Oxidative coupling of methane over Li/MgO catalysts prepared by sol-gel and impregnation methods

Article abstract of DOI:10.1080/24701556.2017.1357582

Oxidative coupling of methane (OCM) to C₂ ⁺ hydrocarbons was studied over a series of lithium-promoted magnesium oxide catalysts, prepared using different precursor salts for MgO, Li, and catalyst preparation methods (namely wet impregnation and sol-gel), under different reaction temperatures ranging from 675 to 825°C. The prepared catalysts were characterized by a variety of techniques, including X-ray diffraction, temperature programmed desorption of carbon dioxide, the Brunauer–Emmett–Teller, and scanning electron microscopy. Catalytic activity and selectivity toward desired products over the Li/MgO catalysts depended on the precursors used, catalyst preparation method, and nature of synthesized catalysts.

Full text of DOI:10.1080/24701556.2017.1357582

Inorganic and Nano-Metal Chemistry
ISSN: 2470-1556 (Print) 2470-1564 (Online) Journal homepage: http://www.tandfonline.com/loi/lsrt21
Oxidative Coupling of Methane over Li/MgO
Catalysts Prepared by Sol-Gel and Impregnation
Methods
Hanieh Habibpoor, Majid Taghizadeh & Fereshteh Raouf
To cite this article: Hanieh Habibpoor, Majid Taghizadeh & Fereshteh Raouf (2017): Oxidative
Coupling of Methane over Li/MgO Catalysts Prepared by Sol-Gel and Impregnation Methods,
Inorganic and Nano-Metal Chemistry, DOI: 10.1080/24701556.2017.1357582
To link to this article: http://dx.doi.org/10.1080/24701556.2017.1357582
Accepted author version posted online: 28
Jul 2017.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=lsrt21
Download by: [Australian Catholic University]
Date: 31 July 2017, At: 21:12

ACCEPTED MANUSCRIPT
Oxidative Coupling of Methane
Oxidative Coupling of Methane over Li/MgO Catalysts Prepared by Sol-Gel and
Impregnation Methods
a
a,*
b
HANIEH HABIBPOOR , MAJID TAGHIZADEH , FERESHTEH RAOUF
a
Chemical Engineering Department, Babol Noshirvani University of Technology, Babol, Iran
b
Chemical Engineering Department, University of Guilan, Rasht, Iran
*
Address correspondence to Majid Taghizadeh, Chemical Engineering Department, Babol
University of Technology, P.O. Box 484, Babol 4714871167, Iran. E-mail:
m_taghizadehfr@yahoo.com.
ABSTRACT
+
2
Oxidative coupling of methane (OCM) to C hydrocarbons was studied over a series of lithium
promoted magnesium oxide catalysts, prepared using different precursor salts for MgO, Li, and
catalyst preparation methods (namely, wet impregnation and sol-gel), under different reaction
temperatures ranging from 675 to 825°C. The prepared catalysts were characterized by a variety
of techniques, including X-ray diffraction (XRD), temperature programmed desorption of carbon
dioxide (CO -TPD), the Brunauer--Emmett--Teller (BET), and scanning electron microscopy
2
(SEM). Catalytic activity and selectivity toward desired products over the Li/MgO catalysts
depended on the precursors used, catalyst preparation method and nature of synthesized
catalysts.
Keywords
oxidative coupling of methane (OCM), Li/MgO catalyst, sol-gel, wet impregnation, lithium
loading
1
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Introduction
Natural gas as a raw material has become more important in the chemical industry, because of
the ever-increasing consumption of crude oil. C-H bond activation is a subject of immense
practical and fundamental interest and the conversion of methane into more valuable chemicals
and fuels is a challenge for the future.
Ethylene is conventionally produced in petrochemical industries by steam cracking of
different liquid feedstock especially naphtha. Due to the rising price of feedstock that is induced
by the global oil demand and the particular importance of ethane and ethylene, there has been
always a need to seek other alternatives for their production. Oxidative coupling of methane is
[1--5]
one of these methods which is significantly useful because it uses natural gas.
A number of metal oxides are considered to be effective catalysts for the OCM process and
[
6]
different active sites based on different oxygen species have been suggested. In summary, the
first group of catalysts contains reducible metal oxides, which are able to store and release
oxygen cyclically between multiple valence states. These materials are more efficient in terms of
their redox kinetics and are suitable for use with alternating feeds of methane and oxygen
(
known as the redox mode). Some examples are SnO --SnO and BiO--Bi where the surface
2
2
−
[7]
lattice oxygen (O ) was proposed to be the active site. The second group of catalysts contains
irreducible metal oxides including alkali/alkaline earth metal oxides, such as magnesium oxide
[
8]
[9--11]
doped with lithium oxide (Li/MgO) and most rare earth metal oxides.
[
12]
[8]
Lunsford and Ito et al. found that Li/MgO is a selective catalyst for the reaction. They


