Research Article

Cite This: ACS Catal. 2018, 8, 10446−10456 pubs.acs.org/acscatalysis

Oxygen Vacancy Promoting Dimethyl Carbonate Synthesis from CO2

and Methanol over Zr-Doped CeO2 Nanorods
Bin Liu,† Congming Li,*,† Guoqiang Zhang,† Xuesi Yao,‡ Steven S. C. Chuang,‡ and Zhong Li*,†
†
Key Laboratory of Coal Science and Technology, Ministry of Education and Shanxi Province, Taiyuan University of Technology,
Taiyuan, Shanxi 030024, China
‡
Department of Polymer Science, The University of Akron, 170 University Avenue, Akron, Ohio 44325, United States
*
S Supporting Information
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ABSTRACT: The synthesis of dimethyl carbonate (DMC) from CO2 and methanol by
Downloaded via NATURAL RESOURCES CANADA on March 12, 2019 at 14:37:02 (UTC).

Zr-doped CeO2 nanorods with different ratios of Zr/Ce has been studied at 6.8 MPa and 140 °C.
The catalysts were characterized extensively by TEM, XRD, N2 adsorption, Raman spectroscopy,
UV−vis spectroscopy, XPS, CO2-TPD, and in situ FTIR techniques. Doping of Zr atoms into the
ceria lattice produced a fluorite-like solid solution, promoting the formation of oxygen vacancy
sites. Zr-doped CeO2 nanorods exhibited significantly more oxygen vacancy sites than pure CeO2
nanorods. Zr0.1Ce nanorods which exhibited DMC synthesis activity also possess the highest concentration of oxygen vacancy sites.
In situ FTIR studies further revealed that CO2 can adsorb on the oxygen vacancy to form bidentate carbonate and as intermediate
to participate in the reaction. This study presents a strategy to design a high-efficiency CeO2-based catalysts by controlling the
concentration of the surface oxygen vacancies.
KEYWORDS: Zr-doped CeO2 nanorods, oxygen vacancy, CO2 adsorption, dimethyl carbonate, bidentate carbonate

1. INTRODUCTION In our previous research,20 we also found the surface oxygen

Dimethyl carbonate (DMC) is considered as an environmentally vacancies favored CO2 adsorption and improved the catalytic
benign chemical in the emerging area of “green chemistry”.1 The performance for the synthesis of DMC from CO2 and methanol.
DMC molecule contains a number of organic functional groups, The morphology of CeO2 catalyst has a strong influence on
methoxy, carbonyl, and methyl groups, which can be used as a the exposed crystal planes and the concentration of their sur-
potential building block in the methylation and carbonylation face oxygen vacancies. The latter has been shown to govern a
reactions for replacing toxic precursors such as dimethyl sulfate number of catalytic reactions.21−24 CeO2 has been successfully
(DMS) and phosgene.2 DMC has also been used as a fuel synthesized with different morphologies: nanorods, nanocubes,
additive because of its high octane number (105) and oxygen and nano-octahedra. Among these different morphologies, the
content (53%).2 surface of the nanorods has the highest concentration of oxy-
The direct synthesis of DMC from methanol/CO2 has been gen vacancy sites because of their enclosed (110) and (100)
gaining importance because of the abundance of CO2, its crystal planes.25 Wu and co-workers have studied the activity
environmentally benign nature, and the high atom efficiency of of CeO2 nanocrystal with well-defined surface planes for CO
the synthesis process. Catalysts such as Cu−Fe bimetal cata- oxidation reaction, which followed the sequence of: rods >
lyst, ionic liquid, zirconia, CeO2, and CeO2−ZrO2 have been cubes > octahedra.26 Wang et al. studied the morphology
studied for the direct synthesis of DMC from methanol and effects of CeO2 on the direct synthesis of DMC with
CO2.3−7 Among these reported catalysts, ceria-based catalysts 2-cyanopyridine as a dehydration agent, and they found that
showed promising selectivity and activity.8−11 However, the CeO2 nanorods exhibited higher activity than nanocubes and
activation of CO2 is still a great challenge because CO2 is a nano-octahedra because of their high density of defect sites
fully oxidized, thermodynamically stable, and chemically inert and acid−base sites.27 Compared to pure CeO2, doping the
molecule.12 It is well known that the nature of catalytic sites for transition metal oxides such as Y2O3, ZrO2, and TiO2 into the
CO2 adsorption is an important factor affecting the efficiency CeO2 could vary the surface acid−base property and oxygen
of CO2 activation. The surface oxygen vacancy sites of a vacancies.5,28,29 Chen et al. demonstrated that the CeO2−CuO
catalyst could promote the adsorption and activation of CO2.13 nanorods exhibited strikingly high catalytic activity for CO
Oxygen vacancies have been proposed to serve as the active oxidation, which was ascribed to the introduction of CuO
sites which could promote the CO2 conversion in the CO2 species into CeO2 nanorods generating oxygen vacancies.30 Liu
methanation.14−16 It has also been reported that oxygen vacan- et al. reported that ZrO2-doped CeO2 nanorods possessed high
cies were acid active centers active in CO2 hydrogenation and activities for the selective oxidation of styrene to styrene oxide
stabilize the thermodynamically unstable metal oxides.17,18
A number of studies found that the surface oxygen vacancies Received: January 31, 2018
could provide a means of stabilizing the end products by filling Revised: September 12, 2018
one oxygen atom of the CO2 molecule to the oxygen vacancy.19 Published: October 1, 2018

