Synergy effects of Al2O3 promoter on a highly ordered mesoporous heterogeneous Rh-g-C3N4 for a liquid-phase carbonylation of methanol

Article abstract of DOI:10.1016/j.apcata.2019.117209

The effects of Lewis acid Al₂O₃ on Rh-incorporated mesoporous g-C₃N₄ (AlRh-CN) were investigated to verify an enhanced activity and stability in a liquid-phase carbonylation of methanol. The novel AlRh-CN with 2.0 wt%Al (Al(2)Rh-CN) revealed a higher yield of acetic acid (AA) and methyl acetate (MA) with a small Rh leaching. The enhanced activity on the optimal Al(2)Rh-CN was attributed to the highly dispersed and partially reduced Rh nanoparticles by preserving initial oxidation states as well as by maintaining its stronger interaction on the Al-modified g-C₃N₄ surfaces. Its superior activity was originated from the highly dispersed Lewis acid Al³⁺ sites assigned to Al₂O₃ phases due to an increased reverse esterification reaction rate. The higher activity on the Al(2)Rh-CN was originated from synergy effects such as a larger Lewis acid Al₂O₃ sites acting as a pillaring material to increase specific surface area of Al(2)Rh-CN and to increase a reverse esterification reaction.

Full text of DOI:10.1016/j.apcata.2019.117209

AppliedCatalysisA,General585(2019)117209
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Synergy effects of Al2O3 promoter on a highly ordered mesoporous
heterogeneous Rh-g-C3N4 for a liquid-phase carbonylation of methanol
Da Mi Kima,1, A Rong Kima,1, Tae Sun Changb, Hyun Mo Kooa,c, Jung Kyu Kima,⁎
,
Gui Young Hana, Chae-Ho Shinc, Jong Wook Baea,⁎
a School of Chemical Engineering, Sungkyunkwan University (SKKU), 2066 Seobu-ro, Jangan-gu, Suwon, Gyeonggi-do, 16419, Republic of Korea
b Korea Research Institute of Chemical Technology (KRICT), 141 Gajeong-ro, Yuseong-gu, Daejeon, 34114, Republic of Korea
c Department of Chemical Engineering, Chungbuk National University, Chungdae-ro 1, Cheongju, Chungbuk 28644, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Keywords:
Liquid-phase carbonylation
Methanol
Rh nanoparticles
Mesoporous graphitic carbon nitride (mpg-
C3N4)
The effects of Lewis acid Al2O3 on Rh-incorporated mesoporous g-C3N4 (AlRh-CN) were investigated to verify an
enhanced activity and stability in a liquid-phase carbonylation of methanol. The novel AlRh-CN with 2.0 wt%Al
(Al(2)Rh-CN) revealed a higher yield of acetic acid (AA) and methyl acetate (MA) with a small Rh leaching. The
enhanced activity on the optimal Al(2)Rh-CN was attributed to the highly dispersed and partially reduced Rh
nanoparticles by preserving initial oxidation states as well as by maintaining its stronger interaction on the Al-
modified g-C3N4 surfaces. Its superior activity was originated from the highly dispersed Lewis acid Al3+ sites
assigned to Al2O3 phases due to an increased reverse esterification reaction rate. The higher activity on the Al(2)
Rh-CN was originated from synergy effects such as a larger Lewis acid Al2O3 sites acting as a pillaring material to
increase specific surface area of Al(2)Rh-CN and to increase a reverse esterification reaction.
Lewis acid Al2O3promoter
1. Introduction
[1,2,9,10]. Among the heterogenized homogeneous catalysts, the basic
graphitic carbon nitride (g-C3N4), which has been widely used for a
A typical liquid-phase carbonylation reaction of methanol has been
widely investigated to selectively produce acetic acid (AA) using var-
ious metal-organic complexes such as Rh or Ir-based homogeneous
catalysts [1–3]. Those homogeneous metal-organic complexes have
been successfully commercialized by Monsanto [1] and Cativa process
[2] as well. However, those commercialized homogeneous metal-or-
ganic complexes revealed the difficult separation problems from liquid
products by forming precipitated metal complexes, and it further
caused an excessive energy consumption as well as cost for their se-
parations. Therefore, other alternative methods to synthesize AA from
various feedstocks have been largely investigated by using various
heterogeneous catalysts through co-activation of CH4 and CO2 [4,5] or
oxidation of hydrocarbons such as ethylene or ethane feedstock [6,7].
To overcome the disadvantages of the homogeneous metal-organic
complexes and novel routes by using heterogeneous catalysts, various
heterogenized homogeneous catalysts such as active metals supported
on some inorganic metal oxides or polymers [8] have been largely in-
vestigated for an easy separation as well as high AA selectivity. Among
them, Acetica process using homogeneous Rh-complexes immobilized
on the specific polymer support has been recently commercialized
photo-catalysis [11,12], bio-imaging [13], metal-free catalytic reac-
tions [14], and porous supporting material [15] has been newly pro-
posed from our recent research [10] as a novel basic supporting ma-
terial for the preparation of Rh-incorporated heterogeneous catalysts
for a liquid-phase carbonylation reaction of methanol. The g-C3N4
generally contains the various basic nitrogen-derived functional groups
with high electron density, and it can effectively donate electrons to the
incorporated transition metals [9,15]. In addition, the highly ordered
mesoporous and chemically stable g-C3N4 structures with its larger
specific surface area can also enhance the electron-rich and basic sur-
faces, which can act as new active sites for a facile adsorption of re-
actants for liquid-phase reactions.
In the present investigation, the highly ordered mesoporous g-C3N4
(mpg-C3N4) having a specific surface area above 900 m2/g was suc-
cessfully synthesized by using
a hard silica template of SBA-15
[10,16,17], and the heterogeneous basic mpg-C3N4 was further in-
corporated with Rh metal and Al metal oxide, which increased the
dispersion of active Rh nanoparticles and generated additional Lewis
acid Al2O3 sites. The surface properties of the Rh nanoparticles as well
as acid natures of Al2O3 promoter, which was well distributed in the
⁎ Corresponding author.
E-mail addresses: legkim@skku.edu (J.K. Kim), finejw@skku.edu (J.W. Bae).
1 Authors were equally contributed.
https://doi.org/10.1016/j.apcata.2019.117209
Received 17 May 2019; Received in revised form 2 August 2019; Accepted 15 August 2019
Availableonline16August2019
0926-860X/©2019ElsevierB.V.Allrightsreserved.

D.M. Kim, et al.
