Synthesis and catalytic properties of macroporous SiO2-coated CNT-sieve-composite-supported 12-tungstophosphoric acid catalysts with dual pore structure for the Baeyer–Villiger oxidation of cyclic ketones under microwave irradiation

Article abstract of DOI:10.1016/j.jcat.2019.01.007

Carbon nanotube sieves (CNTs-S) with a flutelike shape were synthesized by cutting and perforating using a chemical oxidation etching method. Using the obtained CNTs-S as a hard template, SiO₂-coated CNTs-S dual pore structures (CNTs-S@SiO₂) were formed. Finally, tungstophosphoric acid was supported on CNTs, CNTs-S, and CNTs-S@SiO₂ matrices by the incipient wetness method to obtain the corresponding supported catalysts PW₁₂/CNTs, PW₁₂/CNTs-S, and PW₁₂/CNTs-S@SiO₂. The results showed that obtained 30%PW₁₂/CNTs-S and 30%PW₁₂/CNTs-S@SiO₂ catalysts in 30 wt% dipping solution had the best catalytic performance for Baeyer–Villiger oxidation of cyclic ketones by H₂O₂. The conversion of cyclohexanone was 94% and 91%, respectively, in 5 min under microwave irradiation. Cyclopentanone and adamantanone approached quantitative conversion and the selectivity of corresponding lactones reached 100%. The dual pore structure catalyst showed very excellent cyclic stability.

Full text of DOI:10.1016/j.jcat.2019.01.007

Journal of Catalysis 371 (2019) 196–206
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Synthesis and catalytic properties of macroporous SiO2-coated CNT-
sieve-composite-supported 12-tungstophosphoric acid catalysts with
dual pore structure for the Baeyer–Villiger oxidation of cyclic ketones
under microwave irradiation
Yuli Wei a,b, Zhiwang Yang b, Guohu Zhao a, Wu Yang b,
⇑
a Provincial Key Laboratory of Gansu Higher Education for City Environmental Pollution Control, School of Chemistry and Chemical Engineering, Lanzhou City University,
Lanzhou 730070, China
b College of Chemistry and Chemical Engineering, Northwest Normal University, Lanzhou 730070, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Carbon nanotube sieves (CNTs-S) with a flutelike shape were synthesized by cutting and perforating
using a chemical oxidation etching method. Using the obtained CNTs-S as a hard template, SiO2-coated
CNTs-S dual pore structures (CNTs-S@SiO2) were formed. Finally, tungstophosphoric acid was supported
on CNTs, CNTs-S, and CNTs-S@SiO2 matrices by the incipient wetness method to obtain the correspond-
ing supported catalysts PW12/CNTs, PW12/CNTs-S, and PW12/CNTs-S@SiO2. The results showed that
obtained 30%PW12/CNTs-S and 30%PW12/CNTs-S@SiO2 catalysts in 30 wt% dipping solution had the best
catalytic performance for Baeyer–Villiger oxidation of cyclic ketones by H2O2. The conversion of cyclo-
hexanone was 94% and 91%, respectively, in 5 min under microwave irradiation. Cyclopentanone and
adamantanone approached quantitative conversion and the selectivity of corresponding lactones reached
100%. The dual pore structure catalyst showed very excellent cyclic stability.
Received 13 December 2018
Revised 2 January 2019
Accepted 3 January 2019
Keywords:
Carbon nanotube sieves
Dual pore structure
Tungstophosphoric acid
Baeyer–Villiger oxidation
Lactone
Ó 2019 Elsevier Inc. All rights reserved.
1. Introduction
practical applications. Thus, solid-matrix-supported catalysts are a
necessary choice. Porous materials are preferred matrix materials
The Baeyer–Villiger oxidation reaction (B-V reaction), one of the
most important organic reaction types, has been intensively
studied and applied [1]. This reaction allows efficient synthesis
of important esters or lactones directly from ketones [2].
Especially, lactones are important intermediates for preparing
many valuable polymers, medicines, insecticides, and natural
products [3–8].
As is known, the most commonly used oxidants in B-V reactions
are organic peroxide acids [9,10], but these substances usually are
flammable and explosive and their use poses a safety risk. Further,
use of peroxide acids will produce equimolar carboxylic acid
waste. Therefore, scientists have been trying to use H2O2 to replace
organic peroxide acids as environmentally friendly oxidants
[3,5,6,11–13].
owing to their high specific surface area and high affinity.
The main drawbacks of microporous materials are low adsorp-
tive capacity and high mass transfer resistance [14]. Macroporous
materials possess high permeability and high mass transfer rates,
but poor selectivity for solute molecules [15]. Mesoporous struc-
tures have relatively higher specific surface area, medium aperture,
and controlled hole-wall composition by comparison, but its per-
meability for larger quest molecules is unsatisfactory. It can be pre-
dicted that catalysts with mesopore–macropore dual structure can
effectively improve catalytic activity and selectivity.
There has been a report of the use of carbon nanotubes (CNTs)
to catalyze B-V reactions [16]. But conveniently obtained CNTs
have sealed ends and dense hole-walls; when they are used as cat-
alyst supporters, the channels are very easily blocked so that the
catalytic activity will change for the worse. Thus the structural
modification for CNTs is very necessary.
12-Tungstophosphoric acid (H3PW12) is a commonly used
highly effective catalyst for the B-V reaction, but the poor separa-
tion and cycling performance of the homogeneous catalyst limit its
In this paper, we first prepared flutelike carbon nanotube sieves
(CNTs-S) with medium hole aperture of the CNTs (ꢀ20 nm) as raw
materials by cutting and drilling holes in walls by a chemical etch-
ing method; then we synthesized macroporous silicon oxide (i.d.
⇑
Corresponding author.
E-mail address: yangw@nwnu.edu.cn (W. Yang).
https://doi.org/10.1016/j.jcat.2019.01.007
0021-9517/Ó 2019 Elsevier Inc. All rights reserved.

Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
197
50 nm) as a coating for CNTs-S composite (CNTs-S@SiO2); and last
we successfully obtained composite-supported 12-
heating rate of 10 °C/min under N2. XRD patterns were acquired by
a Shimadzu XRD 6000 X-ray diffractometer (Japan) with CuK irra-
a
a
tungstophosphoric acid catalyst by a simple dipping method. The
catalyst showed excellent catalytic performance for B-V oxidation
of cyclic ketones. The results also displayed that microwave irradi-
ation could remarkably improve the reactions.
diation (k = 1.5406 nm), tube voltage 40 kV, and tube current
30 mA. X-ray photoelectron spectroscopy (XPS) was carried out
with a Physical Electronics FHI-5702/ESCA/SAM multifunctional
X-ray photoelectron spectrometer (USA) with MgK
a as excitation
source and C1s binding energy 284.6 eV as internal standard. The
Brunauer–Emmett–Teller (BET) surface area and pore size distribu-
tion of the samples were measured by an N2 adsorption–desorp-
tion technique using a Micromeritics ASAP2010 surface area and
porosity analyzer (USA). A JEOL JSM-6701F ultra-high-resolution
FE scanning electron microscope (SEM, Japan) was employed to
observe sample surface morphologies and to obtain surface com-
positions (EDS). Pore morphologies of the samples were observed
by a JEOL JEM 1200 EX transmission electron microscope (TEM,
Japan).
2. Experimental details
2.1. Reagents and apparatus
Multiwall carbon nanotubes (MWCNTs) were obtained from
Shenzhen Nanoscale Harbor Co., with pipe diameter 10–30 nm
and length 0.1–200 lm. Tetraethoxysilane (TEOS, AR) was pur-
chased from BASF, Tianjin. 3-Aminopropyltriethoxysilane (APTES,
AR) was bought from Tianjin Chunjing Chemical Co. Cyclohex-
anone (AR) was acquired from Yaohua Chemical Regents Co., Tian-
jin. Cyclobutanone (98%), cyclopentanone (99%), 2-methyl
cyclohexanone 95%), 4-methyl cyclohexanone (98%), 2- tert-butyl
cyclohexanone (98%), 4-tert-butyl cyclohexanone (98%), adaman-
tanone (98%), and menthone (98%) were purchased from ABCR
GmbH & Co. KG, Germany. Di-isopropylketone (AR) and di-
isobutylketone (AR) were obtained from Union Chimique Belge,
Belgium. Hexadecyltrimethylammonium bromide (CTAB, AR),
mesitylene (TMB, CP), and H2O2 (30%, AR) were bought from Sino-
pharm Chemical Reagent Co. H3[PW12O40]ÁxH2O (H3PW12, AR) and
Na3[PW12O40]ÁxH2O (Na3PW12, AR) were acquired from Aladdin,
Shanghai. Other reagents were of analytical grade and used with-
out further purification.
A SRJX2229 tubular furnace (Experimental Electric Furnace Fac-
tory of Changsha, China) was used for template removal of the
materials and
a Midea microwave stove (Guangzhou) was
exploited to improve the catalytic reaction.
A Hewlett–Packard GC6890/MS5793 gas chromatograph mass
spectrometer was used to identify the oxidation products of the
B-V reaction (30 m  0.53 mm  0.25 mm column, mass detector,
He 1.0 lL/min, injection temperature 280 °C, programmed temper-
ature rate 15 °C/min). A Shimadzu GC-2010 gas chromatograph
(15 m  0.53 mm  0.25 Rtx-1 column, FID detector, N2
30 mL/min, air 400 mL/min, sample size 0.2 L) was used to eval-
lm
l
uate the catalytic activity of the synthesized catalysts.
IR spectra were measured by a Digilab Merlin FTS 3000 FT-IR
spectrometer (USA). Raman spectra were observed with a Bruker
RFS100/S FT Raman spectrometer (Germany) with an excitation
wavelength of 1065 nm. Thermogravimetric analysis (TG) and dif-
ferential thermal analysis (DTA) curves were recorded by a Perkin
Elmer Pyris Diamond TG/DTA 6300 thermal analyzer (USA) with a
2.2. Preparation of CNTs and CNTs-S supported tungstophosphoric acid
catalysts
2.2.1. Purification of MWCNTs and preparation of CNTs-S
According to Ref. [27], 0.5 g of MWCNTs was put into a sand
core crucible, which was transferred into a sealed Teflon-lined
Scheme 1. Preparation procedures for the catalysts PW12/CNTs-S and PW12/CNTs-S@SiO2.

198
Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
autoclave with 5 mL concentrated HNO3, and then the temperature
was kept at 160 °C for 5 h. After it was washed with distilled water
and absolute ethanol and dried at 80 °C in a vacuum oven for 5 h,
purified CNTs were obtained. Similarly, by keeping the tempera-
ture at 200 °C for 3 h, CNTs-S would be obtained.
2.3. Preparation of SiO2-coated CNTs- and CNTs-S-supported
tungstophosphoric acid catalysts
2.3.1. Preparation of SiO2-coated CNTs and CNTs-S composites
A quantity of 0.1 g of CNTs (or CNTs-S) and 1 g of CTAB were
mixed, added into 30 mL deionized water, and ultrasonically dis-
persed for 1 h; then the mixture was added into 80 mL absolute
ethanol and continually ultrasonically dispersed for 2.5 h more. A
quantity of 2 mL of ammonia and TEOS solution (1 mL TEOS
+ 40 mL ethanol) were successively dripped into the above mixture
and continuously stirred to react for 13 h. After the reaction was
finished, the mixture was centrifugally separated, washed three
times with ethanol, and dried at 80 °C in a vacuum oven for 12 h.