proposed a mechanism in which surface Li O species constitute the active sites at which the
reaction is initiated by the abstraction of a hydrogen from methane; ethane is then produced by
2
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
•
[13]
coupling two of the CH radicals.
The subsequent (oxidative) dehydrogenation of ethane
3
produces ethylene, while carbon monoxide and carbon dioxide are formed by the attack of
oxygen on the CH radicals. The challenges that limit the commercialization of the OCM process
3
+
are: (i) high temperature (973 to 1173K) to achieve high C2 yield, (ii) the active sites in the
+
2
coupling catalysts also activate the C-H bond in C , resulting in the formation of CO by
2
+
combustion, (iii) limitation on methane conversion (<45%) and C yield (<27%) imposed by the
2
explosion limit of oxygen concentrations in the feed, (iv) low concentrations of ethylene in the
product, making the separation of the product stream uneconomical, (v) low selectivity at higher
conversion, making the achievement of simultaneous good selectivity and conversion extremely
[
14]
difficult, and (vi) short lifetime for catalysts containing volatile promoters.
The OCM reaction over Li/MgO is the archetype of radical-mediated gas--solid catalytic
reactions. The so-called “Lunsford mechanism” assumes that active sites create methyl radicals
[
13,15,16]
that dimerize in the gas phase to form ethane.
[
17]
Ito et al. demonstrated a possible mechanism for the oxidative coupling of methane over


Li O centres on the catalyst and CH
radicals in the gas phase. Based on these observations as
3
[18]
well as the studies carried out by Olson et al., they proposed that the active centers are formed
+
2+
by substitution of a Li ion for a Mg ion in the MgO matrix, creating an oxygen vacancy.
2

Gaseous oxygen molecules react with this vacancy, resulting in
O
ions and positive holes. A


2
+
Li O centre is produced by a hole being trapped at
O
ions which are adjacent to Li ions.

Thus, it is suggested that the
O
ion is then capable of abstracting a hydrogen atom from the
methane, resulting in the formation of a CH radical and a reduced active site. Then ethane is
3
formed by the coupling of two CH radicals, in the gas phase near the surface. Subsequently
3
3
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ethylene produced from the ethane by oxidative dehydrogenation by surface
O
ions. Finally
total oxidation products, CO and CO proceeds through the oxidation of the CH radicals by a
2
3
2