AppliedCatalysisA,General585(2019)117209
matrices of the mpg-C3N4 by enhancing reverse esterification reaction
of methyl acetate (MA), were characterized to explain an enhanced
catalytic activity during a relatively shorter carbonylation reaction
duration of 1 h and structural stability by verifying a less leaching of
active Rh metal species with the help of an optimal amount of Lewis
acid Al2O3 promoter on the basic mpg-C3N4 structures.
well as to confirm the ordered mesoporous structures. N2 adsorption-
desorption analysis was performed by using a Tristar II 3020 instrument
(Micromeritics) at a liquid nitrogen temperature of −196 °C and va-
cuum level of 10−6 Pa to measure specific surface area (Sg, m2/g),
average pore diameter (Dp, nm) and pore volume (Pv, cm3/g) of the
fresh AlRh-CN. The surface area was calculated by Brunauer-Emmett-
Teller (BET) method and the pore size distribution was also determined
by Barrett–Joyner–Halenda (BJH) method. X-ray photoelectron spec-
troscopy (XPS) analysis on the fresh and used AlRh-CN was performed
by using a VG Multilab 2000 instrument with a monochromatic Al Kα
X-ray source to measure the variations of binding energy (BE) of Rh
3d5/2, N 1s and O 1s peaks with the surface oxidation states of Rh
species before and after reaction. The surface oxidation states were
further calculated on the fresh AlRh-CN by comparing the intensity
ratios of the metallic Rh0 appeared at 307.3 eV and partially oxidized
Rhn+ at 309.6 eV (I(Rh0)/I(Rhn+)). All BEs on the AlRh-CN from the
surface intensive XPS spectra were previously corrected by using a re-
ference BE of C 1s at 286.4 eV. Thermogravimetric analysis (TGA) using
a TGA Q50 instrument was carried out under air environment to con-
firm the thermal stability of carbon nitride by observing the weight loss
of the pristine mpg-C3N4.
CO chemisorption was performed by using an ASAP 2020 instru-
ment with an assumption of stoichiometry number of CO/Rh = 1 at a
fixed adsorption temperature of 190 °C to measure the surface area of
metallic Rh (Rh0, m2/gRh) and its dispersion. Before the analysis, the
fresh AlRh-CN sample was pretreated at 100 °C for 3 h under Ar flow to
remove any impurity and water adsorbed. The amounts of surface
acidic sites (mmol-NH3/g) and its strengths on the fresh AlRh-CN were
analyzed by temperature programmed desorption of NH3 (NH3-TPD)
with a BELCAT-M instrument equipped with a thermal conductivity
detector (TCD). About 30 mg of the sample was used for NH3-TPD at the
temperature range of 50–350 °C after its in-situ pretreatment at 350 °C
for 1 h under He environment. Transmission electron microscopy (TEM)
analysis on the fresh AlRh-CN was also conducted by using a TECHNI
G2 T-20S instrument operating at a voltage of 200 kV to verify the
surface morphology and the metal distributions on the AlRh-CN was
characterized by using an energy dispersive spectrometer (TEM-EDS)
analysis as well. The leaching amount (ppm) of Al and Rh metal on the
AlRh-CN after the liquid-phase carbonylation reaction duration for 3 h
were measured by using an inductively coupled plasma-auger electron
spectroscopy (ICP-OES) with a CIROS CCD ICP spectrometer. X-ray
absorption near edge structures (XANES) and extended X-ray absorp-
tion fine structures (EXAFS) of the Rh K-edge on the fresh and used
AlRh-CN after the reaction time for 3 h were measured in a 7D XAFS
beamline at Pohang Light Source (PLS) facility. The storage energy ring
electron beam was 3.0 GeV with its current of ∼360 mA. The incident
X-ray source was monochromatized by two sets of Si(111) crystals, and
the Rh K-edge spectra were obtained in a fluorescence mode using a
passivated implanted planar silicon (PIPS) detector. Two typical re-
ference spectra such as Rh foil and Rh2O3 were further measured to
elucidate the oxidation state of the Rh nanoparticles on the AlRh-CN as
well.
2. Experimental section
2.1. Synthesis of AlRh metal-incorporated mesoporous g-C3N4 and activity
measurement
A highly ordered mesoporous graphitic carbon nitride (mpg-C3N4)
[16,17] was synthesized by using a hard silica template of SBA-15 and
hexamethylenetetramine (HTM) as a precursor of the mpg-C3N4. The
active Rh and Al metals were simultaneously incorporated into the
mpg-C3N4 structures by simultaneous impregnating the mixed solution
of Rh nitrate (Rh(NO3)3), Al nitrate (Al(NO3)3) and HTM precursor in a
highly ordered mesopores of SBA-15, which has a surface area of
650 m2/g to synthesize the Al-promoted as well as Rh metal-in-
corporated mpg-C3N4 structures. For more details, after completely
pillaring the metal precursors into the SBA-15 mesopores, which were
previously well-dispersed in aqueous HTM solution, the excess solvent
was dried overnight at 80 °C and the sample was carburized at 750 °C
for 5 h under N2 environment. The obtained black solid powders were
subsequently treated with a 2 M NaOH solution several times to remove
the hard template of SBA-15. And, the as-prepared sample was washed
again with deionized water for several times. The pristine dark-colored
sample was further dried at 80 °C overnight. A nominal concentration of
Rh-metal based on a total weight of the heterogeneous catalyst was
fixed to 2.0 wt% (20,000 ppmw) and that of Al metal was varied from 0
to 2.5 wt% (0 to 25,000 ppmw), where the final concentrations of Rh
and Al species in the g-C3N4 were found to be similar with their nom-
inal contents with an error range of 5% confirmed by an inductively
coupled plasma (ICP) analysis on the fresh AlRh-CN. As-prepared or-
dered mesoporous heterogeneous Al-promoted and Rh metal-in-
corporated mpg-C3N4 were denoted as Al(x)Rh-CN, where x represents
wt% of Al metal and CN for mpg-C3N4, respectively.
Catalytic activity for a liquid-phase carbonylation of methanol to
AA and MA was measured by three successive repetitions in a batch
reactor equipped with a glass liner with a reactor volume of 150 ml at
the reaction conditions of T = 190
5 °C, CO pressure =4.0 MPa
(90 mol% CO balanced with N2), methanol/methyl iodide/water molar
ratio = 46/28/26 (CH3I/CH3OH = 0.61 and H2O/CH3OH = 0.57) with
0.4 g of the heterogeneous AlRh-CN catalyst without any reductive
treatment. After a liquid-phase carbonylation reaction for the shorter
reaction duration of 1–3 h, the corrected liquid product samples were
analyzed by using an off-line gas chromatography (GC, Agilent in
6890 N) equipped with a flame ionization detector (FID) connected to a
DB-WAX capillary column. The conversion and product distribution
were calculated based on the carbon balance. In addition, the recycle
tests of the used AlRh-CN were carried out for three successive times
separately just after filtering and drying the used catalyst at 80 °C
without any chemical or thermal treatments to clearly confirm the
stability of the novel heterogenous AlRh-CN.
