoxide selec-
tivity on Co-ZSM-5(L) catalysts has been 100%, while the substrate
conversions are relatively low with about 41.8–43.0 mol%, possibly
assigned to the diffusion limitation of larger cyclooctene molecules
in small pores of ZSM-5 zeolite and to the difficult activation of
these catalysts for cycloalkenes with molecular oxygen [27]. Gen-
erally, the activity of three Co-ZSM-5(L) catalysts can be ordered as
follows: Co-ZSM-5(L1) > Co-ZSM-5(L2) > Co-ZSM-5(L3), relevant to
the difference of electron-donating characteristics of three Schiff-
base ligands. A comparison of our results with the earlier data
reported for the epoxidation of alkenes is summarized in Table 6,
where further shows the excellent catalytic activity of the present
aerobic epoxidation system.
3.2.2. Effect of various oxidants, solvents and temperature over
Co-ZSM-5(L1)
Table 3 compares the effect of various oxidants on the epox-
idation of ␣-pinene catalyzed by Co-ZSM-5(L1) under identical
conditions. When TBHP is used as the oxidant, the conversion of
␣-pinene is 31.1 mol% in 5 h, with an epoxide selectivity of 78.8%.
Without the addition of TBHP, dry air oxidizes 26.6 mol% of ␣-
pinene at 90 ◦C, with an epoxide selectivity of 77.4%. Also, singular
TBHP (0.5 mmol) as the oxidant, the conversion of ␣-pinene and
the selectivity of epoxide are only 7.6 mol% and 79.7%. However,
once TBHP in small amounts is added into the catalytic system, the
conversion of ␣-pinene and the selectivity of epoxide are largely
elevated to 95.3 mol% and 88.4%. Compared with air, other oxidants
like NaClO, NaIO4 and H2O2 are less efficient for the epoxidation of
␣-pinene under the present experimental conditions. H2O2 con-
verts 12.6 mol% of ␣-pinene (76.1% the epoxide selectivity), while
NaIO4 achieves only 9.8 mol% of conversion (77.6% the epoxide
selectivity); however, NaClO merely oxidizes a trace of ␣-pinene.
Table 4 presents the effect of various solvents on the epox-
idation of ␣-pinene with air catalyzed by Co-ZSM-5(L1), in
which the solvents tested include DMF, DMA, dioxane, cyclo-
hexanol and toluene. Under our experimental conditions, the
conversion of ␣-pinene can be arranged in the ascending
order of toluene (6.3 mol%) < cyclohexanol (22.0 mol%) < dioxane
(22.1 mol%) < DMA (42.8 mol%) < DMF (95.3 mol%), inconsistent
3.4. Recycling studies
The catalyst Co-ZSM-5(L1) is used for recycling studies. It can be
observed from Fig. 2, the activity of the catalyst shows an apparent
reduction in the former three recycles, but remains basically unaf-
fected in the latter five recycles. The conversion of ␣-pinene and
the epoxide selectivity decrease from 95.3 to 90.6 mol% and from
88.4 to 86.1% in the former three recycles; however, in the latter
five recycles both values have maintained more or less constants of
ca. 89.0 mol% and 86.0%. ICP analyses of reaction mixture and the
used catalysts did detect the leaching of cobalt from Co-ZSM-5(L1),
leading to the reduction of the cobalt content from 1.16 to 1.05 wt%,
which could be due to the leaching of soluble Co species trapped in

48
B. Qi et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
Table 2
Epoxidation of ␣-pinene over various catalysts.a
Catalyst
␣-Pinene conversion (mol%)
Selectivity (%)
TON.
Epoxide
Verbenol
Verbenone
Other
Blank
Co(Ac)2
Co-ZSM-5
Co-L1
Co-ZSM-5(L1)
Co-L2
Co-ZSM-5(L2)
Co-L3
Co-ZSM-5(L3)
Co-Y(L1)
Co-A(L1)
10.9
20.1
68.3
68.7
95.3
53.9
92.4
52.3
90.2
75.5
62.3
67.4
79.9
71.4
90.3
88.4
89.5
84.8
87.8
86.3
84.3
85.1
9.5
6.2
7.7
1.0
2.3
1.6
4.1
2.1
3.0
3.7
3.5
18.9
11.4
15.6
5.9
6.7
6.1
9.5
7.4
7.7
9.8
4.2
2.5
5.3
2.8
2.6
2.8
1.6
2.7
3.0
2.2
2.7
–
19.5
100.1
74.2
145.4
62.8
141.1
59.0
137.6
107.8
103.0
8.7
a
Reaction conditions: Solvent, DMF(10 g); ␣-pinene, 3 mmol; catalyst, 0.01 g for homogeneous and 0.1 g for heterogeneous; tempt., 90 ◦C; time, 5 h; TBHP, 0.5 mmol; air
flowrate, 30 ml/min.
