Catalytic synthesis of glycerol carbonate from biomass-based glycerol and dimethyl carbonate over Li-La2O3 catalysts

Article abstract of DOI:10.1016/j.apcata.2018.07.032

The trans-esterification of glycerol and dimethyl carbonate (DMC) for the synthesis of glycerol carbonate was investigated over a series of Li-La₂O₃ catalysts with different Li contents under mild reaction conditions. Li-La₂O₃ catalysts were prepared by simple precipitation and impregnation methods, and characterized using ICP, XRD, BET, XPS, CO₂-TPD and TEM techniques. It was found that the basicity of La₂O₃ was obviously changed and the interaction between Li and La₂O₃ occurred after doping Li component, and the catalytic activity of Li-La₂O₃ was highly relevant to the strength and content of basic sites on the catalysts. Li doping influenced the crystal phase growth of La₂O₃, and also changed the exposed crystal planes of the Li-La₂O₃ catalysts. The highest catalytic performance was obtained over 3.50Li-La₂O₃ catalyst calcined at 600 °C for 5 h. The glycerol conversion of 94.4% and glycerol carbonate selectivity of 92.1% were obtained over this catalyst under the proper reaction conditions of 85 °C, 3 h, and the molar ratio of glycerol and DMC being 1/3 with using 0.1 g 3.50Li-La₂O₃ catalyst.

Full text of DOI:10.1016/j.apcata.2018.07.032

Accepted Manuscript
Title: Catalytic Synthesis of Glycerol Carbonate from
Biomass-based Glycerol and Dimethyl Carbonate over
Li-La2O3 Catalysts
Authors: Yajin Li, Jiaxiong Liu, Dehua He
PII:
S0926-860X(18)30371-5
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2018.07.032
APCATA 16758
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
25-4-2018
5-7-2018
24-7-2018
Please cite this article as: Li Y, Liu J, He D, Catalytic Synthesis of Glycerol Carbonate
from Biomass-based Glycerol and Dimethyl Carbonate over Li-La2O3 Catalysts,
Applied Catalysis A, General (2018), https://doi.org/10.1016/j.apcata.2018.07.032
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.

Catalytic Synthesis of Glycerol Carbonate from Biomass-based
Glycerol and Dimethyl Carbonate over Li-La2O3 Catalysts
Yajin Li, Jiaxiong Liu, Dehua He*
Innovative Catalysis Program, Key Lab of Organic Optoelectronics & Molecular
Engineering of Ministry of Education, Department of Chemistry, Tsinghua University,
Beijing 100084, China
*Corresponding author:
E-mail: hedeh@mail.tsinghua.edu.cn ,
Full postal address:
Department of Chemistry,
Tsinghua University,
Qinghua Yuan 1, Haidian
Beijing 100084, China
Tel/Fax: +86-010-62773346
1

Graphical Abstract
Highlights:
 xLi-La2O3 catalysts were prepared and employed in the trans-esterification of
glycerol and DMC to glycerol carbonate.
 Li dopping into La2O3 increased the catalytic performance of Li-La2O3 for the
yield and selectivity of glycerol carbonate.
 Doped Li could enter the lattice of La2O3, and caused the lattice expansion of (002)
crystal plane of La2O3.
 The strong basic property of Li-La2O3 and the interaction between Li and La2O3
played an important role on the catalytic performance.
Abstract
The trans-esterification of glycerol and dimethyl carbonate (DMC) for the
synthesis of glycerol carbonate was investigated over a series of Li-La2O3 catalysts
with different Li contents under mild reaction conditions. Li-La2O3 catalysts were
prepared by simple precipitation and impregnation methods, and characterized using
ICP, XRD, BET, XPS, CO2-TPD and TEM techniques. It was found that the basicity
2

of La2O3 was obviously changed and the interaction between Li and La2O3 occurred
after doping Li component, and the catalytic activity of Li-La2O3 was highly relevant
to the strength and content of basic sites on the catalysts. Li doping influenced the
crystal phase growth of La2O3, and also changed the exposed crystal planes of the
Li-La2O3 catalysts. The highest catalytic performance was obtained over
3.50Li-La2O3 catalyst calcined at 600 °C for 5 h. The glycerol conversion of 94.4%
and glycerol carbonate selectivity of 92.1% were obtained over this catalyst under the
proper reaction conditions of 85 °C, 3 h, and the molar ratio of glycerol and DMC
being 1/3 with using 0.1 g 3.50Li-La2O3 catalyst.
Keywords: Trans-esterification, Glycerol carbonate, Base site, Li-La2O3, Dimethyl
carbonate
1. Introduction
The utilization of renewable resource, such as biodiesel from biomass, has
attracted many attentions. As a byproduct of biodiesel production, glycerol becomes
more and more surplus along with the production and application of biodiesel. Owing
to the requirement of sustainable development, the conversion of glycerol to
high-value added chemicals becomes extremely urgent. The upgrading of glycerol has
various pathways, such as selective oxidation, dehydration, acetylation,
trans-esterification, steam reforming, hydrogenolysis and etherification [1]. Glycerol
carbonate (GLC), which is one of the glycerol derivatives, has several functional
groups, such as the dioxolane ring with the three carbon atoms and the hydroxyl
group [2], which allow glycerol carbonate to be used as chemical intermediates to
produce lots of chemical products. Several routes have been employed for upgrading
glycerol into GLC [3]. The carbonation routes of synthesizing GLC from glycerol
with carbon monoxide or carbon dioxide had been investigated. However, the carbon
monoxide had been deprecated due to its high toxicity [4]. Several approaches have
been reported by using metal oxides [5], such as ZnO, SnO2, and CeO2, as the
catalysts in the direct reaction of glycerol and CO2 to GLC, while the conversion of
3