3. Results and discussion
3.1. Bulk and surface characteristics of AlRh-CN catalysts
2.2. Catalyst characterization
The bulk properties of the fresh AlRh-CN and its highly ordered
mesoporous structures were confirmed by XRD and SAXS analysis, and
the results are displayed in Fig. 1(A) as well as in supplementary Fig.
S1(A). The characteristic mpg-C3N4 peaks were clearly observed at the
diffraction peak of 2θ = 26.1° on all the AlRh-CN, which was attributed
to the characteristic interlayer structures of the g-C3N4 [16,17]. The
intense SAXS peak appeared at around 1° can be originated from the
highly ordered mesoporous structures of the mpg-C3N4, which can be
attributed to the highly ordered mesoporous structures of the hard
Wide angle powder X-ray diffraction (XRD) and small-angle X-ray
scattering (SAXS) analysis of the fresh and used AlRh-CN were carried
out by using a Brucker D8 Advance X-ray diffractometer and an ANTON
PAAR instrument with a Cu Kα irradiation (λ =0.15406 nm) generated
from Ge(111) monochromator operating at 40 kV and 40 mA, sepa-
rately. The characteristic diffraction peaks of the AlRh-CN were sepa-
rately measured at a scanning rate of 4 °C/min in the range of 2θ = 5 –
90° and 0.5 – 5° for XRD and SAXS to verify their crystalline phases as
2

D.M. Kim, et al.
AppliedCatalysisA,General585(2019)117209
Fig. 1. Characteristics of the fresh AlRh-CN catalysts; (A) Small-angle X-ray scattering (SAXS) patterns, high resolution TEM (HRTEM) images of the (B) Rh-CN, (C) Al
(2)Rh-CN with their surface distributions of carbon, nitrogen and Rh species measured by TEM-EDS analysis, and (D) XPS spectra of N 1s.
Table 1
Bulk and surface characteristics of the Al-modified ordered mesoporous Rh-incorporated mpg-C3N4 and catalytic activity for a liquid-phase carbonylation of me-
thanol.
Catalysta
N2 sorptionb
Sg (m2/g)
CO chem.c
XPSd
NH3-TPDe (mmol/g)
Conv. of methanol/rate
[mol%/mmol/(cm3·h)]f
Selectivity (mol%)
AA/MA/others
Pv (cm3/g)
Dp (nm)
Sg (Rh0, m2/gRh
)
I(Rh0)/I(Rhn+
)
Rh-CN
1058
923
974
1027
951
0.84
0.50
0.52
0.62
0.54
3.4
2.6
2.6
2.8
2.7
18.8
6.6
58.2
76.8
16.3
0.14
0.16
0.24
0.28
0.13
0.132
0.359
0.501
0.393
0.280
87.5 / 10.80
85.6 / 10.57
83.3 / 10.28
89.0 / 10.99
80.1 / 9.89
47.5/51.3/1.2
54.3/44.6/1.1
59.8/39.5/0.7
62.9/36.5/0.6
47.1/51.8/1.1
Al(1)Rh-CN
Al(1.5)Rh-CN
Al(2)Rh-CN
Al(2.5)Rh-CN
a
Al(x)Rh-CN denotes the 2.0 wt%Rh-incorporated mesoporous graphitic-C3N4 (mpg-C3N4) with different content of Al2O3 promoter (x) from 0 to 2.5 wt%Al,
where the Rh and Al precursor were simultaneously introduced to the mpg-C3N4 structures using a hard template of SBA-15.
b
Specific surface area (Sg, m2/g), pore volume (Pv, cm3/g) and average pore diameter (Dp, nm) were measured by N2 adsorption-desorption analysis of the fresh
AlRh-CN.
c
CO chemisorption was carried out at 190 °C and the metallic surface area of Rh species (Rh0, m2/gRh) was estimated with an assumption of the stoichiometry of
CO/Rh = 1, which was corrected by multiplying the relative degree of reduction of Rh nanoparticles on the fresh AlRh-CN using the ratios of I(Rh0)/I(Rhn+
)
measured by XPS analysis.
d
The surface ratio of the Rh 3d5/2 peak was calculated using the integrated area of each Rh peak assigned to the metallic Rh0 appeared at 307.3 eV and oxidized
Rhn+ at 309.6 eV on the fresh AlRh-CN.
e
The acidic sites on the fresh AlRh-CN were represented by the total amounts of NH3 uptakes.
f
Catalytic activity of the AlRh-CN was measured at T = 190
5 °C, CO partial pressure =4.0 MPa (90 mol% CO balanced with N2), molar ratio of CH3OH/CH3I/
H2O = 47/28/25 (molar ratio of CH3I/CH3OH = 0.61 and H2O/CH3OH = 0.57) for the reaction duration of 1 h with a total reactant volume of 20 ml at a fixed
stirring speed of 300 rpm. The main products by a liquid-phase carbonylation of methanol were found to be acetic acid (AA) and methyl acetate (MA) with a small
amount of propionic acid and gaseous CH4, CO2 and dimethyl ether less than 1.2 mol%. The rate of methanol consumption (mmol/(cm3·h)) was defined as the
consumed amount of methanol reactant (mmol) per volume of reactants and hour.
template of SBA-15 [18]. However, the simultaneous addition of Lewis
acid Al promoter with Rh species on the mpg-C3N4 caused the structural
disintegrations of those ordered mesoporous structures by in-
corporating into the main frameworks of the mpg-C3N4. Those struc-
tural instability and collapses were found to be more significant with an
increase of Al content above 2 wt%Al by showing a much smaller peak
intensity of SAXS spectra as shown in Fig. 1(A). Interestingly, the or-
dered mesoporous structures of the mpg-C3N4 were not significantly
changed even after an acidic liquid-phase carbonylation reaction as
clearly revealed by the observed XRD patterns on the used AlRh-CN as
shown in supplementary Fig. S1(B). The results of N2 adsorption-des-
orption analysis on the fresh AlRh-CN are further summarized in
Table 1. The extremely larger specific surface areas were characterized
on all the fresh AlRh-CN in their area ranges of 923–1058 m2/g with a
maximum surface area of 1058 m2/g on the Rh-CN and medium one of
1027 m2/g on the Al(2)Rh-CN. The pore volumes and average pore
3

D.M. Kim, et al.
