Methane coupling and hydrogen evolution induced by palladium-loaded gallium oxide photocatalysts in the presence of water vapor

Article abstract of DOI:10.1016/j.jcat.2021.03.024

Non-oxidative coupling of methane (NOCM, 2CH₄ → C₂H₆ + H₂) is a reaction that can directly produce ethane and hydrogen at the same time, and gallium oxide (Ga₂O₃) powder has been reported as an effective photocatalyst for NOCM at room temperature. In this study, we investigated the reaction conditions for Pd-loaded Ga₂O₃ photocatalysts to improve the production rate of C₂H₆ and H₂. We found that the 0.1 wt% Pd/Ga₂O₃ exhibited high selectivity of C₂H₆ (75.8%, carbon-based) under the conditions of steam reforming of methane. Photocatalytic NOCM seems to proceed in the presence of small amount of water. An increase in water vapor pressure (P_H2O) was essential for the steady production of C₂H₆ and H₂. The C₂H₆ production rate was 0.79 μmol min^?1 for 50 mg of Pd/Ga₂O₃ powder at P_H2O = 3.6 kPa. The apparent quantum efficiency (AQE) for C₂H₆ production was 5.1%, which is much higher than that of conventional photocatalytic NOCM in the absence of water vapor. The importance of water adsorbates on the photocatalyst surface was suggested by water vapor adsorption isotherm and Fourier transform infrared (FT-IR) spectroscopy. It is revealed that multilayered water molecules adsorbed on the photocatalyst surface play a role as a reaction field that promotes the dehydrogenative coupling of CH₄.

Full text of DOI:10.1016/j.jcat.2021.03.024

Journal of Catalysis 397 (2021) 192–200
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Methane coupling and hydrogen evolution induced by palladium-loaded
gallium oxide photocatalysts in the presence of water vapor
a
b,c
Mizuki Ishimaru a, Fumiaki Amano a,b, , Chiho Akamoto , Seiji Yamazoe
⇑
a Department of Chemical and Environmental Engineering, The University of Kitakyushu, 1-1 Hibikino, Wakamatsu, Kitakyushu, Fukuoka 808-0135, Japan
b Precursory Research for Embryonic Science and Technology (PRESTO), Japan Science and Technology Agency (JST), 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan
c Department of Chemistry, Graduate School of Science, Tokyo Metropolitan University, 1-1 Hachioji-shi, Tokyo 192-0397, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Non-oxidative coupling of methane (NOCM, 2CH4 ? C2H6 + H2) is a reaction that can directly produce
ethane and hydrogen at the same time, and gallium oxide (Ga2O3) powder has been reported as an effec-
tive photocatalyst for NOCM at room temperature. In this study, we investigated the reaction conditions
for Pd-loaded Ga2O3 photocatalysts to improve the production rate of C2H6 and H2. We found that the
0.1 wt% Pd/Ga2O3 exhibited high selectivity of C2H6 (75.8%, carbon-based) under the conditions of steam
reforming of methane. Photocatalytic NOCM seems to proceed in the presence of small amount of water.
An increase in water vapor pressure (PH2O) was essential for the steady production of C2H6 and H2. The
Received 24 January 2021
Revised 18 March 2021
Accepted 22 March 2021
Available online 31 March 2021
Keywords:
Photocatalysis
Nonoxidative coupling of methane
Gallium oxide
Wide bandgap semiconductor
Adsorbed Water layer
C2H6 production rate was 0.79 l
mol minꢀ1 for 50 mg of Pd/Ga2O3 powder at PH2O = 3.6 kPa. The apparent
quantum efficiency (AQE) for C2H6 production was 5.1%, which is much higher than that of conventional
photocatalytic NOCM in the absence of water vapor. The importance of water adsorbates on the
photocatalyst surface was suggested by water vapor adsorption isotherm and Fourier transform infrared
(FT-IR) spectroscopy. It is revealed that multilayered water molecules adsorbed on the photocatalyst
surface play a role as a reaction field that promotes the dehydrogenative coupling of CH4.
Ó 2021 Elsevier Inc. All rights reserved.
1. Introduction
tion [5]. To solve these problems, a single-step process that acti-
vates CH4 at lower temperatures is considered necessary.
Methane (CH4) is of considerable interest as energy resources
due to the enormous amounts of unconventional natural gases
such as shale gas and methane hydrate [1,2]. It is desired to use
CH4 more effectively as a carbon resource for chemicals as a substi-
tute for petroleum [2,3]. There are indirect catalytic processes that
convert CH4 into chemical products, but overall energy efficiencies
are not high because of the stepwise thermal reactions [4]. There-
fore, single step (direct) conversion of CH4 into chemical products
such as C2 hydrocarbons (ethane and ethylene) is desirable. Oxida-
tive coupling of methane (OCM), which activates CH4 at high tem-
peratures in the presence of oxygen and converts it to C2
hydrocarbons in a single step, has been investigated as the most
potential catalytic method [5,6]. However, there are problems such
as the sequential oxidation of the target molecule to CO2 at high
temperatures and the carbon deposition causing catalyst deactiva-
One way to activate CH4 at low temperatures is to use heteroge-
neous photocatalysts [7]. The photocatalysts excited by light
energy proceed uphill reactions with reaction Gibbs energy
DrG > 0 even at room temperature [8]. Photoexcited electrons
(eꢀ) and hole (h+) generated in semiconductor materials such as
metal oxide particles induce reductive and oxidative reactions on
the solid surface. Although the photocatalytic reaction mecha-
nisms have not been fully understood, many solid photocatalysts
have been reported for CH4 conversion reactions [9,10]. Photoelec-
trochemical CH4 conversions using semiconductor electrodes have
also been investigated, but there are few reports [11,12].
Photocatalytic non-oxidative coupling reaction of methane
(NOCM, 2CH4 ? C2H6 + H2) can convert CH4 to ethane and hydro-
gen in a single step at room temperature in the absence of oxidants
such as oxygen and water [9,10]. Many heterogeneous photocata-
lysts have been developed for NOCM [9,10,13–17]. Wang and
Zhang et al. reported that
2 wt% Pt-loaded and Ga-doped
TiO2-SiO2 photocatalyst showed CH4 conversion of 6.24% and
C2H6 selectivity (SC2H6) of 90.1% (on molar carbon basis) in a batch
reactor under ultraviolet (UV) light irradiation [13]. However, the
apparent quantum efficiency (AQE) was very low, 1 ꢁ 10ꢀ4 %, a
⇑
Corresponding author at: Department of Chemical and Environmental Engi-
neering, The University of Kitakyushu, 1-1 Hibikino, Wakamatsu, Kitakyushu,
Fukuoka 808-0135, Japan.
E-mail address: amano@kitakyu-u.ac.jp (F. Amano).
https://doi.org/10.1016/j.jcat.2021.03.024
0021-9517/Ó 2021 Elsevier Inc. All rights reserved.

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
wavelength of 350 nm. Ordomsky and Khodakov et al. reported a
stoichiometric photoreaction with AQE of 3.5% at 362 nm and
SC2H6 of 90% on TiO2 photocatalyst loaded with silver ions and het-
eropoly acids (HPW) [14].
reaction temperature suggested the importance of water adsorp-
tion on the photocatalytic surface. Therefore, we measured Fourier
transform infrared (FT-IR) spectroscopy and water vapor adsorp-
tion isotherm to discuss the effect of water molecules adsorbed
on the photocatalyst surface. We also compared the RC2H6 in the
presence of water vapor and liquid water.