glycerol and the yield of GLC were still very poor because of the thermodynamic
limitation. The transcarbonation routes for using phosgene, urea and dialkyl carbonate
also had some applications. Phosgene had been abandoned for the same reason as
carbon monoxide, while another method by using urea and glycerol to synthesize
GLC could obtain a favorable GLC yield [6], but the byproduct NH3 became the
limitation for this reaction, which required the reaction system being well designed to
move NH3 for shifting the equilibrium of the reaction. On the other hand, the
trans-esterification of glycerol with dialkyl carbonate, such as dimethyl carbonate
(DMC), which could be catalyzed by both acidic and basic catalysts, is an ideal route
for the synthesis of GLC. In comparison with acidic catalysts, the basic catalysts
could give higher GLC yield and glycerol conversion [7].
Several basic catalysts, such as KOH, NaOH and K2CO3 etc, had been
investigated for the trans-esterification of glycerol and DMC to GLC [7, 8]. However,
using KOH, NaOH and K2CO3 etc had the problem of separating products and the
catalysts. Some enzymes, like Novozym 435, could also catalyze the
trans-esterification of glycerol and DMC, but this system suffered many drawbacks,
such as high cost reactants, poor catalytic activity and long reaction time [9].
Meanwhile, some solid catalysts had also been investigated for the synthesis of GLC
from glycerol and DMC. Ochoa-Gómez et al [7] investigated the basic properties of
the catalysts on the catalytic activities, and found that the activities of some alkali
earth metal oxides for the trans-esterification of glycerol and DMC were in the order
of CaO > Ca(OH)2 > MgO > CaCO3. However, the exposure of CaO catalyst in air
during the reaction runs caused the deactivation and agglomeration of CaO, and the
recycle test results of the CaO revealed a quick decrease in both glycerol conversion
and GLC yield. The Ca-Al-O hydrocalumite was demonstrated having 93% of
glycerol conversion under mild condition [10], and the proportion of strong basic sites
in the Ca-Al-O catalysts could influence the catalytic performance. The catalyst with
Ca12Al14O33 phase and CaO phase was prepared by calcining the hydrocalumite
precursor, and Ca12Al14O33 phase was stable in recycle tests of the trans-esterification,
while CaO phase was easy loss in reuse experiments, which led to the deactivation of
4

Ca-Al-O catalysts [10]. Simanjuntak et al [11] investigated Mg-La catalysts for the
trans-esterification of glycerol and DMC, and they found that the suitable molar ratio
of Mg and La could deeply influence the basic site concentration and catalytic activity
of the catalysts. Singh et al [12] used coprecipitation method to prepare ZnO/La2O3
catalysts with different Zn/La molar ratios, and investigated their catalytic activity for
the synthesis of glycerol carbonate. However, owing to the adsorption of the
carbonaceous by-products on the catalytic basic sites, the activity of ZnO/La2O3
decreased during recycle reaction runs. The investigation of Mg/Al/Zr mixed oxides
demonstrated that the catalyst activities depended on the pretreatment temperatures,
catalyst structures and basic sites on the catalysts, and the suitable molar ratio of
Mg/Al/Zr for the maximum activity was 3:1:1 with calcinating at 650 °C [13].
According to these investigation results of using various catalysts, the basic
property of catalysts was an important factor for trans-esterification of glycerol and
DMC to GLC. Owing to this reason, alkali metals were often used to modify the
catalysts and improve their activities. Hu et al [14] reported the KNO3/CaO catalytic
system for the trans-esterification of glycerol and DMC, and found that the basicity of
KNO3/CaO catalyst depended on the loading amounts of KNO3 and the calcination
temperatures. The tests of recycle use also revealed 94.95% of glycerol conversion
after five runs, which was much higher than that of pure CaO. Bai et al [15] employed
the NaOH/γ-Al2O3 as the catalyst for the synthesis of glycerol carbonate, and obtained
high conversion of glycerol under mild reaction conditions. They also found that the
Na doped catalysts showed much better catalytic performance than some
homogeneous catalysts, such as K2CO3. Also, the Li/ZnO catalytic system was
investigated by Song et al [16]. They reported that the basic properties of catalysts
were greatly influenced by the Li loading amounts, and the interaction between Li+
and Zn2+ led to the transition of basic sites on the catalyst surface. In general, doping
alkali metal could not only improve the basic properties of catalysts, but also change
the structures of supports [17]. Compared to other alkali based catalysts, Li/ZnO
catalyst presented a distinctive performance. The activities of the Li doped catalysts
were highly relevant to the natures of the supports [18]. Therefore, it is important to
5

investigate other metal oxide supports and explore the interaction between Li and the
carrier metal oxides.
Recently, we explored some kinds of metal oxides, as the supports of Li, for
preparing Li-MxOy catalysts, and found that Li-La2O3 catalysts behaved the best
catalytic performance among the tested Li-MxOy catalysts. In this paper, the Li doped
La2O3 catalysts, here marked as Li-La2O3, were focused and investigated for their
catalytic performance in the trans-esterification of glycerol and dimethyl carbonate to
glycerol carbonate. After doping Li, the Li-La2O3 catalysts presented a favorable
glycerol conversion, GLC selectivity and GLC yield under mild reaction conditions.
The properties of Li-La2O3 catalysts were identified with Inductively coupled
plasma-atomic emission spectrometry (ICP-AES), X-ray diffractometry (XRD), Low
temperature nitrogen adsorption/desorption, X-ray photoelectron spectroscopies
(XPS), Temperature-programmed desorption of CO2 (CO2-TPD) and Transmission
electron microscopy (TEM). The interaction between lithium and La2O3 support was
also studied and discussed.
2. Experimental section
2.1. General
Lanthanum nitrate (La(NO3)3·6H2O) was purchased from Aladdin Industrial
Corporation. Ammonium bicarbonate (AR) and Lithium nitrate (LiNO3) were
purchased from Sinopharm Chemical Reagent Beijing Company. Dimethyl carbonate
(99.5%) was obtained from J&K Scientific. Glycerol (HPLC grade) was supplied by
Alfa Aesar Company and glycerol carbonate (90%) was afforded by TCI Chemicals.
All the reagents were used as received.
2.2. Catalyst preparation
La2O3 support was prepared by a simple precipitation method [19]. In a modified
procedure, 0.12 mol ammonium bicarbonate was dissolved in 160 mL deionized water
and the obtained solution was dropped into an equal volume of 0.25 mol/L lanthanum
nitrate (La(NO3)3·6H2O) aqueous solution under continuous stirring at room
temperature. Then, the mixture was continuously stirred for 6 h, and then separated by
centrifugation. Finally, the obtained solid paste was dried at 110 °C for 24 h, and
6