AppliedCatalysisA,General585(2019)117209
Fig. 2. Characteristics of the fresh and used AlRh-CN catalysts; (A) XPS spectra of Rh 3d5/2 peak on the fresh AlRh-CN, (B) Rh K-edge k3-weight Fourier transformed
EXAFS spectra of the fresh and used AlRh-CN, (C) AA yield with respect to Al content (wt%) on the AlRh-CN catalysts at the reaction conditions of T = 190 5 °C,
CO partial pressure =4.0 MPa (90 mol% CO balanced with N2), molar ratio of CH3OH/CH3I/H2O = 47/28/25 with the respective reaction duration of 1 h and 3 h,
and (D) proposed schematic reaction mechanisms on the AlRh-CN in terms of the separate contributions of the metallic Rh nanoparticles and Lewis acid Al2O3 sites.
diameters on the RhAl-CN were found to be in the range of 0.50 –
0.84 cm3/g and 2.6–3.4 nm, respectively. Except for those largest values
on the Al-unpromoted Rh-CN, a relatively larger pore volume and its
diameter of 0.62 cm3/g and 2.8 nm were observed on the Al(2)Rh-CN.
It seems to be mainly attributed to the highly dispersed Al species and
Rh nanoparticles in the matrices of the mpg-C3N4 up to the Al content
of 2 wt%Al. While adding larger amount of the Al promoter on the
AlRh-CN above 2 wt%Al such as the Al(2.5)Rh-CN, the specific surface
area (Sg) and pore volume (PV) were slightly decreased due to the
partial blockage of the ordered g-C3N4 mesoporous structures [19,20]
or structural collapses [21] due to an excess formation of the Lewis acid
Al2O3. The selective formations of the ordered mesoporous structures
on the AlRh-CN were further confirmed by TEM analysis, and the TEM
images of the selected Rh-CN and Al(2)Rh-CN are also displayed in
Fig. 1(B), (C) and supplementary Fig. S2. The replica structures of the
highly ordered mpg-C3N4 with an average pore diameter of 2.5 nm
[19,20], which was attributed to the inversed structures of the ordered
mesoporous hard template structures of SBA-15, were observed with
showing no larger aggregated metal nanoparticles. The obvious meso-
porous structures were characterized on the Rh-CN compared to the Al
(2)Rh-CN because the thermodynamically stable Al2O3 phases on the
basic graphitic-C3N4 at an excess amount [21] can act as an impurity to
partially disintegrate the ordered mesoporous structures of the g-C3N4
due to an excess amount of the incorporated acidic Al3+ species [21],
where the Al2O3 can further play important roles by increasing the
specific surface area as a pillaring material as well as Lewis acid sites.
This observation is also in line with the results of SAXS analysis as
shown in Fig. 1(A). However, the elements such as C and N species as
well as partially reduced Rh metal nanoparticles were found to be
highly dispersed in the matrices of the mpg-C3N4 on both the Rh-CN
and Al(2)Rh-CN as confirmed by the surface-intensive TEM-EDS ana-
lysis (supplementary Fig. S3).
The larger metallic surface area of the well-dispersed Rh nano-
particles on the fresh AlRh-CN was observed in the range of
6.6–76.8 m2/gRh, and CO chemisorption results are summarized in
Table 1. A relatively larger surface area of the metallic Rh nanoparticles
with the value of 76.8 m2/gRh was observed on the Al(2)Rh-CN and
smaller one on the pristine Rh-CN with the value of 18.8 m2/gRh, which
could be mainly attributed to the larger specific surface area of the
highly ordered mesoporous g-C3N4 structures. The phases of Rh nano-
particles were found to be in the metallic Rh as well as Rh2O3 oxides as
confirmed by XPS analysis. This observation further revealed that the
Al2O3 promoter at an optimal 2 wt%Al on the AlRh-CN played the
crucial roles such as formation of Lewis acid sites and enhancement of
Rh dispersion by preventing the aggregation of Rh nanoparticles in the
matrices of the Al-promoted mpg-C3N4 structures. In general, the dis-
persions and electronic states of active metals can be enhanced on the
bimetallic systems by selectively anchoring on the surface functional
groups such as -NH2 or -NH species on the mpg-C3N4 as well as by
strongly interacting between two metals [11,22–24]. Therefore, the
much lower surface area of metallic Rh nanoparticles with an excess
mount of Lewis acid Al2O3 species on the Al(2.5)Rh-CN as well as the
Rh-CN with the respective values of 16.3 and 18.8 m2/gRh compared to
that of 76.8 m2/gRh on the Al(2)Rh-CN seems to be possibly attributed
to their heterogeneous distributions of Rh and Al species on the mpg-
C3N4 surfaces. The XPS spectra of the N 1s and O 1s on the fresh AlRh-
4

D.M. Kim, et al.
AppliedCatalysisA,General585(2019)117209
CN are displayed in Fig. 1(D) and supplementary Fig. S4, respectively.
Three characteristic N 1s peaks were clearly observed at BE of 398.2 eV
assigned to the pyridinic N species, that of 399.6 eV for pyrrolic N
species and that of 400.9 eV for graphitic N species [16]. The BEs of N
1s were slightly shifted to higher ones on the unmodified Rh-CN com-
pared to that of the Al-modified Rh-CN due to the reductive natures of
the closely located Al2O3 promoter [21], which can further increase the
reducibility of Rh nanoparticles on the mpg-C3N4 due to the intimately
located Al2O3 species to the smaller Rh nanoparticles. This observation
can further support the formation of the highly ordered mpg-C3N4
structures with an insignificant structural and chemical deformation.
Furthermore, the O 1s peak appeared at the BE of 530.5 eV can be
originated from the partially oxidized Rh-O species [25] as shown in
supplementary Fig. S4. Those BEs were found to be slightly higher on
the Rh-CN at 532.0 eV compared to those of the fresh Al-modified RhAl-
CN in the range of 531.6–531.9 eV, which can be attributed to domi-
nant presence of Rh2O3 phases with a little contribution by Al2O3
promoter possibly. The main phases of Al species on the fresh Al(2.5)
Rh-CN with an excess amount of Al promoter also revealed the pre-
ferential presence of Al3+ species assigned to Al2O3, which was also
supported by the observed BE position at 74.5 eV [21]. Interestingly,
the BEs of the N 1s and O 1s on the used AlRh-CN as displayed in
supplementary Fig. S5 were not significantly altered even after an
acidic liquid-phase carbonylation reaction, which strongly suggests the
stable preservation of the surface g-C3N4 structures during the reaction.
The outermost surface ratios of I(Rh0)/I(Rhn+) on the fresh and used
AlRh-CN were calculated by deconvoluting the Rh 3d5/2 peak as shown
in Fig. 2(A) and the results are summarized in Table 1 and in supple-
mentary Table S1. The BEs of the Rh 3d5/2 peak appeared at 307.3 eV
generally indicate a zero-valence metallic Rh species (Rh0) and that of
309.6 eV suggests the presence of the oxidized Rhn+ species [19,26].