surface
O
ion or a gas phase O molecule, this occurring parallel to the formation of C H .
2 2 6
Recently, nanocatalysts had attracted much attention. The higher activity and better selectivity
of nanocatalysts over traditional catalysts are attributed to their large specific surface area, high
percentage of surface atoms and special crystal structures. The method of preparation has a
significant influence on the structure, size distribution, and the morphology of the nanocatalysts,
[
19]
which in turn greatly affects the properties of the catalysts.
Although the catalyst shape, size, active sites, oxides surface, and morphology also have
[
20--23]
influence on catalytic properties of Li/MgO catalyst,
but, less works have been done on the
nanocatalysts and many questions should be answered about the effect of using nanocatalyst on
[
14]
the OCM reaction.
[
19]
For instance, He et al. synthesized La O /BaCO nanocatalysts from the coupling route of
2
3
3
homogeneous precipitation with micro emulsion under pulsed microwave heating. They
compared the performance of La O /BaCO nanocatalysts with that of La O /BaCO catalysts
2
3
3
2
3
3
prepared by conventional homogeneous precipitation. The authors found that by using
La O /BaCO nanocatalysts, the reaction could take place at a lower temperature. The
2
3
3
conversion of CH was higher, but the carbon deposition on the catalysts was more serious and
4
+
[19,24,25]
there was no distinct improvement on the selectivity of C .
2
In this work, a nanosized Li/MgO catalyst prepared by sol-gel method was compared with
catalyst made by conventional wet impregnation method for the OCM reaction. It has been
+
shown that the nanocatalyst is more active and selective to C2 products, than an equivalent
4
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
catalyst prepared by the impregnation method. Nanocatalyst activity in terms of methane
+
conversion, C2 selectivity, and yield was compared to ordinary prepared catalysts and some
improvements were observed. Although different Li loading on synthesized Li/MgO
nanocatalyst were investigated.
Experimental
Catalyst preparation
Catalyst samples were prepared using the sol-gel method. A solution of magnesium methoxide in
methanol (Aldrich, 6 to 10 wt %) was refluxed under constant stirring rate for 20 min. The
mixture of LiNO (Merck, 98%) and deionized water was added to the resultant gradually till
3
hydrolysis rate is controlled and the obtained gel aged at 60°C for 24 h, and was finally dried at
6
5°C for 12 h and calcined at 500°C for 1 h in air flow with a heating rate of 5°C/min.
Catalyst samples were prepared using conventional wet impregnation method, lithium chloride
LiCl) was used as Li precursor, and ethanol was used as liquid media. MgO powder (99%
(
Merck) was impregnated with the solution of LiCl (99% Merck) by immersion of MgO powder
in pure ethanol (99% Merck) followed by gradual addition of lithium chloride solution in
deionized water at 60°C. The suspension was stirred and followed by its evaporation to dryness
at 100°C. The resulting mixture was dried at 120°C for 16 h before being calcined in air at 850°C
for 4 h.
Catalytic activity test
Catalytic tests were carried out in a laboratory-scale continuous flow system with a fixed quartz
reactor (ID = 20 mm) operating at atmospheric pressure. The tubular reactor was filled with
catalyst (25 to 30 mesh) and it was activated in situ by heating at 550°C for 1 h under nitrogen
5
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
flow. The reactant feed gases used in this study were high purity methane (99.9%), oxygen
(99.99%) and nitrogen (99%). The inlet volumetric flow rate of each gas was controlled using an
individual volumetric flow controller and the total flow rate of the reactant gas mixture was 150
mL/min with a feed composition of CH :O :N = 4:2:4. The catalyst mesh and weight were
4
2
2
similar in all experiments (1.5 g with mesh 25--30). A Hewlett Packard (HP) 6890 series online
gas chromatograph equipped with a thermal conductivity detector (TCD) and a flame ionization
detector (FID) were used to analyze the effluent gases of the reactor.
Catalyst characterization
Siemens D5000 with Fe anode (FK 60-04) was used for XRD analysis of the fresh catalyst.
Wavelength of 1.93604 Å at 35 kV and 25 mA radiations in step mode between 5° and 80° with
step size of 0.02° was used. The crystalline phases in the catalyst were identified by comparing
diffraction lines of the samples with those of instruments in the literature. The specific surface
area of the catalysts was determined by the BET method using a CHEMBET-3000 with N2
adsorbent at liquid nitrogen temperature. The morphology of the catalysts was determined using
scanning electron microscopy (SEM) model VEGA2 Tscan instrument USA.
Results and Discussion
X-Ray diffraction
The comparison of 5%Li/MgO catalyst XRD patterns synthesized by conventional wet
impregnation method and nanosized 5%Li/MgO synthesized with sol-gel method is shown in
Figure 1. XRD analysis of catalysts showed that the catalysts were predominantly mixtures of
MgO and Li compounds. It could be observed from these patterns that catalysts possessed sharp
diffraction lines at 2θ values of 43 and 62° that corresponded with MgO. The weak diffraction
6
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
line at 2θ value of 37° represented the presence of Li CO . The observed Li CO compound
2
3
2
3
formation might presumably be due to peak results from decomposition of the LiCl precursor in
the catalysts. The change of phase transformation over XRD analysis is not clearly distinguished
and all the species are in stable phase and the peaks position remained unchanged. However, the
intensity of the peaks makes some differences between the catalysts prepared. Obviously, the
intensity of MgO especially at 2θ of 43 and 62° is higher as compared to the catalysts made
using impregnation method. It can be concluded that there are increment and reduction of species
intensity with preparation change. The average crystal size of 5%Li/MgO samples was estimated
by the broadened peak width using the Scherrer formula (Eq. (1)):
K
cos
L 
(1)