calcined at 800 °C for 6 h in air.
A series of Li doped catalysts (Li-La2O3) were prepared by impregnation and
solid-phase reaction methods, and the procedures were as follows. A certain amount
LiNO3 was dissolved in deionized water and then a given mass of La2O3 powder was
added into the LiNO3 aqueous solution. The resulted mixture was then stirred at room
temperature for 12 h. After impregnation step, the slurry was evaporated by a rotary
evaporator to remove water and then the obtained solid was dried at 110 °C for 24 h,
followed by the solid-phase reaction (calcination) at appropriate temperatures
(400 °C-700 °C) in air. The obtained catalysts were labeled as xLi-La2O3-T, where x
was the mass percentage of Li2O in Li-La2O3 catalysts, and T was the calcination
temperature. The samples without “T” were defaulted to 600 °C calcination. The
actual Li2O mass percentages in a series of Li-La2O3 samples were measured by
ICP-AES, and the obtained catalysts were 0.74%Li2O/Li-La2O3 (denoted as
0.74Li-La2O3), 1.58%Li2O/Li-La2O3 (denoted as 1.58Li-La2O3), 3.50%Li2O/Li-La2O3
(denoted as 3.50Li-La2O3) and 7.32%Li2O/Li-La2O3 (denoted as 7.32Li-La2O3).
Several other oxides were also used as the supports to prepare Li doping catalysts, and
the obtained catalysts were 3.51%Li2O/Li-ZnO (denoted as 3.51Li-ZnO),
3.51%Li2O/Li-Al2O3 (denoted as 3.51Li-Al2O3) and 3.51%Li2O/Li-SBA-15 (denoted
as 3.51Li-SBA-15), in which the actual Li2O mass percentages were also measured by
ICP-AES.
2.3. Catalyst characterization
The actual Li loading amounts on xLi-La2O3 catalysts were analyzed by
Inductively coupled plasma-atomic emission spectrometry (ICP-AES) operated on a
VARIN VISTA-MPX instrument. The samples were dissolved in concentrated HCl
solution and diluted to a proper concentration before the measurements.
The crystal phases of the xLi-La2O3 and 3.50Li-La2O3-T (T=400, 500, 600 and
700 °C) were characterized by powder X-ray diffraction (XRD), and the patterns were
recorded on a Rigaku D/Max-2500 diffractometer with Cu Kα radiation (λ=1.5406 Å)
and a nickel filter (voltage 40 kV, current 200 mA). The scanning range (2θ) was from
10° to 90°, with a slit of 1° at a scanning rate of 10° min-1. The average crystallite
7

sizes of xLi-La2O3 and 3.50Li-La2O3-T samples were also calculated by the line
broadening method using Scherrer equation. (d=kλ/βcosθ, k: Scherrer constant taken
as 0.89; λ: the wave length of X-ray; β: the width of the peak at half-maximum height;
θ: diffraction angle)
The specific surface areas of xLi-La2O3 samples were measured by BET method,
based on low temperature nitrogen adsorption/desorption isotherms, using a
Micromeritics Tristal II 3020M analyzer. The samples were first degassed at 200 °C in
vacuum for 1 h, and then measured their isotherms. The BET surface areas were
calculated by Brunauer-Emmett-Teller equation using the adsorption branch.
The binding energies of the elements in Li-La2O3 were measured by X-ray
photoelectron spectroscopy (XPS), which was performed on a PHI Quantera Scanning
X-ray Microprobe spectrometer of ULVAC-PHI Inc. The binding energies of the
elements in Li-La2O3 were referenced with respect to the carbonaceous C1s line at
284.8 eV.
The basic property of xLi-La2O3 was characterized by Temperature-programmed
desorption of CO2 (CO2-TPD), with a Cat-Lab instrument (BEL-A, Japan) equipped
with a GAM 200 mass spectrometer of Inprocess Instruments. In a typical procedure,
a sample was pretreated at 400 °C in Ar with a flow rate of 40 mL/min for 30 min,
then cooled to 100 °C. The adsorption of 20.1% CO2/Ar was performed at 100 °C for
30 min. After the adsorption, the sample was purged in Ar, then the desorption process
was carried out from 100 °C to 800 °C with a heating rate of 15 °C/min in Ar stream
(80 mL/min), and stayed for 30 min at 800 °C. The quadrupole mass spectrometer
was employed to monitor the desorbed CO2 and related molecular ions (m/z=12, 16,
18, 28, 40, 44) in the desorbed flow. After the measurements, the amounts of desorbed
CO2 from the catalysts were calculated by the integrated desorption area of CO2-TPD
peaks with a calibration of the standard gases.
The morphologies of xLi-La2O3 samples were characterized by transmission
electron microscopy (TEM) on a JEM 2010 TEM (JEOL Ltd.) with an acceleration
voltage of 120 kV. Before measurements, the samples were suspended in ethanol
solution and dispersed by ultrasonic dispersion for 30-60 min. Then, the suspensions
8

were dropped onto copper grid for TEM measurements.
2.4. Catalytic performance
The trans-esterification of glycerol and DMC over various xLi-La2O3 catalysts
was carried out in a 100 mL stainless-steel autoclave. The size of the reactor body is
25 × 210 mm and there is a stainless steel blind tube with a thermocouple for
temperature measurement of reaction mixture inside the reactor. The reaction mixture
is stirred with a magnetic stirrer during the trans-esterification.
In a typical experiment, 3.0 g (33 mmol) of glycerol, 9.0 g (100 mmol) of DMC
and 0.1 g of Li-La2O3 catalyst were added into the autoclave. After sealing the reactor,
the mixture in the autoclave was heated to a desired reaction temperature (60, 70, 85,
100 or 120 °C) for a certain time (1, 1.5, 2, 3 or 4 h) under consecutive stirring. After
the reaction, the reactor was cooled down in ice-water bath and then a given mass of
internal standard substance (diethylene glycol monomethyl ether, shorted as DEGME)
was added into the reaction mixture, meanwhile, 10 mL DMF was also added for
diluting the reaction mixture. Then, the mixture was separated by centrifugation, and
obtained liquid phase was analyzed by a gas chromatography (Lunan-SP 6890, China),
equipped with a flame ionization (FID) detector and a PEG-20M (30 m × 0.25 mm)
column. The quantitative analysis of the reaction products was made by Gas
chromatography -Mass.
3. Results and discussion
3.1. Reaction results of Li doped metal oxides
Firstly, the catalytic performance of several Li doped oxides (La2O3, ZnO, Al2O3
and SBA-15) in the trans-esterification of glycerol and DMC was investigated, and
the results are shown in Table 1. Among them, the Li-MxOy catalysts with base
supports (Li-La2O3 and Li-ZnO) showed high conversions of glycerol (91-94%),
while the catalyst with acid support (Al2O3) displayed relatively low conversion
(59%), and the selectivity of GLC over Li-Al2O3 was not so high. On the other hand,
the catalyst with a neutral support (SBA-15, i.e. SiO2) revealed somewhat high
conversion (82%) and good selectivity (90%), but the catalytic performance of
Li-SBA-15 was still less than that of Li-La2O3. For Li-ZnO catalyst, although it
9

