Developing reactive catalysts for low temperature oxidative coupling of methane: On the factors deciding the reaction performance of Ln2Ce2O7 with different rare earth A sites

Article abstract of DOI:10.1016/j.apcata.2018.01.004

To develop more efficient catalysts that can be operated at low temperature region for oxidative of methane (OCM), a series of Ln₂Ce₂O₇ compounds with different A sites (Ln = La, Pr, Sm and Y) have been prepared. It is revealed by XRD and Raman techniques that a defective cubic fluorite phase has been formed in all the catalysts. As a consequence, all the Ln₂Ce₂O₇ catalysts possess much stronger surface basicity and more abundant electrophilic oxygen species in comparison with individual CeO₂, which is beneficial to OCM reaction. It is believed that the concerted interaction between surface intermediate basic sites and selective electrophilic oxygen species is the predominant reason controlling the reaction performance of the catalysts. Furthermore, Ln₂Ce₂O₇ catalysts own more abundant mesopores and higher surface areas than pure CeO₂, which could also be favorable for the contacting of the reactants with the active sites. Due to the optimal synergistic interaction of these factors, La₂Ce₂O₇ exhibits the best performance among all the catalysts, on which the highest C₂ yield of 16.6% is achieved at 750 °C. In comparison with Mn/Na₂WO₄/SiO₂, the most promising catalyst at present, La₂Ce₂O₇ display much improved reaction performance at low temperature region (<750 °C).

Full text of DOI:10.1016/j.apcata.2018.01.004

Accepted Manuscript
Title: Developing Reactive Catalysts for Low Temperature
Oxidative Coupling of Methane: On the Factors Deciding the
Reaction Performance of Ln2Ce2O7 with Different Rare Earth
A sites
Authors: Junwei Xu, Liang Peng, Xiuzhong Fang, Ziyang Fu,
Wenming Liu, Xianglan Xu, Honggen Peng, Renyang Zheng,
Xiang Wang
PII:
S0926-860X(18)30010-3
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2018.01.004
APCATA 16511
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
3-11-2017
29-12-2017
3-1-2018
Please cite this article as: Xu J, Peng L, Fang X, Fu Z, Liu W, Xu X, Peng H,
Zheng R, Wang X, Developing Reactive Catalysts for Low Temperature Oxidative
Coupling of Methane: On the Factors Deciding the Reaction Performance of
Ln2Ce2O7 with Different Rare Earth A sites, Applied Catalysis A, General (2010),
https://doi.org/10.1016/j.apcata.2018.01.004
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Developing Reactive Catalysts for Low Temperature Oxidative
Coupling of Methane: On the Factors Deciding the Reaction
Performance of Ln2Ce2O7 with Different Rare Earth A sites
Junwei Xu1, Liang Peng1, Xiuzhong Fang1, Ziyang Fu1, Wenming Liu1, Xianglan Xu1,
Honggen Peng1, Renyang Zheng2, Xiang Wang1, *
1 Institute of Applied Chemistry, College of Chemistry, Nanchang University,
Nanchang, 330031, PR China
2 Research Institute of Petroleum Processing(RIPP), SINOPEC,18 Xueyuan Road,
Haidian district, Beijing 100083, China
 Corresponding author: E-mail: xwang23@ncu.edu.cn (X. Wang)
GRAPHICAL ABSTRACT
1

Highlights
 Ln2Ce2O7 with different Ln elements form mainly a defective cubic fluorite phase
 Ln change alters the surface oxygen property and alkalinity of the catalysts
 Surface alkaline and electrophilic oxygen sites control the reaction performance
 La2Ce2O7 shows the best performance due to its richest alkaline and oxygen sites
 La2Ce2O7 displays better performance than Mn/Na2WO4/SiO2 catalyst below 750 oC
Abstract: To develop more efficient catalysts that can be operated at low temperature
region for oxidative of methane (OCM), a series of Ln2Ce2O7 compounds with
different A sites (Ln=La, Pr, Sm and Y) have been prepared. It is revealed by XRD
and Raman techniques that a defective cubic fluorite phase has been formed in all the
catalysts. As a consequence, all the Ln2Ce2O7 catalysts possess much stronger surface
basicity and more abundant electrophilic oxygen species in comparison with
individual CeO2, which is beneficial to OCM reaction. It is believed that the concerted
interaction between surface intermediate basic sites and selective electrophilic oxygen
species is the predominant reason controlling the reaction performance of the
catalysts. Furthermore, Ln2Ce2O7 catalysts own more abundant mesopores and higher
surface areas than pure CeO2, which could also be favorable for the contacting of the
reactants with the active sites. Due to the optimal synergistic interaction of these
factors, La2Ce2O7 exhibits the best performance among all the catalysts, on which the
highest C2 yield of 16.6% is achieved at 750 oC. In comparison with
Mn/Na2WO4/SiO2, the most promising catalyst at present, La2Ce2O7 display much
improved reaction performance at low temperature region (<750 oC).
Keywords: oxidative coupling of methane; A2B2O7 catalyst; “A” site replacement
effect; defective cubic fluorite; surface electrophilic oxygen; surface alkaline sites.
1. Introduction
2

Oxidative coupling of methane (OCM) into ethylene is one of the important routes
for the value-added transformation of natural gas. Though three decades have passed
since the early work of Keller and Bhasin on the OCM [1], there is still some way to
go for the successful industrialization of this process. It is commonly accepted that
improving C2 selectivity at relatively high methane conversion to achieve one-way C2
product yield above 30% is still the core issue to be solved. The mechanism of OCM
over different catalysts has been investigated by many scientists over past years [2-5].
It is found that suitable alkaline sites and reactive oxygen species on the surface of a
catalyst can activate the potent C-H bonds in methane molecules to form gas phase
CH3· radicals [2,6,7]. During this process, the abstracted hydrogen atom will react
with the surface active oxygen sites to form hydroxyl groups, which can react further
with each other to form water and generate oxygen vacancies at the same time [7,8].
On the other hand, two of the formed methyl radicals in gas phase will unite with each
other to produce C2H6, which will be further dehydrogenated into C2H4 [9]. At the
same time, some of the methyl radicals adsorbed on the catalyst surface, and the
formed ethylene product will be oxidized deeper into CO and CO2 [7,9]. For a good
catalyst for OCM, suitable alkaline sites, selective mobile oxygen sites and high
thermal stability are required.
Several catalyst systems such as Mn/Na2WO4/SiO2, Li/MgO, rare-earth oxides and
perovskite composite oxides have been investigated for this reaction. However,
Li/MgO catalyst deactivates rapidly due to the vaporization of Li+ cations, which is in
fact a structure modifier of MgO, during the high temperature OCM reaction [10-12],
Therefore, poor stability is still a problem for this catalyst. Although pure rare-earth
oxides or rare-earth oxides doped by alkaline earth metal oxides, such as lanthanide
oxide, exhibit high activity, they generally have low C2 yield due to the low C2
selectivity [8,13,14]. In comparison with these two types of catalysts, relatively fewer
studies have been reported on perovskite composite oxide catalysts, although it is
found that they also display certain activity for this reaction [15,16].
3