Table 3
Effect of various oxidants on the epoxidation of ␣-pinene.a
Oxidant
␣-Pinene conversion (mol%)
Selectivity (%)
Epoxide
Verbenol
Verbenone
Other
Airb
26.6
31.1
95.3
7.6
0.7
9.8
77.4
78.8
88.4
79.7
80.4
77.6
76.1
3.5
6.4
2.3
6.0
0
13.7
11.5
6.7
11.8
16.1
12.7
16.2
5.4
3.3
2.6
2.5
3.5
6.3
3.6
TBHPc
Air + TBHPd
TBHPe
NaClOc
c
NaIO4
H2O2
3.4
4.1
c
12.6
a
Reaction conditions: solvent, DMF(10 g); ␣-pinene, 3 mmol; catalyst: Co-ZSM-5(L1), 0.1 g, tempt., 90 ◦C; time, 5 h.
Air flowrate, 30 ml/min.
Oxidant, 3 mmol.
Air flowrate, 30 ml/min; TBHP, 0.5 mmol.
Oxidant, 0.5 mmol.
b
c
d
e
Table 4
Effect of various solvents on the epoxidation of ␣-pinene.a
Solvent
␣-pinene Conversion (mol%)
Selectivity (%)
Epoxide
Verbenol
Verbenone
Others
DMF
DMA
Dioxane
95.3
42.8
22.1
34.3
97.4
22.0
6.3
88.4
64.6
71.2
75.0
83.7
27.8
45.0
62.4
2.3
6.4
14.1
8.4
2.7
6.0
6.7
19.5
12.1
15.2
10.6
60.1
32.2
18.1
2.6
9.5
2.6
1.3
3.0
6.1
5.2
4.8
Dioxane + DMFb
Dioxane + IBAc
Cyclohexanol
Toluene
17.6
14.6
Toluene + DMFb
20.1
a
Reaction conditions: solvent, 10 g; ␣-pinene, 3 mmol; TBHP, 0.5 mmol; catalyst, Co-ZSM-5(L1) 0.1 g; tempt., 90 ◦C; time, 5 h; flowrate of air, 30 ml/min.
9 mmol of DMF.
9 mmol of isobutylaldehyde(IBA) as the reductant (IBA/alkene = 3 molar ratio).
b
c
Table 5
Epoxidation of various alkenes over Co-ZSM-5(L) composite catalysts.a
Run
Alkene
Catalyst
Alkene Conv. (%)
Selectivity (%)
Epoxide
A
B
Others
1
2
3
4
5
6
7
8
␣-Pineneb
Co-ZSM-5(L1)
Co-ZSM-5(L2)
Co-ZSM-5(L3)
Co-ZSM-5(L1)
Co-ZSM-5(L2)
Co-ZSM-5(L3)
Co-ZSM-5(L1)
Co-ZSM-5(L2)
Co-ZSM-5(L3)
Co-ZSM-5(L1)
Co-ZSM-5(L2)
Co-ZSM-5(L3)
95.3b
92.4
90.2
93.9c
92.6
83.7
100d
99.7
100
88.4
84.8
86.3
89.7
91.1
91.7
74.1
67.9
66.1
100
2.3
4.1
3.0
5.7
5.1
4.8
16.5
22.9
23.9
0
6.7
9.5
7.7
3.6
3.2
2.7
9.4
9.2
10.0
0
2.6
1.6
3.0
1.0
0.6
0.8
0
0
0
0
0
Styrenec
␣-Methyl styrened
Cyclooctenee
9
10
11
12
43.0
41.9
41.8
100
100
0
0
0
0
0
a
Reaction conditions: solvent, DMF(10 g); alkene, 3 mmol; TBHP, 0.5 mmol; catalyst, 0.1 g; tempt., 90 ◦C; time, 5 h; flowrate of air (30 ml/min).