showed high glycerol conversion, but the selectivity of GLC over Li-ZnO was not
high. In addition, the reactions of glycerol and DMC over these Li-MxOy catalysts
also produced small amount of glycidol (GD) and trace amounts of others (identified
as 4-(methoxymethyl)-1,3-dioxolan-2-one and Methyl (oxiran-2-ylmethyl) carbonate
according to GC-Mass qualitative analysis and the reference [20]). Among these
catalysts, the catalytic reactions over Li-La2O3 had minimal byproduct formation and
highest GLC yield. Therefore, Li-La2O3 catalysts were further investigated for their
properties and catalytic performance in the trans-esterification reaction, and the
influence factors on the reaction were also examined.
3.2. Catalyst characterizations of xLi-La2O3
Crystal phases of various xLi-La2O3 samples with different Li contents, which
were calcined at 600 °C for 5 h, were investigated by XRD, as presented in Fig. 1.
According to the XRD patterns, all xLi-La2O3 samples were well crystalline materials
with hexagonal lanthanum oxide structures (P-3m1, JCPDS 05-0602). However, no
signals assigned to the phases of lithium oxides were observed from the XRD patterns
of xLi-La2O3 samples. Because lithium is a lighter element than lanthanum, it is
difficult to detect the crystalline phases of lighter element species for XRD
measurement, due to weak diffraction of the heavier atoms, as reported by Mishra et
al [21].
The main diffraction peaks in the XRD pattern of pure lanthanum oxide were
observed at 2θ=26.1, 29.1, 30.0, 39.5, 46.1 and 52.1°, corresponding to (100), (002),
(101), (102), (110) and (103) crystal planes, respectively. The intensities of diffraction
peaks of the xLi-La2O3 samples increased with the Li loading amounts, indicated that
higher contents of Li doping on La2O3 were favorable to the crystal growth, compared
to pure La2O3. On the other hand, the XRD patterns also showed crystal phase peaks
of La(OH)3 (P63/m, JCPDS 36-1481), which were positioned at 2θ=15.7, 27.3, 28.0,
48.3 and 48.6°. The appearance of La(OH)3 crystal phase might be due to the
impregnation of La2O3 with an aqueous solution of LiNO3, as also reported by Bernal
et al [22]. The average crystallite sizes of xLi-La2O3 are shown in Table S1 (shown in
SM). It can be seen that, after Li was doped on La2O3, the particle size of Li-La2O3
10

catalysts increased somewhat (from 51.4 nm to about 64.1 nm), and the BET surface
areas of Li-La2O3 also decreased (Table S1, shown in SM), compared to pure La2O3.
However, for the different Li loading amounts (Li2O mass percentage in the range of
0.74-3.50%), the particle sizes of the xLi-La2O3 samples changed little, and only
higher Li loading amount sample 7.32Li-La2O3 had a larger particle size (64.1 nm).
The XRD patterns in the region of 2θ=25-32° were enlarged to further
investigate the Li doping effect on the structures of Li-La2O3 samples (Fig. 1B). For
xLi-La2O3 catalysts, the diffraction peaks of (002) and (101) crystal planes shifted to
higher 2θ angles with increasing Li doping amounts, compared to pure La2O3, and
3.50Li-La2O3 catalyst had the maximum shift in the peak positions. The same
phenomenon was also observed in the region of 2θ=38-58° (Fig. 1C). As it has been
reported by Kung et al [23], the substitution of La3+ ions (1.03 Å) by Li+ ions (0.76 Å),
which has different ionic radii from La, would cause asymmetry in the crystal
structure of La2O3. The 7.32Li-La2O3 catalyst had almost similar 2θ diffraction angles
with 3.50Li-La2O3, and this indicated that the Li+ ions entering into La2O3 lattice had
a limitation in quantities [24]. On the other hand, for the samples with Li loading
amounts in the range of 1.58-7.32% Li2O, the diffraction peak identity of (002) crystal
plane increased and became the main diffraction peak. The similar phenomenon in the
case of ZnO doped with Li was also observed and reported by Kung et al [23], and
they indicated that the peak intensity of (0002) crystal plane of ZnO was gradually
decreased after doping Li on ZnO. Based on the reference and our results, we
supposed that Li on La2O3 could influence the recrystallization of Li-La2O3 catalysts
under high calcination temperatures. The growth of (101) crystal plane of La2O3 was
restrained by Li doping on La2O3, while the growth of (002) crystal plane of La2O3
was promoted due to the Li doped in xLi-La2O3 catalysts. Li et al reported a
Li1.2N0.2Mn0.6O2 (LNM) material and investigated its crystal phase by XRD. They
also found that the intensity of the (110) peak increased with the (018) peak
decreasing in the LNM sample, which meant the preferential growth of the crystal
plane appeared. They inferred the spacious layered interspace was beneficial for Li+
diffusion, and led to the intensity change of these two diffraction peaks [25]. These
11

results obtained from XRD characterizations illustrated the relatively stronger
interaction between Li and La2O3 caused by the doping Li into La2O3.
The XPS spectra of xLi-La2O3 samples are presented in Fig. 2 and the binding
energies are listed in Table S2 (shown in SM). Fig. 2A shows the survey spectra of
pure La2O3 and xLi-La2O3 samples. No N1s signal was observed in the survey XPS
spectra, and this was consistent with the result of N1s spectra (Fig. S1, shown in SM),
indicating that no N element was remained in xLi-La2O3 samples. According to these
results, it could validate that the impregnated LiNO3 was completely decomposed,
and no NOx remained on the samples due to the high calcination temperatures. The
polluted C signal was probably because the adsorption of CO2 on the surface of
Li-La2O3 catalysts [16].
Fig. 2B shows the O1s spectra, which consisted of two overlapping peaks with
binding energies at 531.1 and 528.5 eV. The main peak (531.1 eV) was corresponded
to the oxygen ions O2- in the La2O3 crystal lattice, while the peak at B.E. 528.5 eV
with lower intensity could be assigned to oxygen ions in the hydroxyl group [26] of
La2O3 support. Meanwhile, it can be seen from the spectra of Li 1s (Fig. 2C) that the
intensity of the peak attributed to Li species with the B.E. at ~55 eV increased clearly
with the Li loading, implying the successful impregnation of Li. However, for pure
Li2O sample, it’s B.E. of Li1s was centered at 55.6 eV [27], while the B.E. of Li1s of
xLi-La2O3 decreased from 55.3 eV to 55.0 eV with doping Li amounts from
0.74%Li2O to 7.32%Li2O, indicating the change of Li chemical environment in
xLi-La2O3. The peaks composed of La3d3/2 and La3d5/2 due to spin-orbit coupling
were assigned to La species in La2O3 (Fig. 2D). Meanwhile, the La3d3/2 spectra
demonstrated a double peak with main B.E. of 850.9 eV and a satellite peak with B.E.
of 854.9 eV, and the La3d5/2 spectra revealed a main peak with B.E. of 834.2 eV and a
satellite peak with B.E. of 837.9 eV. These XPS results were in good agreement with
La2O3 XPS spectra reported in literature [26]. In addition, the B.E. of La3d5/2 of
xLi-La2O3 slightly shifted to higher binding energy (834.3~834.5 eV) after doping Li
on La2O3 compared with pure La2O3 (834.2 eV). For example, for 0.74Li-La2O3 and
3.50Li-La2O3, the B.E. of La3d5/2 slightly shifted to 834.3 eV and 834.5 eV,
12

