Continuous methane to ethane conversion using gaseous oxygen on ceria-based Pd catalysts at low temperatures

Article abstract of DOI:10.1016/j.apcata.2021.118245

Methane upgrading into more valuable chemicals has been a promising challenge. It was previously reported that highly oxidized Pd/CeO₂ could produce ethane from methane. Here, we enhanced the ethane productivity by controlling oxygen mobility of the catalyst support and adding water adsorbents. Pd/Ce_1-xPd_xO_2-y catalyst, in which Pd atom was doped at the ceria support, showed enhanced oxygen activation and oxygen transfer, improving the ethane productivity from 0.04 mol_C2H6/mmol_SurfacePd/h in Pd/CeO₂ to 0.57 mol_C2H6/mmol_SurfacePd/h in Pd/Ce_1-xPd_xO_2-y. In order to remove the water produced from methane oxidation, zeolite 13X was physically mixed with Pd/Ce_1-xPd_xO_2-y. The water adsorbent could increase the ethane production significantly, especially at low temperatures. The ethane yield of 0.30 % could be obtained at 290 °C using gaseous O₂ as an oxidant. This work can provide useful ways to enhance the yield of value-added products in oxidative methane conversion.

Full text of DOI:10.1016/j.apcata.2021.118245

Applied Catalysis A, General 623 (2021) 118245
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Continuous methane to ethane conversion using gaseous oxygen on
ceria-based Pd catalysts at low temperatures
Gihun Kwon, Gunjoo Kim, Hyunjoo Lee*
Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, Daejeon 34141, South Korea
A R T I C L E I N F O
A B S T R A C T
Keywords:
Methane upgrading into more valuable chemicals has been a promising challenge. It was previously reported that
highly oxidized Pd/CeO2 could produce ethane from methane. Here, we enhanced the ethane productivity by
controlling oxygen mobility of the catalyst support and adding water adsorbents. Pd/Ce1-xPdxO2-y catalyst, in
which Pd atom was doped at the ceria support, showed enhanced oxygen activation and oxygen transfer,
improving the ethane productivity from 0.04 molC2H6/mmolSurfacePd/h in Pd/CeO2 to 0.57 molC2H6/mmolSurfa-
cePd/h in Pd/Ce1-xPdxO2-y. In order to remove the water produced from methane oxidation, zeolite 13X was
physically mixed with Pd/Ce1-xPdxO2-y. The water adsorbent could increase the ethane production significantly,
especially at low temperatures. The ethane yield of 0.30 % could be obtained at 290 ◦C using gaseous O2 as an
oxidant. This work can provide useful ways to enhance the yield of value-added products in oxidative methane
conversion.
Methane
Ethane
Oxygen transfer
Water adsorbent
Heterogeneous catalyst
1. Introduction
methane and O2, the ethane production was saturated at low O2 con-
centrations on the Pd/CeO2 catalyst, whereas the ethane production
The direct conversion of methane to value-added chemicals has high
industrial potential as an efficient route using natural gas or shale gas
[1–3]. This process has received much attention, but direct methane
conversion has been very difficult [4]. Non-oxidative route to convert
methane to C2/C3 chemicals and aromatics requires high reaction tem-
perature (> 1000 ◦C), necessitating huge energy use [5–7]. Oxidative
methane conversion to produce ethane and ethylene has been reported
using O2 as an oxidant, typically allowing lower reaction temperature (>
700 ◦C) [8,9]. However, various side-products were formed together due
to radicals produced at such high temperatures. The poor selectivity
causes additional cost for separation [10].
increased linearly up to high concentrations of methane. It indicates that
O2 activation is limiting the ethane production.
Many CeO2-based composites are known to be an excellent oxygen
carrier. CeO2-Fe2O3 showed greater and faster oxygen storage/release
property than CeO2 or Fe2O3 [12]. When CeO2 nanoparticles were
deposited on Cu(111), they provided a fast oxygen channel for CO
oxidation, resulting in higher activity than original Cu(111) [13].
Especially, ceria-based CexM1-xO2-y (M = Cu, Fe, Zr, Ru, Mn, Pd) solid
solution, in which other metal atoms are substituted in the ceria lattice,
has shown improved oxygen storage or transfer, comparing to original
CeO2. Ce0.9Ru0.1O1.94 presented higher oxygen storage capacity (OSC)
than CeO2 [14]. Ce0.5Zr0.5O2 showed the reducibility superior to CeO2 or
ZrO2 [15]. Ce0.9Fe0.1O2 or Ce1-xCuxO (x = 0.1~0.3) had oxygen
desorption peak at lower temperatures than CeO2 on O2-temperature
programmed desorption (TPD) analysis, indicating high oxygen
mobility [16,17]. Due to their facile oxygen transfer, the ceria-based
solid solutions have been used as a good support to maintain the
oxidation state of surface PdO. Pd on MnOx-CeO2 solid solution
remained more oxidized during CO oxidation than bare CeO2 support
[18]. PdO on kappa-CeZrO, which had better oxygen mobility than
tetragonal-CeZrO, showed more stable redox cycle on CH4 and O2 pulse
We previously showed that highly oxidized Pd/CeO2 catalysts can
catalyze oxidative methane conversion into ethane below 400 ◦C [11].
Density functional theory (DFT) calculation results showed that the
– –
– –
O Ce, enabled the oxidative methane
Pd
O
Pd sites, not Pd
– –
–
H
coupling. Although the Pd
O
Ce sites preferred subsequent C
bond cleavage, eventually leading to CO2 formation, the CeO2 support
was still required because it helped the Pd domain to maintain a highly
oxidized state in spite of reductive methane flow. However, the
maximum ethane yield obtained in the Pd/CeO2 catalysts was below 0.1
%. When ethane production was monitored for various concentrations of
* Corresponding author.
E-mail address: azhyun@kaist.ac.kr (H. Lee).
https://doi.org/10.1016/j.apcata.2021.118245
Received 12 March 2021; Received in revised form 1 June 2021; Accepted 5 June 2021
Available online 8 June 2021
0926-860X/© 2021 Elsevier B.V. All rights reserved.