Up to date, Mn/NaWO4/SiO2 is regarded as the most promising catalyst system for
OCM, on which the highest yield around 25% has been achieved [17]. In addition, it
also possesses good reaction and thermal stability. Therefore, it has caught much
attention over recent twenty years. For instance, different supports such as Mg-Ti
mixed oxides, SBA-15, COK-12, SiC have been used to substitute the original SiO2
support to improve the performance of the catalyst [18-20]. To get better catalysts,
people also tried to replace the Na+ cations in the structure with other alkali metal
oxides or chlorides [21-24]. Moreover, promoting Mn/Na2WO4/SiO2 with other metal
oxides such as SnO2, La2O3 and CeO2 have been investigated [25-27]. It is proved that
the stability, selectivity and C2 yield can be improved by the adopting of these
methods. However, over all of these catalysts, the highest C2 yield obtained is still
lower than the minimum yield requirement of 30%. In addition, high temperature
o
above 800 C is generally required to precede the reaction. Therefore, for the
implementation of this reaction industrially, there is still strong motivation to seek
novel catalysts with higher reaction performance, especially at lower temperatures
[28]. Notably, It is reported recently by Lu and co-workers that by using TiO2 to
modify the Mn/Na2WO4/SiO2 catalyst, the OCM reaction temperature can be evident
decreased [28].
Pyrochlore composite oxides, which have a formula of A2B2O7, usually have high
melting points and very good thermal stability. As demonstrated by our former works,
pyrochlores with different chemical compositions display good performance for
several high temperature reactions [29-31]. For example, it is reported that La2Zr2O7
o
has a high melting point of 2300 C [32], which enables it to stand high
temperatures. An ideal pyrochlore A2B2O7 structure belonging to space group Fd 3
m, whose formula can also be written as A2B2O6 O', indicating the presence of two
types of lattice oxygen anions with different chemical environments in its crystal
structure. Each of the six lattice equal oxygen ions (O) is coordinated to two A and
two B sites cations, and the left one lattice oxygen ions (O') is coordinated to four A
sites cations [31,33-35]. Compared with ordered cubic fluorite structure, one lattice
4

oxygen ions is absent, thus resulting in inherent oxygen vacancies in the pyrochlore
compounds, which improves the oxygen mobility of the pyrochlore catalysts
[29,31,36]. Furthermore, the thermal stability and oxygen mobility of pyrochlore
compounds can be tuned by replacing or partly substituting the A and B sites
[29,30,37,38]. Most interestingly, it has been revealed that the phase composition of
the A2B2O7 compounds can also be altered with varied combination of the A and B
sites cations. In detail, if the rA3+/rB4+ ratio of a compound is in the range of 1.46 to
1.78, a stable and ordered cubic pyrochlore crystalline phase will be formed. However,
with the decrease of this rA3+/rB4+ ratio below 1.46, the initial ordered cubic
pyrochlore phase will subsequently become a defective cubic fluorite phase, which
now belongs to space group Fm 3 m and the oxygen mobility improves due to the
randomly exchange of the A and B site cations [31].
Therefore, it seems that the requirements for a good OCM catalyst can be mostly
matched with A2B2O7 compounds. However, up to now, using these compounds as
OCM catalysts has been rarely investigated and lacks systematic understanding on its
structure and reactivity relationship. Over the past 30 years, less than 20 literatures
have been reported on using A2B2O7 catalysts for this reaction [39-43]. In 1989,
Ashcroft and co-workers prepared a series of Ln2Sn2O7 pyrochlores by varying the
rare earth metals for OCM, and revealed that they are active catalysts for this reaction
[39]. On a Pr2Sn2O7 catalyst prepared by them, 10.4% C2+ product yield can be
o
obtained at 870 C. In 1994, Roger and co-workers reported that Sm2Sn2O7 catalysts
doped by Ba and Cl displayed very good reaction performance for OCM. Over the
best catalyst, 17% C2 yield can be obtained at 775 °C [40]. It was found by
Kiennemann and co-workers that on pyrochlores with different chemical
compositions (A = rare earth, B = Ti, Zr, Sn), C2 selectivity and yield are strongly
influenced by the B-O bond energy. Typically, the lower the B-O bond energy, the
higher the C2 yield [41]. Lewandowski and co-workers synthesized a series of
Bi2Sn2-xBixO7-x/2 (0≤x≤0.86) pyrochlore oxides and used for OCM, in which the B site
cations were partially substituted by excess Bi cations. With the increasing of the
5

amount of Bi cations in the B sites, the C2 selectivity becomes higher but at the
expense of activity [42]. In addition, Kiennemann and co-workers prepared a series of
Sm2Sn2O7 catalysts with a sol-gel like method for OCM, which lead to the formation
of tin-deficient Sm2Sn2O7 pyrochlores. It is concluded that the tin-deficiency in the B
sites can improve the C2 yield evidently [43].
Although these former works have demonstrated that A2B2O7 compounds with
varied chemical compositions could be promising for OCM, the relationship between
catalytic performance and the structure have not been clearly elucidated. Therefore, as
an endeavor to develop improved catalysts with potential industrial applications, in
this work, a series of Ln2Ce2O7 composite oxides with a fixed Ce B site cation but
with altered rare earth A sites have been prepared with a sol-gel method. It has been
found that in comparison with the formerly reported catalyst systems, La2Ce2O7
displays much better reaction performance at low temperature region, on which 15.3%
o
C2 yield can be obtained even at 650 C. Notably, Sm2Ce2O7 exhibits superior low
temperature reaction performance, on which 10.3% C2 product yield can be obtained
o
even at 550 C. With different characterization techniques, the change of the
crystalline phase structure and surface property of the catalysts by the replacement of
A site have been elucidated, which has been correlated with the reaction performance.
2. Experimental
2.1 Catalyst preparation
Synthesis of Ln2Ce2O7 catalysts
A series of Ln2Ce2O7 catalysts with a fixed Ce B site cation but with varied A site
cations (Ln=La, Pr, Sm and Y) have been prepared by citric acid sol-gel method,
using La(NO3)3.6H2O (99.99%), Pr6O11 (99.9%), Sm2(CO3)3 (99.9%), Y(NO3)3.6H2O
(99.99%), Ce(NO3)3.6H2O (99.99%) and citric acid (99.8%) as the starting materials.
While La(NO3)3.6H2O, Y(NO3)3.6H2O and Ce(NO3)3.6H2O can be directly dissolved
into deionized water to prepare stable solutions, Pr6O11 and Sm2(CO3)3 were dissolved
6