where L is the average crystal size (approximation), K is the crystal shape factor (typically ~ 0.9)
and β is the full width at half-maximum (FWHM) of a selected peak at 2θ diffraction angle. The
average crystal sizes of selected samples are summarized in Table 1.
The XRD spectra of spent catalyst indicated a loss of lithium carbonate. The loss of activity
of the Li/MgO catalysts is mainly due to the loss of lithium, which evaporates in the form of
lithium hydroxide. When the catalysts were operated at 725°C for 4 h, the lithium carbonate
peaks had a weak intensity. The other diffraction patterns were similar to the fresh samples,
which confirms the absence of any new phase in the used catalyst with slightly less intense.
Scanning electron microscopy (SEM)
Figure 2 shows the SEM images of both synthesized 5%Li/MgO fresh catalysts prepared by sol-
gel and impregnation methods. The results revealed that the general morphological aspect and
7
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
the size of the produced catalysts are very strongly influenced by the catalyst precursors and the
method of catalyst preparation. 5%Li/MgO catalyst prepared by impregnation method
demonstrated the microstructure made from small porous particles that stuck to each other with
smooth amorphous particles. The SEM image of 5%Li/MgO nanocatalyst prepared by sol-gel
method could be observed as shown in Figure 2 (b). As it can be seen, the nanocatalyst in such
scale showed smaller particle size against conventional 5%Li/MgO. The synthesis of the sol-gel
system led to the formation of particles clusters of spheroidal shape. The average particle size of
5
%Li/MgO (SG) was estimated to be 50--100 nm.
The catalysts used after treatments at 725°C (SG) and 750°C (WI) for 4 h in time on steam were
characterized by SEM (Figure 3). Few changes could be observed in the surface configuration of
the used catalysts after the OCM reaction. Under the operating conditions, there was little
sintering which caused the particles to stick together and the total porosity to be reduced.
Temperature programmed desorption of carbon dioxide (CO -TPD)
2
Using CO -TPD peak temperatures and the amount of desorbed CO , the strength and density of
2
2
surface basicity of the catalysts were measured. CO -TPD profiles of the catalysts are shown in
2
Figure 4, which revealed the base strength distribution in these catalysts. Carbon dioxide was
used as a probe molecule to measure the surface basicity and numbers of weak, medium and
strong sites. To facilitate the discussion, the basic sites were divided to weak, moderate and
strong sites, of which total basicity refers to the total amount of desorbed CO . According to this
2
division for base strength distribution, their difference can be roughly classified based on the
temperature at which desorption peaks was recorded in the three temperature range, weak
[
26--28]
(<200°C), medium (200--400°C) and strong (>400°C) basic sites.
For better difference
8
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
between strong and very strong basic sites in unusual classification, desorption peak revealing
temperature range greater than 600°C was assumed to belong to the very strong basic sites (so
[
28]
strong sites refer to the 400--600°C). The medium basic sites were associated with desorption
peak in 200--400°C and there was no CO desorption peaks in the region of medium
2
temperature. As shown in Figure 4, the desorption peaks in the temperature range from 150 to
2
00°C appeared in nanosized 5%Li/MgO sample, which showed that a site with lower basic
strength was present in this catalyst. At higher temperature ramp for conventional 5%Li/MgO,
the peak was demonstrated at 512°C. In the case of 5%Li/MgO (SG), two peaks at 600°C were
clearly visible and they referred to very strong basic sites. The CO -TPD results in Table 2 show
2
that both 5%Li/MgO samples exhibited significant desorption at temperatures above 500°C (very
strong sites). The basicity of conventional 5%Li/MgO catalyst was less than 1.3 mmol/g. In
contrast, 5%Li/MgO nanocatalyst showed desorption peaks at both low and high temperatures.
More interestingly, the total basicity of 5%Li/MgO nanocatalyst was around 3.3 mmol/g.
BET surface area
The catalysts were characterized with respect to finding the specific surface area by nitrogen
adsorption at 77K using an accelerated surface area and porosity apparatus. Prior to the analysis,
0
.5 g of the catalyst was degassed at 120°C for 3 h in order to desorb moisture and other weakly
adsorbed residues.
Figure 5 shows the N adsorption-desorption isotherms of two 5%Li/MgO samples. The
2
synthesized samples displayed type IV isotherm with a H hysteresis loop according to the
3
IUPAC classification. The hysteresis loop was due to the capillary condensation taking place in
the confined mesopore. More importantly, the nitrogen uptake of 5%Li/MgO (SG) increased
9
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
sharply at high relative pressure (P/P0> 0.7) compared to 5%Li/MgO (WI), suggesting its highly
porous structure. However, 5%Li/MgO (WI) exhibited a very low N adsorption quantity and
2
non-porous structure.
The BET surface areas of the 5%Li/MgO catalysts are shown in Table 3. The specific surface
2
area of the catalysts before and after the reaction showed a difference of 2.6 m /g for the
2
nanocatalyst and 1.5 m /g for the ordinary catalyst. It can be seen that the samples were
relatively sintered at operating temperatures (725°C and 750°C, respectively) but for 5%Li/MgO
2
nanocatalyst it can be said to have remained almost constant (24.6 to 21.9 m /g with 10%
decrease) for the sintering time 4 h at 725°C.
Catalytic performance of nanosized 5%Li/MgO compared to conventional 5%Li/MgO catalyst
The results of methane conversions in OCM reaction over ordinary and nanosized 5%Li/MgO in
different range of temperature are shown in Figure 6 (a). The conversion of methane over the
5