A = verbenol, B = verbenone.
A = benzaldehyde, B = phenylacetaldehyde and benzoic acid as others.
A = phenylacetaldehyde, B = acetophenone.
Tempt., 125 ◦C.
b
c
d
e

B. Qi et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
49
Table 6
Comparison of different Co-containing catalysts for the epoxidation of alkene.
Catalyst
Subsa
Reaction condition
Tempt. (◦C)
Conv (%)
Sele. (%)
Ref.
Oxidant
P (atm)
NaCoY93
Co3O4
I
I
100
90
100
100
100
90
100
90
90
O2
5.4
1
1
1
1
1
4
1
1
45
63.8
100
45
45
91.8
47
95.3
93.9
43.0
71
92.1
83
65
62
90.1
100
88.4
89.7
100
26
42
25
23
24
33
27
b
Air, 30 ml min−1
O2, 6–8 ml min−1
O2, 3 ml min−1
O2, 3 ml min−1
Air, 30 ml min−1
O2
BaCoX15
II
II
II
II
III
I
Co2+-X
Co-MCM-41
Co-ZSM-5(L2)b
NaCoX96
Co-ZSM-5(L1)b
Co-ZSM-5(L1)b
Co-ZSM-5(L1)b
Air, 30 ml min−1
Air, 30 ml min−1
Air, 30 ml min−1
Present work
Present work
Present work
II
III
125
1
a
I = ␣-pinene, II = styrene, III = cyclooctene.
TBHP as the initiator.
b
the pores of ZSM-5. For the latter five recycles, no further leaching
of cobalt was detected.
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
In order to further verify whether the observed catalysis is truly
heterogeneous or not, the following control experiment is carried
out according to the method described by articles [38]. We investi-
gate the effect of reaction time on the epoxidation of ␣-pinene over
Co-ZSM-5(L1) at 90 ◦C, as illustrated in Fig. 3. Clearly, the conversion
of ␣-pinene is stepwise increased from 4.5 mol% in 15 min to 95.3%
in 5 h. When the reaction is initially run for 1 h, the obtained conver-
sion is 25.1 mol%. Subsequently, the solid catalyst is filtered off from
the reaction mixture, and the reaction is continued with the result-
ing filtrate for another 4 h. Finally, the conversion is determined to
be only 36.5 mol% with a low increment of 11.4 mol% (attributed
to the contribution of trace amount of Co2+ ions leaching from Co-
ZSM-5(L1) and the autooxidation of ␣-pinene by air). This result
testifies the heterogeneity of the composite catalyst Co-ZSM-5(L1).
conv.
conv'
sele.
control experiment
0
6
0
1
2
3
4
5
t/ h
3.5. Possible reaction mechanism
Fig. 3. Control experiment on the epoxidation of ␣-pinene. ␣-Pinene, 3 mmol; DMF,
10 g; Co-ZSM-5(L1), 0.1 g; TBHP, 0.5 mmol; flowrate of air, 30 ml/min; tempt., 90 ◦C.
In the present work, the Co-L1 complex is used to study the
catalytic epoxidation mechanism of ␣-pinene, in which the reac-
tion mixture is monitored by UV–vis spectrum as shown in Fig. 4.
The Co-L1 complex in DMF shows three strong bands centered at
275, 302 and 402 nm, appreciably different from weak signals in
C2H5OH, indicating the occurrence of the coordination between
Co(II) ions and DMF molecules [24,25]. The introduction of air into
the Co-L1 complex in DMF solution results in four clear bands
at 277, 334, 402 and 420 nm, which could be attributed to the
interaction of Co-L1 complex, DMF and molecular oxygen. This
probably means that molecular oxygen is activated by Co2+ cations
to form Co(II)-(O2−) or Co(III)-(O2−), as revealed by the litera-
ture [25,39–42]. However, the addition of ␣-pinene into the above
d
c
b
a
250
350
450
550
λ/nm
Fig. 4. UV–vis spectra of mixtures: (a) Co-L1 (in ethanol), (b) Co-L1 + DMF, (c) Co-
1 + DMF + air, and (d) Co-L1 + DMF + air + ␣-pinene.
Fig. 2. Recycling studies of Co-ZSM-5(L1) catalyst.