Gallium oxide (Ga2O3), which is a wide bandgap semiconductor
(Eg = ~4.4 eV), has also been reported as photocatalysts that can
promote NOCM under UV light irradiation[15,17]. Yoshida et al.
reported NOCM with a CH4 conversion rate of 0.5
l
mol gꢀ1 hꢀ1
2. Experimental method
and AQE of 0.01% at 254 nm using a Ga2O3 photocatalyst [15]. They
also reported that the loading of 0.5 wt% Pd cocatalyst increased
the C2H6 formation rate (RC2H6) of bare Ga2O3 three times [17].
The ratio of C2H6 to H2 produced was 0.96, suggesting that NOCM
is proceeding with high selectivity with few byproducts. However,
the RC2H6 and AQE were not high in the NOCM system using
Ga2O3-based photocatalysts. Therefore, it is necessary to improve
the photocatalytic activity for NOCM dramatically.
Photocatalytic H2 production by water splitting is known to be
enhanced by the presence of CH4 because of the progress of photo-
catalytic steam reforming of methane (photo-SRM), defined as the
stoichiometric reaction of CH4 + 2H2O ? CO2 + 4H2 [18,19]. For
photo-SRM, we found that Pt cocatalyst-loaded Ga2O3 photocata-
lyst (Pt/Ga2O3) shows not only a high rate of H2 production but also
a high RC2H6 when methane partial pressure (PCH4) is high [20].
RC2H6 increased monotonically with the increase in PCH4 and
reached 35 mmol hꢀ1 for 50 mg of Pt/Ga2O3 at PCH4 = 300 kPa.
The SC2H6 was 67% on a carbon basis, which indicating that photo-
catalytic NOCM partly proceeds under the condition of photo-SRM
with water vapor. Based on the kinetic analysis and ESR measure-
ment, we proposed the reaction mechanism via hydroxyl radical
(ꢂOH) for C2H6 production by Pt/Ga2O3 photocatalyst in the pres-
ence of water vapor as follows [20].
2.1. Preparation of Pd/Ga2O3 photocatalysts
Ga2O3 powder (purity 99.99%) was obtained from Kojundo
Chemical Laboratory. The BET specific surface area (SBET) was
9.9 m2 gꢀ1 from the N2 adsorption isotherm measurement.
The Pd cocatalyst was loaded by impregnation method accord-
ing to the following procedure. Palladium chloride (PdCl2) aqueous
solution and deionized water (600
lL) were mixed with 1.5 g of
Ga2O3 powder dried at 100 °C for 1 h. The sample was dried again
at 100 °C for 1 h to obtain Pd(Imp)/Ga2O3 powders. The loading
amount was 0.1 wt% as Pd metal for Ga2O3.
We carried out the photodeposition method according to the
following procedure. Ga2O3 powder, 2.0 g, was added to a mixed
solution of 10 vol% ethanol (180 mL of water, 20 mL of ethanol)
and a PdCl2 solution and ultrasonicated for 5 min. The loading
amount was 0.1 wt% or 0.5 wt% Pd for Ga2O3. After stirring with
a magnetic stirrer for 30 min in the dark, UV light was irradiated
by a 40 W low-pressure mercury lamp (ASM401N, Asumi Giken)
for 1 h from the top of the beaker with magnetic stirring. Then,
Pd/Ga2O3 particles were precipitated by centrifugation for 10 min
and washed with 200 mL of deionized water under ultrasonication
for 5 min. The precipitate was dried at 100° C for 1 h to obtain a Pd
(PD)/Ga2O3 powder.
Pt=Ga2O3 þ hv ! hþ þ eꢀ
ð1Þ
2.2. Photocatalytic reaction in a flow reactor
H2O þ hþ ! ꢂOH þ Hþ
2Hþ þ 2eꢀ ! H2
ð2Þ
ð3Þ
ð4Þ
ð5Þ
We carried out photocatalytic CH4 conversion reactions using
a fixed-bed flow reactor with an irradiation area of 25 cm2
(Fig. S1). Photocatalytic powder, 50 mg, was applied onto a glass
substrate using deionized water, dried at room temperature, and
placed in a stainless-steel reactor (volume 25 cm2 ꢁ 0.025 cm).
CH4 gas with water vapor was continuously supplied at a flow
rate of 20 mL minꢀ1 at room temperature of 25 °C. The PH2O
was measured using a dew point transmitter for high humidity
environments (EE33, Tekne). A 40 W low-pressure mercury lamp
(ASM401N, Asumi-Giken) was used as the light source for deep
UV light irradiation. The incident light intensity (I0) was mea-
sured by a photometer (H9535-254, Hamamatsu Photonics) and
14 mW cmꢀ2 at a wavelength of 254 nm. The reaction tempera-
ture was controlled by an electronic cooler (HMC-17F-2700,
Hayashi-Repic) installed the bottom of the stainless reactor.
When the cooler was set to 25.0 °C, the temperature of the fixed
bed was increased to 27.0 °C under UV light irradiation. The sat-
urated vapor pressure is 3.56 kPa at 27.0 °C [23]. The stainless-
steel tubes connected to the reactor were warmed by ribbon hea-
ter to prevent water vapor’s condensation. The reactants and
products were analyzed by gas chromatography using packed col-
umns (GC-8A with MS-5A, Shincarbon ST, and Polapak-Q, Shi-
madzu). H2 was quantified by a thermal conductivity detector
(TCD) with an argon carrier. CO2 and CO were quantified by
TCD with a helium carrier. Lower hydrocarbons such as C2H6
and C3H8 were quantified by a flame ionization detector (FID)
with a nitrogen carrier.
ꢂOH þ CH4 ! ꢂCH3 þ H2O
ꢂCH3 þ ꢂCH3 ! C2H6
Yu and Li et al. reported direct production of C2H6 and H2 in a
gas–liquid-solid system using Pt/TiO2 photocatalysts suspended
in water [21]. The byproducts were CO2 and CO, and SC2H6 was
61.7%. The AQE in H2 production was 4.7% at a wavelength of
254 nm. They also reported that Pd/TiO2 photocatalyst reduced
AQE to 2.83% but improved SC2H6 to 72.6% [22]. Yoshida et al. also
reported that Pd cocatalyst was useful for NOCM among various
metals (Pt, Pd, Ag, Ni, Fe, Zn, In, Rh) [17]. These reports suggest that
Pd cocatalysts are more advantageous for C2H6 production than Pt
cocatalyst.
In this study, we investigated the characteristics of the Pd/
Ga2O3 photocatalyst using a flow reactor to improve the amount
of C2H6 produced in the CH4 conversion reaction in the presence
of small amount of water. The Pd cocatalyst was loaded by the
impregnation and the photodeposition methods. We examined
the difference between Pt and Pd and the effect of the amount of
Pd loading. Then, we investigated the effects of photocatalytic con-
ditions such as water vapor pressure (PH2O), PCH4, light intensity,
and reaction temperature. The dependence of RC2H6 on PH2O and
The AQE was calculated as the ratio of the number of electrons
consumed for the photocatalytic reaction to the number of incident
photons using the following equation.