%Li/MgO nanocatalyst prepared by the sol-gel method (SG) was apparently higher than the
ordinary catalyst which was prepared using the wet impregnation method (WI) at all experiment
temperatures, especially at lower temperatures. In the former case, high methane conversion of
the OCM reaction could almost be achieved at temperature in the range of 675 to 800°C and the
dependency of catalyst activity toward temperature is low in this experiment range. The greatest
difference between WI and SG performances can be seen at 675°C and SG nanocatalyst shows
9
4% activity improvement in comparison to WI. The higher activity of the nanocatalyst could be
attributed to its smaller particle size, larger specific surface area, and stronger intensity of surface
basic sites. To find a sensible reason for this matter, specific surface area and active sites of both
10
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
5
%Li/MgO catalysts were investigated. The reported BET surface area and XRD crystalline
phase were in good agreement.
+
The difference in C selectivity of two catalysts (Figure 6 (b)) was more at low temperatures
2
while it became less at higher temperatures. The reactivity of a catalyst often depends very
sensitively on the microscopic structure of the catalyst, e.g. on particle size and morphology. The
presence of specific active centers or an active metal-support interface and the role of surface
+
species during the reaction are of great importance. Therefore, high difference between C2
selectivity of 5%Li/MgO nanocatalyst and 5%Li/MgO ordinary catalyst at low temperatures can
be explained by such events. Nanocatalysts with a smaller size have a higher activity as a
consequence of their size (Figure 6 (a)). This result may be reasoned as follows, while the
average size was estimated under the hypothesis that the nanocatalyst had spherical morphology
and smooth surface, the nanocatalyst was not strictly spherical and had many defects on its
[
29--31]
surface.
The surface defects, such as crush, distortion, entanglement, and collapse, played a
[
32,33]
critical role in the activity of catalyst.
The average size of nanocatalyst could not describe
such defect in satisfactory details; therefore, it is difficult to find a separate relationship between
the average size of nanocatalyst and the conversion of methane. Since the variation of the
catalyst particle is narrow, it is necessary to carry out more experiments to verify the above
deduction.
+
For selectivity towards the C hydrocarbon and process yield over 5%Li/MgO, nanocatalyst
2
had high similarity with the methane conversion trend and increased slightly with temperature
until it reached the optimum temperature (725°C here), then decreased with mild slope.
11
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
C H /C H ratio versus temperature on WI and SG catalysts is shown in Figure 6 (d). In this
2
4
2
6
case, the best ratio is seen in the SG catalyst at 725°C and there are big visible differences
between this point and the nearest temperature.
Throughout the temperature range over SG nanocatalyst, C H was the main C compound,
2
4
2
but in the case of WI ordinary catalyst, C H was the main C compound up to 725°C. After this
2
6
2
point, C H was also produced in considerable amounts. The former is considered to be a more
2
6
valuable chemical than the latter; thus, it is desirable to convert C H into C H .
2
6
2
4
In the very high temperature region, CO and CO , increased steeply, while above 750°C the
2
increase in conversion was mainly due to CO and CO formation. Ratios of C H to C H in the
2
2
4
2
6
products were approximately 1.4 in the high temperature region over both catalysts.
Catalytic activities of the OCM reaction over two different Li/MgO catalysts are summarized in
Table 3. Among the different Li/MgO catalysts investigated for OCM, some are relatively
progressive in activity and selectivity which are listed in Table 4.
Catalytic performance of Li/MgO nanocatalyst
+
The conversion of CH and selectivity of C2 hydrocarbons over Li/MgO nanocatalyst with
4
different Li loadings as a function of operation temperature at atmospheric pressure are shown in
Figure 7, while the other operation parameters were similar. With nanocatalyst, the surface and
active area is increased and rate of surface reactions becomes more important as compared to the
gas phase reaction.
The selectivity, conversion and yield increased by increasing the temperature from 650 to
7
25°C simultaneously. As shown in Figure 7(b), selectivity toward desired products is almost
similar without an impressive difference in the product selectivity in three different loading of Li
12
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
over synthesized Li/MgO nanocatalyst. These catalysts were selective at 700 to 750°C and the
optimum operation temperature in the mentioned conditions was 750°C (for 3%Li/MgO and
7
%Li/MgO) and 725°C for 5%Li/MgO with considering methane conversion on total desired
product yield. Further increases in temperature caused more CO formation that had a visible
x
impact on the total selectivity.
The addition of Li into MgO increases the basicity of the catalyst, thereby producing good
OCM activity. This is the major reason for the large use of alkali metal compounds additives,
because they result in loss of surface area caused by sintering. It was also noted that in oxidation
catalyst, too large surface area favors consecutive reactions, thereby leading to deep oxidation.
Therefore, during the OCM reaction over the nanocatalyst, low temperatures favor the
interaction between carrier and active component to form more active sites and also new sites are
formed.
+
Hydrocarbons and the ratio of C H /C H were enhanced with the content of Li in the
2
4
2
6
nanocatalyst from 3 to 5% wt, which might be due to the fact that more oxygen vacancies on the
+
catalyst surface were formed when the Li content was increased, the oxygen vacancies reacted
2

