Selective oxidation of methacrolein to methacrylic acid on carbon catalysts

Article abstract of DOI:10.1016/j.catcom.2019.04.019

Different carbon structures (activated carbon, carbon nanotubes, graphene and graphite)were investigated for the replacement of heteropoly catalysts for the oxidation of methacrolein to methacrylic acid. Activated carbon showed the best catalytic performance with higher catalytic activity than molybdovanaphosphoric acid at lower temperatures. The catalytic performance of activated carbon was further improved by the addition of heteroatoms, such as P, B, N and S. The latter occupy the electrophilic oxygen functional groups thus preventing further oxidation of methacrylic acid and improving reaction selectivity. The best catalytic performance results were obtained on activated carbon with 10 wt% P, where methacrolein conversion, methacrylic acid selectivity and yield at 270 °C were 40.1, 70.8 and 30.3%, respectively. The results of the present work show a novel way to design non-metal catalysts for the selective oxidation of methacrolein to methacrylic acid.

Full text of DOI:10.1016/j.catcom.2019.04.019

Accepted Manuscript
Selective oxidation of methacrolein to methacrylic acid on carbon
catalysts
Lilong Zhou, Ye Sun, Bin Li, Zhengjie Li, Zhikun Zhang, Jimmy
Yun, Runjing Liu
PII:
S1566-7367(19)30133-5
DOI:
Reference:
https://doi.org/10.1016/j.catcom.2019.04.019
CATCOM 5687
To appear in:
Catalysis Communications
Received date:
Revised date:
Accepted date:
8 December 2018
22 April 2019
24 April 2019
Please cite this article as: L. Zhou, Y. Sun, B. Li, et al., Selective oxidation of methacrolein
to methacrylic acid on carbon catalysts, Catalysis Communications, https://doi.org/
10.1016/j.catcom.2019.04.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Selective Oxidation of Methacrolein to Methacrylic Acid on Carbon
Catalysts
a
Lilong Zhou , Ye Suna, Bin Lic, Zhengjie Lia, Zhikun Zhanga, Jimmy Yun*a,b
,
Runjing Liua
a
College of Chemical and Pharmaceutical, Hebei University of Science &
Technology, Shijiazhuang 050018, Hebei province, P. R. China
b School of Chemical Engineering, The University of New South Wales, Australia,
Sydney, NSW 2052, Australia
c
Hebei Key Laboratory of Heterocyclic Compounds, College of Chemical
Engineering and Materials, Handan University, Handan 056005, Hebei Province, P. R.
China
 Corresponding author. Tel: +86 0311 81668386. E-mail: lanruohe@126.com (Lilong Zhou);
jimmy.yun@unsw.edu.au (Jimmy Yun)

ACCEPTED MANUSCRIPT
Abstract:
Different carbon structures (activated carbon, carbon nanotubes, graphene and
graphite) were investigated for the replacement of heteropoly catalysts for the
oxidation of methacrolein to methacrylic acid. Activated carbon showed the best
catalytic performance with higher catalytic activity than molybdovanaphosphoric acid
at lower temperatures. The catalytic performance of activated carbon was further
improved by the addition of heteroatoms, such as P, B, N and S. The latter occupy the
electrophilic oxygen functional groups thus preventing further oxidation of
methacrylic acid and improving reaction selectivity. The best catalytic performance
results were obtained on activated carbon with 10wt% P, where methacrolein
o
conversion, methacrylic acid selectivity and yield at 270 C were 40.1, 70.8 and
30.3%, respectively. The results of the present work show a novel way to design
non-metal catalysts for the selective oxidation of methacrolein to methacrylic acid.
Keywords: activated carbon, oxidation, heteropoly catalysts, methacrolein,
methacrylic acid