first into a suitable amount of nitric acid solution to prepare Pr(NO3)x and Sm(NO3)3
solutions. Then, the stoichiometric molar ratio of Ln nitrates: Ce(NO3)3.6H2O: citric
acid = 1:1:1.2 were dissolved into a certain amount of deionized water and stirred
thoroughly for 2 h for uniform mixing. NH3.H2O solution (22 ~ 25 wt%) was slowly
added into the solution mixture to adjust the pH value around 2. Afterwards, the
excess water was evaporated under constant stirring in a water bath controlled at 80
oC until viscous gel was formed. The gel was aged at 130 oC for 12 h, which was then
o
calcined at 800 C for 4 h in a muffle furnace in air atmosphere to remove the citric
acid and obtain the final Ln2Ce2O7 (Ln=La, Pr, Sm, Y) catalysts. Using the same
method and procedure, a CeO2 comparison sample was also prepared with a
Ce(NO3)3.6H2O: citric acid molar ratio of 1:1.2.
Preparation of Mn/Na2WO4/SiO2 comparison catalyst
For comparison purpose, a Mn/Na2WO4/SiO2 catalyst was also prepared with
incipient-wetness impregnation method according to the references [2,44]. In detail,
manganese (II) nitrate solution (50 wt%) and hydrated sodium tungstate (99.8%) were
dissolved together into a certain amount of deionized water to achieve aqueous
solution. Under constant stirring, a commercial SiO2 support (Aladdin, purity 99.8%)
with a surface area of 200 m2 g-1 was then added into a certain amount of the solution,
o
which was followed by drying at 105 C overnight. Afterwards, the dried solid was
o
calcined in an air atmosphere in a muffle furnace controlled at 800 C for 4 h to get
the final catalyst, which contains 2% weight percentage of Mn and 5% weight
percentage of Na2WO4. XRD technique was used to analyze its phase composition.
As shown in Fig. S1, Na2WO4, Mn2O3 and α-cristobalite phases are detected,
confirming the successful preparation of the desired catalyst [44].
2.2 Catalyst characterization
To analyze the bulk phase composition of the fresh and spent catalysts, X-ray
diffraction ( XRD) patterns were collected on a Bruker AXS D8 Focus diffractometer
7

using a Cu target and Kα-ray irradiation, which is operated at 40KV and 30mA. Scans
were collected over a 2θ rangefrom 10° to 90° with step of 0.02° s-1.
Nitrogen adsorption-desorption experiments of the samples were performed at 77 K
on ASAP2020 instrument. The specific surface areas of the catalysts were calculated
using the Brunauer-Emmett-Teller (BET) method in a relative pressure (P/P0) range
of 0.05 ~ 0.25. The pore size distributions of samples were calculated with the
Barrett-Joyner-Halenda (BJH) method. The average pore sizes were obtained from the
peak positions of the distribution curves. Total pore volumes were accumulated at a
relative pressure of P/P0 = 0.99.
Raman spectra of the catalysts were recorded on a Renishaw inVia Raman
spectrometer equipped with an argon laser excitation source operating at 532 nm and
a Renishaw RenCam CCD detector. The scanned Raman shift range is from 100 to
1500 cm−1.
Hydrogen temperature programmed reduction (H2-TPR) was performed on
FINESORB 3010C instrument. Generally, 0.05 g samples were used for the
experiments. Before the experiments, the catalysts were calcined again in a high
purity air flow at 120 oC for 30 min to remove any possible impurities. After purging
with an ultra-high purity Ar flow at room temperature for 30 min, the sample was then
o
heated from room temperature to 900 oC with a ramp of 10 C min-1 in a 30 mL
min-1 10% H2/Ar gas mixture flow. A thermal conductivity detector (TCD) was
employed to monitor the H2 consumption. To quantify the H2 uptake amount, a 99.99%
CuO standard sample was used for calibration.
CO2 temperature programmed desorption (CO2-TPD) was performed with a
micromeritics Auto Chem 2920 apparatus. Generally, 0.05 g catalyst was used for the
o
test in a 30 ml min-1 He flow. Prior to the test, the sample was heated to 400 C and
kept for 30 min to remove any possible impurities. Afterwards, the sample was cooled
o
to 50 C and exposed to a 30 ml min-1 CO2 flow for 1 h to saturate the surface
8

completely, which was followed by purging with a 30 ml min-1 ultra-high purity He
flow to remove the any physically absorbed CO2 for 30 min. After all these
pretreatments, the catalyst was heated from 50 to 900 oC at a rate of 10 oC min-1.
X-ray Photoelectron Spectroscopy (XPS) tests were carried out on a PHI-5000C
ESCA system using a single MgKα X-ray source operating at 300 W and 15 kV of
voltage. The spectra were obtained at ambient temperature with an ultrahigh vacuum.
The binding energies were calibrated by using the C 1s peak of graphite at 284.6 eV
as a standard.
2.3 Activity evaluation
The catalytic performance of the catalysts was carried out in a fixed-bed quartz
micro-reactor with an inner diameter of 10 mm under 1 atm. typically, 0.2 g catalyst
was placed in the middle of reactor, and then heated to the target temperature. The
reaction feed consists of 40% methane and 10% oxygen with 50% nitrogen as a
balance gas, with the flow rates controlled by three mass flow controllers separately.
The CH4 to O2 volume ratio of 4/1 is adopted for all the experiments. The total flow
rate of the feed gas is 60 mL min-1, which results in a weight hourly space velocity of
18000 mL h-1 g-1. The measurement of the activity and selectivity generally starts at
o
o
o
550 C and the temperature is increased by a 50 C gap until 750 C. In order to get
steady state kinetic data, all the injections were taken after 1 h stabilization at a certain
reaction temperature. The stability of the best catalyst was further studied for 100 h at
the optimum reaction temperature. The methane coupling products were analyzed by
two online gas chromatographs, with a GC9310 chromatograph equipped with a
TDX-01 column and a TCD detector for the analysis of CO, CO2 and CH4; and a
GC9310II chromatograph equipped with a Propak Q column and a FID detector for
the analysis of CH4, C2H6, C2H4, C3H8, C3H6.
The methane conversion (XCH4), C2 selectivity (SC2) and C2 yield (YC2) in this study
are calculated with the following equations [16,45]:
9