193

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
Rp ꢁ ne ꢁ NA
Ga2O3 particles with a width of 500 nm to 1
lm and a length of 2–
h
i
%AQE ¼
ꢁ 100
ð6Þ
ðI0 ꢁ A ꢁ 60 s minꢀ1 Þ=ðhc=kÞ
4
lm (Fig. S3). The backscattered electron detector (BED) image,
which shows the difference in the elemental composition, indi-
cates the presence of Pd particles as bright spots with a size of sev-
eral tens of nanometers or less. TEM images of 0.5 wt% Pd(PD)/
Ga2O3 revealed that the photodeposited Pd particles were about
10–20 nm in diameter (Fig. S4).
where Rp, ne, NA, I0, A, h, c, and k stand for the rate of molecule
production (in mol minꢀ1), the number of electrons involved in
the reaction, the Avogadro constant, light irradiance (in W cmꢀ2),
irradiation area (A = 25 cm2), the Planck constant, the speed of
light, and the wavelength of incident light (k = 254 nm). The AQE
for H2 was calculated using ne = 2 assuming a two-electron process
(Eq. 3).
We performed Pd K-edge XAFS measurement to investigate the
difference in the chemical state of the Pd cocatalysts prepared by
the photodeposition and the impregnation methods. The X-ray
absorption near edge structure (XANES) spectrum of 0.5 wt% Pd
(PD)/Ga2O3 shows that the photodeposited Pd species was in a
metallic state (Fig. S5). On the other hand, the XANES spectrum
of Pd(Imp)/Ga2O3 was similar to PdCl2. Similar results were
obtained from extended X-ray absorption fine structure (EXAFS)
analysis (Fig. S6). The results of the curve fitting of the EXAFS spec-
tra are summarized in Table S1. The coordination number of Pd–Pd
bonds was 9.7 for Pd(PD)/Ga2O3. In the case of Pd(Imp)/Ga2O3, Pd–
Pd bonds were hardly observed, and the coordination number of
Pd-Cl bonds was 3.8, indicating that the local structure of the
impregnated Pd species is like PdCl2.
2.3. Characterization of photocatalyst powders
X-ray diffraction (XRD) pattern was measured on a SmartLab
diffractometer (Rigaku) using Cu K
tron microscopy (TEM) image was obtained using a JEM-3010
(JEOL) to observe the particles fixed on a microgrid. For scanning
electron microscope (SEM) image, JSM-7800F (JEOL) was used to
observe the particles fixed on the surface of carbon tape. The dif-
a radiation. Transmission elec-
fuse reflectance UV–visible spectrum was measured using
a
UV2600 spectrophotometer with an ISR-2600Plus integrating
sphere accessory (Shimadzu) using barium sulfate as a reference
sample.
The diffuse reflectance spectrum of FT-IR was measured using
IRSpirit-T (Shimadzu) with a diffuse reflection chamber HC1000
(ST Japan) under humid CH4 flow of 20 mL minꢀ1. An aluminum
mirror was used as a reference sample. Water vapor adsorption
3.2. Effect of cocatalyst on the photocatalytic activity
The photocatalytic CH4 conversion was investigated for 0.1 wt%
Pt(Imp)/Ga2O3 and Pd(Imp)/Ga2O3 prepared by the impregnation
method. Fig. 1 shows the time course of at PCH4 = 98 kPa and
PH2O = 3.3–3.6 kPa under UV irradiation. For Pt(Imp)/Ga2O3, the
amount of CO2 produced was higher than that of C2H6, suggesting
that photo-SRM proceeded predominantly. On the other hand,
C2H6 was produced more than CO2 for Pd(Imp)/Ga2O3. The selectiv-
ity to C2H6 calculated on a molar carbon basis was 75.8%, which
was much higher than SC2H6 of Pt(Imp)/Ga2O3, indicating that Pd
is a more suitable cocatalyst for C2H6 production than Pt. In the
photocatalytic CH4 conversion on Pd(Imp)/Ga2O3, CO and a small
amount of C3H8 was detected as byproducts. We also confirmed
the corresponding decrease of CH4 inlet under photoirradiation
(Fig. S7), indicating that CH4 is a carbon source of C2H6 and CO2.
and desorption isotherm was measured at 25 °C using
a
BELSORP-max (MicrotracBEL). The Ga2O3 powder (1.0 g) was evac-
uated at 200 °C for 2 h before the measurement. Electron paramag-
netic resonance (EPR) spectrometry was performed on Magnettech
ESR5000 (Bruker) at room temperature. The Pd(Imp)/Ga2O3 pow-
der (20 mg) was dispersed in an aqueous solution of 5,5-
dimethyl-l-pyrroline-N-oxide (DMPO, 0.1 mol Lꢀ1) as a radical
trapping agent. The suspension was irradiated by UV light emitted
from a 200-W mercury-xenon lamp (L9588-02A, Hamamatsu Pho-
tonics). The X-band EPR parameters were magnetic field of 330–
340 mT, modulation amplitude of 0.1 mT, modulation frequency
of 100 kHz, and microwave power of 10 mW.
The X-ray absorption fine structure (XAFS) of the Pd K-edge was
measured at the beamline BL01B1 of the SPring-8 facility (Japan
Synchrotron Radiation Research Center, Hyogo) with a Si (311)
double-crystal monochromator at room temperature. The refer-
ence samples were measured by the transmission method using
ionization chambers, and the photocatalyst samples were mea-
sured by the fluorescence method using a 19-element germanium
solid-state detector. The XAFS spectrum was analyzed using
1.2
(a) Pt/Ga2O3
4
3
2
1
0
1
0.8
0.6
0.4
0.2
0
H2
CO2
C2H6
xTunes software [24]. The
v spectrum as a function of photoelec-
CO
tron wavenumber k was weighted by k3, and Fourier transformed
in the range of k = 3.0–15.8 Åꢀ1. The R range of 1.4–2.8 Å was
inverse Fourier transformed and curve-fitted. The phase shift and
backward scattering amplitude functions of Pd–Pd and Pd–Cl were
extracted from Pd metal and PtCl2 using the FEFF8 program.
0
60 120 180 240 300
Time on stream / min
1.2
1
(b) Pd/Ga2O3
4
C2H6
0.8
0.6
0.4
0.2
0
3
2
1
0
H2
3. Results and discussion
CO2
CO
3.1. Morphology and structure of Pd/Ga2O3 photocatalysts
We performed XRD measurement to analyze crystalline compo-
nents of Pd(PD)/Ga2O3 powder. The XRD pattern exhibited the
peaks assigned to b-Ga2O3 (Powder Diffraction File 01–087-1901)
(Fig. S2). The diffraction peak of Pd metal was not confirmed for
0.1 wt% and 0.5 wt% Pd(PD)/Ga2O3. This is due to the high disper-
sion of Pd and the overlapping with Ga2O3 peaks.