ACCEPTED MANUSCRIPT
1. Introduction
Methacrylic acid (MAA) is an important chemical intermediate, which is used
for the production of methyl methacrylate, functional plastics, synthetic rubber and
paints [1]. It is traditionally produced by the Acetone Cyanhydrin process that needs
toxic HCN, corrosive H2SO4, and in which vast scale of NH4SO4 is produced as
solid waste [2]. Oxidation of methacrolein (MAL) to MAA over heteropoly
compounds, such as molybdophosphoric and molydbovanadylphosphoric is
considered as a green production process of MAA [2-9]. However, heteropoly
compounds used as catalysts have poor thermal stability, low catalytic activity and
high cost. Metal-free compounds are emerging as catalysts for selective oxidation,
oxidative dehydrogenation and oxygen reduction reactions in recent years [10-14].
Carbon-based catalysts show a promising future perspective for oxidative
dehydrogenation reactions [13-15]. For oxidation reactions, some carbon materials,
such as carbon nanotubes, graphene and nanodiamond have been used for the
oxidation of acrolein to acrylic acid and ethylbenzene to aldehyde and carboxylic acid
[16-22]. However, very little has been reported on ways to improve the catalytic
performance of carbon-based catalysts for selective oxidation reactions. If oxidation
of MAL to MAA could be efficiently catalyzed by carbon-based catalysts, such as
activated carbon (AC), the cost of catalysts will be sharply reduced and thermal
stability problems of catalysts will also be avoided.
In this work, various structures of carbon, namely: activated carbon, graphite,
graphene oxide, carbon nanotubes, and oxidized graphite were investigated for the

ACCEPTED MANUSCRIPT
oxidation of methacrolein to methacrylic acid. The carbon catalysts were also
modified by doping some nonmetal elements. Their catalytic performance was also
compared with H4PMo11VO40 (HPAV), which is a heteropoly catalyst commonly used
for the reaction of interest.
2. Experimental
2.1 Catalysts preparation
Activated carbon 1 and 2 (Sinopharm Chemical Reagent Co.) were washed by
deionized water before usage. Graphite was purchased from Sinopharm Chemical
Reagent Co., Ltd. Graphene oxide and multi-wall carbon nanotubes were purchased
from Nanjing XFNANO Materials Tech Co. Ltd.
Glucose synthesized activated carbon was prepared by the following process. 50
g glucose (Sinopharm Chemical Reagent Co., Ltd, AC) was loaded in a corundum
crucible, and the crucible was covered by tinfoil. It was calcined at 200 oC for 3 h, and
the temperature was then increased to 450 oC and kept at this temperature for 4 h.
H4PMo11VO40 (HPMV) was prepared as reported [17]. 10 g MoO3 (Sinopharm
Chemical Reagent Co., Ltd.) and 0.574 g V2O5 (Tianjin Guangfu Fine Chemical
Research Institute) were added to 200 mL deionized water, and the mixture was
subsequently refluxed. Then, 0.728 g of 85% phosphoric acid (Sinopharm Chemical
Reagent Co., Ltd.) was added to the suspension. The mixture was refluxed and
vigorous stirred for 5 h. A deep-orange solution was formed. The insoluble solid was
filtered. The solution was evaporated at 80 oC, and a bright dark red solid (HPAV) was

ACCEPTED MANUSCRIPT
obtained.
Cs2.6H1.4PMo11VO40 was prepared as previously reported [23]. 5 g HPAV was
dissolved into 10 mL deionized water and 1.42 g CsNO3 was dissolved into 10 mL
o
deionized water. The two resulting solutions were mixed and stirred for 2 h at 25 C.
This suspension was evaporated under stirring to dryness. The obtained orange solid
o
was dried at 80 C overnight. HPAV and Cs2.6H1.4PMo11VO40 were calcined under air
o
flow at 360 C before catalytic performance tests. These two catalysts were
characterized by FT-IR and powder XRD (see Figs S1, S2, ESI).
2.2 Activated carbon modification
N-doped activated carbon was prepared by the method described below. Firstly,
0.5 g melamine (Sinopharm Chemical Reagent Co., Ltd) was added into 20 mL of
o
deionized water and dissolved at 80 C. 5 g activated carbon 1 was then added, and
the mixture was stirred for 2 h. Then the water solvent in the mixture was evaporated
o
o
at 80 C. The solid obtained was dried in oven for 12 h and calcined at 450 C in a
close crucible for 4 h. N-doped activated carbon was prepared.
B-, P- and S-doped activated carbon were prepared by a similar process using
HBO3, H3PO4 and H2SO4, respectively (Sinopharm Chemical Reagent Co., Ltd). Acid
was dissolved in some deionized water, and some activated carbon was added, while
o
the mixture was stirred for 2 h and the water solvent was evaporated to dry at 80 C.
o
The solid was then dried in oven for 12 h and calcined at 350 C in a close crucible
for 4 h. The doped activated carbon was prepared and named B/AC, S/AC and xP/AC