The presynthesized products were extracted twice with 60 mL of
ethanol–HCl mixture solution (5 d 2 M HCl + 60 mL ethanol), cen-
trifugally separated, washed three times with deionized water,
and dried at 80 °C in a vacuum oven for 12 h. The obtained prod-
ucts were respectively labeled as CNTs@SiO2 and CNTs-S@SiO2.
2.2.2. Preparation of CNTs- and CNTs-S-supported tungstophosphoric
acid catalysts
The incipient wetness method [28] was used to prepare the tar-
get catalysts. The typical preparation process was as follows: A
weight of 0.5 g of CNTs or CNTs-S was dried at 100 °C in a vacuum
oven overnight in a glass dish, into which tungstophosphoric acid
solution with a given concentration (10, 20, 30, and 40%) was
dripped until the samples were just covered. The mixtures were
allowed to stand for 4–5 h; then the redundant solution was blot-
ted carefully with filter paper and allowed to stand for 48 h more.
Last, the samples were dried at 100 °C overnight and calcined at
300 °C for 4 h in a tubular furnace to obtain the supported cata-
lysts, which were respectively labeled as w%PW12/CNTs and w%
PW12/CNTs-S (w% was the mass concentration of tungstophospho-
ric acid).
2.3.2. Preparation of CNTs@SiO2- and CNTs-S@SiO2-composite-
supported tungstophosphoric acid catalysts
NTs@SiO2 and CNTs-S@SiO2 (0.5 g) were dried at 100 °C in a
vacuum oven overnight, and then treated them with given concen-
trations of tungstophosphoric acid solution (10%, 20%, 30%, and
40%) respectively by a procedure similar to that in Section 2.2 to
A
a
887
795
b
get the final catalysts, named w%PW12/CNTs@SiO2 and w%PW12
/
CNTs-S@SiO2. The preparation procedures for the catalysts are
shown in Scheme 1.
980
1072
c
2.4. Baeyer–Villiger oxidation reaction of ketones
1705
1543
3418
2361
1065
After 0.1 mmol ketone, 40 mL 30% H2O2, and 3 mL of solvent
were mixed in a 10-mL glass flask equipped with a reflux con-
denser, a given amount of catalyst was added. Then the reaction
was carried out under microwave irradiation at a desired power
for different times without stirring. The reaction products were
analyzed by GC and GC/MS to evaluate the structures of the prod-
ucts, the conversion, and the selectivity.
For comparison, the catalytic reaction was also carried out at
80 °C using conventional oil-bath heating and magnetic stirring.
It was found that conversion of cyclohexanone could reach 90%
after 6 h with 30%PW12/CNTs@SiO2 or 30%PW12/CNTs-S@SiO2 as
catalyst.
1566
517
500
4000
3500
3000
2500
2000
1500
1000
Wavenumber (cm-1)
695
801
B
d
e
951
679
6000
D band
1420
1640
463
887
4000
988
G band
810
1088
1080
2000
c
a
b
0
4000
3500
3000
2500
2000
1500
1000
500
Wavenumber (cm-1)
0
200 400 600 800 1000 1200 1400 1600 1800 2000
Wavenumber (cm-1)
Fig. 1. IR spectra of the samples. (A) CNTs (a), CNTs-S (b), and 30%PW12/CNTs-S (c);
(B) CNTs-S@SiO2 (d) and 30%PW12/CNTs-S@SiO2 (e).
Fig. 2. Raman spectra of CNTs(a), CNTs-S (b), and 30%PW12/CNTs-S(c).

Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
199
60000
B
A
C1s
C1s
40000
C=C
O1s
W4p3
OKLL
C-C
W4d3/2
W4f
30%PW12/CNTs-S
W4d5/2
20000
C-O/C=O
20%PW12/CNTs-S
10%PW12/CNTs-S
800
COOH
0
1000
600
400
200
0
294
292
290
288
286
284
282
280
Binding energy (eV)
Binding energy(eV)
2050
2000
1950
1900
D
C
P 2p
W 4f7/2
W 4f5/2
4000
2000
145
140
135
130
45
40
35
30
Binding energy(eV)
Binding energy(eV)
Fig. 3. XPS survey spectrum of w%PW12/CNTs-S (A) and core-level spectra of C1s (B), W4f (C), and P2p (D) of 30%PW12/CNTs-S.
Table 1
Relationship between amounts of surface H3PW12 supported on the different substrates and their catalytic activities.
Catalyst
Concentration of H3PW12 (%)
XPS area ratio
W content by EDS (wt.%)
Conversion of cyclohexanone (%)
AW4f/AC=C
AW4f/ASi2p
10%PW12/CNTs-S
20%PW12/CNTs-S
30%PW12/CNTs-S
10%PW12/CNTs-S@SiO2
20%PW12/CNTs-S@SiO2
30%PW12/CNTs-S@SiO2
10
20
30
10
20
30
0.544
0.649
0.859
4.11
9.83
15.31
15.13
29.19
34.21
<10
65
94
<10
73
90
0.954
1.01
1.10
Because we used the gas oxidation method with HNO3 as sole oxi-
dant, the oxidation degree was low and the content of the obtained
surface functional groups, especially carboxyl, was relatively low.
Some weak peaks at 2900–3000 cmÀ1 indicated the existence of
alkyl groups. When H3PW12 was supported on CNTs-S, all of the
characterization peaks of the H3PW12 Keggin structure appeared.
The symmetrical stretching vibration of the PO4 tetrahedron was
located at 1072 cmÀ1, the asymmetrical stretching vibration of
the W@Od bond at 980 cmÀ1, the asymmetrical stretching vibra-
tion of the W-Ob-W bond linked with the corner-sharing oxygen
bridge 887 cmÀ1, and the asymmetrical stretching vibration of
the WAOcAW bond linked with the edge-sharing oxygen bridge
3. Results and discussion
3.1. Characterization of the catalysts
3.1.1. FT-IR spectra of the catalysts
Fig. 1 shows IR spectra of the samples. IR spectra of CNTs
(Fig. 1A(a)) and CNTs-S (Fig. 1A(b)) were very similar. –OH and
adsorbed H2O stretching vibrations appeared near 3420 cmÀ1
.