0
60 120 180 240 300
Time on stream / min
Fig. 1. Time course of the production rate in the photocatalytic conversion of CH4
(PCH4 = 98 kPa) with H2O vapor: (a) 0.1 wt% Pt(Imp)/Ga2O3 (PH2O = 3.3 kPa), (b)
0.1 wt% Pd(Imp)/Ga2O3 (PH2O = 3.6 kPa). Photoirradiation (14 mW cmꢀ2) was
performed for 60–240 min on stream. The reactor temperature was set to 25 °C. The
We observed the morphology of Pd(PD)/Ga2O3 particles by SEM
and TEM. The SEM image of 0.5 wt% Pd(PD)/Ga2O3 shows the large
total flow rate was 20 mL minꢀ1
.
194

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
We compared the impregnation and the photodeposition meth-
ods to examine the effect of Pd cocatalyst loading. Fig. 2 shows the
CH4 conversion time course for 0.5 wt% Pd(Imp)/Ga2O3 and Pd(PD)/
Ga2O3. There was an induction period in the photocatalytic product
formation for Pd(Imp)/Ga2O3 since the initial oxidation state is
Pd2+ species. UV–visible diffuse reflection spectra revealed that
PdCl2-like species in Pd(Imp)/Ga2O3 was reduced to Pd0 during
the reaction (Fig. S8). This suggests that photoexcited electrons
reduce the Pd species at the beginning of light irradiation, and
the Pd metal particles exit during the photocatalytic reaction. In
the case of 0.1 wt% Pd loading, there was no induction period in
the photocatalytic reaction (Figure S9), suggesting the fast reduc-
tion of the supported PdCl2-like species. The Pd(Imp)/Ga2O3 with
in-situ formed Pd0 species showed RC2H6 slightly higher than Pd
(PD)/Ga2O3.
1.2
1
3
(a)
H2
C2H6
0.8
0.6
0.4
0.2
0
2
1
0
CO2
CO
0
0.05
Pd loading / wt%
0.1
0.5
100
80
60
40
20
0
(b)
C2H6
Fig. 3 shows the effect of Pd cocatalyst loading on the product
formation over Pd(Imp)/Ga2O3 after 3-h photoirradiation. The bare
Ga2O3 shows the low H2 and CO2 production rate, suggesting that
photo-SRM is sluggish in the absence of Pd cocatalyst. The Pd
cocatalyst significantly enhanced the reaction rate of photo-SRM
CO2
CO
and also the formation of C2H6. The rate of H2 production (RH2
)
0
0.05
Pd loading / wt%
0.1
0.5
was significantly increased by loading few Pd (less than
a
0.1 wt%), suggesting that metallic Pd cocatalyst provides a catalytic
site for H2 production. The CO2 formation rate (RCO2) was not
dependent on the Pd loading when the amount is higher than
0.025 wt%. On the other hand, the RC2H6 was gradually increased
with the Pd loading, indicating that the Pd cocatalyst also affects
the selectivity on the oxidative reaction side. The RC2H6 and SC2H6
Fig. 3. Effect of Pd loading amount for Pd(Imp)/Ga2O3 on (a) the production rate at
3-h photoirradiation and (b) the carbon-based selectivity in the photocatalytic CH4
conversion with H2O vapor (PCH4 = 98 kPa, PH2O = 3.0–3.6 kPa, Temperature = 25 °C,
Flow rate = 20 mL minꢀ1, Light intensity = 14 mW cmꢀ2).
were 0.89 l
mol minꢀ1 and 81.7% for 0.5 wt% Pd(Imp)/Ga2O3.
diation is summarized in Figure S10. Under the NOCM conditions
without the oxidant (PH2O = 0.09 kPa), photocatalytic activity was
very low since photo-SRM is difficult in the absence of water vapor.
The addition of water vapor (PH2O = 1.5 kPa) increase RH2 and RCO2
due to the progress of photo-SRM. We also found a significant
increase of RC2H6 by water vapor, indicating that water molecule
is useful not only for photo-SRM but also for C2H6 production (ap-
parently NOCM).
3.3. Effect of water vapor on the photocatalytic reaction
We investigated the effect of adding water vapor on the CH4
conversion on a 0.1 wt% Pd(Imp)/Ga2O3 photocatalyst. Fig. 4 shows
the time course of photocatalytic reactions in the CH4 stream at
different PH2O (0.1–3.6 kPa). In the case of PH2O = 2.9 kPa, the pro-
duction rate was degraded after 1-h photoirradiation. The details of
this degradation will be discussed in the following sections. The
effect of PH2O on each product’s formation rate after 3-h photoirra-
The CO as a byproduct was detected under the conditions of
PH2O ꢃ 2.9 kPa (Fig. 4d). The CO production from CH4 can be
expressed by steam reforming of methane (SRM: CH4 + H2O ? CO +
1.2
1
0.8
0.6
0.4
0.2
0
3
(a) 0.5 wt% Pd/Ga2O3
(b)
PD
Imp
Imp
2
1
0
PD
0
60 120 180 240 300
Time on stream / min
0
60 120 180 240 300
Time on stream / min
1.2
1
0.6
0.4
0.2
0
(c)
(d)
0.8
0.6
0.4
0.2
0
PD
Imp
PD
Imp
0
60 120 180 240 300
Time on stream / min
0
60 120 180 240 300
Time on stream / min
Fig. 2. Time course of the production rate of (a) C2H6, (b) H2, (c) CO2, and (d) CO in the photocatalytic CH4 conversion on 0.5 wt% Pd(Imp)/Ga2O3 and Pd(PD)/Ga2O3 with H2O
vapor (PCH4 = 98 kPa, PH2O = 3.3 kPa, Temperature = 25 °C, Flow rate = 20 mL minꢀ1, Photoirradiation 60–240 min, Light intensity = 14 mW cmꢀ2).
195

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
1.2
1
3
2
1
0
PH2O / kPa
2.9
(a)
(b)
PH2O / kPa
3.6
0.8
0.6
0.4
0.2
0
3.6
2.9
2.5
1.5
1.5
2.5
0.1
0.1
0
60 120 180 240 300
Time on stream / min
0
60 120 180 240 300
Time on stream / min
1.2
1
0.6
0.4
0.2
0
(c)
(d)
0.8
0.6
0.4
0.2
0
PH2O / kPa
2.9
3.6
PH2O / kPa
1.5
3.6
2.5
0.1
2.9
0
60 120 180 240 300
Time on stream / min
0
60 120 180 240 300
Time on stream / min
Fig. 4. Effect of PH2O on the production rate of (a) C2H6, (b) H2, (c) CO2, and (d) CO in the photocatalytic CH4 conversion on 0.1 wt% Pd(Imp)/Ga2O3 with H2O vapor (PCH4 = 98–
101 kPa, PH2O = 0.1–3.6 kPa, Temperature = 25 °C, Flow rate = 20 mL minꢀ1, Photoirradiation 60–240 min, Light intensity = 14 mW cmꢀ2).
3H2) defined in an industrial catalytic process. We evaluated the
material balance between the reduction product of H2 and the oxi-
dation products, assuming that C2H6, CO, and CO2 are derived from
the dehydrogenation of CH4 (Fig. S11). The quantitative error was
less than 10% at PH2O ꢃ 1.5 kPa, guaranteeing few missing products.