ACCEPTED MANUSCRIPT
(x=wt%P).
2.3 Characterization of catalysts
AC was analyzed by conducting thermogravimetry and differential thermal
analysis (TG/DTA, DTG-60H, Shimadzu, Japan) on a simultaneous DTA-TG
apparatus allowing both TGA and DTA curves to be recorded simultaneously. The
heated program was from room temperature to 800 oC at the heating rate of 5 oC/min.
FT-IR spectra of catalysts were collected on a FT-IR spectrometer (Nicolet iS5,
Thermal Scientific). Powder X-ray diffraction (XRD) was performed on an X-ray
diffractometer with a step of 0.2o (Rigaku SmartLab X-ray powder diffractometer, Cu
K α radiation, 60 kV, 300 mA). The surface area and pore structure of catalysts were
measured by a Surface Area and Pore Analyzer (Mocromeritics ASAP 2460, US).
The elements concentration in solid catalysts was determined by X-ray Fluorescence
(AXIOS-MAX). The X-ray photoelectron spectroscopy (XPS) spectra were recorded
using a Kratos Axis Ultra DLD spectrometer with monochromatic Al Kα radiation.
31P NMR was performed on a Bruker Avance III (400M). Raman spectroscopy was
conducted using a laser Raman spectrometer (HR800 UV, Horiba Scientific Co. Ltd)
with a 100 mV laser, equipped with a wavelength of 514 nm and CCD detector.
2.4 Catalytic performance
Catalytic oxidation of MAL to MAA was performed under atmosphere pressure.
0.8 mL catalyst was loaded in a fix-bed micro-reactor, and the contact time was about

ACCEPTED MANUSCRIPT
2 s. Temperature was increased to reaction temperature at first, then reactants
mixtures of MAL (4.4vol.%), O2 (11.1vol.%), H2O (17.8vol.%) and N2 (balance) were
fed into the reactor to start the reaction. After the various reactions reached
steady-state rates, the products were collected and analyzed by GC (7890B, Agilent
Technologies Co. Ltd) which was equipped by a flame ionization detector, thermal
conductivity detector and a capillary column (DB-FFAP) and packed column
(TDX-01) for MAL, acetic acid and CO2, and CO detection, respectively. MAL
conversion, MAA selectivity and yield were calculated by the following Eqs. 1, 2 and
3, respectively. The carbon mass balance and byproducts selectivity were calculated
via the Eqs. S1-S4 (see ESI).The carbon mass balance was better than 98%, and the
selectivity of byproducts obtained is shown in Tables S3 and S4 (see ESI).
MAL conversion= (mole of MAL in feed gas—mole of MAL in tail gas)/ mole of
MAL in feed gas×100%
(1)
(2)
(3)
MAA selectivity=mole of MAA/mole of converted MAL×100%
MAA yield=MAA selectivity×MAL conversion
3. Results and discussion
3.1 Characterization of catalysts
3.1.1 Thermal stability of catalysts
Firstly, AC was tested for its tolerance of hash reaction conditions, so the thermal
stability of AC was tested by TG-DTA in air at 1 atm. As can be seen in Fig. S1 (ESI),
o
AC was stable when temperature was lower than 600 C. Hence, the thermal stability

ACCEPTED MANUSCRIPT
of AC was good enough for this reaction, because the reaction temperature for the
oxidation of MAL to MAA is much lower than 400 oC.
3.1.2 XRD patterns, Raman and FT-IR spectra of carbon catalysts
The XRD patterns of AC1, B/AC, S/AC and xP/AC show similar patterns, where
o
two peaks emerge at 24.5 and 43.5 o, respectively (Fig. S4, ESI), which suggest that
all these carbon catalysts belong to amorphous carbon. The XRD patterns of graphite,
graphene oxide and CNTs show the characteristic diffraction peaks at 26.5 o, 11.5o and
25.6 o (Fig. S5, ESI). The Raman spectra of AC1 and P/AC show that the modification
of AC by H3PO4 would slightly increase the disorder degree (ID/IG increased from
0.91 to 0.93) [24]. However, AC prepared from glucose has the lowest disorder degree
(ID/IG =0.73). To investigate the functional groups on carbon catalysts, FT-IR was
performed. The main functional groups emerged in the spectrum of AC 1 are vs(O-H)
(3672.5 cm-1), vs(C=O) (2975.6, 2898 cm-1), vs(-COOH) (1392 cm-1), vs(C-O-C)
(1052 cm-1) and vs(epoxy group) (870, 1229 cm-1) (Fig. S6, ESI) [25, 26]. After the
loading of H3PO4, two new IR vibration bands emerged at 1260 and 1073 cm-1 which
can be assigned to vs(P=O) and vs(P-O-P), respectively [27, 28]. It suggests that P was
successfully loaded on AC. The vibrational IR band of vs(epoxy group) shifted to
885.7 cm-1 and vs(C-O-C) was almost disappeared. Graphene oxide and graphite show
similar vibrational IR bands with AC in FT-IR spectra (Fig. S6, ESI). The vibrational
IR bands for the C-C in aromatic nucleus (1580 cm-1) and unsaturated hydrocarbon
(1626.7 cm-1), vs(-CH3) (1460, 1404 cm-1) appear in the spectra of CNTs [26]. It