G. Kwon et al.
Applied Catalysis A, General 623 (2021) 118245
[19]. Pd/Ce1-xPdxO2-y catalysts have shown the improved catalytic
performance in CO oxidation [20–22].
2.2. Catalytic reactions
Methane oxidation is a highly exothermic reaction, forming hot spots
easily within the catalyst bed, which often leads to catalyst degradation.
The formation of hot spots can be minimized through thermal dilution
by adding an inert diluent along with the catalyst [23–25]. The methane
oxidation also produces water, promoting or hindering further reaction.
Adding water increased methanol productivity in direct methane
oxidation by blocking over-oxidation of methane over CeOx-Cu2O
composites [26,27]. CO oxidation on PdO or Pd/Al3O3 catalysts was
promoted when water was introduced [28,29]. On the other hand,
adding water deactivated Pd/CeO2 catalysts for methane combustion
The catalytic performance of the catalysts was examined in a U-
shaped quartz glass fixed bed flow reactor at atmospheric pressure. The
catalyst 10 mg was placed inside the reactor. The inlet gas was intro-
duced with 6.8 sccm of O2 (99.995 %), 8.4 sccm of N2 (99.999 %), and
90 sccm of CH4 (99.999 %). All data about the catalytic activity were
obtained when the temperature was held for 1 h to establish steady-state
condition. The product gases (CO2 and C2H6) were analyzed with a gas
chromatograph (GC-6100 series, Younglin) with Molecular Sieve 5A and
Porapak N columns (Sigma-Aldrich) equipped with thermal conductiv-
ity detector (TCD) and flame ionization detector (FID). To maximize the
ethane yield, Pd/CePdO catalysts were physically mixed with a diluent
such as silica (Junsei, Quartz sand), molecular sieves 13X (Thermo
Scientific), or calcium oxide (Sigma, anhydrous). C2H6 selectivity, CH4
conversion, and C2H6 yield were estimated with the following equations
using molar flow rates:
–
[30,31]. Water retarded C H activation on PdO and oxygen transfer
between PdO and support, such as Al2O3 or SnO2, suppressing the
methane combustion [32,33]. Removing water during the methane
oxidation could increase the yield of the desired product.
In this study, Pd/Ce1-xPdxO2-y catalyst was prepared by solution
combustion method and used for direct methane oxidation to produce
ethane using gaseous O2 in a continuous flow reactor. The oxygen
transfer and ethane productivity of Pd/Ce1-xPdxO2-y were compared
with those of Pd/CeO2. The hot spots were formed within the catalyst
bed, easily degrading the catalyst, and they were minimized by adding
diluents. The water, which was formed during the methane oxidation,
was removed using zeolite 13X as water adsorbent to increase the ethane
yield as much as possible at low temperatures.
ꢀ 1
C2H6 selectivity (%) = 2C2H6(out) × (CO2(out) + 2C2H6(out)
)
× 100
ꢀ 1
CH4 conversion (%) = (CO2(out) + 2C2H6(out)) × CH
C2H6 yield (%) = (CH4 conversion × C2H6 selectivity) × 100ꢀ 1
× 100
4(feed)
2.3. Characterizations
2. Experimental section
The actual Pd amount in the catalysts was measured with an
inductively coupled plasma optical emission spectrometer (ICP-OES,
ThermoFisher Scientific). Specifically, 30 mg of the powder sample was
introduced in an open Teflon-coated vessel containing 2.5 mL nitric
acid, 2.5 mL hydrogen peroxide and 0.1 mL sulfuric acid. The solution
was heated to 180 ◦C. When the amount of solution decreased, it was
cooled to room temperature, 3 mL of nitric acid was added and heated at
180 ◦C again. The process was repeated to obtain a clear yellow or or-
ange solution in which all the powders were dissolved completely. The
crystalline structure was analyzed by powder X-ray diffractometer
(XRD, RIGAKU). High angle annular dark field-scanning TEM (HAADF-
STEM) images and energy-dispersive X-ray spectroscopy (EDS) mapping
images were measured using a Titan cubed G2 60 ꢀ 300 (FEI) with an
accelerating voltage of 300 kV. X-ray absorption spectroscopy (XAS) was
measured at 10 C Wide XAFS beamline of the Pohang Light Source (PLS).
The Pd K-edge spectra were observed in a fluorescence mode using a
passivated implanted planar silicon (PIPS) detector (Canberra). Each
sample was calibrated by measuring the reference Pd foil concurrently.
The XAS data were processed and fitted using the ARTEMIS and
ATHENA software. A coordination number was calculated by fixing the
amplitude reduction factor (S20) to the values obtained from the refer-
2.1. Catalyst syntheses
Pd/Ce1-xPdxO2-y (x = 0ꢀ 0.05) catalyst was synthesized by a solution
combustion method. [20,34] (NH4)2Ce(NO3)6 (Sigma-Aldrich, ≥ 99.99
%) 1 g was dissolved in 0.8 mL of deionized water. Pd(NO3)2 (Sig-
ma-Aldrich, ≤ 100 %) 9 mg and 340 mg of glycine (C2H5NO2,
Sigma-Aldrich, ≥ 99 %) were dissolved in 0.6 mL of deionized water,
then this Pd solution was mixed with the Ce solution. The mixture was
stirred to make a homogenous solution and moved to an alumina cru-
cible. The crucible was transferred into a furnace maintained at 350 ◦C.
The solution was boiled and burned with a flame to form solid product.
The solid was ground in a mortar and calcined at 650 ◦C for 16 h in air.
We named this samples as ‘Pd/CePdO’. The surface Pd was removed by
leaching the Pd/CePdO with nitric acid to prepare CePdO support.
Specifically, 0.1 g of Pd/CePdO powder was immersed in 5 mL nitric
acid (SAMCHUN, 60 %) in a Teflon-coated vessel and the lid was closed
to prevent evaporation. The vessel was heated at 250 ◦C for 1 h. When
the temperature decreased to room temperature, the vessel was opened
and the powder was washed with a copious amount of deionized water.
The remaining solid was dried at 80 ◦C for 12 h.