The photocatalytic CH4 conversion over 0.1 wt% Pd(Imp)/Ga2O3 can
be expressed in the combination of three reactions: NOCM, SRM to
evolve CO2, and SRM to evolve CO. Noted that thermogravimetry
differential thermal analysis (TG-DTA) of the photocatalyst after
the reaction indicates that there is a trace amount of carbon depo-
sition on the surface of the used photocatalyst (Fig. S11).
cates that the cause of deactivation at PH2O = 2.9 kPa is not the irre-
versible deactivation of the Pd/Ga2O3 photocatalyst but the
reversible change in the reaction field. This phenomenon might
be related to the amount of water vapor adsorbed on the surface,
which is dependent on PH2O. The production rate was very stable
at the relative humidity (RH) close to 100% (PH2O = 3.6 kPa at
27 °C). The temperature inside the reactor was increased to 27 °C
under UV irradiation even though the electric cooler temperature
was set to 25 °C.
In the case of 0.1 wt% Pd(Imp)/Ga2O3 photocatalyst, the AQE for
H2 evolution was 11.6% at PH2O = 3.6 kPa, and the SC2H6 was 75.8%
hꢀ1
ꢀ1
The product time courses (Fig. 4) showed characteristic behav-
ior at PH2O = 2.9 kPa. The RC2H6 was high at the initial stage (60–
on carbon basis. The RC2H6 reached 951
(0.793
l
mol
g
cat
l
mol minꢀ1 for 50 mg photocatalyst). The high performance
120 min on stream) but decreased to 0.23
l
mol minꢀ1 after 3-h
was stable only at high RH, suggesting that the state of water
adsorbed on the surface was related to the reaction field of the
Pd/Ga2O3 photocatalyst. To investigate the amount of water
adsorbed, we measured the water vapor adsorption and desorption
isotherm of Ga2O3 powder at 25 °C (Fig. S12). The amount of water
adsorbed on the Ga2O3 surface was increased with an increase in
the PH2O. The adsorbed amount was significantly increased at
PH2O = 2.9 kPa (RH = 91%). This type II isotherm is characteristic
of multi-layer adsorption in non-porous materials [25]. The
amount of water adsorbed at PH2O = 2.9 kPa was 12.8 cm3 gꢀ1 as
an ideal gas under standard conditions (0 °C, 1 atm). Since the SBET
of Ga2O3 was 9.9 m2 gꢀ1, the water molecules adsorbed was
3.5 ꢁ 1015 cmꢀ2. The number of water molecules in one monolayer
is about 1 ꢁ 1015 cmꢀ2 calculated from liquid water density. There-
fore, approximately 3.5 monolayers (about 1 nm thickness) are
formed on Ga2O3 at PH2O = 2.9 kPa and 25 °C. The multilayered
water molecules with nanometer-scale thickness were necessary
to induce stable CH4 conversion on the Pd/Ga2O3 photocatalyst’s
surface.
photoirradiation. To investigate the cause of this deactivation, we
conducted a stepwise reaction by changing PH2O after the reaction
at 2.9 kPa. Fig. 5 shows the time course when the deactivated pho-
tocatalyst was repeatedly used at PH2O = 3.6 kPa. The degradation
was observed at PH2O = 2.9 kPa like in Fig. 4, but the production rate
was recovered when PH2O was changed to 3.6 kPa. The result indi-
Light on
off
Light on
off
1.2
1
4
3
2
1
0
PH2O = 2.9 kPa
PH2O = 3.6 kPa
C2H6
0.8
0.6
0.4
0.2
0
H2
CO2
CO
0
120
240
360
480
600
3.4. Effect of reaction temperature on water adsorption
Time on stream / min
Fig. 5. Time course of the photocatalytic CH4 conversion on 0.1 wt% Pd(Imp)/Ga2O3
with different H2O vapor pressure (PCH4 = 97–98 kPa, PH2O = 2.9 and 3.6 kPa,
Temperature = 25 °C, Flow rate = 20 mL minꢀ1, Light intensity = 14 mW cmꢀ2).
Photoirradiation was performed for 60–240 min and 360–540 min on stream. PH2O
was changed from 2.9 kPa to 3.6 kPa at 300 min on stream.
In general, the amount of water adsorption decreases at high
temperatures and increases at low temperatures. Therefore, the
amount of water adsorbed on the Pd/Ga2O3 surface can be con-
trolled by changing the reactor temperature. Consequently, we
196

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
investigated the effect of reactor temperature on the photocatalytic
CH4 conversion reaction. The time courses on a 0.1 wt% Pd(Imp)/
Ga2O3 photocatalyst are shown in Fig. S13. Fig. 6 shows each pro-
duct’s formation rate after 3-h photoirradiation as a function of the
reaction temperature. The RC2H6 was maximized at room tempera-
ture (25 °C), and decreased as the temperature increased. The RC2H6
dropped significantly to 0.05
l
mol minꢀ1 at 65 °C. Yoshida et al.
reported thermal acceleration of electron migration in Ga2O3 pho-
tocatalysis at temperatures less than 70 °C [19]. Therefore, the eꢀ–
h+ recombination is not the reason for the decreased RC2H6. The
decreased activity would be due to the decrease in the amount of
reactants adsorbed on the Ga2O3 photocatalyst at higher
temperatures.
To investigate the change in the amount of reactants adsorbed,
we performed diffuse reflectance FT-IR measurement of the Ga2O3
powder at different temperatures (Fig. 7). The FT-IR spectra were
recorded for the powder under the CH4 flow with water vapor of
3.0 kPa. The absorption peak with a wavenumber of 3010 cmꢀ1
is assigned to the CH stretching vibration of CH4 in the gas phase.
The adsorption of CH4 is very weak at around room temperature.
On the other hand, we observed broad absorption bands at
3460 cmꢀ1 and 1636 cmꢀ1 due to the adsorbed water. These bands
are attributed to the OH stretching vibration (3460 cmꢀ1) and the
H–O–H bending vibration (1636 cmꢀ1) of the hydrogen-bonded
water molecules, respectively [26,27]. Increasing the temperature
from 25 °C to 65 °C decreased the intensity of the OH stretching
and H–O–H bending bands. This result indicates that the amount
of water adsorbed on the Ga2O3 photocatalyst decreases when
the reaction temperature is higher than room temperature. There-
fore, we can conclude that the decrease in photocatalytic activity at
35 and 65 °C is due to the decrease in the adsorbed water.
When the photocatalytic reaction was performed at low tem-
Fig. 7. Diffuse reflectance FT-IR spectra of Ga2O3 powders under the stream of CH4
with H2O vapor (PCH4 = 98 kPa, PH2O = 3.0 kPa) at different temperatures (25–65 °C).