ACCEPTED MANUSCRIPT
suggests that some raw materials used for the fabrication of CNTs were not totally
carbonized. The vibrational IR bands of vs(C-O-C) and vs(epoxy group) were
negligible in the spectrum of CNTs. .
3.1.3 XPS and 31P NMR analysis of carbon catalysts
Table 1. C 1s binding energy and mole ratios of different kinds of carbon in various types of
carbon structures
π-π* COOR/ C-O/ C=C/ C-C/
eV eV eV eV eV
291.3 288.7 286.3 285.0 284.6
291.2 288.3 286.3 -- 284.8
Mole ratio of carbon
atom, %
Catalysts
AC 1
AC 2
14:18:19:30:30
21:13:15:0:52
21:13:0:23:44
18:11:0:23:48
15:16:19:37:13
13:19:10:0:57
18:16:0:24:42
19:13:0:28:40
16:11:23:0.2:50
17:0:17:9:55
P/AC
291.3 287.9
291.0 287.9
--
--
285.6 284.8
285.7 284.8
B/AC
Used AC 1
Used AC 2
Used P/AC
Used B/AC
292.1 288.8 286.3 285.0 284.5
292.3 288.3 286.1 -- 284.8
292.1 288.4
291.6 288.4
--
--
285.7 284.8
285.6 284.8
Glucose prepared AC 291.2 288.6 286.2 285.1 284.7
Graphite
290.6
291.5 287.7
291.7 288.2 286.0 284.8 284.6
--
286.1 285.4 284.8
Carbon nanotubes
Graphene oxide
-- 285.5 284.8
16:8:0:21:38
17:14:27:25:17
In order to find out the functional groups and compositions on the different
carbon structures catalysts surfaces, XPS analysis were performed. Five different
kinds of carbon atoms were found on the surface of these types of carbon structures
studied (Table 1, Fig. S7). The kind of carbon with the highest binding energy
corresponds to carbon atoms with π-π*. The kind of carbon with binding energy of
288.3~290.7 eV can be assigned to carbon atoms with two or three oxygen atoms
(COOR) in lactone, carboxylic acid, carboxylic anhydride, quinine and ketone [10].
The kind of carbon emerged ~286 eV is carbon atoms with one oxygen atom in ether,
phenol and lactone [10]. The other kinds of carbon species with two carbon atoms

ACCEPTED MANUSCRIPT
refer to C=C (~285.2 eV) and C-C (~284.6 eV) in basal plane of sp2 carbon exposing
the hexagonal arrangement of carbon atoms [10]. The mole ratio of these five kinds of
carbon atoms in AC 1 was found to be 1:1.3:1.4:2.2:2.2.
With the addition of P or B, the mole ratio of COOR kind of carbon increased,
and that of C-O disappears, result similar to that seen by FT-IR. It suggests that P or B
combined with C-O as reported previously [13]. The mole ratio of the kinds of carbon
atoms with oxygen atoms increased obviously in AC 1, while these carbon atoms
mole ratios decreased significantly in P/AC and B/AC catalysts. It suggests that AC
was oxidized during assessment, while P and B could suppress the electrophilic
groups and prevent the further oxidation of carbon catalysts during evaluation [13].
No C=C was found in AC 2 (Fig. S9, ESI). The mole ratio of C-O was much higher
than that in AC 1. For AC prepared by glucose, more carbon atoms with oxygen
atoms were detected than AC 1. Only C-O was found on graphite, which suggests that
only phenol groups existed on the surface of graphite. Two kinds of carbon atoms
with oxygen atom on carbon nanotubes have similar mole ratio with these on AC.
More oxygen contenting functional groups were found in graphene. The mole ratio of
O-C=O on graphene was higher than that on AC 1 and carbon nanotubes, lower than
that on P/AC, and similar with B/AC (Fig. S10-13, ESI). After the evaluation, the
binding energy of different carbon species and their ratios changed slightly, which
indicates that carbon catalysts retain their structure under reaction conditions.
The binding energy of P in P/AC and used P/AC is about 134.4 and 134.0 eV,
respectively, which is lower than the binding energy of P in P2O5 and H3PO4,