The absorption band at 1543 cmÀ1 was assigned to the E1u vibra-
tion mode, indicating the existence of a graphite structure of CNT
walls [17]. At 2360 cmÀ1 was the absorption peak of CO2 in the
environment. The band at 1630 cmÀ1 resulted from the C@C
stretching vibration of five-member rings or seven-member rings
at the bend and sealing of CNTs [18]. However, when CNTs were
oxidized, these peaks disappeared, suggesting that CNTs had been
cut off, which was advantageous to guest transfer. In addition, the
IR spectra of CNTs-S showed a weak, but well-defined C@O band
near 1700 cmÀ1, which was assigned to aldehyde or carboxyl.
at 795 cmÀ1
.
When the surfaces of CNTs-S were coated with SiO2 (Fig. 1B),
AOH stretching vibration peaks from adsorbed water, silanol
groups, and tungstophosphoric acid appeared near 3410 cmÀ1
,
and related bend vibrations at 1640 cmÀ1. The strong peak at
1088 cmÀ1 was assigned to asymmetrical stretching vibration of

200
Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
B
A
C1s
40000
30%PW12/CNTs-S@SiO2
20%PW12/CNTs-S@SiO2
20000
0
10%PW12/CNTs-S@SiO2
CNT-S@SiO2
295
290
285
280
1000
800
600
400
200
0
Binding energy(eV)
Binding energy (eV)
C
D
5200
5100
5000
4900
7000
Si 2p
P2p
6000
5000
110
105
100
95
140
135
130
125
Binding energy(eV)
Binding energy(eV)
16000
12000
8000
E
W4f7/2
W 4f5/2
4000
45
40
35
30
25
Binding energy(eV)
Fig. 4. XPS survey spectra of w%PW12/CNTs-S@SiO2 (A) and core-level spectra of C1s (B), Si2p (C), P2p (D), and W4f (E) of 30%PW12/CNTs-S@SiO2.
the SiAOASi bond, the peak at 801 cmÀ1 was its symmetrical
mode, and the peak at 463 cmÀ1 was its bend vibration. These
results indicated that the materials had been coated successfully.
In IR spectra of 30%PW12/CNTs-S@SiO2, all of the characterization
peaks of tungstophosphoric acid appeared and their intensities
were very high, indicating that a high content of tungstophospho-
ric acid was supported on the CNTs-S@SiO2 substrate.
100
a
b
c
90
80
70
d
3.1.2. Raman spectra of CNTs-S and 30%PW12/CNTs-S
Fig. 2 shows Raman spectra of CNTs (a), CNTs-S (b), and 30%
PW12/CNTs-S (c). The D band and G band of CNTs were respectively
located at 1348.4 and 1594.6 cmÀ1 and ID/IG = 1.285. When CNTs
were oxidized as CNTs-S, the intensity ratio of the D band at
1351.5 cmÀ1 and the G band at 1586.8 cmÀ1 increased to 1.643,
indicating that defects and disorder increased. But when
tungstophosphoric acid was supported, the ratio decreased slightly
0
200
400
600
800
1000
T(oC)
Fig. 5. TG curves of CNTs (a), CNTS-S (b), 30%PW12/CNTs-S (c), and 30%PW12/CNTs-
S@SiO2 (d).

Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
201
peak area, carbon from carboxyl was only 7.4% of total carbon,
which indicated that the HNO3 gas-phase oxidation employed here
produced only a small number of carboxyl groups, so that the sur-
face of the resulting CNTs-S was relatively hydrophobic and favor-
able to assembly of the surfactant molecules. In addition, at
134.81 eV appeared a P2p peak, and at 36.0 eV and 38.13 eV
respectively appeared strong W4f7/2 and W4f5/2 spin–orbit cou-
pling peaks. These results were very strong evidence of
tungstophosphoric acid support.
With increasing concentration of the tungstophosphoric acid
dipping solution (10, 20, and 30 wt%), the AW4f/AC=C area ratio grad-
ually increased (0.544, 0.649, 0.859), indicating that the surface
loading amounts of tungstophosphoric acid increased successively,
which was consistent with EDS results (Table 1).
600
300
0
30%PW12/CNTs-S
30%PW12/CNTs-S@SiO2
0.0
0.5
1.0
Fig. 4 shows the survey spectrum of w% PW12/CNTs-S@SiO2 and
the core-level spectra of C1s, Si2p, P2p, and W4fs. The C1s spec-
trum was in accordance with CNTs-S as mentioned above. Si2s
and Si2p BEs were respectively located at 158.4 and 103.86 eV,
assigned to Si4+–O. P2p BE at 134.50 eV and W4f7/2 and W4f5/2
BEs at 36.36 and 38.36 eV were respectively attributable to P5+–O
and W6+–O. As the concentration of the tungstophosphoric acid
solution increased from 10 to 20 and 30 wt%, the AW4f /ASi2p area
ratio increased from 0.964 to 1.01 and 1.10, which accorded with
EDS results (Table 1).
Relative pressure/(p/p0)
Fig. 6. N2 adsorption and desorption isotherms of 30%PW12/CNTs-S and 30%PW12
/
CNTs-S@SiO2.
to 1.564. Further observation showed that the intensities of the D
band and G band of CNTs-S were obviously higher than those of
their parent CNTs and the G band red-shifted 7.8 cmÀ1, suggesting
that the length–diameter ratio and tube length were decreased
[19]; that is, CNTs had been successfully cut off after oxidation.