L

50
B. Qi et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
O
N
Cl
O
0.5
0.0
a
O
CoII
2
a
(O2)
DMF
-0.5
-1.0
-1.5
b
O
N
DMF(O2)
O
O
O
CoII
Cl
DMF
O
(I)
CoIII
Cl
N
O
(IV)
Initiator
-1.5
-1.0
-0.5
0.0
0.5
E/V
0.5
0.3
0.0
0.3
0.5
0.8
1.0
1.3
1.5
1.8
2.0
2.3
2.5
2.8
b
O-O
O-O
O
DMF
O
DMF
a
b
CoIII
Cl
CoIII
Cl
(II)
N
O
(III)
O
N
Scheme 2. Possible reaction mechanism for the epoxidation of alkene.
c
with O2 by the UV spectrum [33], appreciably revealing the initial
point of activation of molecular O2 under the present experimental
conditions.
Although the mechanism for the catalytic epoxidation of alkenes
with aerobic oxygen using Co-ZSM-5(L) is not understood very
well, the possible synergic coordination of DMF and O2 molecules
to Co(II) sites on Co-ZSM-5(L) or Co-L has been observed by
the UV–vis spectra (Fig. 4). Based on above UV–vis spectra and
CV analyses as well as the results reported earlier [23–27], one
possible reaction mechanism has been postulated in Scheme 2.
Firstly, the Co-L1 complex interacts with DMF and molecular
oxygen to form a species (I), where O2 is adsorbed onto the
surface of (I). Then, O2 is activated synergically by Co2+ com-
plex and TBHP to form active peroxy species (II), and Co2+ is
oxidized to Co3+ simultaneously. Finally, alkene molecules consec-
utively interact and coordinate with (II) to form cyclic intermediate
(IV), which subsequently releases epoxide molecules, accompanied
with that species (IV) is reduced to species (I), to end one reaction
circulation.
-1.5
-1.0
-0.5
0.0
0.5
E/V
Fig. 5. (A) Cyclic voltammetry of Co-L1 in (TEAP in DMF) solution (0.1 M): (a) in
N2 atmosphere; (b) in O2 atmosphere, potential sweep rate of 0.1 V/s. (B) Cyclic
voltammetry of mixtures: (a) 0.1 M (TEAP in DMF) solution in N2 atmosphere; (b)
0.1 M (TEAP in DMF) solution in O2 atmosphere; (c) 0.1 M (TEAP in ethanol) solution
in O2 atmosphere, potential sweep rate of 0.1 V/s.
solution does not obviously change the wavelengths of observable
absorption peaks.
To study the redox behavior of Co species in the Co-complex (Co-
L1), cyclic voltammetry (CV) analysis is conducted in an unstirred
solution (0.1 M TEAP in DMF) at a potential sweep rate of 10 mV/s,
where the range of potential is from 0.6 to −1.6 V vs. Pt. As shown
in Fig. 5(A), one reversible redox peak emerges at −0.27 V, cor-
responding to the transfer of Co(II)/Co(III) oxidation states. This
reversible change between Co oxidation states is important for
the catalytic oxidation of hydrocarbons involving molecular oxy-
gen [40]. Once air is introduced into the mixture consisting of
Co-L1 complex and 0.1 M (TEAP in DMF) solution, the redox peak at
−0.27 V disappears and one new peak emerges at −0.85 V, which
can be attributed to the reduction from coordinative state Co-L1-
DMF-O2 to Co-L1-DMF-O-O−.
Also, we study the interaction between DMF and molecular oxy-
gen using cyclic voltammetry analysis (Fig. 5(B)). Apparently, there
is no any peak when N2 is introduced into 0.1 M (TEAP in DMF)
solution. However, once air is bubbled into 0.1 M (TEAP in DMF)
solution, one reduction peak appears at −0.81 V, attributable to
the reduction from coordinative state DMF-O2 to DMF-O-O−. It
is noteworthy that the replacement of DMF by ethanol results in
the disappearance of aforesaid reduction peak. These results are in
agreement with the earlier observations on the interaction of DMF
4. Conclusions
Three tridentate or bidentate Schiff-base ligands inclusive of
salicylaldehyde benzoylhydrazone (L1), vanillic aldehyde benzoyl-
hydrazone (L2) and 4-methyl benzaldehyde benzoylhydrazone
(L3) have been synthesized and employed to coordinate with
Co2+ ions and ion-exchanged Co-ZSM-5. The composite catalyst
s Co-ZSM-5(L) were readily prepared by simply reacting Co-ZSM-5
and Schiff-base ligands in hot ethanol. The structures of Schiff-base
ligands have been characterized by IR spectra, TG, elemental analy-
sis, 1H-NMR and UV–vis spectroscopy. Co-ZSM-5(L) exhibits a high
activity for the epoxidation of several alkenes with air under atmo-
spheric pressure (using TBHP in small amounts as the initiator).