ACCEPTED MANUSCRIPT
respectively (Fig. S11, ESI). It suggests that these P atoms interacted with
oxygen-containing functional groups on the surface of AC and some H3PO4 was not
degraded during calcination. The binding energy of B in B/AC and used B/AC can be
assigned to B2O3 [14]. The binding energy of Mo in AC 1 was about 233.6 eV, which
can be related to Mo in MoO3 (Fig. S7, S8, ESI). To further investigate P species in P
doped AC, 31P NMR was performed. As seen in Fig. S14 (ESI), three chemical shift
peaks emerged at -9.4, -23.34 and -37.39 ppm which correspond to P in
pyrophosphoric acid, metaphosphoric acid and P2O5, respectively [29, 30]. All these
phosphorus species were generated from the dehydration of H3PO4 during calcination.
3.2. Catalytic performance of different carbons and other materials
Table 2. Catalytic performance results of different carbons materials at 290 oC
BET surface
Catalysts
Conversion%
Selectivity%
Yield%
acrea/m2·g-1
1018.1
972.0
AC 1
AC 2
40.9
13.2
23.0
9.5
33.3
21.0
24.8
3.6*
97.3*
13.6
2.8
5.7
0.3
0.7
Glucose prepared AC
Graphite
2.6
10.2
Graphene oxide
Multi-wall carbon
nanotubes
296.7
0.7
218.1
0
0
0
H4PMo11VO40
Cs2.6H1.4PMo11VO40
Blank
3.3
167.6
--
6.7
84.8
0
39.8
44.2
0
2.7
37.5
0
*No MAL conversion on graphite and graphene at 290 oC. The conversion of MAL and selectivity
to MAA on these two catalysts correspond to 310 oC.
As shown in Table 2, three different kinds of AC, namely: AC 1, AC 2 and
glucose synthesized AC, show good catalytic performance for the oxidation of MAL

ACCEPTED MANUSCRIPT
to MAA. It should be noted that some metal elements exist in activated carbon. The
XPS results showed that there are about 0.69 atom% Mo, 2.2 atom% Co and 0.18
atom% Bi in AC 1, and 0.03 atom% Mo, 0.14 atom% Co and 0.01 atom% Bi in AC 2.
According to the XPS results, Mo was in the form of MoOx and Mo atoms in carbon
framework, and Bi and Co were mainly in the form of their metal oxides [31-33]. AC
2 contains less metal atoms than AC 1, and exhibits lower MAL conversion (13.2%)
and much lower MAA yield (2.8%) than AC 1. It was speculated whether these metal
elements are active sites for this reaction. To probe for this, activated carbon was
prepared by glucose to make sure that no metal elements were contained. This glucose
prepared AC resulted in a higher MAL conversion (24.8%) and MAA yield (5.7%)
than AC 2 (Table 2). These results illustrate that these specific metal atoms in
activated carbon should not be considered as active sites for this reaction. Graphite
shows low MAL conversion and MAA yield on activated carbon, which might be due
rather to its inert surface than its low surface area. Graphene shows also low MAL
conversion (0.7%), but the highest MAA selectivity (Table 2). No MAL conversion
was detected on carbon nanotubes. The reason might be the absence of epoxy groups
on CNTs which was suggested to be the active site for the oxidation of acrolein to
acrylic acid.
It was reported [17, 19] that oxygen-containing functional groups are active sites
for oxidation reactions. HPMV shows much lower MAL conversion (6.7%) and
slightly higher MAA yield (2.7%) than the activated carbon, whereas the catalytic
performance of activated carbon is much lower than that of Cs2.6H1.4PMo11VO40 at