The surface ratios of I(Rh0)/I(Rhn+) on the fresh AlRh-CN were found
to be in the range of 0.13 – 0.28 with a maximum value of 0.28 on the
Al(2)Rh-CN and lower one of 0.13 - 0.14 on the Rh-CN and Al(2.5)Rh-
CN. These observations also clearly suggest that the partially reduced
Rh nanoparticles can be selectively formed on the Al(2)Rh-CN due to
the reductive natures of the mpg-C3N4 with their higher dispersion in
the matrices of the mpg-C3N4 by their stronger interactions with mpg-
C3N4 as well as Al2O3 promoter [11]. In addition, the BEs of the Rh 3d5/
2 peak on the used AlRh-CN as displayed in supplementary Fig. S6 were
slightly shifted to lower ones in the range of 306.9–307.1 eV. The ob-
servations suggest the partial reduction of the oxidized Rhn+ species to
metallic Rh0 during a liquid-phase carbonylation reaction. Based on
those XPS results, the partially reduced Rh nanoparticles seem to be
active sites for a liquid-phase carbonylation of methanol to AA and MA
by completing the cyclic processes between the Rh0 and Rhn+ species
[27].
These XPS results were further supported by the results of XANES
analysis as shown in supplementary Fig. S7. The results revealed the
similar XANES spectra patterns with that of the reference Rh2O3, and a
slightly smaller intensity rather than that of reference Rh foil on the
fresh AlRh-CN can be attributed to the copresence of the mixed Rh
phases such as the metallic Rh0 and Rh oxides [19,28,29]. In addition,
Rh K-edge k3-weight Fourier transformed EXAFS spectra on the fresh
and used AlRh-CN as shown in Fig. 2(B) further revealed that the
partially reduced Rh nanoparticles before the acidic carbonylation re-
action were further reduced to metallic Rh species due to reductive
reaction conditions. The relative degree of reduction was found to be
more significant on the Al(2)Rh-CN compared to that of the Rh-CN by
showing a relatively larger peak intensity appeared at 0.25 nm. As
summarized in supplementary Table S2, the bond lengths of Rh-O on
the fresh AlRh-CN were found to be similar with that of the Rh2O3
reference with its length of 0.16 nm, and the bond lengths of Rh-Rh on
the used AlRh-CN were found to be slightly smaller than that of Rh-foil
(0.26 nm) with the value of 0.24 nm as well as smaller Rh-O bond
lengths of 0.14 nm, which was more clearly observed on the used Al(2)
Rh-CN. In addition, the oxidation states of Rh species on the fresh and
used AlRh-CN were compared by using their white line area (WL,
Gaussian area) in supplementary Table S2, which revealed the Rh
species even after a liquid-phase carbonylation reaction mainly existed
in the highly oxidized states with their values of WL of 3.59–3.94 on the
used AlRh-CN from those of 4.44–4.96 on the fresh AlRh-CN with much
larger values on the most active Al(2)Rh-CN. In addition, these ob-
servations strongly suggest the preferential formation of the much
smaller and partially reduced Rh nanoparticles on the Al(2)Rh-CN
compared to that of the unmodified Rh-CN. These results are in line
with the results of XPS and CO chemisorption analysis by showing a
larger metallic surface area of the partially reduced Rh nanoparticles on
the most active Al(2)Rh-CN.
3.2. Roles of Al2O3 promoter for an enhanced catalytic activity on AlRh-CN
catalysts
The catalytic activity for the liquid-phase carbonylation
(CH3OH + CO ⇔ CH3COOH) such as conversion of methanol and
product distribution on the AlRh-CN are summarized in Table 1 and
supplementary Table S3 at the reaction conditions of T = 190
5 °C,
P(CO) =4.0 MPa, molar ratio of CH3OH/CH3I/H2O = 47/28/25 for a
short reaction duration of 1 h, which was similar with commercial re-
action conditions [9,10]. The comparable methanol conversion of
89.0% (rate of 10.99 mmol/(cm3·h)) with a higher AA selectivity of
62.9% and lower MA selectivity of 36.5% was observed on the Al(2)Rh-
CN, where the AA selectivity was found to be much higher than the
other AlRh-CN in the range of 47.1–59.8%. With an increase of Al
content up to 2 wt% on the Al(2)Rh-CN, AA selectivity was steadily
increased from 47.5% on the unmodified Rh-CN to 62.9% on the Al(2)
Rh-CN, and it was decreased again to 47.1% on the Al(2.5)Rh-CN at an
excess amount of Al2O3 promoter. On all the AlRh-CN, the similar
methanol conversions of 80.1–89.0% corresponding to the methanol
consumption rates in the range of 9.89–10.99 mmol/(cm3·h), which are
found to be similar with the previous work [2], were observed with an
inverse selectivity trend of the main MA intermediate in the range of
36.5–51.8%. Interestingly, other gaseous and liquid byproducts such as
CH4, CO2, dimethyl ether and propionic acid were formed insignif-
icantly on the Al(2)Rh-CN with their total selectivity below 0.6%. This
lower byproduct formation was mainly originated from the lower ac-
tivity for a gas-phase reforming of oxygenates and/or water-gas shift
(WGS) reaction [2,6,30]. Furthermore, the leached amount of Rh metal
after the reaction for 1 h on the Al(2)Rh-CN was found to be 7 ppm in
liquid products as summarized in Table S3, which was smaller than that
of the Rh-CN with 10 ppm due to the formation of strongly interacted
smaller and partially reduced Rh nanoparticles on the Al(2)Rh-CN with
the help of their intimately distributed Lewis acid Al2O3 promoter.
However, the leached amount of Al2O3 promoter was found to be much
higher than that of Rh species with 34 ppm on the Al(2)Rh-CN. This
observation can be attributed to an intrinsically much weaker interac-
tion of the Lewis acid Al2O3 species with the surface functional groups
of the mpg-C3N4 [21] than that of the Rh nanoparticles.
A small amount of byproducts less than 1.2% on the AlRh-CN was
further confirmed by an analysis of gaseous and liquid products with
the bench-scale Al(1.5)Rh-CN supported on the activated carbon (AC)
sphere (1 mm in diameter) after the carbonylation reaction for two
separate times for 1 and 2 h, and the results are summarized in sup-
plementary Table S4. The main byproducts at the conversion of me-
thanol above 95% were found to be gaseous CO2, CH4 and dimethyl
ether (DME) (below 1.8 mol%) with an insignificant amount of a liquid
propionic acid. The yields of AA for 1 h and 3 h reaction duration are
respectively plotted in Fig. 2(C) in terms of the Al content on the AlRh-
CN with three repetitions, and the maximum yield of 56.0% for 1 h
reaction and 65.4% for 3 h reaction was observed on the most active Al
(2)Rh-CN by showing a typical volcano-pattern of AA yield. We believe
that the observed higher activity on the Al(2)Rh-CN was mainly
5

D.M. Kim, et al.