3.1.4. Thermal stability of the catalysts
3.1.3. XPS analysis of the catalysts
Fig. 5 is the TG curves of the samples. CNTs(a) hardly had any
weight loss below 730 °C. CNTs-S(b) was stable below 220 °C, but
had a continuous weight loss process from 220 to 730 °C with a
weightlessness rate of ꢀ 8%, attributed to removal of substituent
groups on CNTs-S. After 730 °C, the CNTs-S skeleton started to
decompose. These results indicated that the CNTs-S skeleton was
sufficiently stable. CNTs-S-supported catalyst (c) had about 4%
weight loss below 170 °C, attributed to removal of adsorbed water,
Fig. 3 shows The XPS survey spectrum of CNTs-S with different
loading amounts of tungstophosphoric acid and high-resolution
elemental scans of C1s, W4f, and P2p of 30%PW12/CNTs-S. Fig. 3A
shows that the catalyst contains C, O, and W elements. The C1s
core-level spectrum included four components with BEs at
284.37, 284.94, 285.77, and 288.33 eV (Fig. 3B), attributable to
C@C, CAC, CAO/C@O, and COOH, respectively. According to the
Fig. 7. SEM images of CNTs (A), CNTs-S (B), and CNTs-S@SiO2 (C).

202
Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
Fig. 8. TEM images of CNTs (A, B), CNTs-S (C, D), and CNTs-S@SiO2 (E, F).
and two small weight loss ranges between 200 and 730 °C, corre-
sponding to decomposition of organic groups on CNTs-S and
removal of partial structural water of tungstophosphoric acid. In
the TG curve of SiO2-coated CNTs-S-supported catalyst (d), the
weight loss process below 350 °C resulted from removal of
adsorbed water and silica gel dehydration, and after that, it would
successively experience residual template removal, hydrous silica
and tungstophosphoric acid structural water removal, and CNTs-
S structure decomposition. Overall, CNTs-S- and CNTs-S@SiO2-
supported catalysts basically remained stable below 500 °C and
displayed sufficient thermal stability.
3.1.6. Surface morphologies of the catalysts
Fig. 7 is SEM images of CNTs (A), CNTs-S (B), and CNTs-S@SiO2
(C). Purified CNTs were very pure and possessed a very smooth sur-
face and a higher length–diameter ratio (A). After CNTs oxidation,
the diameter became greater and the surface was rougher. At the
same time, many breaches appeared on the wall and some tubes
got shorter. These characteristics made them very advantageous
for diffusion and transfer of the quest molecules when they were
used as a catalyst matrix. When CNTs-S was coated with SiO2 the
diameter became larger and better distributed.
Fig. 8 is TEM morphology images of CNTs (A, B), CNTs-S (C, D),
and CNTs-S@SiO2 (E, F). The diameter of CNTs was about 20 nm;
after oxidation, it became 25 nm, and breaches were observed.
When coated with SiO2, the composite tube had an outer diameter
of some 100 nm and an inner diameter of some 50 nm, and CNTs-S
was distinctly enveloped in an SiO2 nanotube to form a hierarchical
structure. The designed purpose had been achieved.
3.1.5. N2 Adsorption–desorption isotherms of the catalysts
Fig. 6 is the N2 adsorption–desorption isotherms of 30%PW12
/
CNTs-S and 30%PW12/CNTs-S@SiO2. It could be seen that these
two isotherms were very similar and corresponded to type IV with
an obvious hysteresis loop. The BET surface areas were calculated
to be 164.2420 and 203.8966 m2/g, and the BJH pore volumes were
1.026232 and 0.595129 cm3/g, respectively. The isotherm of 30%
PW12/CNTs-S@SiO2 still showed a sharp uptake below P/P0 = 1.0,
indicating the existence of mesopores and macropores.
Fig. 9 displays the morphologies of CNTs-S- and CNTs-S@SiO2-
supported catalysts. The catalytic active component tungstophos-
phoric acid was well dispersed on the inner and outer surfaces of
CNTs-S and CNTs-S@SiO2. This structure was very favorable to con-
tact of the reactant molecules with the active sites of the catalyst.

Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
203
Fig. 9. Morphologies of the prepared catalysts: (A, B) SEM and TEM images of 30%PW12/CNTs-S; (C, D) SEM and TEM images of 30%PW/CNTs-S@SiO2.
3.2. Catalytic activities of the catalysts for Baeyer–Villiger oxidation of
cyclic ketones
because SiO2 coating decreased the freedom of the substrate mole-
cules and prevented them from entering CNTs and contacting
active sites. However, due to a relatively high surface loading
H3PW12 amount and large specific surface area, w%PW12/CNTs-
S@SiO2 still maintained satisfactory catalytic activity; for example,
30%PW12/CNTs-S@SiO2 showed 90% conversion.
3.2.1. Optimization of the reaction conditions
In this paper, catalytic activities of the prepared catalysts for
Baeyer–Villiger oxidation of cyclic ketones with H2O2 as oxidant
under microwave irradiation were investigated. First, the experi-
mental conditions were optimized and the results are listed in
Table 2.
From Table 2, it can be found that B-V oxidation of cyclohex-
anone did not happen without catalyst or oxidant. While 30%
PW12/CNTs-S and 30%PW12/CNTs-S@SiO2 displayed excellent cat-
alytic performance, the conversion was comparable to that of the
homogeneous catalyst H3PW12 and reached 94% and 90%,
respectively.
CNTs, CNTs-S, and the corresponding SiO2-coated products
CNTs@SiO2 and CNTs-S@SiO2 did not show any catalytic properties.
But when tungstophosphoric acid was supported, all of the cata-
lysts displayed catalytic abilities to a certain extent, indicating that
tungstophosphoric acid was the catalytically active component. w%
PW12/CNTs-S had higher activity than w%PW12/CNTs, which was
because more catalytically active components could enter CNTs
through sieve pores distributed on the inner and outer surfaces
of the walls and sufficiently contacted with freely entered reactant
molecules, whereas in CNTs the substrate molecules could only
enter through the terminal opening, which limited free diffusion
of the substrate molecules. For the same reason, the catalytic activ-
GC–MS analysis results of the products of cyclohexanone con-
version indicated that the byproduct was 6-hydroxycaproic acid,
suggesting that produced
lyzed to acid (Scheme 2).
e-caprolactone was continually hydro-
This mechanism is possibly relevant to the strong acidity of
H3PW12. It ensures high catalytic activity of the catalyst, on one
hand; on the other hand, it enhances the hydrolysis efficiency of
the product to decrease the lactone selectivity. From Table 2, we
could find that with increasing reaction time, conversion of cyclo-
hexanone increased, but lactone selectivity decreased and degree
of hydrolysis of the lactone increased.