Pd/CeO2 was prepared using a conventional deposition-precipitation
method [11]. First, CeO2 support was synthesized; 1 g of Ce(NO3)3⋅6H2O
(Kanto chemical, 99.99 %) was added in 25 mL deionized water and the
aqueous ammonia (Duksan, 25~30 % NH4OH) was injected dropwise
into the solution until pH reached ~8.5. After stirring for 1 h, the so-
lution was filtered with deionized water. The collected precipitate was
dried at 80 ◦C for 10 h. The solid product was calcined at 500 ◦C for 5 h
in air. Then, CeO2 powder 380 mg was dispersed in 5 mL of deionized
water. H2PdCl4 solution was obtained by dissolving PdCl2 (Sigma-Al-
drich, 99 %) in HCl (SAMCHUN, 35~37 %) solution with 1: 2 molar
ratio. The H2PdCl4 solution containing an appropriate amount of Pd was
injected dropwise into the CeO2 solution under rapid stirring. The
Na2CO3 solution was introduced together to make the pH of the solution
~9. The final solution was stirred for 2 h and aged without stirring for 2
h at room temperature. The slurry was filtered with deionized water.
The obtained precipitate was dried at 80 ◦C for 6 h, and the final solid
product was calcined at 750 ◦C for 25 h in air. We named this samples as
‘Pd/CeO2’.
ence Pd foil. The Debye-Waller factor (
σ
2) was fixed to a reasonable
value to 3.0⋅10ꢀ 3 in order to limit independent parameters for fitting
[35,36]. The oxidation state of Pd was investigated by X-ray photo-
electron spectroscopy (XPS, K-Alpha, Thermo VG Scientific). The bind-
ing energies were calculated using the maximum intensity of the
advantageous C 1s signal at 284.8 eV as a reference.
Pd dispersion was measured with a pulsed CO adsorption, which was
modified from Takeguchi’s method [37]. First, the Pd/CePdO catalyst
30 mg was heated in the 5% O2/He gas at 300 ◦C for 10 min followed by
cooling to 50 ◦C with He gas for 5 min for purging. The sample was
heated in 4.9 % H2/Ar gas to 200 ◦C for 15 min. After cooling to 50 ◦C,
the catalyst was treated under following conditions; 1) He gas for 5 min,
2) 5% O2/He gas for 5 min, 3) CO2 gas for 10 min, 4) He gas for 20 min,
5) 4.9 % H2/Ar gas for 5 min. CO2 was introduced to form carbonates on
ceria surface to avoid overestimating the Pd dispersion. Finally, CO gas
was pulsed every 1 min in He stream repeatedly until the adsorption of
CO onto the catalyst was saturated.
In-situ diffuse reflectance infrared Fourier transform spectroscopy
2

G. Kwon et al.
Applied Catalysis A, General 623 (2021) 118245
Fig. 1. The C2H6 yield on (a) Pd/CePdO and (b) Pd/CeO2 with various Pd contents. All Pd contents was identified by ICP-OES analysis. (c) The comparison of C2H6
yields on the 1 wt% Pd/CePdO and the 4 wt% Pd/CeO2 and (d) their turnover frequency (TOFC2H6). The number of surface Pd atom was estimated from CO uptake.
The reaction was performed using 10 mg of catalysts and 105 sccm of total feed flow with 86 % CH4 and 6.5 % O2.
(DRIFTS, Thermo Scientific) was performed with a MCT detector and a
diffuse reflectance assembly chamber having a KBr window. The 5 mg
sample was mixed with 95 mg KBr powder and pretreated at 100 ◦C for 1
h under Ar gas flow to remove water and impurities. The mixture was
cooled to room temperature and a background spectrum was measured.
To observed CO adsorption, 1% CO/Ar gas flowed over the sample for 10
min to saturate CO atmosphere. The spectra were obtained during CO
desorption by Ar flow with evacuation for 10 min at room temperature.
Finally, adsorbed-CO spectra onto the sample was obtained.
ethane from direct methane oxidation [11]. However, O2 activation was
rather poor in the Pd/CeO2 catalyst, limiting the ethane productivity.
The Pd/CePdO catalyst was prepared to enhance the O2 activation by
using a solution-combustion method. When a high concentration of CH4
(86 %) was introduced with gaseous O2 (6.5 %) at elevated temperature,
the C2H6 yield was compared by adjusting the Pd contents of Pd/CePdO
to 1, 3.5, 8 wt%. The Pd contents in Pd/CeO2 catalysts was also varied to
1, 4, 8 wt%. The conventional deposition-precipitation method was used
to synthesize the Pd/CeO2. The C2H6 yields obtained were shown in
Fig. 1a and b. The 1 wt% Pd/CePdO and 4 wt% Pd/CeO2 showed the
highest C2H6 yield of 0.084 % and 0.054 % at 350 ◦C and 390 ◦C,
respectively, as shown in Fig. 1c. The C2H6 selectivity was also shown in
Fig. S1. When the Pd content of Pd/CePdO increased over 1 wt%, large
PdO domains were formed, and PdO peak was observed in the XRD
patterns as shown in Fig. S2. Turnover frequency for ethane production
(TOFC2H6) was estimated per surface Pd atom to compare the catalytic
activities on the 1 wt% Pd/CePdO and the 4 wt% Pd/CeO2 as shown in
Fig. 1d. The maximum TOFC2H6 could reach 0.57 molC2H6/mmolSurfa-
cePd/h at 350 ◦C in the 1 wt% Pd/CePdO, while the 4 wt% Pd/CeO2 had
the maximum TOFC2H6 value of 0.04 molC2H6/mmolSurfacePd/h at 390 ◦C.
These results show that low-temperature C2H6 production could be
much increased in Pd/CePdO catalyst. When the reaction temperature
was higher than 350 ◦C, the C2H6 production suddenly decreased. X-ray
photoelectron spectroscopy (XPS) results showed that the oxidic state of
the surface Pd maintained after the reaction at 350 ◦C as shown in
Fig. S3. But the Pd surface was reduced after the reaction at 430 ◦C,
indicating that C2H6 production did not occur at metallic Pd surface.