ACCEPTED MANUSCRIPT
290 oC.
3.3 The effect of temperature on the catalytic performance of activated carbon, HPMV
and Cs2.6H1.4PMo11VO40 solids
Figure 1. MAL conversion (%), MAA selectivity (%) and MAA yield (%) on AC 1,
HPAV and Cs2.6PMo11VO40 solids at different reaction temperature (230-370 oC)
To effect of reaction temperature on the catalytic performance of HPAV,
Cs2.6H1.4PMo11VO40 and AC was evaluated in the 270-370 oC range as depicted Fig. 1.
o
The results show that AC starts to catalyze this reaction at 230 C, while HPAV and
Cs2.6H1.4PMo11VO40 show no activity and much lower activity, respectively, at 250 oC.
More precisely, MAA yield is 1.3% on HPAV and 3.0% on Cs2.6H1.4PMo11VO40
compared to 8.8% on AC. With the increase of reaction temperature, MAL conversion

ACCEPTED MANUSCRIPT
over these 3 catalysts increases. MAL conversion over AC and HPAV increases from
3.4 to 52.4% and 0 to 50.2%, respectively, with increasing reaction temperature from
o
230 to 370 C. In the case of Cs2.6H1.4PMo11VO40, MAL conversion increases from 0
o
to 100% with increasing reaction temperature from 230 to 330 C. Meanwhile, MAA
o
o
selectivity on HPAV increases from 0% (230 C) to 83.8% (330 C). The highest
MAA selectivity was obtained over AC and Cs2.6H1.4PMo11VO40, ca. 33.3 and 44.2%,
respectively, which correspond to 290 oC. With further increasing reaction
temperature, the MAA selectivity over AC and Cs2.6H1.4PMo11VO40 was found to
decrease. Regarding MAA yield, AC shows higher values than HPAV and
o
Cs2.6H1.4PMo11VO40 when the reaction temperature is lower than 320 and 270 C,
respectively. The results shown in Fig. 1 clearly indicate that AC shows better
catalytic activity towards MAA than heteropoly acid catalysts at T≤270 oC.
3.4 Modification of activated carbon by different heteroatoms
Table 3. Catalytic performance at 290 oC over activated carbon doped by different elements
Catalysts doped by
BET surface area/m2·g-1
Conversion% Selectivity% Yield%
different elements
Un-doped AC
N-AC
1018.1
936.5
68.8
40.9
36.7
19.5
34.1
50.2
33.3
21.0
49.6
64.6
40.8
13.6
7.7
B-AC
9.7
P-AC
929.4
1008.1
22.0
20.5
S-AC
To enhance the catalytic performance of AC, some non-metal elements were
used to modify AC. Oxygen containing functional groups mainly include two kinds,
namely, nucleophilic (ketone, quinine and carboxylic anhydride) and electrophilic

ACCEPTED MANUSCRIPT
groups (such as phenol, ether, lactone, carboxylic acid and aldehyde) [11]. The
electrophilic groups have stronger oxidizing ability than nucleophilic groups. The
electrophilic groups tend to further oxidize the reactants [14]. Hence, we tried to
suppress these electrophilic groups by the addition of some heteroatoms, such as N, B,
P and S, to increase MAA selectivity. According to the results shown in Table 3, with
the addition of N, both MAL conversion and MAA selectivity decrease. The addition
of N in carbon catalysts can increase their charge and spin density, which would
reduce the oxidizability and promote the formation of reactive oxygen species, which
may lead to the further oxidation of MAA [17]. The MAL conversion and MAA yield
decrease to 19.5% and 9.7% with the addition of B, respectively (Table 3). In the case
of B addition, this can react with the electrophilic groups and suppress their reaction
with MAL or MAA, so MAA selectivity would increase and MAL conversion
decrease [13]. The addition of P has also the same ability [13]. Surprisingly, MAL
conversion decreases by ~ 6%, and MAA yield increases to 22% with the addition of
P in AC (Table 3). XPS results showed that the amount of electrophilic groups in B-
and P- doped AC decreased.
Sulfur (S) was also doped into AC and results are presented in Table 3. MAL
conversion (50.2%), MAA selectivity (40.8%) and MAA yield (20.5%) on S/AC were
all improved. The results demonstrate that the addition of P and S enhance the
catalytic performance, and obviously P improves MAA selectivity. Based on this
behavior, P was selected for further studies, aiming to optimize its loading for
optimum catalytic performance.