[
(
)
(
)
8
] × 100%
(1)
2∗ C H +C H +3∗ C H +C H +CO+CO
2
4
2
6
3
6
3
2
XCH
=
4
(
)
CH
4 in
(
)
2∗ C H +C H
2 6
2
4
Sc2 = 2∗ C H +C H +3∗ C H +C H +CO+CO × 100%
(2)
(3)
(
)
(
)
8
2
4
2
6
3
6
3
2
YC = XCH × SC2 × 100%
2
4
It is particularly noted here that CH4, C2H4, C2H6, C3H6, C3H8, CO and CO2 in the
equations stands for the molar amount of the corresponding reactants and products,
which has already calibrated by the integrated areas and response factors.
3. Results and discussion
3.1 XRD phase analysis of the La2Ce2O7 catalysts
As presented in Fig. 1, XRD technique was used to identify the phase composition
of the catalysts with different A sites. Fig. 1 (a) displays the XRD patterns of all the
freshly calcined catalysts. The sharp peaks of all the samples testify that they are well
o
crystallized after calcination at 800 C for 4 h in air atmosphere. For comparison
purpose, the diffraction pattern of the comparison CeO2 sample is also included,
which shows the typical diffraction feature of a cubic fluorite CeO2 crystalline phase.
Interestingly, all the Ln2Ce2O7 samples exhibit very similar diffraction features to the
cubic CeO2 phase with only slight shift of the 2θ values, implying that these
compounds have very similar bulk phase structure to cubic CeO2. The samples after
high temperature reaction have also been investigated by XRD, with the results shown
in Fig. S2. Compared with the fresh catalysts, no evident difference can be discerned,
indicating that these Ln2Ce2O7 compounds with varied A sites are both physically and
chemically stable for the high temperature OCM reaction.
10

111
111
(a)
220
220
200
311
222
La Ce O
7
420
331
2
2
400
311
222
200
400
331
Pr Ce O
7
420
400
2
2
222
440
440
622
444
Sm Ce O
7
662
800
800
840
840
2
2
622
444
222 400
111 200
662
Y Ce O
7
2
2
311
222
220
331
420
400
CeO
2
10
20
30
40
50
2 ()
60
70
80
(b)
Ln=La
Ln=Pr
Fluorite type
Rare earth C-type
431
411
332
Ln=Sm
Ln=Y



411
431
332



CeO
2
30
32
34
36
38
40
42
44
2 ()
Fig. 1 XRD patterns of the freshly calcined Ln2Ce2O7 catalysts, (a) Complete patterns
from 10 ~ 80 o, (b) Magnified patterns from 30 ~ 44 o.
As formerly reported, for A2B2O7 compounds, the bulk phase structures are
determined by the rA3+/rB4+ ratios [31,34,46-49]. In detail, if the rA3+/rB4+ ratio of a
compound is in the range of 1.46 to 1.78, a stable and ordered cubic pyrochlore phase
will be formed. With the decrease of this rA3+/rB4+ ratio below 1.46, the initial ordered
cubic pyrochlore crystalline phase will subsequently become a defective cubic fluorite
phase. For all the Ln2Ce2O7 compounds in this study, the radius of the 6-fold
coordinated Ce4+ at B site is 0.87 Å, and the radii of the 8-fold coordinated La3+, Pr3+,
11

Sm3+ and Y3+ at A site are 1.16, 1.13, 1.08 and 1.02 Å in sequence. Therefore, the
calculated rA3+/rB4+ ratios of La2Ce2O7, Pr2Ce2O7, Sm2Ce2O7 and Y2Ce2O7 are 1.33,
1.30, 1.24 and 1.17, respectively. Apparently, all these values are below the threshold
of 1.46, indicating the possible formation of a defective fluorite phase in the catalysts.
XPS technique was used to analyze the surface composition of the catalysts, with
the spectra shown in Fig. 2. The typical binding energies testify that all the A site rare
earth metal cations are present as 3+ valence state [50-52] except for Pr in Pr2Ce2O7,
for which both Pr3+ and Pr4+ cations are co-existing, as shown in Fig. 2(C). In detail,
the peaks at the lower binding energy of 928.5 eV and 933.3 eV in the Pr 3d5/2 state
are ascribed to Pr3+, the peaks at the higher binding energy of 930.5 eV and 936.3 eV
in the Pr 3d5/2 state are assigned to Pr4+. In the Pr 3d3/2 state, the Pr3+ peaks are
observed at 949.4 eV and 953.2 eV, while Pr4+ peaks are located at 946.0 eV , 952.0
eV and 954.8 eV. It can thus be concluded that both Pr3+ and Pr4+ cations are present
in the Pr2Ce2O7 sample [53].
Ce4+3d3/2
Ce3+3d3/2
900.3
(a) Ce 3d
Ce3+3d5/2
(b) La 3d
Ce4+3d5/2
La3+ 3d3/2
897.6
855.7
La3+ 3d5/2
835.2
915.8
852.0
838.6
881.7
920
910
900
Binding Energy (eV)
890
880
860
855
850
845
840
835
830
Binding Energy (eV)
(c) Pr 3d
(d) Sm 3d
1083.5
Pr 3d5//2
Sm3+3d5/2
Pr 3d3/2
Sm3+3d3/2
1110.8
Pr3+
Pr3+
Pr4+
Pr4+
Pr4+
1120
1110
1100
1090
1080
1070
960
955
950
945
940
935
930
925
Binding Energy (eV)
Binding Energy (eV)
Ce 3d Ln Ce O
7
(f)
(e) Y 3d
Y3+3d5/2
2
2
Y3+3d3/2
158.8
Ce4+3d5/2
Ce3+3d3/2
Ce4+3d3/2
Ce3+3d5/2
Y2Ce2O7
156.8
Sm2Ce2O7
Pr2Ce2O7
La2Ce2O7
CeO2
170
165
160
155
150
930
920
910
900
890
880
870
Binding Energy (eV)
Binding Energy (eV)
Fig. 2 XPS spectra of the metal cations on the surface of the corresponding Ln2Ce2O7
12

catalysts. (a) Ce 3d on CeO2, (b) La 3d on La2Ce2O7, (c) Pr 3d on Pr2Ce2O7, (d) Sm
3d on Sm2Ce2O7, (e)Y 3d on Y2Ce2O7, (f) Ce 3d on Ln2Ce2O7 with different A sites.
It is particularly noted here that for the Pr-modified CeO2, besides the formation of
defective cubic Pr2Ce2O7 fluorite phase, an unstable solid solution structure might
have also been formed [54-56]. As discussed above, in defective cubic Pr2Ce2O7
fluorite phase, the Ce4+ is 6-fold coordinated with a radius of 0.87 Å. In individual
cubic fluorite CeO2, the Ce4+ is 8-fold coordinated with a radius of 0.97 Å. However,
in this study, it is found by XPS that both Pr3+ and Pr4+ cations are present in the
Pr-modified CeO2 sample. If a Pr4+ cation has an 8-fold coordination, its radius is 0.96
Å, which is nearly same to that of 8-fold coordinated Ce4+ cations in the lattice of
cubic fluorite CeO2 matrix. Therefore, there is high possibility for the observed Pr4+
cations to be dissolved into the crystal matrix of cubic fluorite CeO2 to form an
unstable solid solution phase. When studying Pr-modified ceria microspheres for
methane combustion, Guo and other authors also found that the incorporation of Pr4+
cations into the fluorite lattice of CeO2 to form nonstoichiometric and unstable
Pr-Ce-O solid oxide solution [54,56], which is consistent with what we observed in
this study. Whereas, the observed Pr3+ cations is believed to be present as A sites in
the structure of defective cubic Pr2Ce2O7 fluorite phase. Therefore, it is reasonable to
deduce that the Pr-modified CeO2 sample is composed of both cubic Pr-Ce-O solid
solution and defective Pr2Ce2O7 fluorite phases. for which our H2-TPR results in the
following section will provide more evidence. Indeed, this explains the reason that all
the Ln2Ce2O7 compounds exhibit similar diffraction patterns to cubic fluorite CeO2.
With the decreasing of the rA3+/rB4+ ratios, the 2θ values of the Ln2Ce2O7 catalysts
shift gradually to those of cubic CeO2 fluorite, as shown in Fig. 1(b). Started from
Sm2Ce2O7, besides the cubic fluorite type (F-type) phase, a new rare-earth cubic-type
(C-type) superstructure can be observed, as evidenced by the (411), (332) and (431)
diffraction peaks in the enlarged Fig. 1(b) [57-61]. Table S1 demonstrates that with
the formation of C-type structure, the cubic lattice parameter of these two samples is
13