CO-DRIFT spectra were obtained to observe whether the CePdO
support, which was prepared by immersing the 1 wt% Pd/CePdO in
nitric acid (60 %) at 250 ◦C for 1 h, has surface Pd sites, as shown in
Fig. S4 [20]. The CO molecules adsorbed onto single Pd sites show a
O2-temperature programmed desorption (O2-TPD) spectra was per-
formed on a BETCAT-B (BEL, Japan) equipped with high sensitivity
TCD. A water trap was used to exclude the signals from water. Each
catalyst 0.1 g was pretreated for 30 min at 100 ◦C in 5% O2/He gas flow
and cooled down to 50 ◦C by purging with He gas flow. The catalyst was
exposed to He gas flow for 1 h and stabilized under the flow for 30 min.
Then, the temperature increased from 30 to 900 ◦C with a ramping rate
of 10 ◦C minꢀ 1 to obtain O2-TPD spectra. The temperature increased
from 30 to 900 ◦C with a ramping rate of 10 ◦C minꢀ 1. H2-temperature
programmed reduction (H2-TPR) spectra were obtained using BEL-CAT-
II (BEL Japan Inc.) equipped with a TCD. Each catalyst 50 mg was heated
for 1 h at 200 ◦C in Ar flow and cooled down to -90 ◦C using a cryogenic
apparatus using liquid nitrogen. Then, the catalysts were exposed to a
4.9 % H2/Ar gas flow and stabilized under the flow for 10 min. The
temperature increased from -90 to 600 ◦C with a ramping rate of 10 ◦C
minꢀ 1
.
3. Results and discussion
3.1. Catalytic activity of Pd/CePdO and Pd/CeO2 catalysts
In our previous work, we showed that Pd/CeO2 catalyst can produce
3

G. Kwon et al.
Applied Catalysis A, General 623 (2021) 118245
as shown in Fig. 2a. Red dots (Pd) were evenly dispersed and several
clustered Pd red dots indicated the presence of Pd nanoparticles on the
surface of the CePdO support. The structures of the 1 wt% Pd/CePdO
and the 4 wt% Pd/CeO2 were compared in Pd K-edge extended X-ray
absorption fine structure (EXAFS), as shown in Fig. 2b. Fig. S6 and
Table S1 showed the EXAFS fitting results. While the Pd/CeO2 has PdO
nanoparticles on CeO2 support, the Pd/CePdO has PdO nanoparticles on
CePdO support. Fig. S7 showed X-ray diffraction (XRD) patterns of the
Pd/CePdO, Pd/CeO2, CePdO support and CeO2 support. The XRD peaks
for only CeO2 phase were observed. The XRD peaks for PdO or Pd are not
observed, indicating fine dispersion of Pd species. The shift in the CeO2
XRD peaks upon Pd substitution was not clear in the Pd/CePdO or
CePdO, probably due to low Pd content of 0.5 %.
Table 1
Comparison in Pd dispersion, size, and BET surface area in the 1 wt% Pd/CePdO
and the 4 wt% Pd/CeO2.
1 wt% Pd/CePdO
4 wt% Pd/CeO2
Pd dispersion (%)
Pd size (nm)
29.6a
38.1
2.9
3.8
BET surface area (m2 gꢀ 1
)
7.8
32.4
a
When Pd content was measured after removing the surface Pd, which was
performed by immersing the 1 wt% Pd/CePdO in nitric acid (60 %) at 250 ◦C for
1 h, the residual Pd content was 0.5 wt%. The Pd dispersion was estimated based
on the surface Pd content of 0.5 wt%. The amount of Pd atoms exposed at the
surface was measured by pulsed CO chemisorption.
Fig. 2c shows the O2-TPD results of 1 wt% Pd/CePdO and 4 wt% Pd/
CeO2, indicating the oxygen transfer. The Pd/CePdO showed the oxygen
desorption peak at 270 ◦C while the Pd/CeO2 had the peak at much
higher temperature of 620 ◦C, indicating that the surface oxygen species
can be desorbed much more easily on the Pd/CePdO. The lattice oxygen
started to be desorbed at 450◦C in the Pd/CePdO catalyst. The change in
oxygen vacancy sites was investigated using Ce 3d and O 1s XPS on the
Pd/CePdO and Pd/CeO2, as shown in Fig. S8. The Ce 3d XPS peaks were
deconvoluted and the percentages of Ce3+ and Ce4+ were estimated
from the peak area ratio. The ratio of Ce3+ was higher in the Pd/CePdO
(15.1 %) and CePdO (16.5 %) than Pd/CeO2 (9.7 %) and CeO2 (10.4 %).
The higher Ce3+ ratio indicates more oxygen vacancy sites, resulting in
higher oxygen activation [20,41,42]. Similarly, O 1s XPS peaks were
also deconvoluted to lattice oxygen (Olatt) and surface oxygen adsorbed
on the defect sites (Oads). The Pd/CePdO had more Oads than Pd/CeO2,
with 27.8 % versus 19.8 %. The Pd/CePdO possesses more surface ox-
ygen species. The H2-TPR results in Fig. 2d show that PdO nanoparticles
peak at 2000–2200 cmꢀ 1, while the CO adsorbed onto ensemble Pd sites
show a peak at 1800–2000 cmꢀ 1 [38]. The Pd/CePdO clearly showed
the peaks for adsorbed CO, but the CePdO support showed no CO peak,
indicating that Pd was not exposed at the CePdO support surface. When
the Pd content was measured by ICP-OES, it was 0.5 wt% in the CePdO
support. Table 1 showed the Pd dispersion, size and BET surface area of
the 1 wt% Pd/CePdO and 4 wt% Pd/CeO2. The Pd dispersion and size
were estimated based on surface Pd contents (0.5 wt% for Pd/CePdO
and 4 wt% for Pd/CeO2). The Pd size was similar as 3.8 nm for the
Pd/CePdO and 2.9 nm for the Pd/CeO2. Their TEM images were shown
in Fig. S5. The Pd/CePdO has much larger ceria domain than the
Pd/CeO2. The Pd/CePdO catalyst has the surface area of 7.8 m2 gꢀ 1
,
much lower than the Pd/CeO2 with 32.4 m2 gꢀ 1. The solid solution
synthesized by the solution combustion method usually had low surface
area of <10 m2 gꢀ 1 [39,40].