ACCEPTED MANUSCRIPT
3.5 Optimization of P amount in AC
P-doped AC showed the best catalytic performance. With the increase of P to 2
wt%, MAL conversion increases from 38.4 to 48.4%, and it decreases to 16.1% with
o
P content further increasing to 20 wt% at 270 C (Fig. 2). MAA selectivity increases
o
from 30.3 to 80.8% with the increase of P content from 0 to 20% at 270 C, whereas
MAA yield increases from 11.6% (AC 1) to 30.3% (10%P/AC) and then decreases to
13% (20%P/AC) at 270 oC. MAL conversion and MAA yield on these catalysts show
o
similar trends at 290 and 310 C. The only difference is that the highest MAL
o
conversion and MAA yield are obtained on 5 wt%P/AC at 290 and 310 C,
respectively. According to Frank et al. [13, 19], oxidation of acrolein to acrylic acid
includes three steps. Acrolein is first adsorbed by ketone like groups, then interacts
with electrophilic groups and is get oxidized to acrylic acid, and finally acrylic acid is
desorbed from the catalyst’s surface and the electrophilic groups are get re-oxidized
by O2. Therefore, the decrease of MAL conversion with increasing P content after a
given level is suggested to be due to the following reasons. P firstly reacts with
electrophilic groups and reduces MAL conversion due to the activity of these groups.
By the increase of P content, the electrophilic groups interact with P, and the excess of
it interacts with the nucleophilic groups leading to the further decrease of MAL
conversion due to these groups [13].
XPS results showed that much more electrophilic groups than nucleophilic
groups were present in AC, for instance, the molar ratio of electrophilic to

ACCEPTED MANUSCRIPT
nucleophilic groups was 4.3:1 in AC 1. Therefore, P firstly is combined with
electrophilic groups to suppress the reaction between them and MAL, and the further
oxidation of MAL to byproducts is avoided, which leads to the increase of MAA
selectivity. With the increase of P content, more and more electrophilic groups are
combined with P. After most of these groups interact with P compounds, the latter
start to interact with nucleophilic groups, which lead to the decrease of effective
adsorption sites of MAL. The latter then reacts with excess electrophilic groups, and
the MAA selectivity decreases. With the increase of P content, MAL conversion
increased, which also suggests that dopant P might provide new active sites for this
reaction [28]. More research work is needed to better understand the mechanism of
how P compounds improve the catalytic performance of carbon catalysts for the
oxidation of an unsaturated aldehyde to the unsaturated carboxylic acid.

ACCEPTED MANUSCRIPT
Figure 2. MAL conversion (%), MAA selectivity (%) and yield (%) on AC with
different P content at 270, 290 and 310 oC
4. Conclusions
Methacrolein was oxidized to methacrylic acid by activated carbon and graphite.
P-doped activated carbon showed much higher catalytic activity than H4PMo11VO40
and much higher MAA selectivity than Cs2.6H1.4PMo11VO40. The methacrylic acid
selectivity can be significantly improved by doping activated carbon with P, B and S.
The best catalytic performance was obtained on activated carbon doped by 10 wt% P,
on which methacrolein conversion methacrylic acid selectivity and yield were 42.8%,
70.8% and 30.3%, respectively. The mechanism of this reaction catalyzed by carbon

ACCEPTED MANUSCRIPT
catalysts and the improvement of catalytic performance by the addition of P, B and S
still needs further investigations for better understanding.
Acknowledgements
The authors gratefully acknowledge the financial support by the national natural
science foundation of China (NO. 21106032), Natural Science Foundation of Hebei
Province (No. B2016109028) and the education department of Hebei province (NO.
ZD2016064; NO. 2017102002381).
References
[1] E.C. Hess, R. Schlögl, Nanostructured Catalysts - Selective Oxidations, RSC 2011,
chapter 7.
[2] K. Nagai, Appl. Catal. A-Gen. 221 (2001) 367–377.
[3] M. Misono, Appl. Catal. 64 (1990) 1-30.
[4] N. Mizuno, M. Misono, Chem. Rev. 98(1998) 199-217.
[5] M. Misono, Chem. Comm. (2001) 1141-1152.
[6] L. Zhou, L. Wang, S. Zhang, R. Yan, Y. Diao, J. Catal. 329 (2015) 431-440.
[7] H. Zhang, R. Yan, L. Yang, Y. Diao, L. Wang, S. Zhang, Ind. Eng. Chem. Res.
52(2013) 4484-4490.
[8] L. Zhou, L. Wang, Y. Diao, S. Zhang, R. Yan, Molecular Catal. 433(2017)
153-161.
[9] L. Zhou, L. Wang, Y. Cao, Y. Diao, R. Yan, S. Zhang, Molecular Catal. 438(2017)