respectively. At the same time, the B.E. of La3d3/2 of xLi-La2O3 also slightly changed
(851.0~851.2 eV), compared to pure La2O3 sample (850.9 eV). When the loading
amounts of Li varied from 0.74%Li2O to 3.50%Li2O, the addition of Li led to the
increase B.E. of La3d, which meant an electron transfer from lanthanum atoms to
lithium atoms [28]. But for the 7.32Li-La2O3 catalyst, the surplus Li caused surface
enrichment of lithium, meanwhile, the binding energies of La3d decreased with
binding energy of O increased. This might due to the the surface enrichment of Li
caused redistribution of electrons, influenced the sharing of electrons between La and
O and the covalent component of the La-O bonds [29], the similar phenomenon was
also reported by Song et al [16]. The 3.50Li-La2O3 sample showed the highest B.E.
shifts, implying the relatively strong interaction of La and Li.
The basic properties of various xLi-La2O3 samples calcined at 600 °C were
examined by CO2-TPD, as shown in Fig. 3, and the amounts of basic sites of
xLi-La2O3 samples obtained from CO2-TPD are listed in Table 2. For the TPD profile
of La2O3 support, there was a big CO2 desorption peak at around 500-750 °C, which
indicated that the existent of basic sites on the surface of La2O3 [30]. Besides the large
CO2 desorption peak located at around 500-750 °C, another small peak also appeared
in 0.74Li-La2O3 sample at 800 °C, which stood for much stronger basic sites after
loading Li on La2O3, and the similar phenomenon was also observed for 1.58Li-La2O3
and 3.50Li-La2O3 samples.
0.74%Li2O doped on La2O3 could improve the base site amount (578.5 μmol/gcat)
of 0.74Li-La2O3 catalyst, compared to pure La2O3 with 283.9 μmol/gcat (Table 2). And
further increasing Li contents would cause the increase of the amount of strong base
sites, but decreasing the amount of moderate strong base sites. Up to 7.32%Li2O
doping, the moderate strong desorption peak disappeared, meanwhile, the total
amount of base sites on 7.32Li-La2O3 decreased to 362.3 μmol/gcat, and the desorption
of CO2 became more difficult (The initial desorption temperature shifted to 700 °C for
7.32Li-La2O3, and still incompletely desorbed after keeping 800 °C for 30 min).
These results suggested that Li doped on La2O3 could increase the amount of base
sites on Li-La2O3 and produce much strong basic sites, caused the complete
13

desorption of CO2 to be very difficult. From the characterization results of CO2-TPD,
it can be understood that Li doping could increase the strength of Li-La2O3 and
improve the total amount of base sites on the surface of Li-La2O3, and 3.50Li-La2O3
had the highest amount of total base sites, with 335.9 μmol/gcat of much strong basic
sites.
Fig. 4 shows the transmission electron microscope (TEM) images of pure La2O3,
0.74Li-La2O3 and 3.50Li-La2O3 catalysts. The morphology of pure La2O3 (Fig. 4 A,
A’) showed a layer structure. After doping Li and calcined at 600 °C, the particle
aggregations were observed for 0.74Li-La2O3 and 3.50Li-La2O3 catalysts, as shown in
Fig. 4 (B, B’ and C, C’). It has been reported that doping alkali metals could cause the
deformation of metal oxide structures due to the high temperature solid-phase reaction
[31, 32]. The sintering effect at high temperature (600 °C) not only led to the particle
growth but also decreased the surface area of support oxide, as shown in XRD and
BET (Table S1, in SM). The exposed crystalline planes of pure La2O3 were (100) (002)
and (101), and these results were in accordance with XRD results. However, after
0.74 and 3.50Li doped, TEM showed that the d002 of La2O3 increased from 0.306 nm
(pure La2O3) to 0.312 nm (0.74Li-La2O3) and 0.316 nm (3.50Li-La2O3), respectively,
and d101 reduced from 0.298 nm (pure La2O3) to 0.295 nm (0.74Li-La2O3) and 0.291
nm (3.50Li-La2O3), respectively. Combined the XRD characterization results, it can
be believed that the doping of Li on La2O3 could suppress the growth of (101) crystal
plane of La2O3, and consequently cause the lattice expansion of (002) crystal plane.
This phenomenon might due to the preferred orientation of Li+ in La2O3 lattice. For
the lattice expansion of (002) crystal plane, it could be inferred that Li+ could diffuse
into the lattices of La2O3 and cause the growth of (002) plane of La2O3. Li addition
might influence the distribution of electrons and atoms in the crystal lattice, and
finally led to the crystal spacing of (002) crystal plane expansion. Li et al [29] also
reported the Li-CeO2 system and found that Li induced the redistribution of electrons
and then influenced the Ce/O atomic arrangement in the crystal lattice. The (110)
plane of CeO2 were distorted by Li and the spacing of this plane increased and caused
the expansion of lattice after lithiation [29]. Based on these results, it is reasonable to
14

be considered that under the high temperature calcination, Li ions might enter into the
lattice of La2O3.
3.3. Catalytic performance of xLi-La2O3
3.3.1. Catalytic performance of Li-La2O3 with different Li doping amounts
The catalytic performance of xLi-La2O3 calcined at 600 °C was examined under
the reaction temperature of 85 °C, and the results are shown in Table 3. All xLi-La2O3
catalysts with Li loading amounts in the range of 0.74-7.32% Li2O revealed high
glycerol conversions (93%-99%) at 85 °C and 3 h reaction. Over these Li-La2O3
catalysts, the conversions of glycerol increased slightly, but the selectivities of GLC
varied with the Li loading amounts. Among them, 3.50Li-La2O3 showed the highest
selectivity of GLC (92%), while 7.32Li-La2O3 had lower GLC selectivity (78%).
Furthermore, the byproducts produced over 3.50Li-La2O3 were also a bit less. There
was no significant difference in the glycerol conversion over these xLi-La2O3 with
different Li loadings, and this might be due to the relatively high reaction temperature
(85 °C). Therefore, we further carried out the reactions over these xLi-La2O3 catalysts
at a relatively lower reaction temperature (70 °C), and the glycerol conversions,
products selectivities, yields are shown in Table S3 (shown in SM), and the data were
also plotted as shown in Fig. 5. Clearly, the glycerol (GL) conversions were
significantly improved by loading Li on La2O3 support, with increasing from 12.3%
for pure La2O3, to 67.2% for 0.74Li-La2O3 and 81.1% for 3.50Li-La2O3, respectively.
7.32Li-La2O3 showed up the highest glycerol conversion but lower GLC selectivity
with the increase of GD selectivity. In general, 3.50Li-La2O3 showed the best catalytic
performance among these xLi-La2O3 catalysts even at 70 °C reaction. The
1.58Li-La2O3 and 3.50Li-La2O3 catalysts had similar amounts of moderate strong
basic sites, while the amount of strong basic sites of 3.50Li-La2O3 was higher than
that of 1.58Li-La2O3 based on the CO2-TPD results (Table 2). Combined with reaction
results, it is implied that strong basic sites would be more favorable to facilitate the
trans-esterification of glycerol and DMC. However, too much amount of Li loading
(7.32Li-La2O3) would not benefit the selectivity of GLC, but increased the byproduct
GD. According to the CO2-TPD results, the 7.32Li-La2O3 catalyst had the strongest
15