EDS mapping image was obtained for the 1 wt% Pd/CePdO catalyst,
Fig. 2. (a) EDS mapping image of 1 wt% Pd/CePdO (Pd: red, Ce: green). White circle indicates Pd nanoparticles on the surface. (b) Pd K edge EXAFS spectra of the 1
wt% Pd/CePdO and 4 wt% Pd/CeO2, Pd foil, and PdO. The lines indicate experimental data, and dots indicate fitting results. (c) O2-TPD results and (d) H2-TPR
results. The O2-TPD and H2-TPR peaks were normalized by the catalyst weight.
4

G. Kwon et al.
Applied Catalysis A, General 623 (2021) 118245
Fig. 3. Effect of the catalyst amount on the (a) C2H6 yield and temperature difference (ΔT) when 5, 10, 25, 50 and 100 mg Pd/CePdO catalysts were loaded in the
reactor, respectively. ΔT is the value of (reactor temperature with catalyst - reactor temperature without catalyst). The red dashed box indicates the hot spot for-
mation. (b) Effect of the catalyst amount on the TOFC2H6 plotted with actual catalyst temperature. Effect of silica diluent on the (c) C2H6 yield and (d) TOFC2H6. Silica
15 g was mixed with 5, 10, 25, 50 and 100 mg of Pd/CePdO catalysts, respectively. The reaction was performed with 105 sccm of total feed flow with 86 % CH4 and
6.5 % O2.
were reduced to Pd metal at 60 ◦C in the Pd/CePdO. A shoulder at higher
temperatures (~110 ◦C) was ascribed to the reduction of Pd2+ in the
lattice of CeO2 on Pd/CePdO catalyst [43,44]. PdO nanoparticles
deposited on CeO2 showed the reduction peak at lower temperature of
30 ◦C. These results inform that the oxidic Pd was preserved better on
the Pd/CePdO under reductive condition. In our previous work, the
24]. When the hot spots are formed, the catalyst could be easily
degraded, resulting in low C2H6 production rate. The temperature dif-
ference between the reactor temperature with and without the catalyst
was measured for various amounts of the Pd/CePdO catalysts, as shown
in the lower figure in Fig. 3a. Fig. 3b shows TOFC2H6 estimated for the
various catalyst amounts, plotted versus the actual catalyst temperature.
TOFC2H6 was nearly the same irrelevant to the catalyst amount, but the
catalyst was degraded at lower temperatures as the catalyst amount
increased.
– –
O Pd sites are the active sites to produce C2H6 at low
oxidic Pd
– –
O Pd
temperature [11]. The Pd/CePdO could maintain the oxidic Pd
better than the Pd/CeO2, resulting in better C2H6 productivity.
The effect of CH4 and O2 content was examined on the Pd/CePdO
and Pd/CeO2 as shown in Fig. S9. The C2H6 productivity increased lin-
early as the CH4 content increased on both Pd/CePdO and Pd/CeO2
catalyst, indicating that CH4 was easily activated on the catalysts. The
C2H6 productivity became saturated as the O2 content increased on the
Pd/CeO2, whereas the C2H6 productivity kept increasing on the Pd/
CePdO. Clearly, better oxygen activation and better oxygen transfer on
the Pd/CePdO catalyst could increase the C2H6 production.
Silica sand was used as a diluent to buffer the thermal gradient on
oxidative methane conversion, as shown in Fig. 3c and d. The highest
C2H6 yield increased from 0.02 % to 0.18 % for 100 mg catalyst, from
0.09 % to 0.16 % for 50 mg catalyst, and from 0.09 % to 0.12 % for 25
mg catalyst when each Pd/CePdO catalyst was physically mixed with 15
g of silica diluent. The 5 and 10 mg Pd/CePdO catalysts showed the
similar C2H6 yield when diluted with silica sand, indicating that the 5
mg or 10 mg of the catalyst would not form the hot spots much. Fig. S10
showed that the C2H6 selectivity was hardly affected by the silica
diluent. Fig. S11 also showed that the C2H6 selectivity had little de-
pendency on the space velocity. TOFC2H6 kept increasing at higher
temperature in Fig. 3d, although the larger catalyst amounts still showed
some deviation from the case of 5 mg or 10 mg. The deviation implies
that there might be another factor suppressing the C2H6 production. It
3.2. Effect of thermal dilution
It was attempted to increase the C2H6 production further by con-
trolling the reaction condition. The C2H6 yield and TOFC2H6 were
observed by controlling the catalyst weight from 5 mg to 100 mg, as
shown in Fig. 3. When the amount of Pd/CePdO catalysts increased, the
C2H6 production stopped at earlier temperature. When 100 mg of the
Pd/CePdO catalyst was used, the catalyst stopped working at 290 ◦C.
This sudden stop in the ethane production was due to the catalyst
reduction [11]. Because methane oxidation is highly exothermic, a large
amount of reaction heat is generated from the catalytic reaction,
creating a hot spot with a large thermal gradient in the catalyst bed [23,
–
has been reported that water can inhibit C H activation on PdO and
oxygen transfer between PdO and the support of Al2O3 or SnO2 [32,33].
The water effect is investigated in the next section.
3.3. Effect of water adsorbents
Fig. 4a shows that the C2H6 yield decreased when water introduced
5

G. Kwon et al.
Applied Catalysis A, General 623 (2021) 118245
Fig. 4. (a) Poisoning effect of water on the C2H6 yield. The reaction was performed using 100 mg of 1 wt% Pd/CePdO catalyst and 15 g of silica sand. 105 sccm of
total feed flow was introduced with 78 % CH4, 13 % O2 and 0 - 3.6 % H2O. (b) TGA result of zeolite 13X in N2 flow. (c) The effect of adsorbed water on the C2H6 yield.
The 1 wt% Pd/CePdO 100 mg was physically mixed with 10 g of zeolite 13X, then pre-treated at 200, 300, and 400 ◦C for 2 h in air flow. 52.5 sccm of total feed flow
was introduced with 73 % CH4 and 18 % O2. (d) The C2H6 yield in various partial pressures of CH4 and O2 (balance N2). The Pd/CePdO 100 mg was mixed with 10 g
of zeolite 13X, and pre-treated at 400 ◦C.