ACCEPTED MANUSCRIPT
47-54.
[10] D.S. Su, S. Perathoner, G. Centi, Chem. Rev. 113(2013) 5782-5816.
[11] D.S. Su, J. Zhang, B. Frank, A. Thomas, X. Wang, ChemSusChem, 3(2010)
169-180.
[12] Z. Guo, Z.-H. Huang, M. Wang, F. Kang, Catal. Sci. Technol. 5(2015) 827-829.
[13] B. Frank, J. Zhang, R. Blume, R. Schlögl, D.S. Su, Angew. Chem. Int. Ed.
48(2009) 6913-6917.
[14] J. Zhang, X. Liu, R. Blume, A. Zhang, R. Schlögl, D.S. Su, Science,
322(5898)( 2008) 73-77.
[15] B. Frank, M. Morassutto, R. Schomöcker, R. Schlögl, D.S. Su, ChemCatChem,
2(2010) 644-648.
[16] B. Frank, R. Blume, A. Rinaldi, A. Trunschke, R. Schlögl,
DGMK-Tagungsbericht 2011-2.
[17] Y. Gao, G. Hu, J. Zhong, Z. Shi, Y. Zhu, D.S. Su, J. Wang, X. Bao, D. Ma,
Angew. Chem. Int. Ed. 52(2013) 2109-2113.
[18] G. Centi, S. Perathoner, Catal. Today, 143 (2009) 145-150.
[19] B. Frank, R. Blume, A. Rinaldi, A. Trunschke, R. Schlögl, Angew. Chem. Int. Ed.
50(2011), 10226-10230.
[20] H. Yu, F. Peng, J. Tan, X. Hu, H. Wang, J. Yang, W. Zheng, Angew. Chem.
123(2011) 4064-4068.
[21] B. Zhong, H. Liu, X. Gu, D.S. Su, ChemCatChem, 6(2014) 1553-1557.
[22] A. Fortunya, J. Fontb, A. Fabregat, Appl. Catal. B-Environ. 19 (1998) 165-173.

ACCEPTED MANUSCRIPT
[23] L. Zhou, L. Wang, Y. Diao, R. Yan, S. Zhang. Mol. Catal. 433 (2017) 153-161.
[24] Y. Guo, Z. Zeng, Y. Zhu, Z Huang, Y. Cui, J. Yang. Appl. Catal. B-Environ. 220
(2018) 635-644.
[25] L. A. Silva, S. M. S. Borges, P. N. Paulino, M. A. Fraga, S. T. Oliva, S. G.
Marchetti, M. C. Rangel. Catal. Today 289 (2017) 237-248.
[26] Y. Yan, J. Miao, Z. Yang, F. X. Xiao, H. B. Yang, B. Liu, Y. Yang. Chem. Soc.
Rev. 44 (2015) 3295-3346
[27] M. A. Patel, F. Luo, M. R. Khoshi, Rabie E., Zhang Q., Flach C. R., Mendelsohn
[28] R., Garfunkel E., Szostak M., He H. ACS Nano. 10(2)(2016) 2305-2315.
[29] R. Ye. Master thesis, Guangdong University of Technology 2011.
[30] C. Zhai, L. Liu, W. Lu. Chemical Reagent 6(1981) 324-332.
[31] B. Brox, I. Olefjord, Surf. Interface Anal. 13(1988) 3-6.
[32] Z.M. Hanafi, F.M. Ismail, S.A. Nasser, Phys. Inorg. Chemie (Leipzig), 267(1986),
667-669.
[33] J.P. Bonnelle, J. Grimblot, A. D'huysser, J. Electron Spectrosc. Relat. Phenom.
7(1975), 151-162.

ACCEPTED MANUSCRIPT
Highlight
 Oxidation of methacrolein to methacrylic acid can be catalyzed by carbon
 Catalytic performance of activated carbon can be improved by the addition of P
 Activated carbon showed the best catalytic activity than other carbon materials