AppliedCatalysisA,General585(2019)117209
Table 2
Methanol conversion and product distribution for 3 successive recycle tests on the Rh-CN and Al(2)Rh-CN with the leached amount of Rh and Al2O3 species.
Catalyst
Cycle numbera
Conv. of methanol (mol%)
Selectivity (mol%)
Yield of AA (mol%)
Leaching amount (ppm)b
AA
MA
others
Rh
Al
Rh-CN
1
2
3
1
2
3
93.9
94.6
94.4
95.3
95.3
94.6
57.4
57.2
62.7
68.6
67.9
68.3
41.3
41.5
36.5
30.9
31.1
31.3
1.3
1.3
0.8
0.5
1.0
0.4
53.9
54.1
59.2
65.4
64.7
64.6
16
11
15
6
8
12
–
–
–
57
64
37
Al(2)Rh-CN
a
Catalytic activity and stability of the selected Rh-CN and AlRh-CN were measured at T = 190
5 °C, CO partial pressure = 4.0 MPa (90 mol% CO balanced with
N2), molar ratio of CH3OH/CH3I/H2O = 47/28/25 (molar ratio of CH3I/CH3OH = 0.61 and H2O/CH3OH = 0.57) for reaction duration of 3 h with the total reactant
volume of 20 ml at a stirring speed of 300 rpm with three successive repetitions just after filtering and drying the used catalysts.
b
The leached amounts (ppm) of the Rh and Al species from the AlRh-CN surfaces were measured by ICP-OES by using the liquid products after separate cycle tests
for 3 h reaction.
originated from the large surface area of the active Rh metal (CO
chemisorption) as well as the larger amount of acidic sites originated
from the Lewis acid Al2O3 species. In addition, the highly dispersed
Al2O3 particles further played an important role as a pillaring material
to increase the specific surface area of the RhAl-CN at an optimal Al2O3
content of 2 wt%Al. Those results were further confirmed by NH3-TPD
analysis and the results are summarized in Table 1 (detailed data in
supplementary Table S5) and in supplementary Fig. S8. The char-
acteristic NH3 desorption peaks on the acidic sites appeared below and
above the temperature regions of 180 °C, which can be separately as-
signed to the weak and strong acidic sites originated from Lewis acid
Al2O3 surfaces [30,31], were found to be slightly higher on the most
active Al(2)Rh-CN compared to other AlRh-CN (Fig. S8(A)). Interest-
ingly, the desorption peak at around 200 °C was found to be much
smaller on the most active Al(2)Rh-CN, and the result suggests the more
stable preservation of the surface pyridinic and pyrrolic N species by
intimately interacting with the highly dispersed Lewis acid Al2O3 pro-
moter [10,16,21]. Those results were also supported from TGA profiles
by showing a larger weight loss of the pristine mpg-C3N4 above 500 °C
as shown in supplementary Fig. S9. These findings strongly suggest the
preferential presence of the much stronger acidic sites on the Al(2)Rh-
CN due to the stronger interactions of the highly dispersed Lewis acid
Al2O3 and Rh nanoparticles on the mpg-C3N4 surfaces. The total
amounts of acidic sites on the AlRh-CN were found in the range of 0.132
– 0.501 mmolNH3/g, and the largest amounts of the acidic sites in the
range of 0.393 – 0.501 mmolNH3/g on the Al(1.5)Rh-CN and Al(2)Rh-
CN were characterized. The observed smaller amount of acidic sites
with 0.280 mmolNH3/g on the Al(2.5)Rh-CN can be possibly attributed
to heterogeneous distributions of relatively larger Lewis acid Al2O3
particles. The relatively larger amounts of acidic sites with 0.132
mmolNH3/g on the Rh-CN seem to be originated from the acidic natures
of Rh2O3 species [10,21] due to insignificant amounts of acidic sites
observed on the pristine mpg-C3N4 as shown in supplementary Fig.
S8(B). Based on our previous work [31], the Al2O3 coated with g-C3N4
with their different weight ratios revealed the characteristic FT-IR
spectra of the Al2O3 and g-C3N4 as shown in supplementary Fig. S10.
The characteristic absorption peaks appeared at the wavenumbers of
1200 - 1600 cm−1 can be assigned to the Al-O bond starching bands
[32] and the characteristic vibration modes of the g-C3N4 appeared at
800 - 1600 cm−1 can be originated from skeletal vibrations of heptazine
heterocyclic rings of the g-C3N4 with that of 2500 - 3500 cm−1 for NeH
band stretching modes [33,34]. These FT-IR peaks observed on the
Al2O3 coated with g-C3N4 strongly suggest the copresences of Lewis
acid Al2O3 as well as basic g-C3N4. Therefore, the highly dispersed and
partially reduced smaller Rh nanoparticles, which were strongly inter-
acted with the mpg-C3N4 at an optimal amount of Lewis acid Al2O3
sites, were possibly responsible for a higher catalytic activity for a li-
quid-phase carbonylation of methanol to selectively form the AA and
MA intermediate. We believe that the Lewis acid Al2O3 sites assigned to
Al3+ can transform MA intermediate to AA product by a reverse es-
terification reaction (CH3COOCH3 + H2O ⇔ CH3OH + CH3COOH) as
well [35–38].
In addition, the effects of liquid-phase reaction times from 1 to 3 h
on the active Al(2)Rh-CN were further investigated, and the results are
summarized in supplementary Table S5. Conversion of methanol was
increased from 89.0% for 1 h reaction time to 95.3% for 3 h reaction
time with a simultaneous increase of AA selectivity from 62.9% to
68.6% (AA yield from 55.8% to 65.4%) with the concomitant decreases
of MA selectivity and small amounts of byproducts below 0.6%. The
continuous increases of AA selectivity with increasing reaction times
suggest the possible reverse esterification of MA to AA on the Lewis acid
Al2O3 sites as proposed. Furthermore, the leached amount of the Rh and
Al species on the Al(2)Rh-CN with increasing reaction times was not
significantly altered by maintaining the amounts of Rh metal in the
range of 6–9 ppm and that of Al2O3 promoter in the range of
34–57 ppm, which was found to be much smaller than the unmodified
pristine Rh-CN. Furthermore, the catalytic stability of the re-
presentative Rh-CN and Al(2)Rh-CN was verified by carrying out suc-
cessive recycle tests for three repetitions for each reaction duration of
3 h by using the used AlRh-CN catalysts just after filtering and drying at
80 °C, and the results are summarized in Table 2. On both Rh-CN and Al
(2)Rh-CN, the methanol conversions were insignificantly altered during
three recycle tests by maintaining their initial conversion and se-
lectivity to AA without significant loss of their initial activities. How-
ever, the higher and more stable yield of AA above 64% with the by-
product selectivity below 1.0% was observed on the most active Al(2)
Rh-CN compared to that of 53.9–59.2% with the byproducts of 1.3% on
the unmodified Rh-CN. The superior stability of the most active Al(2)
Rh-CN seems to be attributed to the less amount of leached Rh species
in the range of 6–12 ppm for three recycle tests compared to those of
the Rh-CN with 22 – 16 ppm. The smaller amounts of Rh leaching on
the Al(2)Rh-CN were possibly attributed to its larger specific surface
area with strongly interacted metallic Rh nanoparticles on the mpg-
C3N4 with the help of the highly dispersed Lewis acid Al2O3 promoter
[10,21,30,31]. Therefore, we believe that the novel heterogeneous
AlRh-CN with an optimal Al promoter (2 wt%Al) can be an alternative
heterogeneous catalytic system to replace the commercialized homo-
geneous Rh-based organic complexes. Furthermore, the Al(2)Rh-CN
preserved an initial activity by maintaining the strongly interacted and
partially reduced Rh nanoparticles in the matrices of the basic mpg-
C3N4 and by enhancing the reverse esterification reaction activity of
MA to AA on the Lewis acid Al2O3 sites [31].