Fig. 5. In-situ mass spectra of CO2, C2H6, H2O obtained at direct methane oxidation when 100 mg of 1 wt% Pd/CePdO catalyst was mixed with (a) 15 g of silica sand
or (b) 10 g of zeolite 13X. The mass signal was stabilized for 1 h at room temperature and the reaction temperature increased to 310 ◦C and each signal was measured
for 4 h at 310 ◦C. The total feed flow with 52.5 sccm was introduced with 73 % CH4 and 18 % O2.
6

G. Kwon et al.
Applied Catalysis A, General 623 (2021) 118245
Table 2
Comparison with literature values for the continuous gas-phase methane oxidation producing valuable chemicals below 400 ◦C.
Space velocity (mL g hꢀ 1
)
Product yield (%)
Production rate (mmol hꢀ 1
)
ꢀ 1
No.
Catalyst
Pt/Y2O3
Fe-ZSM-5
Reaction condition
350 ◦C (1 atm)
300 ◦C (1 atm)
350 ◦C (1 atm)
300 ◦C (50 bar)
270 ◦C (1 atm)
225˚C (1 atm)
290 ◦C (1 atm)
370 ◦C (1 atm)
Catalyst weight (g)
cat
Ref [46]
Ref [47]
0.3
20,000
7500
0.06 (DME)
0.03 (DME)
0.03 (MeOH)
0.01 (DME)
0.009 (MeOH)
0.003 (DME)
0.01 (ethane)
0.003 (ethane)
0.005 (MeOH)
0.002 (MeOH)
0.003 (ethane)
0.15 (ethane)
0.44
Ref [48]
Ref [49]
Ref [50]
Ref [51]
NiO/ Ce0.83Zr0.17O2
(SiO)2Ta-H
0.2
30,000
600
0.1 (ethane)
0.03 (ethane)
< 0.01 (MeOH)
< 0.01 (MeOH)
0.3 (ethane)
0.32 (ethane)
0.3
Cu-CHA
0.25
0.4
6000
6000
Cu-SSZ-13
This work
Pd/CePdO + Zeolite 13X
0.1
31,500
removal of water using zeolite adsorbents and the C2H6 yield, it was
attempted to lower the reaction temperature as much as possible by
diluting the CH4 concentration. Fig. 4d shows that C2H6 yield at 290 ◦C
increased from 0.11 % for the reactant flow of CH4/O2 73 %/18 %
(balance N2) to 0.30 % for CH4/O2 1.5 %/0.38 % (balance N2).
In-situ mass spectra (MS) were measured to observe the signal of
products (C2H6, CO2, H2O) over the physical mixture of 1 wt% Pd/
CePdO catalyst and silica sand or zeolite 13X at 310 ◦C, as shown in
Fig. 5. When the Pd/CePdO catalyst was physically mixed with silica
sand, all the C2H6, CO2 and H2O signals were simultaneously measured
during the direct methane oxidation. When mixed with zeolite 13X,
however, the H2O production signal was delayed while the signals of
C2H6 and CO2 increased significantly at initial stage (~0.3 h). It clearly
shows that the H2O produced during the reaction was adsorbed on
zeolite 13X, leading to H2O-free condition with higher C2H6 production.
The C2H6 yield obtained in this work was compared with the liter-
ature values, as shown in Table 2. The production of valuable chemicals,
such as C2H6, dimethyl ether (DME) and methanol (MeOH), from direct
methane oxidation has been reported using continuous gas-phase
reactor below 400 ◦C. The product yields were lower than 0.1 % in
the most literatures. We could further control the apparent C2H6 selec-
tivity using CaO as an absorbent, as shown in Fig. S13. CaO has been
reported to absorb both H2O and CO2 [45]. The C2H6 yield on the
mixture of 1 wt% Pd/CePdO and CaO was lower than the mixture with
zeolite 13X, but the C2H6 could be obtained with 100 % selectivity
because the side product CO2 was absorbed onto the CaO.
The C2H6 production on the physical mixture of 1 wt% Pd/CePdO
and zeolite 13X was tested for long-term reaction, as shown in Fig. 6.
Both high and low methane concentrations (73 % CH4 and 1.5 % CH4)
were tested at 370 ◦C and 290 ◦C, respectively. The C2H6 yield decreased
from 0.32 % to 0.24 % after 50 h of reaction at 370 ◦C. This is mainly due
to the poisoning from water produced during the reaction. The catalyst
bed was regenerated by treating at 500 ◦C for 2 h under air flow, then the
initial C2H6 yield was recovered. Fig. S14 shows the mass spectra ob-
tained during the regeneration step. The desorbed water was clearly
observed. The water poisoning effect was more dominant for the long-
term reaction at low methane concentration. The C2H6 yield decreased
from 0.30 % to 0.14 % after 50 h of reaction at 290 ◦C. The initial yield
was also recovered after the regeneration step. When C2H6 production
rate was compared, the high methane concentration flow could produce
ethane stably with ~0.1 mmolC2H6/h at 370 ◦C, while the production
rate was much lower as ~0.001 mmolC2H6/h at 290 ◦C for the low
methane concentration flow.
Fig. 6. Long-term reaction results; C2H6 yield (left y axis) and C2H6 production
rate (right y axis) were shown. The 1 wt% Pd/CePdO 100 mg was mixed with
10 g of zeolite 13X. The catalyst bed was pre-treated at 400 ◦C for 2 h under air
flow prior to the reaction. Total flow rate was 52.5 sccm with a (a) high reactant
content of 73 % CH4 and 18 % O2 at 370 ◦C (balance N2) or (b) low reactant
content of 1.5 % CH4 and 0.38 % O2 at 290 ◦C (balance N2). Regeneration was
performed at 500 ◦C for 2 h under air flow.
into reactant feed flow. Surely, water inhibits the C2H6 production,
resulting in low C2H6 yield below 400 ◦C. In order to remove water,
which is formed during methane oxidation, zeolite 13X was used as a
water adsorbent. Instead of silica sand, zeolite 13X was mixed with the
Pd/CePdO catalyst. Fig. 4b shows TGA result of zeolite 13X, showing
that most of the adsorbed water would be removed at 400 ◦C. The zeolite
13X was mixed with the Pd/CePdO catalyst, then treated at 200, 300, or
400 ◦C. No C2H6 was produced on zeolite 13X only. The C2H6 yield
surely increased upon mixing with zeolite 13X, especially at low tem-
peratures. The C2H6 selectivity also increased as shown in Fig. S12.