Solvent types had also a remarkable effect on the catalytic activ-
ity of the catalysts. 1, 2-Dichloroethane was the best solvent for the
B-V reaction of cyclohexanone. In this solvent, 30%PW12/CNTs-S
and 30%PW12/CNTs-S@SiO2 possessed the highest conversion,
94% and 90%. Ethanol was the best protic solvent, providing 62%
conversion. Unfortunately, other solvents did not show acceptable
conversion, especially in CH3CN and DMF, where the reaction could
not occur. In different solvents, the activity order of the catalysts
was
dichloroethane > EtOH > CCl4 > hexane > dioxane > CH3-
CN ꢀ DMF. The roles of different solvents in the catalytic reaction
ity of w%PW12/CNTs-S@SiO2 was higher than that of w%PW12
/
still need to be studied further.
CNTs@SiO2. The results also showed that the obtained catalysts
30%PW12/CNTs-S and 30%PW12/CNTs-S@SiO2 in 30 wt% dipping
solution had the highest catalytic activity in the same series of
catalysts.
From Table 2, the optimal experimental conditions were
selected as 10 mg of catalyst dosage, medium irradiation power
(400 W), 5 min of reaction time, 40 mL of 30 wt% H2O2 dosage,
and 1, 2-dichloroethane as solvent.
w%PW12/CNTs-S serial catalysts possessed higher catalytic
activity than w%PW12/CNTs-S@SiO2 serial catalysts. This was
To evaluate completely the application performance of the pre-
pared catalysts, a comparison of the catalysts with those reported

204
Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
Table 2
Catalytic performance of the prepared catalysts and optimization of the experimental conditions for B-V oxidation of cyclohexanone by H2O2.a
Catalyst
Type
H2O2 dosage (mL)
Solvent
Typec
T (min)
Conversion (%)b
Selectivity (%)b
Added (mg)
Added (mL)
—
—
40
0
40
0
40
0
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCEà
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
CCl4
hexane
CH3CN
DMF
Diox
EtOH
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1
2
3
4
5
6
0
0
91
0
94
0
90
0
0
0
0
0
0
77
0
75
0
75
0
0
0
0
H3[PW12O40]ÁxH2O
H3[PW12O40]ÁxH2O
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S@SiO2
30%PW12/CNTs-S@SiO2
MWCNTsd
CNTs
CNTs-S
CNTs@ SiO2
CNTs-S@ SiO2
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
3
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
0
0
—
10%PW12/CNTs
20%PW12/CNTs
30%PW12/CNTs
40%PW12/CNTs
<10
57
72
47
<10
65
94
68
<10
57
78
34
<10
73
90
52
31
26
0
76
70
72
—
70
75
73
—
73
68
76
—
78
75
77
49
58
—
10%PW12/CNTs-S
20%PW12/CNTs-S
30%PW12/CNTs-S
40%PW12/CNTs-S
10%PW12/CNTs@SiO2
20%PW12/CNTs@ SiO2
30%PW12/CNTs@SiO2
40%PW12/CNTs@ SiO2
10%PW12/CNTs-S@SiO2
20%PW12/CNTs-S@SiO2
30%PW12/CNTs-S@SiO2
40%PW12/CNTs-S@SiO2
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
30%PW12/CNTs-S
0
—
15
62
56
79
94
95
<10
15
37
84
94
93
34
20
74
71
75
87
88
76
75
78
75
82
5
10
15
10
10
10
10
10
10
a
b
c
Cyclohexanone amount: 0.1 mmol; microwave power: 400 W.
Conversion and selectivity measured by GC.
DCE: dichloroethane; Diox: dioxane.
d
MWCNTs: commercially available multiwalled carbon nanotubes.
Scheme 2. Reaction pathway of BV oxidation of cyclohexanone.
recently was also made and the results are given in Table 3. It was
found that our prepared catalysts had comparable or even better
performance, indicating that they had a good application potential.
From Table 4, it was found that the catalyst 30%PW12/CNTs-S
had catalytic activity for B-V oxidation of almost all cyclic ketones
except for 2-tert-butyl cyclohexanone, in which high steric hin-
drance of ortho-tert-butyl of carbonyl limited the occurrence of
the reaction. In addition, the catalyst did not show any catalytic
activity for B-V reactions of chain ketones; the reason was not
clear. However, the behavior suggested that the catalysts had good
selectivity for cyclic ketones.
3.2.2. Catalytic behavior of 30%PW12/CNTs-S for other cyclic ketone
B-V reactions
For investigating applicability of the prepared catalysts, their
catalytic behavior for other ketone B-V reactions were also studied
and typical results were tabulated in Table 4.
It is noteworthy that cyclobutanone, cyclopentanone, and
adamantane displayed nearly quantitative conversion, and lactone

Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
205
Table 3
Comparison of the prepared catalysts with those reported recently.
Catalysts
Reaction conditions
Cyclohexanone conversion (%)
Selectivity (%)
Ref.