Based on the experimental and characterization results on the het-
erogeneous Al(2)Rh-CN, the possible reaction mechanisms of methanol
carbonylation to AA through the formation of MA intermediate in a
liquid-phase reaction are proposed and displayed in Fig. 2(D). The first
possible activation step of methanol seems to be the formation of
CH3COI intermediate by an insertion of CO molecule with the help of
6

D.M. Kim, et al.
AppliedCatalysisA,General585(2019)117209
CH3I promoter on the partially reduced Rh metal surfaces as previously
reported study [24]. Subsequently, the active CH3COI intermediate can
be transformed to AA (CH3COOH) as well as MA (CH3COOCH3) inter-
mediate by selectively reacting with H2O and methanol, respectively.
Therefore, the stable preservations of the partially reduced and strongly
interacted Rh nanoparticles by maintaining its larger metallic surface
area and its less leaching problem are important to get a higher car-
bonylation activity by generating facile redox cycles of the CH3I pro-
moter. We believe that the smaller amount of leached Rh species on the
Al(2)Rh-CN by adding an optimal amount of Al2O3 promoter was
possibly attributed to its stronger interactions with the basic sites of the
mpg-C3N4 surfaces with the help of the well-dispersed Al2O3 particles
by acting as pillaring materials and Lewis acid sites. The additional
roles of the Lewis acid Al2O3 promoter assigned to the Al3+ species
were found to also enhance the reverse esterification reaction rate of
MA intermediate to AA in the presence of water, which seems to be
much faster on the highly dispersed Lewis acid Al2O3 sites in the ma-
trices of the mpg-C3N4 structures with the co-presence of the intrinsic
basic and acidic surface sites [31]. Therefore, the synergy effects of the
Al(2)Rh-CN having the larger metallic surface area of the partially re-
duced Rh nanoparticles and larger amount of Lewis acid sites from the
highly dispersed Al2O3 promoter were mainly responsible for an en-
hanced activity for a liquid-phase carbonylation as well as reverse es-
terification reaction of MA intermediate to final AA product. Those
superior surface properties generated from the highly dispersed Al and
Rh metal-incorporated mpg-C3N4 can increase the catalytic activity and
stability during a liquid-phase carbonylation reaction of methanol to
AA and MA, which can be an efficient novel heterogeneous catalytic
system containing active metal and acidic and basic sites as well.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2019.117209.
References
[1] F.E. Paulik, J.F. Roth, Chem. Commun. (Camb.) (1968) 1578.
[2] A. Haynes, P.M. Maitlis, G.E. Morris, G.J. Sunley, H. Adams, P.W. Badger,
C.M. Bowers, D.B. Cook, P.I.P. Elliott, T. Ghaffar, H. Green, T.R. Griffin, M. Payne,
J.M. Pearson, M.J. Taylor, P.W. Vickers, R.J. Watt, J. Am. Chem. Soc. 126 (2004)
2847–2861.
[3] (a) L. Zhou, S. Li, G. Qi, Y. Su, J. Li, A. Zheng, X. Yi, Q. Wang, F. Deng, Solid State
Nucl. Magn. Reson. 80 (2016) 1–6;
(b) X.F. Wu, H. Neumann, ChemCatChem 4 (2012) 447–458.
[4] R. Shavi, J. Ko, A. Cho, J.W. Han, J.G. Seo, Appl. Catal. B: Environ. 229 (2018)
237–248.
[5] A.M. Rabie, M.A. Betiha, S.E. Park, Appl. Catal. B: Environ. 215 (2017) 50–59.
[6] K. Sano, H. Uchida, S. Wakabayashi, Catal. Surv. Japan 3 (1999) 55–60.
[7] X. Li, E. Iglesia, J. Phys. Chem. C 112 (2008) 15001–15008.
[8] M.A. Betiha, H.M.A. Hassan, E.A. El-Sharkawy, A.M. Al-Sabagh, M.F. Menoufy,
H.E.M. Abdelmoniem, Appl. Catal. B: Environ. 182 (2016) 15–25.
[9] K. Park, S. Lim, J.H. Baik, H. Kim, K.D. Jung, S. Yoon, Catal. Sci. Technol. 8 (2018)
2894–2900.
[10] J.S. Nam, A.R. Kim, D.M. Kim, T.S. Chang, B.S. Kim, J.W. Bae, Catal. Commun. 99
(2017) 141–145.
[11] W.J. Ong, L.L. Tan, Y.H. Ng, S.T. Yong, S.P. Chai, Chem. Rev. 116 (2016)
7159–7329.
[12] (a) B. Kiskan, J. Zhang, X. Wang, M. Antonietti, Y. Yagci, ACS Macro Lett. 1 (2012)
546–549;
(b) Y. Cai, Y. Tang, L. Fan, Q. Lefebvre, H. Hou, M. Rueping, ACS Catal. 8 (2018)
9471–9476.
[13] R. Malik, V.K. Tomer, L. Kienle, V. Chaudhary, S. Nehra, S. Duhan, Adv. Mater.
Interf. 5 (2018) 1701357.
[14] X. Jin, V.V. Balasubramanian, S.T. Selvan, D.P. Sawant, M.A. Chari, G.Q. Lu,
A. Vinu, Angew. Chem. Int. Ed. 48 (2009) 7884–7887.
[15] (a) J. Wang, D. Hao, J. Ye, N. Umezawa, Chem. Mater. 29 (2017) 2694–2707;
(b) M.A.D. Fard, H. Ghafuri, A. Rashidizadeh, Microporous Mesoporous Mater. 274
(2019) 83–93.