(a) OH stretching band region of 4100–2500 cmꢀ1
, and (b) OH deformation
vibration band region of 1800–1400 cmꢀ1. The green curve was obtained under dry
Ar stream at 100 °C. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
l
mol minꢀ1 in the presence of liquid water.
perature (15 °C), RC2H6 was decreased to 0.40
l
mol minꢀ1
decreased to 0.08
The addition of water droplets to Pd/Ga2O3 powder also decreased
RCO and RH2. On the other hand, the RCO2 was not affected by the
presence of liquid water. Therefore, the CO2 selectivity was
increased by the formation of a thick water layer. This is because
hydrophobic CH4 is difficult to approach the photocatalyst surface
covered with liquid water, meanwhile the amount of hydroxy rad-
ical (ꢂOH) is constant. The ꢂOH, which is produced by one-electron
oxidation of water molecule by photogenerated h+, has strong oxi-
dizing power for organic compounds and is known as an active
species for photocatalytic oxidative decomposition into CO2
[29–31]. When the CH4 supply to the Ga2O3 surface is slowed
down, the surface concentration of ꢂOH will increase. The exces-
sive ꢂOH promotes sequential oxidation of CH4 into CO2. The active
species for one-electron oxidation of CH4 to produce ꢂCH3 is
assumed to be ꢂOH (Equation 4). To prove the formation of ꢂOH,
we conducted EPR spin-trapping technique with DMPO
(Figure S15). The signal observed under UV irradiation was consis-
tent to the reported 1:2:2:1 quartet spectrum for DMPO-OH spin
adduct (g = 2.005, AN = 1.49 mT, AH = 1.49 mT) [32,33]. The AN
and AH are hyperfine splittings of the N and b-H atoms of the
DMPO-OH adduct. The ꢂOH to convert CH4 to C2H6 would be
generated in the water layer adsorbed on Pd/Ga2O3 surface
(Figure S16). On the other hand, the complete oxidation to CO2 is
promoted when the surface concentration of ꢂOH is much higher
than the supply of CH4 to the surface.
(Fig. 6). The saturated water vapor pressure at a temperature of
15 °C is 1.7 kPa [23]. Therefore, part of the water vapor condenses
when 3.0 kPa of water vapor is continuously supplied to the reactor
cooled to 15 °C. In fact, we observed that water droplets were
formed in the reactor, and the Pd/Ga2O3 powder was wetted after
the reaction. This suggests that the thick layer of liquid water
formed on the Pd/Ga2O3 surface negatively affects the photocat-
alytic activity. The solubility of hydrophobic CH4 in water is as
low as 1.6 mmol Lꢀ1, and the diffusion coefficient in water is very
small (0.0018 mm2 sꢀ1) [28]. On the other hand, the concentration
of CH4 in the gas phase is 41 mmol Lꢀ1, and the diffusion coefficient
(21 mm2 sꢀ1) is high in the air. This supports the hypothesis that
the thick water layer prevents the supply of CH4 to the photocata-
lyst surface, resulting in low photocatalytic activity.
To confirm the effect of liquid water, we carried out photocat-
alytic reactions using 0.1 wt% Pd(Imp)/Ga2O3 powder wetted with
200
lL of water droplets (Fig. S14). The RC2H6 was significantly
1.2
1
3
2
1
0
H2
C2H6
0.8
0.6
0.4
0.2
0
CO2
CO
25
3.5. Effect of other conditions on the photocatalytic reaction
10
40
55
70
Temperature / °C
We investigated the effect of total pressure on the photocat-
alytic CH4 conversion over 0.1 wt% Pd(Imp)/Ga2O3 photocatalyst
(Fig. 8). When the total pressure was 1 atm or less, RC2H6 and RH2
increased with PCH4. However, the increasing trend of RC2H6 was
not clear at PCH4 ꢃ 100 kPa or more. This result is different from
Fig. 6. Effect of reaction temperature on the production rate in the photocatalytic
CH4 conversion on 0.1 wt% Pd(Imp)/Ga2O3 with H2O vapor (PCH4 = 97–98 kPa,
PH2O = 3.0–3.7 kPa, Temperature = 15–65 °C, Flow rate = 20 mL minꢀ1, Light
intensity = 14 mW cmꢀ2). The data was obtained after 3-h photoirradiation.
197

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
the result observed for Pt/Ga2O3 photocatalyst; The RC2H6 is mono-
tonously increased up to PCH4 = 300 kPa when Pt is cocatalyst [20].
1.2
1
4
3
(a)
The RC2H6 on Pd/Ga2O3 was 0.64
which was much higher than 0.10
l
mol minꢀ1 at PCH4 = 50 kPa,
mol minꢀ1 for Pt/Ga2O3 at
0.8
0.6
0.4
0.2
0
l
C2H6
the same condition [20]. Therefore, we can conclude that Pd/
Ga2O3 is an excellent photocatalyst for selective C2H6 production
even at low PCH4 in the presence of water vapor. The highest
SC2H6 (75.8%) was observed at PCH4 = 100 kPa since CO2 selectivity
slightly decreased at high PCH4. We understand that this is because
high PCH4 increased the surface concentration of ꢂCH3, promoting
coupling into C2H6.
H2
2
1
0
CO2
CO
0
5
10
15
-2
Light intensity / mW cm
Fig. 9 shows the effect of incident light intensity for 0.1 wt% Pd
(Imp)/Ga2O3 photocatalyst. The RCO2 increased linearly with light
intensity. RC2H6 also increased with increasing light intensity, but
the increase became gradual at above 5 mW cmꢀ2. As a result,
the SC2H6 was slightly decreased as the light intensity increased.
When the light intensity is high, the high density ꢂOH promotes
the sequential oxidation into CO2. This implies that a moderate
concentration of surface ꢂOH species is required for highly selec-
tive NOCM. The highest SC2H6 of 81.5% was observed at low light
intensity (5 mW cmꢀ2) for 0.1 wt% Pd(Imp)/Ga2O3.
100
80
60
40
20
0
(b)
C2H6
CO2
CO
Fig. 10 shows the effect of the feed gas flow rate on the photo-
catalytic reaction over 0.1 wt% Pd(Imp)/Ga2O3. We calculated the
conversion rate of CH4 from the ratio of each production rate
(RC2H6, RCO2, RCO) to the flow rate of the CH4 inlet. The formation
rates of each product were constant during photoirradiation. The
RC2H6 was remarkably increased when the CH4 flow rate was
0
5
10
15
Light intensity / mW cm-2
Fig. 9. Effect of light intensity on the production rate in the photocatalytic CH4
conversion on 0.1 wt% Pd(Imp)/Ga2O3 with H2O vapor (PCH4 = 98 kPa, PH2O = 3.6 kPa,
Temperature = 25 °C, Flow rate = 20 mL minꢀ1, Light intensity = 5.0–14 mW cmꢀ2).
The data were obtained after 3-h photoirradiation.
increased. The RC2H6 was 1.11
l
mol minꢀ1, the SC2H6 was 80.4%,
and AQE for H2 evolution was 14.4% at the flow rate of 30 mL minꢀ1
.
The AQE for C2H6 and CO2 was 5.1% and 8.8% assuming a two-
electron process (2CH4 + 2 h+ ? C2H6 + 2H+) and an eight-
electron process (CH4 + 2H2O + 8 h+ ? CO2 + 8H+), respectively.