H-type b-zeolite (Si/Al = 39)
Re complexes
H4O40SiW12/SBA-15
Sn-Beta-HT
Ag/WO3
MgO/SnO2/WO3
30% H2O2, acetonitrile, 30 °C, 3 h
48
24–57
86
30
97
86
30–40
94
52
5–69
99
62
99
94
70–100
75
[20]
[21]
[22]
[23]
[24]
[25]
[26]
1, 2-dichloroethane, 30% H2O2, 70 °C, 6 h
30% H2O2, 1, 2-dichloroethane, 70 °C, 8 h
1, 4-dioxane, 30% H2O2, 80 °C, 3 h
H2O2, acetonitrile, 80 °C, 9 h
50% H2O2 + acetic acid, 50 °C, 3 h
30% H2O2, acetonitrile, 70 °C, 6 h
Mg–Al hydrotalcite-like materials
30%PW12/CNTs-S
1, 2-dichloroethane, 30% H2O2, MW, 5 min
This work
30%PW12/CNTs-S@ SiO2
1, 2-dichloroethane, 30% H2O2, MW, 5 min
90
75
Table 4
B-V oxidation of ketones catalyzed by 30%PW12/CNTs-S.a
Substrate
Conversion (%)
94
Selectivity (%)
75
Product
99
98
>99
>99
98
86
98
>99
25
46
0
0
/
96
32
63
54
0
0
0
0
/
/
a
Conditions: 0.1 mmol ketone, 10 mg catalyst, 3 mL DCE, 5 min irradiation. Conversion and selectivity were measured by GC.
selectivity reached 100%. This is within expectation because five-
membered and six-membered rings, as we know, have the smallest
ring tension and the highest stability among cyclic compounds.
Due to the special polycyclic structure of adamantane lactone, its
stability is also high enough. High stability avoids subsequent
hydrolysis of lactone products.
3.3. Cycle life of the catalysts
Cycle performance of two kinds of catalysts was studied with
cyclohexanone as a model substrate. Used catalysts were centrifu-
gally separated, dried, and then directly used in the next cycle. The
results are shown in Fig. 10.

206
Y. Wei et al. / Journal of Catalysis 371 (2019) 196–206
[3] A. Corma, L.T. Nemeth, M. Renz, S. Valencia, Sn-zeolite beta as a heterogeneous
100
80
60
40
20
100
80
30%pW /CNTs-S
12
chemoselective catalyst for Baeyer-Villiger oxidations, Nature 412 (2001) 423–
425.
30%pW12/CNTs-S@SiO
2
[4] G.J. Ten Brink, I.W.C.E. Arends, R.A. Sheldon, The BaeyerÀVilliger reaction: new
developments toward greener procedures, Chem. Rev. 104 (2004) 4105–4123.
[5] W. Zheng, R. Tan, X. Luo, C. Xing, D. Yin, SnO2 nanoparticle-decorated graphene
oxide sheets efficiently catalyze Baeyer-Villiger oxidation with H2O2, Catal.
Lett. 146 (2016) 281–290.
[6] X. Yang, Y. Jiang, Y. Li, X. Xu, D. Li, K. Lin, Mesoporous silica beads containing
active and stable tin species for the Baeyer-Villiger oxidations of cyclic
ketones, Micropor. Mesopor. Mater. 253 (2017) 40–48.
[7] Z. Zhou, Y. Yu, P. Yu, J. Qin, S. Dai, W. Wu, Ordered mesoporous Sn-TiO2
catalysts via an evaporation induced self-assembly method for the Baeyer-
Villiger oxidation of cyclohexanone by molecular oxygen, React. Kinet. Mech.
Catal. 120 (2017) 295–305.
[8] M. Markiton, A. Szelwicka, S. Boncel, S. Jurczyk, A. Chrobok, Superactive tin(II)
triflate/carbon nanotube catalyst for the Baeyer-Villiger oxidation, Appl. Catal.
A 556 (2018) 81–91.
60
40
20
0
2
4
6
8
10
12
[9] K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hikichi, N. Mizuno,
Efficient epoxidation of olefins with ꢁ99% selectivity and use of hydrogen
peroxide, Science 300 (2003) 964–966.
Cycle time
[10] K. Kamata, K. Yonehara, Y. Nakagawa, K. Uehara, N. Mizuno, Efficient stereo-
Fig. 10. Cycle performances of the catalysts 30%PW12/CNTs-S and 30%PW12/CNTs-
S@SiO2 in catalyzing B-V oxidation of cyclohexanone. Conditions: 0.1 mmol ketone,
10 mg catalyst, 3 mL 1,2-dichloroethane, 5 min irradiation. Conversion and selec-
tivity were measured by GC.
and regioselective hydroxylation of alkanes catalysed by
polyoxometalate, Nat. Chem. 2 (2010) 478–483.
a
bulky
[11] W. Schutyser, S. Van den Bosch, J. Dijkmans, S. Turner, M. Meledina, G. Van
Tendeloo, D.P. Debecker, B.F. Sels, Selective nickel-catalyzed conversion of
model and lignin-derived phenolic compounds to cyclohexanone-based
polymer building blocks, ChemSusChem 8 (2015) 1805–1818.
[12] J.P. Mehta, D.K. Parmar, H.D. Nakum, D.R. Godhani, N.C. Desai, Synthesis,
characterization and catalytic activity of zeolite-Y immobilized oxovanadium
complex on Bayer-Villiger oxidation, Micropor. Mesopor. Mater. 247 (2017)
198–207.
[13] Z.G. Zhu, H. Xu, J.G. Jiang, Y.J. Guan, P. Wu, Sn-Beta zeolite hydrothermally
synthesized via interzeolite transformation as efficient Lewis acid catalyst, J.
Catal. 352 (2017) 1–12.
[14] Y. Tao, H. Kanoh, L. Abrams, K. Kaneko, Mesopore-modified zeolites:
preparation, characterization, and applications, Chem. Rev. 106 (2006) 896–
910.
[15] J.J. Woodford, J.P. Dacquin, K. Wilson, A.F. Lee, Better by design:
nanoengineered macroporous hydrotalcites for enhanced catalytic biodiesel
production, Energy Environ. Sci. 5 (2012) 6145–6150.