[16] Z. Zhao, Y. Dai, J. Lin, G. Wang, Chem. Mater. 26 (2014) 3151–3161.
[17] (a) J. Xu, T. Chen, X. Wang, B. Xue, Y.X. Li, Catal. Sci. Technol. 4 (2014)
2126–2133;
4. Conclusions
A liquid-phase carbonylation reaction of methanol to acetic acid
(AA) as well as methyl acetate (MA) was carried out by using the novel
highly ordered mesoporous heterogeneous Al and Rh metal-in-
corporated mesoporous graphitic-C3N4. The positive and synergistic
roles of Lewis acid Al2O3 promoter were found to act as a pillaring
material of the mesoporous graphitic carbon nitride (mpg-C3N4)
structures by increasing the reverse esterification reaction activity as
well as by preserving the partially reduced and highly dispersed Rh
nanoparticles. The Al(2)Rh-CN with an optimal amount of Al2O3 pro-
moter revealed a higher yield to AA above 65% for a shorter reaction
duration less than 3 h by simultaneously forming MA intermediate. The
enhanced catalytic activity and stability on the Al(2)Rh-CN were
mainly attributed to the stable preservation of the initial chemical
states of the Rh nanoparticles by maintaining its higher dispersion of
the partially reduced Rh metal, which were found to be most active
sites to form CH3COI intermediate for a facile methanol activation. The
larger amount of the highly dispersed Lewis acid Al2O3 sites assigned to
Al3+ species further increased the reverse esterification reaction rate of
MA intermediate to AA by an equilibrated hydrolysis reaction such as
the liquid-phase reverse esterification reaction. The Lewis acid Al2O3
promoter on the heterogeneous AlRh-CN also significantly enhanced
the catalytic stability by preferentially forming the strongly interacted
Rh nanoparticles on the mesoporous Al-modified mpg-C3N4 and by
preserving the active Rh metal surface area with the help of the highly
dispersed Lewis acid Al2O3 pillaring promoter, which also caused a less
leaching of Rh nanoparticles on the optimized Al(2)Rh-CN with the
copresence of intrinsically basic and acidic surface sites.
(b) H.M. Zhao, C.M. Di, L. Wang, Y. Chun, Q.H. Xu, Microporous Mesoporous
Mater. 208 (2015) 98–104.
[18] A. Vinu, Adv. Funct. Mater. 18 (2008) 816–827.
[19] R. Malik, V.K. Tomer, V. Chaudhary, M.S. Dahiya, A. Sharma, S.P. Nehra, S. Duhan,
K. Kailasam, J. Mater. Chem. A Mater. Energy Sustain. 5 (2017) 14134–14143.
[20] J. Zhang, F. Guo, X. Wang, Adv. Funct. Mater. 23 (2013) 3008–3014.
[21] S.Y. Bae, S. Jeon, Y.H. Lee, D.H. Lee, H. Kye, J.W. Bae, RSC Adv. 6 (2016)
20728–20733.
[22] R. Mu, Q. Fu, H. Xu, H. Zhang, Y. Huang, Z. Jiang, S. Zhang, D. Tan, X. Bao, J. Am.
Chem. Soc. 133 (2011) 1978–1986.
[23] B.T. Sneed, A.P. Young, D. Jalalpoor, M.C. Golden, S. Mao, Y. Jiang, Y. Wang,
C.K. Tsung, ACS Nano 8 (2014) 7239–7250.
[24] C. Han, Y. Lu, J. Zhang, L. Ge, Y. Li, C. Chen, Y. Xin, L. Wu, S. Fang, J. Mater. Chem.
A Mater. Energy Sustain. 3 (2015) 23274–23282.
[25] C. Lin, G. Wu, H. Li, Y. Geng, G. Xie, J. Yang, B. Liu, J. Jin, Nanoscale 9 (2017)
1834–1839.
[26] Y. Wang, Z. Song, D. Ma, H. Luo, D. Liang, X. Bao, J. Mol. Catal. A Chem. 149
(1999) 51–61.
[27] (a) A.W. Budiman, J.S. Nam, J.H. Park, R.I. Mukti, T.S. Chang, J.W. Bae, M.J. Choi,
Catal. Surv. Asia 20 (2016) 173–193;
(b) N. Yoneda, S. Kusano, M. Yasui, P. Pujado, S. Wilcher, Appl. Catal. A Gen. 221
(2001) 253–265.
[28] Y. Kwon, T.Y. Kim, G. Kwon, J. Yi, H. Lee, J. Am. Chem. Soc. 139 (2017)
17694–17699.
[29] M. Haneda, T. Kaneko, N. Kamiuchi, M. Ozawa, Catal. Sci. Technol. 5 (2015)
1792–1800.
[30] (a) H. Ham, J. Kim, S.J. Cho, J.H. Choi, D.J. Moon, J.W. Bae, ACS Catal. 6 (2016)
5629–5640;
(b) H. Ham, S.W. Baek, C.H. Shin, J.W. Bae, ACS Catal. 9 (2019) 679–690;
(c) S.M. Pudi, P. Biswas, S. Kumar, B. Sarkar, J. Braz, Chem. Soc. 26 (8) (2015)
1551–1564.
[31] J. Park, F. Zafar, C.H. Kim, J. Choi, J.W. Bae, Fuel Process. Technol. 184 (2019)
36–44.
[32] S. Ramesh, E. Sominska, B. Cina, R. Chaim, A. Gedanken, J. Am. Ceram. Soc. 83
(2000) 89–94.
[33] Y. Cui, J. Zhang, G. Zhang, J. Huang, P. Liu, M. Antoniettib, X. Wang, J. Mater.
Chem. 21 (2011) 13032–13039.
[34] A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J.O. Muller, R. Schloglb,
J.M. Carlsson, J. Mater. Chem. 18 (2008) 4893–4908.
[35] J. Bedard, H. Chiang, A. Bhan, J. Catal. 290 (2012) 210–219.
[36] J.H. Park, C. Pang, C.H. Chung, J.W. Bae, J. Nanosci. Nanotechnol. 16 (2016)
4626–4630.
[37] H.M. Koo, J.H. Lee, T.S. Chang, Y.W. Suh, D.H. Lee, J.W. Bae, Reac. Kinet. Mech.
Catal. 112 (2014) 499–510.
Acknowledgements
The authors would like to sincerely acknowledge the financial
support from the National Research Foundation of Korea (NRF) grant
funded
by
the
Korea
government
(Project
#:
NRF-
[38] N.R. Guha, S. Sharma, D. Bhattacherjee, V. Thakur, R. Bharti, C.B. Reddy, P. Das,
Green Chem. 18 (2016) 1206–1211.
2018M3D3A1A01018009 and NRF-2017R1D1A1B03028214).
7