When the CH4 flow rate was 40 mL minꢀ1, RC2H6 was saturated,
but SC2H6 reached a maximum of 83.3%. The RC2H6 enhanced by
the increased flow rate is probably because of the improvement
1.2
1
4
C2H6
(a)
3.2
2.4
1.6
0.8
0
H2
0.8
0.6
0.4
0.2
0
CO2
CO
1.2
(a)
H2
0
10 20 30 40
3
2
1
0
1
0.8
0.6
0.4
0.2
0
Flow rate / mL min-1
C2H6
CO2
1
0.8
0.6
0.4
0.2
0
100
80
60
40
20
0
(b)
CO
0
100 200 300
P(CH4) / kPa
100
(b)
0
10 20 30 40
Flow rate / mL min-1
C2H6
80
60
40
20
0
Fig. 10. Effect of flow rate of CH4 on (a) the production rate after 3-h photoirra-
diation and (b) the CH4 conversion and the C2H6 selectivity in the photocatalytic
CH4 conversion on 0.1 wt% Pd(Imp)/Ga2O3 with H2O vapor (PCH4 = 97–98 kPa,
PH2O = 3.2–3.7 kPa, Temperature = 25 °C, Flow rate = 5–40 mL minꢀ1, Light
intensity = 14 mW cmꢀ2).
CO2
CO
0
100 200
300
in the diffusivity of ꢂCH3 on the photocatalyst surface to suppress
the sequential decomposition. On the other hand, SC2H6 decreased
with the promotion of CO2 formation when the flow rate was set
to 5 mL minꢀ1. This suggests that the sequential oxidations by
ꢂOH is easily promoted at the slow flow rate. The CH4 conversion
rate was 0.58% when the flow rate was 5 mL minꢀ1, although the
SC2H6 was not high.
P(CH4) / kPa
Fig. 8. Effect of PCH4 on (a) the production rate after 3-h photoirradiation and (b)
the carbon-based selectivity in the photocatalytic CH4 conversion on 0.1 wt% Pd
(Imp)/Ga2O3 with H2O vapor (PCH4 = 25–300 kPa, PH2O = 3.5–3.7 kPa, Tempera-
ture = 25 °C, Flow rate = 20 mL minꢀ1, Light intensity = 14 mW cmꢀ2). Argon was
used as a balance gas when total pressure was less than 101 kPa. The pressure
higher than 101 kPa was regulated by a back-pressure valve.
198

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
4. Conclusion
References
[1] S.Y. Lee, G.D. Holder, Methane hydrates potential as a future energy source,
Fuel Process. Technol. 71 (2001) 181–186, https://doi.org/10.1016/S0378-3820
(01)00145-X.
[2] J. Baltrusaitis, I. Jansen, J.D. Schuttlefield Christus, Renewable energy based
catalytic CH4 conversion to fuels, Catal. Sci. Technol. 4 (2014) 2397–2411,
https://doi.org/10.1039/c4cy00294f.
In this study, we investigated the characteristics of the Pd/
Ga2O3 photocatalyst for CH4 conversion reaction in the presence
of water vapor. Pd/Ga2O3 was effective for selective C2H6 formation
compared with Pt/Ga2O3. The production rate of C2H6 was larger
when the PdCl2-like species was initially loaded by the impregna-
tion method. The Pd species are reduced to a metallic state, which
is the site promoting H2 production, during the photocatalytic
reaction.
[3] H. Song, X. Meng, Z. Wang, H. Liu, J. Ye, Solar-Energy-Mediated Methane
Conversion, Joule.
joule.2019.06.023.
3
(2019) 1606–1636, https://doi.org/10.1016/j.
[4] W. Taifan, J. Baltrusaitis, CH4 conversion to value added products: Potential,
limitations and extensions of a single step heterogeneous catalysis, Appl. Catal.
B Environ. 198 (2016) 525–547, https://doi.org/10.1016/j.apcatb.2016.05.081.
[5] Y. Gambo, A.A. Jalil, S. Triwahyono, A.A. Abdulrasheed, Recent advances and
future prospect in catalysts for oxidative coupling of methane to ethylene: A
review, J. Ind. Eng. Chem. 59 (2018) 218–229, https://doi.org/10.1016/j.
jiec.2017.10.027.
We investigated the effect of adding water vapor on the C2H6
formation rate over the Pd/Ga2O3 photocatalyst. When water was
not added, the amount of C2H6 produced was very small. However,
steady C2H6 formation (RC2H6 = 0.79 l
mol minꢀ1) was confirmed at
[6] P. Schwach, X. Pan, X. Bao, Direct Conversion of Methane to Value-Added
Chemicals over Heterogeneous Catalysts: Challenges and Prospects, Chem.
Rev. 117 (2017) 8497–8520, https://doi.org/10.1021/acs.chemrev.6b00715.
[7] X. Cao, T. Han, Q. Peng, C. Chen, Y. Li, Modifications of heterogeneous
photocatalysts for hydrocarbon C-H bond activation and selective conversion,
Chem. Commun. 56 (2020) 13918–13932, https://doi.org/10.1039/
d0cc05785a.
[8] A. Kudo, Y. Miseki, Heterogeneous photocatalyst materials for water splitting,
Chem. Soc. Rev. 38 (2009) 253–278, https://doi.org/10.1039/b800489g.
[9] K. Shimura, H. Yoshida, Semiconductor Photocatalysts for Non-oxidative
Coupling, Dry Reforming and Steam Reforming of Methane, Catal. Surv. from
Asia. 18 (2014) 24–33, https://doi.org/10.1007/s10563-014-9165-z.
[10] L. Yuliati, H. Yoshida, Photocatalytic conversion of methane, Chem. Soc. Rev. 37
(2008) 1592–1602, https://doi.org/10.1039/b710575b.
PH2O = 3.6 kPa. The AQE for H2 production was 11.6%, and the selec-
tivity of C2H6 was 75.8% on a carbon basis. The photocatalytic
activity was reversibly affected by the PH2O. Also, RC2H6 decreased
significantly under the conditions for condensation of water vapor
to liquid water. The presence of water vapor close to the saturated
water vapor pressure was necessary for inducing the photocat-
alytic dehydrogenative coupling of CH4 to produce C2H6 and H2.
We evaluated the adsorbed water on the photocatalyst surface
by measuring the adsorption isotherm of water vapor and In-situ
FT-IR spectroscopy. The water molecules adsorbed on the surface
provides a reaction field promoting C2H6 production. The multilay-
ered water molecules with a thickness of about 1 nm (3.5 mono-
layers) facilitate the reaction based on the following
mechanisms: 1) increasing the amount of ꢂOH, which activates
CH4 to ꢂCH3, 2) preventing the sequential oxidative decomposition
of the ꢂCH3 by ꢂOH, and 3) increasing the homocoupling of ꢂCH3 to
C2H6. The dependence on the C2H6 production rate on the light
intensity and the flow rate implies that CO2 formation is acceler-
ated when ꢂOH is excessively generated, and C2H6 is easily pro-
duced when the diffusivity of the ꢂCH3 is improved. These new
insights on a hydrophilic reaction field advance the surface engi-
neering controlling the photocatalysis for CH4 conversion.
[11] F. Amano, A. Shintani, K. Tsurui, H. Mukohara, T. Ohno, S. Takenaka,
Photoelectrochemical Homocoupling of Methane under Blue Light
Irradiation, ACS Energy Lett.
acsenergylett.8b02436.
4 (2019) 502–507, https://doi.org/10.1021/
[12] W. Li, D. He, G. Hu, X. Li, G. Banerjee, J. Li, S.H. Lee, Q. Dong, T. Gao, G.W.
Brudvig, M.M. Waegele, D.E. Jiang, D. Wang, Selective CO Production by
Photoelectrochemical Methane Oxidation on TiO2, ACS Cent. Sci. 4 (2018) 631–
637, https://doi.org/10.1021/acscentsci.8b00130.