doubled in comparison with the cubic fluorite phase. However, due to the very low
intensity of the super lattice lines, the formation of C-type superstructure phase could
be incomplete and with only a small amount [57, 58, 60].
3.2 Structure analysis of Ln2Ce2O7 catalysts by Raman spectroscopy
To understand the structure change of the Ln2Ce2O7 catalysts with the replacement
of A sites, the freshly calcined samples were further measured by Raman technique,
with the spectra exhibited in Fig. 3. The Raman spectra of the spent catalysts are also
presented in Fig. S3. Compared with the fresh catalysts, no evident change can be
observed, testifying again that these Ln2Ce2O7 compounds with different A sites are
stable for the high temperature OCM reaction.
F2g
453
256
oxygen vancancy
576
La Ce O
7
2
2
570
586
451
Pr Ce O
7
2
2
476
370
478
260
271
Sm Ce O
7
2
2
585
376
465
Y Ce O
7
2
2
CeO
2
1500
100
200
1400 1200 1000
800
600
400
Raman shift (cm-1)
Fig.3 Raman spectra of the Ln2Ce2O7 catalysts.
For comparison study, the individual CeO2 with a typical cubic fluorite structure
was also measured by Raman technique which displays a peak at 465 cm-1, assigning
to the F2g mode of cubic fluorite structure can be observed [62-65]. In comparison,
while this typical F2g mode peak shifts lower to ~ 450 cm-1 for La2Ce2O7 and
Pr2Ce2O7, it shifts higher to ~ 480 cm-1 for both Sm2Ce2O7 and Y2Ce2O7. In addition,
for all the Ln2Ce2O7 samples, this typical F2g peak is also significantly broadened,
14

which can be explained by the doping of the Ln3+ cations into the cubic fluorite CeO2
crystalline matrix [57,59,66]. While La2Ce2O7, Sm2Ce2O7 and Y2Ce2O7 display very
similar Raman characteristics, the Raman spectroscopy of Pr2Ce2O7 is obviously
different, testifying further that it has different phase structure from other three
Ln2Ce2O7 samples. Indeed, Raman results have demonstrated that with the
replacement of the A sites by different rare earth trivalent cations, the cubic structure
is still retained for all the samples but with some alteration of their crystal lattices.
As formerly reported, an ordered cubic A2B2O7 pyrochlore belongs to space group
Fd 3 m and theoretically has 6 Raman active modes distributed as A1g + Eg + 4F2g,
among which an intense Eg mode at ~ 300 cm-1 will generally be observed [31,67,68].
Apparently, in the spectra of all the Ln2Ce2O7 catalysts, these typical Raman modes
belonging to ordered pyrochlore phase are absent, testifying they consist of defective
cubic fluorite structure, in line with the XRD phase analysis results. For both of
Sm2Ce2O7 and Y2Ce2O7, a new Raman peak at ~ 370 cm-1 is detected, which is not
observed for La2Ce2O7 and Pr2Ce2O7 and can be assigned to rare-earth C-type oxides
[57,59,61].
For all of the Ln2Ce2O7 catalysts, another strong Raman peak at 570 ~ 590 cm-1 is
observed, which is assigned to the oxygen vacancies present on the catalyst surface
[66,69]. As reported previously, if a CeO2 sample calcined at low temperature with
many defects, this oxygen vacancy peak can also be obviously observed [62-64].
o
However, in this study, since all the samples are calcined at 800 C and well
crystallized, it is evident that the surface oxygen vacancies of the individual CeO2
sample has been depleted, thus no corresponding peak can be observed. Whereas,
with the incorporation of La3+, Pr3+, Sm3+ and Y3+ cations to form binary compounds,
the oxygen vacancies can be stabilized even at high temperature. When studying La
modified CeO2, some researchers also revealed that due to the doping of trivalent La3+
into the CeO2, a large amount of oxygen vacancies was created in order to maintain
the overall charge balance [66,69].
15

In summary, the Raman results are well consistent with the XRD measurements,
testifying the replacement of the A site of Ln2Ce2O7 can alter both of the structure and
oxygen properties of the resulted catalysts. However, all of catalysts can still be
considered to consist of defective cubic fluorite phase.
3.3 N2 adsorption-desorption measurement of the Ln2Ce2O7 catalysts
N2 adsorption-desorption technique was used to investigate the texture properties of
the Ln2Ce2O7 compounds containing different A sites, with the results shown in Fig. 4
and Table 1. All the Ln2Ce2O7 samples display Type III isotherms and H3 type
hysteresis loops, indicating they have similar texture structure. In comparison with the
individual CeO2, all the Ln2Ce2O7 catalysts, either fresh or used, have evidently
improved pore volumes. As identified from the pores size distribution profiles (Fig. 4
a' and b'), except for pure CeO2, two set of the mesopores have been observed in the
structure of the samples. The first set below 5 nm is attributed to the intraparticle
mesopores in the bulk structure, and the second set above 10 nm is assigned to the
interparticle mesopores generated from the assembly of the particles.
(a)
(a')
La Ce O
7
CeO
2
2
2
La Ce O
7
2
2
Pr Ce O
7
Pr Ce O
7
2
2
2
2
Sm Ce O
2
2
7
Y Ce O
7
2
2
Sm Ce O
7
2
2
Y Ce O
2
2
7
CeO
2
0.0
0.2
0.4
0.6
0.8
1.0
0
20
40
60
80
100
Relative pressure (P/P0)
Pore size (nm)
(b)
La Ce O
(b')
CeO
2
2
2
7
La Ce O
7
2
2
Pr Ce O
7
Pr Ce O
2
2
7
2
2
Sm Ce O
7
2
2
Y Ce O
7
2
2
Sm Ce O
7
2
2
Y Ce O
7
2
2
CeO
2
0.0
0.2
0.4
0.6
0.8
1.0
0
20
40
60
80
100
Relative pressure (P/P0)
Pore size (nm)
16