Removing water using zeolite adsorbents was not very efficient at high
temperatures, though, because the water sorption capacity was much
smaller. The maximum C2H6 yield obtained increased slightly from 0.27
% to 0.32 % at 370 ◦C upon mixing with zeolite 13X. To maximize the
4. Conclusions
The Pd/Ce1-xPdxO2-y catalyst was compared with the Pd/CeO2
catalyst for direct methane oxidation using O2 in a continuous gas-phase
reactor. The Pd/Ce1-xPdxO2-y could produce C2H6 more than the Pd/
CeO2 with much higher turnover frequency. The O2-TPD and H2-TPR
results confirmed that the Pd/Ce1-xPdxO2-y catalyst has enhanced oxy-
gen transfer, preserving surface oxidic Pd better than the Pd/CeO2
catalyst. The hot spots, which were formed during highly exothermic
7

G. Kwon et al.
Applied Catalysis A, General 623 (2021) 118245
[14] P. Singh, M.S. Hegde, Chem. Mater. 21 (2009) 3337–3345.
methane oxidation, degraded the catalysts easily, thus the diluent
should be mixed with the Pd/Ce1-xPdxO2-y catalyst to increase the C2H6
yield. Water produced during the methane oxidation also behaved as
poisoning species against C2H6 production. Zeolite 13X was added to
remove the water and also as a diluent. Mass spectra clearly showed that
C2H6 production was promoted by removing the poisoning effect of
water. The C2H6 could be produced stably with a rate of ~0.1
mmolC2H6/h at 370 ◦C and ~0.001 mmolC2H6/h at 290 ◦C. The use of a
different diluent, such as CaO, enabled the production of C2H6 only by
absorbing CO2 selectively. This work demonstrates an interesting
example to increase the ethane production in direct methane oxidation
at low temperatures.
[15] L.V. Mattos, F.B. Noronha, J. Power Sources 145 (2005) 10–15.
[16] C. He, Y.K. Yu, C.W. Chen, L. Yue, N.L. Qiao, Q. Shen, J.S. Chen, Z.P. Hao, RSC
Adv. 3 (2013) 19639–19656.
[17] Y. Zuo, X.S. Huang, L.P. Li, G.S. Li, J. Mater. Chem. A 1 (2013) 374–380.
[18] C. Wang, C. Wen, J. Lauterbach, E. Sasmaz, Appl. Catal. B 206 (2017) 1–8.
[19] Y.Q. Ding, Q.Q. Wu, B. Lin, Y.L. Guo, Y. Guo, Y.S. Wang, L. Wang, W.C. Zhan, Appl.
Catal. B 266 (2020), 118631.
[20] L. Meng, A.P. Jia, J.Q. Lu, L.F. Luo, W.X. Huang, M.F. Luo, J. Phys. Chem. C 115
(2011) 19789–19796.
[21] K.R. Priolkar, P. Bera, P.R. Sarode, M.S. Hegde, S. Emura, R. Kumashiro, N.P. Lalla,
Chem. Mater. 14 (2002) 2120–2128.
[22] S.Y. Wang, N. Li, R.M. Zhou, L.Y. Jin, G.S. Hu, J.Q. Lu, M.F. Luo, J. Mol. Catal. A
Chem. 374 (2013) 53–58.
[23] B. Kimmerle, A. Baiker, J.D. Grunwaldt, Phys. Chem. Chem. Phys. 12 (2010)
2288–2291.
[24] E.D. Goodman, A.A. Latimer, A.C. Yang, L.H. Wu, N. Tahsini, F. Abild-Pedersen,
M. Cargnello, ACS Appl. Nano Mater. 1 (2018) 5258–5267.
[25] M. Taniewski, A. Lachowicz, K. Skutil, D. Czechowicz, Chem. Eng. Sci. 51 (1996)
4271–4278.
CRediT authorship contribution statement
G. Kwon and H. Lee conceived the project and wrote the paper. G.
Kim helped the characterizations. All the authors commented on the
manuscript.
[26] Z.J. Zuo, P.J. Ramirez, S.D. Senanayake, P. Liu, J.A. Rodriguez, J. Am. Chem. Soc.
138 (2016) 13810–13813.
[27] Z.Y. Liu, E.W. Huang, I. Orozco, W.J. Liao, R.M. Palomino, N. Rui, T. Duchon,
S. Nemsak, D.C. Grinter, M. Mahapatra, P. Liu, J.A. Rodriguez, S.D. Senanayake,
Science 368 (2020) 513–517.
[28] R. Caporali, S. Chansai, R. Burch, J.J. Delgado, A. Goguet, C. Hardacre,
L. Mantarosie, D. Thompsett, Appl. Catal. B 147 (2014) 764–769.
[29] S.H. Oh, G.B. Hoflund, J. Catal. 245 (2007) 35–44.
[30] M. Danielis, S. Colussi, C. de Leitenburg, L. Soler, J. Llorca, A. Trovarelli, Angew.
Chem. Int. Ed. 57 (2018) 10212–10216.
Declaration of Competing Interest
The authors report no declarations of interest.
Acknowledgements
[31] M. Monai, T. Montini, C. Chen, E. Fonda, R.J. Gorte, P. Fornasiero, ChemCatChem
7 (2015) 2038–2046.
[32] R. Gholami, M. Alyani, K.J. Smith, Catalysts 5 (2015) 561–594.
[33] W. Barrett, J. Shen, Y.F. Hu, R.E. Hayes, R.W.J. Scott, N. Semagina, ChemCatChem
12 (2020) 944–952.
This research was supported by the National Research Foundation of
Korea (NRF-2015M3D3A1A01064899). The experiments at PLS were
supported in part by MSIP and POSTECH.
[34] M.S. Hegde, G. Madras, K.C. Patil, Acc. Chem. Res. 42 (2009) 704–712.