[13] S. Wu, X. Tan, J. Lei, H. Chen, L. Wang, J. Zhang, Ga-Doped and Pt-Loaded Porous
TiO2-SiO2 for Photocatalytic Nonoxidative Coupling of Methane, J. Am. Chem.
Soc. 141 (2019) 6592–6600, https://doi.org/10.1021/jacs.8b13858.
[14] X. Yu, V.L. Zholobenko, S. Moldovan, D. Hu, D. Wu, V.V. Ordomsky, A.Y.
Khodakov, Stoichiometric methane conversion to ethane using photochemical
looping at ambient temperature, Nat. Energy. 5 (2020) 511–519, https://doi.
org/10.1038/s41560-020-0616-7.
The Pd/Ga2O3 photocatalyst shows high efficiency for H2 pro-
duction (AQE = 14.4%) and the C2H6 selectivity higher than 80%
at a CH4 flow rate of 30 mL minꢀ1. The developed photocatalytic
system was much better than previously reported ones for NOCM
in the absence of water vapor. However, the CH4 conversions in the
flow reactor were still low at the industrial level. Therefore, it is
necessary to improve the CH4 conversion rate further while main-
taining high selectivity for C2H6 in the photocatalytic system.
[15] L. Yuliati, T. Hattori, H. Itoh, H. Yoshida, Photocatalytic nonoxidative coupling
of methane on gallium oxide and silica-supported gallium oxide, J. Catal. 257
(2008) 396–402, https://doi.org/10.1016/j.jcat.2008.05.022.
[16] J. Lang, Y. Ma, X. Wu, Y. Jiang, Y.H. Hu, Highly efficient light-driven methane
coupling under ambient conditions based on an integrated design of
a
photocatalytic system, Green Chem. 22 (2020) 4669–4675, https://doi.org/
10.1039/d0gc01608j.
[17] S.P. Singh, A. Anzai, S. Kawaharasaki, A. Yamamoto, H. Yoshida, Non-oxidative
coupling of methane over Pd-loaded gallium oxide photocatalysts in a flow
reactor, Catal. Today. 362 (2021) in Press. https://doi.org/10.1016/
j.cattod.2020.04.023.
[18] K. Shimura, T. Yoshida, H. Yoshida, Photocatalytic activation of water and
methane over modified gallium oxide for hydrogen production, J. Phys. Chem.
C. 114 (2010) 11466–11474, https://doi.org/10.1021/jp1012126.
[19] K. Shimura, K. Maeda, H. Yoshida, Thermal acceleration of electron migration
in gallium oxide photocatalysts, J. Phys. Chem. C. 115 (2011) 9041–9047,
https://doi.org/10.1021/jp110824n.
[20] F. Amano, C. Akamoto, M. Ishimaru, S. Inagaki, H. Yoshida, Pressure-induced
dehydrogenative coupling of methane to ethane by platinum-loaded gallium
oxide photocatalyst, Chem. Commun. 56 (2020) 6348–6351, https://doi.org/
10.1039/d0cc01730b.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
[21] L. Yu, Y. Shao, D. Li, Direct combination of hydrogen evolution from water and
methane conversion in a photocatalytic system over Pt/TiO2, Appl. Catal. B
Environ. 204 (2017) 216–223, https://doi.org/10.1016/j.apcatb.2016.11.039.
[22] L. Yu, D. Li, Photocatalytic methane conversion coupled with hydrogen
evolution from water over Pd/TiO2, Catal. Sci. Technol. 7 (2017) 635–640,
https://doi.org/10.1039/c6cy02435a.
[23] O.A. Alduchov, R.E. Eskridge, Improved Magnus form approximation of
saturation vapor pressure, Appl Meteor. 35 (1996) 601–609, https://doi.org/
10.1175/1520-0450(1996)035<0601:IMFAOS>2.0.CO;2.
[24] H. Asakura, S. Yamazoe, T. Misumi, A. Fujita, T. Tsukuda, T. Tanaka, xTunes: A
new XAS processing tool for detailed and on-the-fly analysis, Radiat. Phys.
Chem. 175 (2020), https://doi.org/10.1016/j.radphyschem.2019.04.020.
[25] N.C. Burtch, H. Jasuja, K.S. Walton, Water stability and adsorption in metal-
organic frameworks, Chem. Rev. 114 (2014) 10575–10612, https://doi.org/
10.1021/cr5002589.
Acknowledgment
This work was supported by JST PRESTO [grant number
JPMJPR18T1] and JSPS KAKENHI [grant number JP20H02525]. The
synchrotron radiation experiment was performed under the
approval of the Japan Synchrotron Radiation Research Institute
(JASRI) as 2020A1410.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2021.03.024.
[26] K. Shirai, T. Sugimoto, K. Watanabe, M. Haruta, H. Kurata, Y. Matsumoto, Effect
of water adsorption on carrier trapping dynamics at the surface of anatase TiO2
199

M. Ishimaru, F. Amano, C. Akamoto et al.
Journal of Catalysis 397 (2021) 192–200
nanoparticles, Nano Lett. 16 (2016) 1323–1327, https://doi.org/10.1021/
acs.nanolett.5b04724.
[31] C.S. Turchi, D.F. Ollis, Photocatalytic degradation of organic water
contaminants: Mechanisms involving hydroxyl radical attack, J. Catal. 122
(1990) 178–192, https://doi.org/10.1016/0021-9517(90)90269-P.
[32] A. Bosnjakovic, S. Schlick, Spin Trapping by 5,5-Dimethylpyrroline-N-oxide in
Fenton Media in the Presence of Nafion Perfluorinated Membranes:
Limitations and Potential, J. Phys. Chem. B. 110 (2006) 10720–10728,
https://doi.org/10.1021/jp061042y.
[33] H. Song, X. Meng, S. Wang, W. Zhou, X. Wang, T. Kako, J. Ye, Direct and
Selective Photocatalytic Oxidation of CH4 to Oxygenates with O2 on
Cocatalysts/ZnO at Room Temperature in Water, J. Am. Chem. Soc. 141
(2019) 20507–20515, https://doi.org/10.1021/jacs.9b11440.
[27] K. Shirai, G. Fazio, T. Sugimoto, D. Selli, L. Ferraro, K. Watanabe, M. Haruta, B.
Ohtani, H. Kurata, C. Di Valentin, Y. Matsumoto, Water-Assisted Hole Trapping
at the Highly Curved Surface of Nano-TiO2 Photocatalyst, J. Am. Chem. Soc. 140
(2018) 1415–1422, https://doi.org/10.1021/jacs.7b11061.
[28] J. Rumble, CRC Handbook of Chemistry and Physics, 100th Edit, CRC Press, Boca
Raton, Florida, 2019.
[29] A. Fujishima, T.N. Rao, D.A. Tryk, Titanium dioxide photocatalysis, J.
Photochem. Photobiol. C. 1 (2000) 1–21, https://doi.org/10.1016/S1389-5567
(00)00002-2.
[30] Y. Nosaka, A.Y. Nosaka, Generation and Detection of Reactive Oxygen Species
in Photocatalysis, Chem. Rev. 117 (2017) 11302–11336, https://doi.org/
10.1021/acs.chemrev.7b00161.
200