Fig. 4 N2 adsorption-desorption isotherms and the pore size distribution profiles of the
Ln2Ce2O7 catalysts. (a), (a') freshly calcined catalysts, (b), (b') spent catalysts.
The BET surface areas of both the fresh and used Ln2Ce2O7 catalysts are shown in
Table 1. Apparently, the incorporation of La3+, Pr3+, Sm3+ and Y3+ cations into CeO2
to form binary compounds can result in samples with higher surface areas, proving the
thermal resistance has been improved. After reaction, the surface areas of all the
Ln2Ce2O7 catalysts have been slightly decreased, indicating these samples are
physically stable during the high temperature reaction process.
In summary, after the combination of La3+, Pr3+, Sm3+ and Y3+ cations into with
CeO2 to form binary compounds, catalysts with more abundant pore structure, higher
surface areas and better thermal stability can be obtained. This, indeed, could be of
advantage for the diffusion of the reactants and their interaction with the active
surface sites, and in favor of the reaction performance of the catalysts.
Table 1. Physicochemical properties of the freshly calcined and spent Ln2Ce2O7
catalysts measured by N2 adsorption-desorption.
Catalysts
BET surface area
(m2. g-1)
Pore volume
(cm3.g-1)
Pore size
(nm)
fresh
23
23
21
9
spent
20
17
20
9
fresh
spent
0.13
0.08
0.12
0.05
0.01
fresh
spent
26.11
18.90
22.26
27.88
32.49
La2Ce2O7
Pr2Ce2O7
Sm2Ce2O7
Y2Ce2O7
CeO2
0.10
0.13
0.13
0.06
0.02
16.41
21.57
21.72
24.32
15.38
4
2
17

3.4 H2-TPR study on Ln2Ce2O7 catalysts
The redox properties of the Ln2Ce2O7 catalysts with different A sites were
investigated by using H2-TPR technique. As shown in Fig. 5 and Table 2, CeO2, the
comparison sample, displays two typical reduction peaks [70,71]. The lower
temperature peak at 425 oC can be assigned to the reduction of the surface active Ce4+
o
to Ce3+; and the higher temperature peak at 780 C is assigned to the reduction of a
fraction of the bulk lattice oxygen. In contrast, all the Ln2Ce2O7 catalysts show very
different reduction behavior from pure CeO2, testifying the incorporation of the
secondary rare earth cations has altered not only the phase structure but also the
oxygen properties of the catalysts. Interestingly, La2Ce2O7, Sm2Ce2O7 and Y2Ce2O7
exhibit very similar reduction behavior. On the profiles of these catalysts, two
separated low intensity reduction peaks are evidently observed, one is positioned at ~
o
o
350 C and the other is positioned above ~ 500 C. Formerly, when investigating
La2Sn2O7 pyrochlore catalysts, it has been found that after the formation of
pyrochlore phase structure, the originally fragile Sn-O bonds become very difficult to
be reduced in the pyrochlore lattice matrix due to the change of the chemical
environment [29,72]. As a result, the Sn4+ in the crystal lattice can only be reduced
o
above 800 C. After all, the same phenomenon could happen in Ln2Ce2O7 catalysts.
Therefore, it is believed that the peak at ~ 350 oC can be assigned to the reduction of
o
the surface loosely bounded oxygen species. However, for the peak above 500 C,
with the change of the A sites, the reduction temperature is different, indicating it
might be assigned to the reduction of a small amount of Ce4+ to Ce3+ in the Ln2Ce2O7
crystal lattice, which is influenced by the variation of the A site. It is particularly
mentioned here that Pr2Ce2O7 shows very different reduction behavior from other
three Ln2Ce2O7 catalysts, which provides extra evidence to XRD and Raman to testify
that this sample has different phase composition. As described above, this sample
consists of both unstable cubic Pr-Ce-O solid solution and defective Pr2Ce2O7 fluorite
phases, in which both Pr3+ and Pr4+ are present. According to the previous studies, its
18

o
two reduction peaks at 453 and 493 C might be attributed to partial reduction of
praseodymia phase similar to Pr6O11 species into Pr2O3 in two steps [73].
453
493
525
351
La2Ce2O7
Pr2Ce2O7
Sm2Ce2O7
561
1/2
584
363
353
559
Y2Ce2O7
780
425
CeO2
200
300
400
500
600
700
800
Temperature (oC)
Fig. 5 H2-TPR profiles of Ln2Ce2O7 catalysts.
Table 2. Quantified H2-TPR results of the Ln2Ce2O7 catalysts.
Catalysts
Temperature
(oC)
H2
Temperature
(oC)
H2
consumption
amount
consumption
amount
(mmol g-1)
0.21
(mmol g-1)
0.24
La2Ce2O7
Pr2Ce2O7
351
453
525
493
0.67
0.65
19

Sm2Ce2O7
Y2Ce2O7
CeO2
363
353
425
0.20
0.23
0.21
559
584
780
0.28
0.21
0.96
3.5 Reaction performance of the Ln2Ce2O7 catalysts
The catalytic performance of the Ln2Ce2O7 catalysts has been evaluated for
methane oxidative coupling, with the results shown in Fig. 6. With the increasing of
o
the reaction temperature from 550 to 750 C, the methane conversion, C2 product
selectivity and yield of all the catalysts increased. In addition, the methane
conversions, C2 product selectivity and yields of the catalysts follow the sequence of
o
La2Ce2O7 > Y2Ce2O7 ≈ Sm2Ce2O7 > Pr2Ce2O7 > CeO2 at 650 ~ 750 C. It is
particularly noted here that Sm2Ce2O7 shows the best reaction performance among all
the catalysts at lower temperature region. Even at 550 °C, 21.8% methane conversion,
47.3% C2 selectivity and 10.3% yield can be obtained. In addition, it is interestingly to
find that the methane conversion, C2 product selectivity and yield of Sm2Ce2O7
catalyst increase slightly in the whole temperature region, implying that the reaction
temperature has only limited influence on its reaction performance.
For the three catalysts of La2Ce2O7, Y2Ce2O7 and Sm2Ce2O7, with sole defective
cubic fluorite phase composition, exhibit significantly better reaction performance
than Pr2Ce2O7 and CeO2, implying this phase composition is favorable for the
reaction [74]. The presence of cubic fluorite CeO2 and Pr-Ce-O solid solution phase
could be harmful for the OCM reaction. Among all the catalysts, La2Ce2O7 shows the
best catalytic performance, on which the highest C2 yield of 16.6% can be obtained at
750 oC.
To obtain deeper understanding on the reaction process, the C2H4 selectivity over
the catalysts has been particularly analyzed in Fig. 6 (d). With the increasing of the
reaction temperature, the C2H4 selectivity increases gradually, indicating the
formation of more C2H4 at higher temperature due to the dehydrogenation of C2H6
[75]. La2Ce2O7 shows always the highest C2H4 selectivity among all the catalysts,
20