basic sites (800 °C peak in CO2-TPD), which contributed to the highest glycerol
conversion. Also, for 7.32Li-La2O3, the excessive Li was enriched on the surface of
the catalyst, as characterized by XPS (Table S4, shown in SM), it was implied that the
excessive Li on the surface of 7.32Li-La2O3 was in connection with the further
conversion of glycerol carbonate to glycidol.
3.3.2. Effect of calcination temperatures on the catalytic performance of
3.50Li-La2O3
It is well known that high temperature solid-phase reaction had an important
influence on the morphology, crystalline and basic property of catalysts. For checking
the effect of calcination temperatures on the catalytic performance, 3.50Li-La2O3-T
catalyst was selected for further investigating under the reaction temperature of 70 °C,
and the results are listed in Table 4. In the range of 400-700 °C, the calcination
temperatures had little influence on the selectivities of GLC over 3.50Li-La2O3
catalyst, but the conversions of glycerol were greatly affected by the calcination
temperatures. The catalysts calcined at 700 °C and 400 °C (3.50Li-La2O3-700 and
3.50Li-La2O3-400) showed the lower activities for the conversion of glycerol, and the
reason might be that the high calcination temperature (700 °C) caused particle
aggregation (Table S5, shown in SM), while the low calcination temperature (400 °C)
couldn’t provide complete growth of crystalline phase of the catalyst (Fig. S2, shown
in SM). 3.50Li-La2O3-600 had higher conversion for glycerol and GLC selectivity
among these four catalysts.
3.3.3. Effect of reaction parameters on the catalytic performance of
3.50Li-La2O3
The 3.50Li-La2O3 catalyst calcined at 600 °C was chosen for further
investigating the influence of reaction parameters on the catalytic performance (Fig.
6-7).
3.3.3.1. Effect of reaction temperatures
The effect of reaction temperatures (from 70 °C to 120 °C) on the conversion and
selectivity over 3.50Li-La2O3 was studied, as shown in Table S6 (shown in SM), and
the data were plotted in Fig. 6. When the reaction was carried out at a lower
16

temperature (60 °C), the conversion of glycerol (GL) was quite low (55.7%), but the
selectivity of GLC was high (about 94%). As the reaction temperature increased from
60 °C to 85 °C, the conversion of glycerol improved. For example, from 70 °C to
85 °C, the glycerol conversion significantly increased from 81.1% to 94.4%,
respectively, while the GLC selectivity (about 92%) kept almost constant. These
results proved that increasing reaction temperatures was favorable for the progress of
trans-esterification, and glycerol mainly converted to GLC during the temperature
from 60 °C to 85 °C. Although the glycerol conversion further increased to about 99%
with continuously increasing reaction temperature to 120 °C, the selectivity of GLC
decreased to 78.1%, and the selectivities of byproducts increased. The decrease of
GLC selectivity was due to the side reactions, and higher temperatures (100-120 °C)
could accelerate the side reactions. Therefore, too high reaction temperatures (above
100 °C) was harmful to the selectivity of GLC, but was beneficial to the formation of
GD and other products. It is also easy to infer that high temperatures would cause the
GLC decarbonylation to GD.
3.3.3.2. Effect of reaction time
The effect of reaction time was studied in the range of 60 min to 240 min (Fig. 7
and Table S7, shown in SM). The conversion of glycerol (GL) was improved with the
increase of reaction time. In the first 60 min of reaction, only 47.9% glycerol
conversion and 43.5% for GLC yield were obtained, this indicated that the reaction
was still far from reaching equilibrium at the first 60 minutes. Although glycerol
conversion was enhanced as the reaction time lengthened, GLC selectivity was
basically the same (about 92%) in the range of 60-180 min reaction. However, when
the reaction time was prolonged to 240 min, the GLC selectivity dropped to 86.2%,
and this might be due to the occurrence of side reactions, such as etherification of
methanol and GLC, and trans-esterification of GD and DMC at long reaction time
[20]. Therefore, in the tested conditions, 180 min was most suitable reaction time for
getting maximum glycerol conversion and GLC yield.
3.3.4. Recycle use of 3.50Li-La2O3
The recycle uses of a catalyst is a challenge for alkali metal promoted catalysts
17

[18]. 3.50Li-La2O3 was chosen for testing the recycle uses under 85 °C and 3 h, and
the results are shown in Fig. 8. After the first run, the 3.50Li-La2O3 catalyst was
separated by centrifugation and recovered, and then the used catalyst was washed
with ethanol for three times to remove the adsorbed organic compounds, and then the
catalyst was dried at 110 °C for 12 h. Above-treated spent 3.50Li-La2O3 catalyst was
then reused for the next run. From the results in Fig. 8, it can be seen that the glycerol
(GL) conversions had no obvious decrease (changing in the range of 92-94%), and
GLC selectivities were also close (in the range of 91-92%) for the first three reaction
runs. But in the fourth run, it showed somewhat decrease in glycerol conversion
(87%), it might be due to some change of the active sites on the catalyst surface after
multi reaction cycles. Therefore, the fresh and spent recycled 3.50Li-La2O3 catalysts
with FT-IR spectra were further characterized, as shown in Fig. S3 (in SM). From Fig.
S3, several IR adsorption bands could be observed. The absorption band at 3606 cm-1
is ascribed to bridged O-H groups from hexagonal La(OH)3, due to the absorption of
moisture in the atmosphere on the surface of La2O3 [33]. While the IR peaks at around
860 cm-1 and 1460 cm-1 belonged to surface carbonate species [34], which was due to
the adsorption of carbon dioxide in atmosphere on the base sites of the catalysts, and
the intensities of peaks at 860 cm-1 and 1460 cm-1 were associated with the basic
property of catalysts [35]. Comparing FT-IR spectra of Fig. S3a and Fig. S3d, it can
be found that the intensity of carbonate species (peak at 1460 cm-1) of “spent
3.50Li-La2O3 after 3rd run” was much lower than that of “spent 3.50Li-La2O3 after
1st run”. This means that spent 3.50Li-La2O3 after 3rd run had less basicity than fresh
catalyst [36]. Based on the IR results, it can be presumed that the decrease in glycerol
conversion after 3 cycles might be due to the loss of the catalyst surface basicity after
multi reaction cycles. [36, 37].
4. Conclusion
A series of Li doped xLi-La2O3 catalysts were prepared by a simple impregnation
method, and employed in the trans-esterification of glycerol and DMC to glycerol
carbonate. The Li-La2O3 catalysts (1.58-3.50Li-La2O3) showed high glycerol
conversion (~94%) and glycerol carbonate selectivity (~92%) under mild conditions
18