with dioxygen molecules at high temperature and produced
O
ions and holes.
The yield of C hydrocarbons and selectivity for ethylene first increased with temperature till
2
the temperature increased to 750°C and then decreased when the temperature exceeded 750°C.
A careful carbon balance was performed for each test to ensure that the analytical results
would not be influenced by a potential absorption of carbon compounds formed by thermal
cracking. Carbon input (methane) and carbon output (such as methane, higher hydrocarbons and
CO ) were measured and analyzed. Table 5 shows the carbon balance for three runs, two at high
x
13
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
+
selectivity and the other at relatively poor selectivity. This corresponds exactly to the C2
selectivity obtained with catalyst in the reactor and indicates that CO was formed thermally and
2
none was absorbed on the catalyst. Excellent balances were obtained in this optimum
temperature range. Some experiments carried out at temperatures above 800°C showed large
differences in carbon balance due to the coke formation.
The activities of both sol-gel and wet impregnated 5%Li/MgO catalysts were investigated
during a typical experiment which is shown in Figure 8. In order to determine catalyst
deactivation in the OCM reaction, the catalytic performances for fresh catalysts were evaluated
+
in separate experiments. The yield of C was measured for 6 hours at best temperature for both
2
sol-gel (725°C) and wet impregnated (750°C) modified catalysts. The yield over the SG catalyst
dropped from about 19 to 18% during the first 3 h time on stream, but after this period very little
change was observed over the next 6 h. This initial deactivation was mainly due to an increase in
the rate of CO formation, which led to a decrease in C selectivity. At 725°C, C H was the
x
2
2
4
main C compound, but C H was also produced in considerable amounts.
2
2
6
The stability of the catalytic activity and selectivity in the OCM process in the presence of
Li-MgO catalyst is attributed mostly to its high CO content which stabilizes the catalyst against
2
sintering and loss of Li during the process. This is consistent with the earlier observation
indicating an increase in the stability of Li-MgO due to the addition of CO at low concentration
2
[
35--37]
in the reaction mixture.
Conclusions
Catalyst preparation methods and precursors had a strong influence on the specific surface area
and affected catalytic activity/selectivity in OCM reaction over Li-promoted MgO catalysts.
14
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Higher basicity was observed for the catalyst prepared using sol-gel method than that
prepared by wet impregnation method. There is a direct relationship between surface basicity and
catalytic activity/selectivity. OCM showed that basicity alone is not responsible for obtaining
high selectivity to C hydrocarbons.
2
The prepared nanocatalyst exhibited different catalytic performance in OCM. This catalyst
had about 6 times greater specific surface area than those of Li/MgO ordinary catalysts and the
+
result exhibited high C2 selectivity, which shows the importance of the surface area and
+
precursors used in the catalyst preparation to obtain high C selectivity in OCM.
2
The amount of Li loading had effect on the Li/MgO properties and its OCM activity, while
the loading does not significantly alter the surface area of catalysts. Then, 3 to 5 wt% doped
Li/MgO catalyst showed an improved methane conversion for the OCM reaction, particularly at
lower temperatures. No significant improvement was observed in C selectivity with the Li
2
increment and high Li loading (>5 wt%) resulted in lower methane conversion. High reaction
temperatures (>750°C) caused a decrease in process yield.
Acknowledgment
This work was financially supported by the Iranian Nanotechnology Initiative Council.
15
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
References
1
2
3
. Balint, I.; Miyazaki, A.; Gingasu, D.; Papa, F. Reac. Kinet. Mech. Cat. 2012, 105, 5--11.
. Raouf, F.; Taghizadeh, M.; Yousefi, M. Int. J. Chem. Reac. Eng. 2014, 12, 1--9.