Especially, Co-ZSM-5(L1) shows the highest activity for the selec-
tive epoxidation of alkenes: 95.3 mol% the conversion and 88.4% the
epoxide selectivity for ␣-pinene, 93.9 mol% and 89.7% for styrene,
100 mol% and 74.1% for ␣-methyl styrene and 43.0 mol% and 100%
for cyclooctene. These catalysts are highly reusable heterogeneous

B. Qi et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 44–51
51
ones, e.g. Co-ZSM-5(L1) does not lose the catalytic activity after
eight reuses appreciably.
[17] (a) B. Bhatia, T. Punniyamurthy, J. Iqbal, J. Org. Chem. 58 (1993) 5518;
(b) T. Punniyamurthy, B. Bhatia, J. Iqbal, J. Org. Chem. 59 (1994) 850;
(c) S. Bhatia, T. Punniyamurthy, B. Bhatia, J. Iqbal, Tetrahedron 49 (1993) 6101;
(d) T. Punniyamurthy, J. Iqbal, Tetrahedron Lett. 38 (1997) 4463;
(e) A. De, P. Basak, J. Iqbal, Tetrahedron Lett. 38 (1997) 8383;
(f) E.N. Prabhakaran, J.P. Nandy, S. Shukla, J. Iqbal. Tetrahedron Lett. 42 (2001)
333.
Acknowledgements
The authors acknowledge the funding supports provided by the
2007 excellent mid-youth innovative project of Hubei Provincial
Education Department of China (no. T200701), by the key project
of Hubei Provincial Science & Technology Department of China (no.
2008CDA030), by the project-sponsored by SRF for ROCS, SEM of
China (no. [2007]24), and by Chen Guang Scheme for Young Scien-
tists of Wuhan City (no. 201050231087).
[18] R.A. Budnik, J.K. Kochi, J. Org. Chem. 41 (1976) 1384.
[19] A. Zombeck, D.E. Hamilton, R.S. Drago, J. Am. Chem. Soc. 104 (1982) 6782.
[20] D.E. Hamilton, R.S. Drago, A. Zombeck, J. Am. Chem. Soc. 109 (1987) 374.
[21] B. Rhodes, S. Rowling, P. Tidswell, S. Woodward, S.M. Brown, J. Mol. Catal. A 116
(1997) 375.
[22] (a) M.K. Lajunen, A.M.P. Koskinen, Tetrahedron Lett. 35 (1994) 4461;
(b) M.K. Lajunen, T. Maunula, A.M.P. Koskinen, Tetrahedron 56 (2000) 8167;
(c) M.K. Lajunen, J. Mol. Catal. A 169 (2001) 33;
(d) M.K. Lajunen, M. Myllykoski, J. Asikkala, J. Mol. Catal. A 198 (2003) 223.
[23] Q. Tang, Y. Wang, J. Liang, P. Wang, Q. Zhang, H. Wan, Chem. Commun. (2004)
440.
[24] Q. Tang, Q. Zhang, H. Wu, Y. Wang, J. Catal. 230 (2005) 384.
[25] J. Sebastian, K.M. Jinka, R.V. Jasra, J. Catal. 244 (2006) 208.
[26] M.V. Patil, M.K. Yadav, R.V. Jasra, J. Mol. Catal. A 277 (2007) 72.
[27] K.M. Jinka, J. Sebastian, R.V. Jasra, J. Mol. Catal. A 274 (2007) 33.
[28] X. Quek, Q. Tang, S. Hu, Y. Yang, Appl. Catal. A: Gen. 361 (2009) 130.
[29] T. Joseph, D.P. Sawant, C.S. Gopinath, S.B. Halligudi, J. Mol. Catal. A 184 (2002)
289.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcata.2010.10.021.