[35] E.K. Dann, E.K. Gibson, R.A. Catlow, P. Collier, T.E. Erden, D. Gianolio,
C. Hardacre, A. Kroner, A. Raj, A. Goguet, P.P. Wells, Chem. Mater. 29 (2017)
7515–7523.
Appendix A. Supplementary data
[36] D.Y. Chung, Y.H. Chung, N. Jung, K.H. Choi, Y.E. Sung, Phys. Chem. Chem. Phys.
15 (2013) 13658–13663.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2021.118245.
[37] T. Takeguchi, S. Manabe, R. Kikuchi, K. Eguchi, T. Kanazawa, S. Matsumoto,
W. Ueda, Appl. Catal. A 293 (2005) 91–96.
[38] C.D. Zeinalipour-Yazdi, D.J. Willock, L. Thomas, K. Wilson, A.F. Lee, Surf. Sci. 646
(2016) 210–220.
References
[39] S. Colussi, A. Gayen, M.F. Camellone, M. Boaro, J. Llorca, S. Fabris, A. Trovarelli,
Angew. Chem. Int. Ed. 48 (2009) 8481–8484.
[40] S. Colussi, A. Gayen, M. Boaro, J. Llorca, A. Trovarelli, ChemCatChem 7 (2015)
2222–2229.
[1] P. Schwach, X.L. Pan, X.H. Bao, Chem. Rev. 117 (2017) 8497–8520.
[2] E.V. Kondratenko, T. Peppel, D. Seeburg, V.A. Kondratenko, N. Kalevaru,
A. Martin, S. Wohlrab, Catal. Sci. Technol. 7 (2017) 366–381.
[3] C. Hammond, S. Conrad, I. Hermans, ChemSusChem 5 (2012) 1668–1686.
[4] P. Tomkins, M. Ranocchiari, J.A. van Bokhoven, Acc. Chem. Res. 50 (2017)
418–425.
[41] C.J. Ma, Y.Y. Wen, Q.Q. Yue, A.Q. Li, J.L. Fu, N.W. Zhang, H.J. Gai, J.B. Zheng, B.
H. Chen, RSC Adv. 7 (2017) 27079–27088.
[42] D.Y. Li, K.Z. Li, R.D. Xu, H. Wang, D. Tian, Y.G. Wei, X. Zhu, C.H. Zeng, L.P. Zeng,
Catal. Today 318 (2018) 73–85.
[43] J.H. Jin, C. Li, C.W. Tsang, B. Xu, C.H. Liang, RSC Adv. 5 (2015) 102147–102156.
[44] Y. Xia, J.R. Ye, D.G. Cheng, F.Q. Chen, X.L. Zhan, Catal. Sci. Technol. 8 (2018)
5137–5147.
[5] P.F. Xie, T.C. Pu, A.M. Nie, S. Hwang, S.C. Purdy, W.J. Yu, D. Su, J.T. Miller,
C. Wang, ACS Catal. 8 (2018) 4044–4048.
[6] X.G. Guo, G.Z. Fang, G. Li, H. Ma, H.J. Fan, L. Yu, C. Ma, X. Wu, D.H. Deng, M.
M. Wei, D.L. Tan, R. Si, S. Zhang, J.Q. Li, L.T. Sun, Z.C. Tang, X.L. Pan, X.H. Bao,
Science 344 (2014) 616–619.
[45] W. Huang, X. Zhang, A.C. Yang, E.D. Goodman, K.C. Kao, M. Cargnello, ACS Catal.
10 (2020) 8157–8167.
[46] V. Vargheese, Y. Kobayashi, S.T. Oyama, Angew. Chem. Int. Ed. 59 (2020)
16644–16650.
[7] S.J. Han, S.W. Lee, H.W. Kim, S.K. Kim, Y.T. Kim, ACS Catal. 9 (2019) 7984–7997.
[8] T.N. Nguyen, T.T.P. Nhat, K. Takimoto, A. Thakur, S. Nishimura, J. Ohyama,
I. Miyazato, L. Takahashi, J. Fujima, K. Takahashi, T. Taniike, ACS Catal. 10 (2020)
921–932.
[47] Y.K. Chow, N.F. Dummer, J.H. Carter, C. Williams, G. Shaw, D.J. Willock, S.
H. Taylor, S. Yacob, R.J. Meyer, M.M. Bhasin, G.J. Hutchings, Catal. Sci. Technol. 8
(2018) 154–163.
[9] D. Kiani, S. Sourav, J. Baltrusaitis, I.E. Wachs, ACS Catal. 9 (2019) 5912–5928.
[10] K.J. Chen, D.G. Madden, S. Mukherjee, T. Pham, K.A. Forrest, A. Kumar, B. Space,
J. Kong, Q.Y. Zhang, M.J. Zaworotko, Science 366 (2019) 241–246.
[11] G. Kwon, D. Shin, H. Jeong, S.K. Sahoo, J. Lee, G. Kim, J. Choi, D.H. Kim, J.W. Han,
H. Lee, ChemSusChem 13 (2020) 677–681.
[48] C. Okolie, Y.M. Lyu, L. Kovarik, E. Stavitski, C. Sievers, ChemCatChem 10 (2018)
2700–2708.
[49] D. Soulivong, S. Norsic, M. Taoufik, C. Coperet, J. Thivolle-Cazat, S. Chakka, J.
M. Basset, J. Am. Chem. Soc. 130 (2008) 5044–5045.
[50] K.T. Dinh, M.M. Sullivan, K. Narsimhan, P. Serna, R.J. Meyer, M. Dinca, Y. Roman-
Leshkov, J. Am. Chem. Soc. 141 (2019) 11641–11650.
[12] M. Machida, T. Kawada, H. Fujii, S. Hinokuma, J. Phys. Chem. C 119 (2015)
24932–24941.
[51] A. Koishybay, D.F. Shantz, J. Am. Chem. Soc. 142 (2020) 11962–11966.
[13] F. Yang, J. Graciani, J. Evans, P. Liu, J. Hrbek, J.F. Sanz, J.A. Rodriguez, J. Am.
Chem. Soc. 133 (2011) 3444–3451.
8