especially at higher temperature region, which implies that it contains more abundant
active sites favoring the dehydrogenation of the initially formed C2H6.
For additional references, both of the COx selectivity and CO/CO2 ratios on the
catalysts are also demonstrated in Fig. S4. With the increasing of the reaction
temperature, while the COx selectivity decreases evidently, the CO/CO2 ratios
increase apparently, which is in accordance with the viewpoint that CO2 can be
favorably formed at lower temperature [75,76]. At higher temperature, more C2
product and CO can be formed. In fact, this phenomenon is attributed to the
thermodynamic equilibrium between the formation of CO and CO2. At high
temperature, the CO2 byproducts from OCM process can react with the left methane
reactant to form CO. On the other hand, with the formation of more C2 products, more
O2 is consumed. The unreacted methane and C2 hydrocarbons thus react with
inadequate O2 to form CO. However, It is proposed that the reforming of methane
with CO2 to produce CO is the main reason of the high CO/CO2 ratio at high
temperature [75,76].
100
CeO
2
(a)
(b)
CeO
La2Ce2O7
Pr2Ce2O7
Sm2Ce2O7
Y2Ce2O7
2
30
20
10
0
La Ce O
7
2
2
80
60
40
20
0
Pr Ce O
7
2
2
Sm Ce O
7
2
2
Y Ce O
7
2
2
550
600
650
700
750
550
600
650
700
750
Temperature (oC)
Temperature (oC)
21

30
20
10
0
50
40
30
20
10
0
CeO
CeO
2
(c)
(d)
2
La Ce O
La Ce O
2 2 7
2
2
7
Pr Ce O
Pr Ce O
2 2 7
2
2
7
Sm Ce O
Sm Ce O
2 2 7
2
2
7
Y Ce O
Y Ce O
2 2 7
2
2
7
550
600
650
700
750
550
600
650
700
750
Temperature (oC)
Temperature (oC)
Fig. 6 The OCM performance of the Ln2Ce2O7 catalysts. (a) Methane conversion, (b)
C2 selectivity, (c) C2 yield, (d) C2H4 selectivity. Reaction conditions: CH4/O2=4:1
(nitrogen as balance gas), WHSV=18000mL-1 h-1 g-1.
To investigate the durability of La2Ce2O7, the catalyst with the best catalytic
o
performance, it was subjected to a long-term stability test at 750 C for 100 h. As
shown in Fig. 7, the methane conversion, C2 product selectivity and yield is very
stable during the time span without any decrease, indicating that this catalyst
possesses very stable reaction performance.
100
Methane conversion
C selectivity
2
80
60
40
20
0
C yield
2
0
20
40
60
80
100
Time on stream(h)
o
Fig. 7 Long-term stability test of La2Ce2O7 catalyst at 750 C for 100 h. Reaction
conditions: CH4/O2=4:1, (nitrogen as balance gas), WHSV=18000 mL-1 h-1 g-1.
22

Up to date, Mn/Na2WO4/SiO2 is still the most promising OCM catalyst for
industrial application, on which the highest one-way yield close to 30% has been
reported by some researchers above 800 oC [17]. However, this catalyst has very low
o
activity below 700 C. Therefore, it is always desirable to find catalysts with
improved low temperature performance for OCM. As described above, in the
temperature region of 650 ~ 750 oC, reaction temperature has limited influence on the
performance of Ln2Ce2O7 catalysts. Therefore, the low temperature reaction
performance of the Ln2Ce2O7 catalysts have been investigated further at temperatures
o
as low as 550 C. In addition, a comparison of the reaction performance between
La2Ce2O7, the best catalyst in this study, and Mn/Na2WO4/SiO2 has been carried out
for more information. As shown in Fig. 8, even at 550 oC, 17.0% methane conversion,
35.0% C2 product selectivity and 6.0% yield have already been achieved over
La2Ce2O7. In comparison, no any reaction happens over Mn/Na2WO4/SiO2 at this
o
temperature. Not until 650 C, does the OCM reaction happen on this catalyst. As
shown in Table S2, at this temperature, only 0.1% methane conversion and 0.01% C2
product yield were obtained on it. In contrast, on La2Ce2O7, 26.6% methane
conversion and 15.3 % C2 product yield have already been achieved. Whereas, at 750
oC, both of the catalysts show close reaction performance. With further increasing the
o
reaction temperature to 800 C, 23.1% C2 product yield can be obtained on
Mn/Na2WO4/SiO2, which is higher than 17.4% on La2Ce2O7, testifying that the
o
former has better reaction performance than the latter above 750 C. These results
have demonstrated that La2Ce2O7 and Sm2Ce2O7 catalysts possess outstanding low
temperature reaction performance for OCM reaction, which may give people new
insight to develop OCM catalysts that can be operated at lower temperature region,
thus facilitating its industrializing process.
23

40
30
20
10
0
100
80
60
40
20
0
(a)
(b)
Mn/Na2WO4/SiO2
La2Ce2O7
Mn/Na2WO4/SiO2
La2Ce2O7
550
600
650
700
750
800
(c)
550
600
650
700
750
800
Temperature (oC)
Temperature (oC)
30
25
20
15
10
5
Mn/Na2WO4/SiO2
La2Ce2O7
0
550
600
650
700
750
800
Temperature (oC)
Fig. 8 The comparison of reaction performance between Mn/Na2WO4/SiO2 and
La2Ce2O7 for OCM. (a) methane conversion, (b) C2 selectivity, (c) C2 yield. Reaction
conditions: CH4/O2=4:1 (nitrogen as dilution gas), WHSV=18000 mL-1 h-1 g-1.
3.6 CO2-TPD analysis of the basic sites on the surface of Ln2Ce2O7 catalysts
The effect of the surface basic sites and reactive oxygen species is considered as the
key role for OCM reaction. It is believed that not only the general basicity on the
catalyst surface, but also the number of surface basic sites can apparently affect the C2
selectivity and C2 yield [9,74,77-79]. While all the basic sites are beneficial for the
activation of methane molecules, the basic sites with intermediate strength are
favorable for improving the C2H4 selectivity [9,77]. Therefore, CO2-TPD is adopted
to probe the surface basicity of the catalysts, with the profiles shown in Fig. 9.
On the profile of individual CeO2, only a small CO2 desorption peak is observed at
o
82 C, indicating the presence of a very small amount of weak basic sites on its
24

surface. In contrast, with the incorporation of La, Pr, Sm and Y cations into CeO2,
multiple CO2 desorption peaks are detected on their profiles, testifying the formation
of much more basic sites with different strength. For clarification, the basic sites are
divided into three groups according to the CO2 desorption temperatures. In detail, it is
defined that the CO2 desorption peaks below 250 oC, between 250 ~ 500 oC and above
o
500 C are corresponding to weak, intermediate, strong basic sites. According to the
quantified results in Table 3, the total amount of basic sites follows the order of
La2Ce2O7 > Y2Ce2O7 > Pr2Ce2O7 > Sm2Ce2O7 > CeO2. However, the amount of the
intermediate basic sites, which are believed to be crucial for the formation of C2
products [9,77], follows the sequence of La2Ce2O7 > Y2Ce2O7 > Sm2Ce2O7 >
Pr2Ce2O7 > CeO2, in good agreement with the C2 product selectivity and yield on the
o
catalysts. For better comparison, the C2 product yield on the catalysts at 750 C is
plotted versus the amount of intermediate surface basic sites in Fig. 9(b). Apparently,
the C2 product yield increases with the increase of the amount of surface intermediate
basic sites, indicating it is critical for the selectivity of the desired product during the
OCM reaction, and could indeed be one of the determining factors influencing the
reaction performance of the Ln2Ce2O7 catalysts for OCM reaction.
25