(85 °C, 3 h). Doped Li could enter the lattice of La2O3, which caused the lattice
expansion of (002) crystal plane of La2O3, and enhanced the interaction between Li
and La2O3. The addition of Li into La2O3 also increased the strength and content of
basic sites of the xLi-La2O3 catalysts. The high activities of Li-La2O3 catalysts were
ascribed to the strong basic property, abundant base sites and the interaction between
Li and La2O3. The Li loading amounts and reaction temperatures had great influence
on glycerol conversion and glycerol carbonate selectivity.
Acknowledgments
We acknowledge the financial support from the Natural Science Foundation of
China (No. 21573120) and Specialized Research Fund for the Doctoral Program of
Higher Education of Ministry of Education of China (20131018984). The authors are
also grateful to Prof. Boqing Xu (Tsinghua University, China) for providing TPD
analysis and Mr. Yu Liang (Tsinghua University, China) for the kind guidance of TPD
operation.
19

References
[1] J.C. Serrano-Ruiz, R. Luque, A. Sepulveda-Escribano, Chem. Soc. Rev. 40 (2011) 5266-5281.
[2] J.R. Ochoa-Gómez, O. Gómez-Jiménez-Aberasturi, C. Ramírez-López, M. Belsué, Org. Process
Res. Dev. 16 (2012) 389-399.
[3] M.O. Sonnati, S. Amigoni, E.P. Taffin de Givenchy, T. Darmanin, O. Choulet, F. Guittard, Green
Chem. 15 (2013) 283-306.
[4] A.-A.G. Shaikh, S. Sivaram, Chem. Rev. 96 (1996) 951-976.
[5] L.P. Ozorio, C.J.A. Mota, ChemPhysChem. 18 (2017) 3260-3265.
[6] H. Nguyen-Phu, C.-y. Park, E.W. Shin, Appl. Catal. A- Gen. 552 (2018) 1-10.
[7] J.R. Ochoa-Gómez, O. Gómez-Jiménez-Aberasturi, B. Maestro-Madurga, A. Pesquera-Rodríguez,
C. Ramírez-López, L. Lorenzo-Ibarreta, J. Torrecilla-Soria, M.C. Villarán-Velasco, Appl. Catal. A- Gen.
366 (2009) 315-324.
[8] G. Rokicki, P. Rakoczy, P. Parzuchowski, M. Sobiecki, Green Chem. 7 (2005) 529-539.
[9] P.-J. Seong, B.W. Jeon, M. Lee, D.H. Cho, D.-K. Kim, K.S. Jung, S.W. Kim, S.O. Han, Y.H. Kim,
C. Park, Enzyme Microb. Tech. 48 (2011) 505-509.
[10] L. Zheng, S. Xia, X. Lu, Z. Hou, Chinese J. Catal. 36 (2015) 1759-1765.
[11] F.S.H. Simanjuntak, V.T. Widyaya, C.S. Kim, B.S. Ahn, Y.J. Kim, H. Lee, Chem. Eng. Sci. 94
(2013) 265-270.
[12] D. Singh, B. Reddy, A. Ganesh, S. Mahajani, Ind. Eng. Chem. Res. 53 (2014) 18786-18795.
[13] M. Malyaadri, K. Jagadeeswaraiah, P.S. Sai Prasad, N. Lingaiah, Appl. Catal. A- Gen. 401 (2011)
153-157.
[14] K. Hu, H. Wang, Y. Liu, C. Yang, J. Ind. Eng. Chem. 28 (2015) 334-343.
[15] R. Bai, Y. Wang, S. Wang, F. Mei, T. Li, G. Li, Fuel Process. Technol. 106 (2013) 209-214.
[16] X. Song, Y. Wu, F. Cai, D. Pan, G. Xiao, Appl. Catal. A- Gen. 532 (2017) 77-85.
[17] V.K. Diez, C.R. Apesteguia, J.I. Di Cosimo, J. Catal. 240 (2006) 235-244.
[18] C.S. Castro, C. Ferreti, J.I. Di Cosimo, J.M. Assaf, Fuel. 103 (2013) 632-638.
[19] L.-h. Wang, X.-d. Yi, J. Fujian Univ. Nat. Sci. Ed.. 28 (2012) 60-63.
[20] S. Guidi, R. Calmanti, M. Noe, A. Perosa, M. Selva, ACS Sustain. Chem. Eng. 4 (2016)
6144-6151.
[21] B.G. Mishra, G.R. Rao, J. Mol. Catal. A-Chem. 243 (2006) 204-213.
[22] S. Bernal, F.J. Botana, R. Garcia, F. Ramirez, J.M. Rodriguez-Izquierdo, J. Mater. Sci. 22 (1987)
3793-3800.
[23] C.Y. Kung, C.C. Lin, S.L. Young, L. Horng, Y.T. Shih, M.C. Kao, H.Z. Chen, H.H. Lin, J.H. Lin,
S.J. Wang, J.M. Li, Thin Solid Films. 529 (2013) 181-184.
[24] B. Wang, L. Tang, J. Qi, H. Du, Z. Zhang, J. Alloys Compd. 503 (2010) 436-438.
[25] Y. Li, Y. Bai, C. Wu, J. Qian, G. Chen, L. Liu, H. Wang, X. Zhou, F. Wu, J. Mater. Chem. A. 4
(2016) 5942-5951.
[26] E. Baskys, V. Bondarenka, S. Grebinskij, M. Senulis, R. Sereika, Lith. J. Phys. 54 (2014)
120-124.
[27] E. Radvanyi, E. De Vito, W. Porcher, S. Jouanneau Si Larbi, J. Anal. At. Spectrom. 29 (2014)
1120-1131.
[28] H. Li, D. Gao, P. Gao, F. Wang, N. Zhao, F. Xiao, W. Wei, Y. Sun, Catal. Sci. Technol. 3 (2013)
2801-2809.
[29] K. Li, X. Zhou, A. Nie, S. Sun, Y.-B. He, W. Ren, B. Li, F. Kang, J.-K. Kim, T.-Y. Zhang, Nano
20