. Papa, F.; Luminita, P.; Osiceanu, P.; Birjega, R.; Akane, M.; Balint, I. J. Mol. Catal. A: Chem.
2
011, 346, 46--54.
4
5
6
7
8
9
1
1
1
1
1
1
. Raouf, F.; Taghizadeh, M.; Yousefi, M. Reac. Kinet. Mech. Cat. 2014, 112, 227--240.
. Raouf, F.; Taghizadeh, M.; Yousefi, M. Reac. Kinet. Mech. Cat. 2013, 110, 373--385.
. Zhang, Zh.; Verykios, X.E.; Baerns, M. Catal. Rev.: Sci. Eng. 1994, 36, 507--556.
. Keller, G.E.; Bhasin, M.M. J. Catal. 1982, 73, 9--19.
. Ito, T.; Wang, J.; Lin, Ch.H.; Lunsford, J.H. J. Am. Chem. Soc. 1985, 107, 5062--5068.
. Tang, L.; Yamaguchi, D.; Wong. L.; Burke, N.; Chiang, K. Catal. Today 2011, 178, 172--180.
0. Kuś, S.; Otremba, M.; Taniewski, M. Fuel 2003, 82, 1331--1338.
1. Tong, Y.; Rosynek, M.P.; Lunsford, J.H. J. Phys. Chem. 1989, 93, 2896--2898.
2. Lunsford, J.H. Angew. Chem. Int. Ed. 1995, 34, 970--980.
3. Lunsford, J.H. Catal. Today 2000, 63, 165--174.
4. Farsi, A.; Moradi, A.; Ghader, S.; Shadravan, V. Chin. J. Chem. Phys. 2011, 24, 70--76.
5. Zavyalov, U.; Geske, M.; Horn, R.; Weinberg, G.; Frandsen. W.; Schuster, M.; Schlogl, R.
ChemCatChem 2011, 3, 949--959.
1
6. Simon, Y.; Baronnet, F.; Côme, G.M.; Marquaire, P.M. Stud. Surf. Sci. Catal. 2004, 147,
5
71--576.
1
1
7. Ito, T.; Wang, J.; Lin, C.H.; Lunsford, J.H. J. Am. Chem. Soc. 1985, 107, 5062--5068.
8. Olson, D.N.; Orera, V.M.; Chen, Y.; Abraham, M.M. Phys. Rev. B 1980, 21, 1258--1263.
16
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
1
2
9. He, Y.; Yang, B.; Cheng, G.; Pan, H. Powder Technol. 2003, 134, 52--57.
0. Myrach, Ph.; Nilius, N.; Levchenko, S.V.; Gonchar, A.; Risse, T.; Dinse, K.P.; Boatner,
L.A.; Frandsen, W.; Horn, R.; Freund, H.J.; Schlçgl, R.; Scheffler, M. ChemCatChem 2010,
2
, 854--862.
2
2
1. Olivier, L.; Haag, S.; Pennemann H.; Hofmann, C.; Mirodatos, C.; van Veen, A.C. Catal.
Today 2008, 137, 80--89.
2. Choudhary, V.R.; Mulla, S.A.R.; Uphade, B.S. J. Chem. Technol. Biotechnol. 1998, 72, 99--
1
04.
2
2
2
3. Tung, W.Y.; Lobban, L.L. Ind. Eng. Chem. Res. 1992, 31, 1621--1625.
4. Gao, Zh.; Shi, Y. J. Nat. Gas Chem. 2010, 19, 173--178.
5. Al Mayman, S.; Al Zeghayer, Y.; AlMasry, W.; AlNashef, I.; Ramay, S.; Jbar, A. Int. J.
Phys. Sci. 2012, 7, 6270--6279.
2
2
2
6. Yaghobi, N.; Ghoreishy, M.H.R.; Eslamimanesh, V. Asia-Pac. J Chem. Eng. 2008, 3, 202--
2
10.
7. Sohrabi, M.; Dabir, B.; Eskandari, A.; Golpasha, R.D. J. Chem. Technol. Biotechnol. 1996,
7, 15--20.
6
8. Sadeghzadeh Ahari, J.; Ahmadi, R.; Mikami, H.; Inazu, K.; Zarrinpashne, S.; Suzuki, Sh.;
Aika, K.I. Catal. Today 2009, 145, 45--54.
2
3
3
3
9. Tang, L.; Dai, M.; Ji, J. Int. J. Nonlin. Sci. 2016, 22, 61--64.
0. Huang, P.; Zhang, Y.; Zhang, J.; Zhu, Y.; Sun, Y. Nanoscale 2013, 5, 10844--10848.
1. He, Y.; Yang, B.; Pan, H.; Li, G. J. Nat. Gas Chem. 2004, 13, 167--171.
2. Lunsford, J.H. Angew. Chem. Int. Ed. 1995, 34, 970--980.
17
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
3
3
3
3
3
3. Korf, S.J.; Roos, J.A.; Veltman, L.J.; van Ommen, J.G.; Ross, J.R.H. Appl. Catal. 1989, 56,
19--135.
4. Bistolfi, M.; Fornasari, G.; Molinari, M.; Palmery, S.; Dente, M.; Ranzi, E. Chem. Eng. Sci.
992, 47, 2647--2652.
5. Choudhary, V.R.; Choudhary, S.T.; Pandit, M.Y. J. Chem. Soc., Chem. Commun. 1991, 17,
158--1159.
6. Korf, S.J.; Roos, J.A.; de Bruijn, N.A.; van Ommen, J.G.; Ross, J.R.H. Appl. Catal. 1990, 58,
31--146.
1
1
1
1
7. Korf, S.J.; Roos, J.A.; de Bruijn, N.A.; van Ommen, J.G.; Ross, J.R.H. J. Chem. Soc., Chem.
Commun. 1987, 19, 1421--1512.
18
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 1. Crystallite size of the catalysts
Catalyst
Crystallite size (nm)
MgO
32
Li CO3
2
5
5
%Li/MgO (WI)
% Li/MgO (SG)
1.8
1.7
19
19
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 2. Temperature programmed data for catalyst sample
Sample
Temperature (°C)
512
Amount of chemisorbed (mmol/g)
Total amount of chemisorbed (mmol/g)
1.26
5
5
%Li/MgO (WI)
%Li/MgO (SG)
1.26
0.97
2.35
1
6
59
00
3.32
20
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 3. Activity and selectivity of the catalysts for oxidative coupling of methane
Specific surface areas
Products distribution at the best temperature
2
BET (m /g)
Best temperature
(°C)
Catalyst
a
Used
C H4
2
SC2