Weak basic sites
(a)
Intermediate basic sites
Strong basic sites
168
82
254
196
291
429
583
La Ce O
2
2
7
7
155
78
276
Pr Ce O
2
2
83
83
82
189
412
294
552
Sm Ce O
7
2
2
169
416
287
560
Y Ce O
2
2
7
CeO
2
100
200
300
400
500
600
700
Temperature (oC)
20
(b)
La Ce O
7
2
2
15
10
5
Y Ce O
2
2
7
Sm Ce O
7
2
2
Pr Ce O
2
2
7
CeO
2
0
26
0
16
18
55
Intermediate basic sites amount (a.u.)
Fig. 9 CO2-TPD analysis of the Ln2Ce2O7 catalysts. (a) CO2-TPD profiles, (b) C2
yield at 750 oC versus intermediate basic sites amount.
26

Table 3 The quantified CO2-TPD results of the Ln2Ce2O7 catalysts.
Catalysts
CO2-TPD results
Weak basic
sites
Intermediate basic
Strong basic
sites
Total basic
sites
sites
amount
(a.u.)
55
amount
(a.u.)
40
amount
amount
(a.u.)
100
62
(a.u.)
La2Ce2O7
Pr2Ce2O7
Sm2Ce2O7
Y2Ce2O7
CeO2
5
9
37
16
25
18
7
50
32
26
14
0
72
6
0
6
3.7 XPS analysis of the oxygen properties on the surface of Ln2Ce2O7 catalysts
It is worth noting here that although the intermediate basic sites are critical for the
OCM reaction, it is not the only reason to decide the reaction performance of the
catalysts. Generally, for OCM reaction, suitable surface oxygen species is also
indispensible for the selectivity to the target products [74,80,81]. It is believed that a
complex relationship between surface basicity and surface oxygen species controls the
OCM performance on a catalyst [74,80-82].
O2-
O2-
(a) CeO
(b) La Ce O
2 2 7
2
O22-
CO32-
O22-
CO32-
O2-
536
534
532
530
528
526
536
534
532
530
528
526
Binding Energy (eV)
Binding Energy (eV)
27

(c)Pr Ce O
(d) Sm Ce O
2 2 7
O2-
2
2
7
O2-
O22-
O22-
CO32-
534
CO32-
536
532
530
528
526
536
534
532
530
528
526
Binding Energy (eV)
Binding Energy (eV)
(e) Y Ce O
7
2
2
O2-
O22-
CO32-
O2-
536
534
532
530
528
526
Binding Energy (eV)
Fig. 10 XPS analysis of O1s peaks for the Ln2Ce2O7 catalysts. (a) CeO2, (b) La2Ce2O7,
(c) Pr2Ce2O7, (d) Sm2Ce2O7, (e) Y2Ce2O7.
-
As formerly reported, surface electrophilic oxygen species such as O-, O2
2-
superoxide and O2 peroxides is beneficial for the C2 product formation [44,80-83].
Whereas, the surface O2- lattice oxygen species usually leads to the complete
oxidation of hydrocarbon products to form COx [81-83]. Therefore, the surface
oxygen properties of the catalysts were investigated by XPS. It is known that the
FWHM (full widths at half maximum) of a single XPS oxygen peak should be
between 1.8 and 2.0 eV [44]. Based on this rule, the O 1s peaks of all the catalysts
have been de-convoluted and quantified, as shown in Fig. 10. According to Fig. S5,
on the surface of all the catalysts, carbonate species has been formed, which displays
a characteristic peak at 289.6 eV [84,85]. Therefore, when analyzing the O1s signals
of the catalysts, the contribution from carbonate should also be considered and
28

omitted. By assuming a single peak having 80% Gaussian plus 20% Lorentzian peak
shape [44], four peaks corresponding to different oxygen species, which can be
-
2-
assigned to O2 (533.0 ~ 533.1 eV), CO32- (531.9 ~ 532.2 eV), O2 (530.9 ~ 531.1 eV)
and O2- (528.6 ~ 529.2 eV) [44,81-83,86], can be obtained.
Table 4. Curve-fitted and quantified XPS O1s peaks for the Ln2Ce2O7 catalysts.
2-
Catalysts
O1s B.E (eV)/ FWHM (eV) / relative amount (%)
(O2 +
O2 )/O2-
-
-
2-
2-
O2
CO3
O2
O2-
La2Ce2O7 533.0/2.0/5.7 532.0/1.8/14.1 531.0/1.7/39.3 528.7/2.0/40.9
1.10
0.30
0.65
0.61
0.17
Pr2Ce2O7
531.9/1.7/13.8 530.9/1.8/19.7 528.6/2.0/66.5
532.1/2.1/13.4 530.9/2.2/34.0 528.7/1.9/52.6
Sm2Ce2O7
Y2Ce2O7 529.9/2.0/6.0 532.0/1.7/13.2 531.1/1.9/28.7 528.7/1.8/52.1
CeO2 532.0/2.0/11.7 530.9/1.6/13.0 529.2/1.9/75.3
Apparently, besides the major O1s peak belonging to surface lattice oxygen (O2-) at
~ 529.0 eV, a shoulder peak attributing to other oxygen species and with varied
integrated areas are found for all the samples. For clarification, by excluding the
2-
-
contribution from carbonate, the ratio of the electrophilic oxygen species (O2 + O2 )
to the lattice oxygen O2- is quantified for each catalyst and listed in Table 4.
2-
-
Interestingly, the surface (O2 + O2 )/O2- ratio achieved on the catalysts follows the
order of La2Ce2O7 > Y2Ce2O7 ≈ Sm2Ce2O7 > Pr2Ce2O7 > CeO2, which are in parallel
to the C2 product selectivity and yield sequence of the catalysts.
2-
-
For clearer comparison, the C2 yield at 750 oC versus surface (O2 +O2 )/O2- ratio of
each catalyst is plotted in Fig. 11, which demonstrates evidently that a catalyst with a
larger amount of electrophilic oxygen species typically possesses a better reaction
29