In the first three cycles, 30%PW12/CNTs-S exhibited higher cat-
alytic activity than SiO2-coated catalyst, but the conversion
decreased rapidly afterward. In the tenth cycle, it was lower than
half of the initial value. In contrast, 30%PW12/CNTs-S@SiO2 dis-
played higher cyclic stability and in the tenth cycle, the conversion
still was nearly 80%, decreased only by 12.2%. The results showed
that SiO2 coating effectively reduced loss of the catalyst.
4. Conclusions
[16] S.Y. Chen, X.T. Zhou, J.X. Wang, R.C. Luo, Q.J. Luo, L.J. Yu, H.B. Ji, Promoting the
aerobic Baeyer-Villiger oxidation of ketones over carboxylic multi-walled
carbon nanotubes, Mol. Catal. 438 (2017) 152–158.
Carbon-nanotube-sieve-supported tungstophosphoric acid and
SiO2-coated-carbon nanotube-sieve-supported tungstophosphoric
acid with dual pore structure were successfully prepared. These
two catalysts showed excellent catalytic activities for Baeyer–
Villiger oxidation of cyclic ketones with hydrogen peroxide as
oxidant. Cyclopentanone and adamantanone approached
quantitative conversion, and lactone selectivity reached 100%.
SiO2 coating effectively reduced loss of the catalytically active
component and ensured long cycle life. The proposed preparation
method provided a new and general strategy for designing and
fabricating Baeyer–Villiger oxidation reaction catalysts. In
addition, with microwave irradiation instead of traditional oil bath
heating, the reaction time dramatically decreased from several
hours to several minutes, which manifested favorable green and
energy-saving characteristics.
[17] X.L. Wu, T. Yue, R.R. Lu, D.Z. Zhu, Z.Y. Zhu, Hydrothermo-assisted
functionalization, FTIR, Raman and XPS spectra characterization of carbon
nanotubes, Spectrosc. Spectr. Anal. 25 (2005) 1595–1598.
[18] S. Jin, W.Z. Qian, F. Wei, J.C. Zhang, Characterization of functionlized carbon
nanotubes by Raman, IR spectroscopy and acid-base titrations, J. Beijing Univ.
Chem. Technol. 47 (6) (2010) 55–59.
[19] R.L. Wu, Z.Z. Shao, S.L. Chang, X.A. Zhang, H.P. Li, X.H. Li, Study on Raman
spectra of multi-walled carbon nanotubes with different parameters,
Spectrosc. Spectr. Anal. 34 (2014) 982–985.
[20] K. Taniya, R. Mori, A. Okemoto, T. Horie, Y. Ichihashi, S. Nishiyama, Role of Al3+
species in beta zeolites for Baeyer-Villiger oxidation ofcyclic ketones by using
H2O2as an environmentally friendly oxidant, Catal. Today 307 (2018) 293–300.
[21] E.C.B.A. Alegria, L.M.D.R.S. Martins, M.V. Kirillov, A.J.L. Pombeiro, Baeyer-
Villiger oxidation of ketones catalysed by rhenium complexes bearing N- or
oxo-ligands, Appl. Catal. A 443–444 (2012) 27–32.
[22] Z.W. Yang, Z.H. Ma, L.Y. Niu, G.F. Ma, H.C. Ma, Z.Q. Lei, Baeyer-Villiger oxidation
of cyclohexanone catalyzed by SBA-15, supported silicotungstic acid, Chin. J.
Catal. 32 (2011) 463–467.
[23] Q.R. Meng, J.X. Liu, G. Xiong, X.Y. Liu, L.P. Liu, H.C. Guo, Aerosol-seed-assisted
hydrothermal synthesis of Sn-Beta zeolite and its catalytic performance in
Baeyer-Villiger oxidation, Micropor. Mesopor. Mater. 266 (2018) 242–251.
[24] S. Ghosh, S.S. Acharyya, R. Singh, P. Gupta, R. Bal, Fabrication of Ag/WO3
nanobars for Baeyer-Villiger oxidation using hydrogen peroxide, Catal.
Commun. 72 (2015) 33–37.
All authors have given approval to the final version of the
manuscript.
Acknowledgment
[25] G.X. Zhang, X.C. Ren, H.B. Zhang, Y. Peng, S.Y. Gui, MgO/SnO2/WO3 as catalysts
for synthesis of
acid, Catal. Commun. 58 (2015) 59–63.
e-caprolactone over oxidation of cyclohexanone with peracetic
The authors are very grateful to the National Natural Science
Foundation of China (21665024) for its financial support.
´
[26] J.E. Olszówka, R. Karcz, B.D. Napruszewska, A. Michalik-Zym, D. Duraczynska, J.
Krys´ciak-Czerwenka, A. Niecikowska, K. Bahranowski, E.M. Serwicka, Effect of
Mg-Al hydrotalcite crystallinity on catalytic Baeyer-Villiger oxidation of
cyclohexanone with H2O2/acetonitrile, Catal. Commun. 107 (2018) 48–52.
[27] J. Ming, Y.Q. Wu, Y.C. Yu, F.Y. Zhao, Steaming multiwalled carbon nanotubes
via acid vapour for controllable nanoengineering and the fabrication of carbon
nanoflutes, Chem. Commun. 47 (2011) 5223–5225.
[28] S.S. Park, S.W. Chu, C. Xue, D. Zhao, C.S. Ha, Facile synthesis of mesoporous
carbon nitrides using the incipient wetness method and the application as
hydrogen adsorbent, J. Mater. Chem. 21 (2011) 10801–10807.
References
[1] A. Baeyer, V. Villiger, Einwirkung des caro’schen reagens auf ketone, Ber. Dtsch.
Chem. Ges. 32 (1899) 3625–3633.
[2] R.A. Michelin, P. Sgarbossa, A. Scarso, G. Strukul, The Baeyer-Villiger oxidation
of ketones: a paradigm for the role of soft Lewis acidity in homogeneous
catalysis, Coord. Chem. Rev. 254 (2010) 646–660.