(%) YC2 (%)

Fresh
X (%)
CH
4
C H6
2
5
5
%Li/MgO (SG) 24.6
%Li/MgO (WI) 4.21
21.9
2.7
725
750
34.21
27.92
55.15
50.69
18.87
14.15
1.84
1.56
Reaction conditions: atmospheric pressure, flow rate 150 mL/min, CH /O = 2, 1.5 g catalyst.
4
2
a
After 4 h time on stream at best temperature.
21
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 4. Reports of Li/MgO catalysts for OCM prepared by wet impregnation and sol-gel
methods
Method
of Li
BET
Components
CH
4
:O
2
:inert X (%) S (%) YC2
(%) Reference
[34]
CH


2
4
C
2
preparation
(%)
(m /g)
Mixing of aqueous
salt
LiCl + MgO
Li/MgO
14
--
4:3:1
22.5
63.4
14.26
Wet impregnation
Wet impregnation
Wet impregnation
3.5
1.3
5
2
4:7.5:1
--
23
54
12.4
16
[26]
[27]
[28]
Mg(OH)
2
+ LiOH
--
--
80
3
Mg(NO ) + LiNO
3
2.9
4:1:3
19.1
55.5
10.6
This
work
This
work
MgO + LiCl
Wet impregnation
Sol-gel
5
5
4.21
24.6
4:2:4
4:2:4
27.92
34.21
50.69
55.15
14.15
18.87
Mg(OCH
3
)
2
+
LiNO
3
22
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 5. Carbon balance for three runs at different conversion and selectivity values
+
Catalyst
Temperature (°C)
Conversion (%)
CO selectivity (%)
C2 selectivity (%)
x
5
5
7
%Li/MgO (WI)
%Li/MgO (SG)
%Li/MgO (SG)
750
725
675
27.9
34.2
26.3
49.3
44.7
54.0
50.7
55.2
45.9
23
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
ꢀ

MgO
Li CO3
2
ꢀ
ꢀ
(a)

(b)
(c)
(d)
1
0
20
30
40
50
60
70
2
theta (degree)
Fig. 1. XRD patterns of 5%Li/MgO (SG) fresh (a) and used (b), 5%Li/MgO (WI) fresh (c) and
used (d) catalysts.
24
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
(a)
(b)
Fig. 2. SEM images of conventional 5%Li/MgO (WI) fresh catalyst (a) and 5%Li/MgO (SG)
fresh nanocatalyst (b).
25
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
(a) (b)
Fig. 3. SEM images of 5%Li/MgO (WI) (a) and 5%Li/MgO (SG) (b) after being used in
reaction.
26
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
5
5
% Li/MgO (SG)
% Li/MgO (WI)
1
00
200
300
400
500
600
700
800
900
Temperature ( C)
Fig. 4. CO TPD spectra of fresh catalysts.
2
27
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
5
5
%Li/MgO (SG)
%Li/MgO (WI)
0
0.5
1
Relative preasure (P/P0 )
Fig. 5. N adsorption-desorption isotherms of two 5%Li/MgO catalysts.
2
28
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
5
4
3
2
1
0
0
0
0
0
0
60
(
a)
(b)
5
4
0
0
SG
WI
SG
WI
30
6
60
680
700
720
740
760
780
800
820
660
680
700
720
740
760
780
800
820
Temperature ( C)
Temperature ( C)
2
2
1
1
5
0
5
0
5
0
2.5
2
(
c)
(
d)
1.5
1
0.5
0
SG
WI
SG
WI
6
60
680
700
720
740
760
780
800
820
660
680
700
720
740
760
780
800
820
Temperature ( C)
Temperature ( C)
+
2
Fig. 6. Change in (a) amount of CH converted, (b) selectivity toward C compounds, (c)
4
process yield, and (d) C H /C H ratio, with respect to an increase in temperature when a
2
4
2
6
reactant mixture containing 20% O was fed over 1.5 g of 5% Li/MgO ordinary (WI) and
2
nanocatalyst (SG), with the flow rate of 150 mL/s.
29
ACCEPTED MANUSCRIPT