Lett. 17 (2017) 1282-1288.
[30] R. Shi, F. Wang, Tana, Y. Li, X. Huang, W. Shen, Green Chem. 12 (2010) 108-113.
[31] T. Berger, J. Schuh, M. Sterrer, O. Diwald, E. Knoezinger, J. Catal. 247 (2007) 61-67.
[32] C.S. Castro, C. Ferreti, J.I. Di Cosimo, J.M. Assaf, Fuel. 103 (2013) 632-638.
[33] D. Tsoutsou, G. Scarel, A. Debernardi, S.C. Capelli, S.N. Volkos, L. Lamagna, S. Schamm, P.E.
Coulon, M. Fanciulli, Microelectron. Eng. 85 (2008) 2411-2413.
[34] M. Nieminen, M. Putkonen, L. Niinisto, Appl. Surf. Sci. 174 (2001) 155-166.
[35] M. Salavati-Niasari, G. Hosseinzadeh, F. Davar, J. Alloys Compd. 509 (2011) 4098-4103.
[36] B.P. Gangwar, S.C. Maiti, S. Sharma, J. Solid State Chem. 256 (2017) 109-115.
[37] C.-y. Park, H. Nguyen-Phu, E.W. Shin, Mol. Catal. 435 (2017) 99-109.
21

Figure captions
Fig. 1 XRD patterns of xLi-La2O3 calcined at 600 °C.
(A) Full pattern, 2θ=10-90°, (B) Partly enlarged pattern, 2θ=25-32°, (C) Partly
enlarged pattern, 2θ=38-58°.
Fig. 2 XPS spectra of xLi-La2O3 calcined at 600 °C.
(A) wide spectra, (B) O1s, (C) Li1s, (D) La3d.
Fig. 3 CO2-TPD profiles of xLi-La2O3 calcined at 600 °C.
Fig. 4 TEM images of samples.
(A) (A’) pure La2O3, (B) (B’) 0.74Li-La2O3, and (C) (C’) 3.50Li-La2O3.
Fig. 5 Effect of Li contents in xLi-La2O3
Reaction conditions: 33 mmol glycerol, 70 °C, 3 h, 0.1 g catalyst, Glycerol/DMC=1:3
Fig. 6 Effect of reaction temperatures
Reaction conditions: 33 mmol glycerol, 3 h, 0.1 g 3.50Li-La2O3 catalyst,
Glycerol/DMC=1:3
Fig. 7 Effect of reaction time
Reaction conditions: 33 mmol glycerol, 85 °C, 0.1 g 3.50Li-La2O3 catalyst,
Glycerol/DMC=1:3
Fig. 8. Reusability of the 3.50Li-La2O3 in the trans-esterification of glycerol and
DMC
Reaction conditions: 0.1 g catalyst, for every run: 85 °C, 3 h, 33 mmol glycerol,
Glycerol/DMC=1:3
22

A
7.32Li-La2O3
3.50Li-La2O3
1.58Li-La2O3
0.74Li-La2O3
1.5
2.5

 La2O3 JCPDS 05-0602 3.0
La2O3
















10
20
30
40
50
2
60
70
80
90
B
 La(OH)3 JCPDS 36-1481
7.32Li-La2O3


3.50Li-La2O3
1.58Li-La2O3
0.74Li-La2O3
La2O3
(1.5
2.5
(3.0
25
26
27
28
29
30
31
32
2
C
7.32Li-La2O3
3.50Li-La2O3
(1.5
2.5
1.58Li-La2O3
0.74Li-La2O3
(3.0
La2O3
38
40
42
44
46
48
2
50
52
54
56
58
Fig. 1 XRD patterns of xLi-La2O3 calcined at 600 °C.
(A) Full pattern, 2θ=10-90°, (B) Partly enlarged pattern, 2θ=25-32°, (C) Partly enlarged
pattern, 2θ=38-58°.
23

7.32Li-La2O3
(A)
(B) O 1s
7.32Li-La2O3
3.50Li-La2O3
3.50Li-La2O3
1.58Li-La2O3
1.58Li-La2O3
0.74Li-La2O3
0.74Li-La2O3
La2O3
La2O3
1200
1000
800
600
B.E.(eV)
400
200
0
538
536
534
532
B.E.(eV)
530
528
526
(C) Li 1s
7.32Li-La2O3
(D) La 3d
7.32Li-La2O3
3.50Li-La2O3
3.50Li-La2O3
1.58Li-La2O3
1.58Li-La2O3
0.74Li-La2O3
0.74Li-La2O3
La2O3
La2O3
870
860
850
B.E.(eV)
840
830
60
58
56
54
52
50
48
B.E.(eV)
Fig. 2 XPS spectra of xLi-La2O3 calcined at 600 °C.
(A) wide spectra, (B) O1s, (C) Li1s, (D) La3d.
24

7.32Li-La2O3
3.50Li-La2O3
1.58Li-La2O3
0.74Li-La2O3
La2O3
800
100 200
400
800
temperature C
600
30min
Fig. 3 CO2-TPD profiles of xLi-La2O3 calcined at 600 °C.
25

Fig. 4 TEM images of samples.
(A) (A’) pure La2O3, (B) (B’) 0.74Li-La2O3, and (C) (C’) 3.50Li-La2O3.
26

100%
80%
60%
40%
20%
0%
GL conversion
GLC selectivity
GD selectivity
GLC yield
0%
2%
4%
6%
8%
x wt% of Li-La2O3 catalyst
Fig. 5 Effect of Li contents in xLi-La2O3
Reaction conditions: 33 mmol glycerol, 70 °C, 3 h, 0.1 g catalyst, Glycerol/DMC=1:3
27

100%
80%
60%
40%
20%
0%
GL conversion
GLC selectivity
GD selectivity
GLC yield
60
70
80
90
100
110
120
Temperature C
Fig. 6 Effect of reaction temperatures
Reaction conditions: 33 mmol glycerol, 3 h, 0.1 g 3.50Li-La2O3 catalyst, Glycerol/DMC=1:3
28

100%
80%
60%
40%
20%
0%
GL conversion
GLC selectivity
GD selectivity
GLC yield
60
120
180
240
Reaction time min
Fig. 7 Effect of reaction time
Reaction conditions: 33 mmol glycerol, 85 °C, 0.1 g catalyst 3.50Li-La2O3 catalyst,
Glycerol/DMC=1:3
29