References
[30] S.N. Rao, K.N. Munshi, N.N. Rao, J. Mol. Catal. A: Chem. 145 (1999) 203.
[31] S.N. Rao, K.N. Munshi, N.N. Rao, J. Mol. Catal. A: Chem. 156 (2000) 205.
[32] S.N. Rao, N. Kathale, N.N. Rao, K.N. Munshi, Inorg. Chim. Acta 360 (2007) 4010.
[33] G. Xu, Q.-H. Xia, X.-H. Lu, H.-J. Zhan, J. Mol. Catal. A: Chem. 266 (2007) 180.
[34] A.A. Aruffo, T.B. Murphy, D.K. Johnson, N.J. Rose, V. Schomaker, Acta Cryst. C 40
(1984) 1164.
[35] C. Chupin, A.C. Veen, M. Konduru, J. Despres, C. Mirodatos, J. Catal. 244 (2006)
208.
[36] V. Mikhani, M. Moghodam, S. Tangestaninejad, B. Bahramian, A.M. Shalamzari,
Appl. Catal. A: Gen. 321 (2007) 49.
[37] T. Fukuda, T. Katsuki, Tetrahedron 53 (1997) 7201.
[38] (a) H. Ding, G. Wang, M. Yang, Y. Luan, Y. Wang, X. Yao, J. Mol. Catal. A 308
(2009) 25;
(b) B. Qi, X.-H. Lu, D. Zhou, Q.-H. Xia, Z.-R. Tang, S.-Y. Fang, T. Pang, Y.-L. Dong,
J. Mol. Catal. A 322 (2010) 73.
[39] T. Punniyyamurthy, M.M. Reddy, S.J.S. Karla, J. Iqbal, Pure Appl. Chem. 68 (1996)
619.
[1] Q.-H. Xia, H.-Q. Ge, C.-P. Ye, Z.-M. Liu, K.-X. Su, Chem. Rev. 105 (2005) 1603.
[2] S. Bhattacharjee, J.A. Anderson, J. Mol. Catal. A 249 (2006) 103.
[3] S.K. Samantaray, K. Parida, Catal. Commun. 6 (2005) 578.
[4] J. Liang, Q. Zhang, H. Wu, G. Meng, Q. Tang, Y. Wang, Catal. Commun. 5 (2004)
665.
[5] T. Punniyamurthy, S. Velusamy, J. Iqbal, Chem. Rev. 105 (2005) 2329.
[6] T.A. Nijhuis, B.J. Huizinga, M. Makkee, J.A. Moulijin, Ind. Eng. Chem. Res. 38
(1999) 884.
[7] T. Hayashi, K. Tanaka, M. Haruta, J. Catal. 178 (1998) 566.
[8] B.S. Uphade, T. Akita, T. Nakamura, M. Haruta, J. Catal. 209 (2002) 331.
[9] M.P. Kapoor, A.K. Sinha, S. Seelan, S. Inagaki, S. Tsubota, H. Yoshida, M. Haruta,
Chem. Commun. (2002) 2902.
[10] A.K. Sinha, S. Seelan, S. Tsubota, M. Haruta, Angew. Chem. Int. Ed. 43 (2004)
1546.
[11] I. Yamanaka, K. Nakagaki, K. Otsuka, J. Chem. Soc. Chem. Commun. (1995) 1185.
[12] B. Meunier, Chem. Rev. 92 (1992) 1411.
[13] M.G. Clerici, P. Inagallina, Catal. Today 41 (1998) 351.
[14] T. Iwahama, G. Hatta, S. Sakaguchi, Y. Ishii, Chem. Commun. (2000) 163.
[15] Y. Liu, K. Murata, M. Inaba, Chem. Commun. (2004) 582.
[16] W.F. Hölderich, F. Lollmer, Pure Appl. Chem. 72 (2000) 1273.
[40] H. Nishide, A. Suzuki, E. Tsuchida, Bull. Chem. Soc. Jpn. 70 (1997) 2317.
[41] J.P. Collman, Acc. Chem. Res. 10 (1977) 265.
[42] X.-H. Lu, Q.-H. Xia, S.-Y. Fang, B. Xie, B. Qi, Z.-R. Tang, Catal. Lett. 131 (2009) 517.