© 2018 American Chemical Society 10446 DOI: 10.1021/acscatal.8b00415

ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

Scheme 1. Schematic Illustration of the Synthesis Process of the Zr-Doped CeO2 Nanorods

compared to pure CeO2 nanorods because of the presence of a Specific surface area measurement was carried out by a
high concentration of oxygen vacancies on the doped CeO2 Beishide 3H-2000PS2 surface analyzer using nitrogen adsorp-
nanorods.31 According to these previous studies, oxygen vacan- tion at liquid-nitrogen temperature (77 K). Before the test, the
cies could play a significant role in the direct synthesis of DMC sample was degassed in situ at 250 °C for 4 h. The surface
on CeO2-based catalysts.32,33 However, the effect of the surface areas of the catalysts were estimated by the Brunauer−
oxygen vacancy on this reaction is poorly understood. Emmett−Teller (BET) method, and the pore size distributions
In this work, the CeO2 nanorods with various Zr content were calculated from analysis of the desorption branch of the
were prepared and their activity in DMC synthesis from CO2 isotherm via the conventional the Barrett−Joyner−Halenda
and methanol was investigated. It was found that the surface (BJH) model.
oxygen vacancy concentration and CO2 adsorption varied with The contents of Ce and Zr were determined by inductively
Zr content, which are crucial to determine the catalytic activity coupled plasma atomic emission spectroscopy (ICP-AES).
of direct synthesis of DMC from methanol and CO2. Before analysis, 2 mL of concentrated HF was used to dissolve
80 mg of the samples, followed by adding 2 mL of 30 wt %
2. EXPERIMENTAL SECTION H2O2, and then the solution was diluted to 1000 mL with
deionized water. Elemental analysis was performed.
2.1. Materials. Cerium(III) nitrate hexahydrate (Ce(NO3)3· Raman spectra were recorded by a Renishaw in ViainVia
6H2O, 99.9%, AR) and zirconium nitrate (Zr(NO3)4·5H2O, micro laser Raman spectrometer (U.K.), Ar+ laser (514.5 nm
99.0%, AR) were purchased from Tianjin Guangfu Fine wavelength), with an output power of 4 mW. The UV−vis
Chemical Research Institute (Tianjin, China); sodium hydrox- spectra of the catalyst samples were obtained by a PerkinElmer
ide (NaOH, 99%, AR) was obtained from Feng Chuan Lambda 900 UV−vis/NIR spectrophotometer.
Chemical Reagent Co, Ltd. (Tian jin, China). All reagents X-ray photoelectron spectroscopy (XPS) data were obtained
were used as received without further purification. by an ESCALab220i-XL electron spectrometer (VG, UK)
2.2. Catalyst Preparation. The nanorods were synthesized using 300 W Al Kα radiation. The samples were compressed
by the hydrothermal process based on a previous report.34 into a pellet of 2 mm thickness and then mounted on a sample
Typically (shown in Scheme 1), Zr(NO3)4·5H2O and holder by utilizing double-sided adhesive tape for XPS analysis.
Ce(NO3)3·6H2O with different mole ratios were dissolved in The sample holder was then placed into a fast entry air load-
30 mL of deionized water under vigorous stirring until com- lock chamber without exposure to air and evacuated under
pletely dissolved. Simultaneously, 84.0 g of NaOH was dis- vacuum (<10−6 Torr) overnight. Finally, the sample holder was
solved in 210 mL of deionized water. Then the two solutions transferred to the analysis chamber for XPS study. The base
were mixed together and kept stirring for 30 min. The obtained pressure inside the analysis chamber was usually maintained at
mixed slurry was transferred into a 300 mL stainless steel <10−10 Torr. The binding energies were referenced to the C 1s
autoclave and kept at 100 °C for 24 h. After the autoclave was line of adventitious carbon at 284.6 eV.
cooled down to room temperature, the products were washed Temperature-programmed desorption (TPD) was carried
with deionized water until the pH reaches 7 and then washed out on a Micromeritics Autochem II 2920 analyzer. A 0.1 g
several times with ethanol. The resulting product was dried at amount of sample was pretreated at 200 °C for 1 h under N2.
80 °C overnight and then calcined at 600 °C for 5 h in a muffle CO2 (NH3) adsorption was performed at 25 °C by switching
furnace to obtain ZrxCe nanorods, where x is the mole ratio of He flow to a stream of 15 vol % CO2 (NH3)−He (40 mL·
Zr to Ce in the catalysts. The Zr/Ce mole ratio was controlled min−1) and maintaining the temperature for 60 min. The
in the range from 0 to 0.3 in order to keep the morphology sample was purged with He (40 mL·min−1) for 2 h in order to
of nanorods. For comparison, the pure ZrO2 were synthesized remove the physically adsorbed CO2 (NH3). TPD test was carried
by the same process. To highlight the important role of hydro- out under the condition of a constant flow of He (40 mL·min−1)
thermal synthesis in the preparation of nanorod catalyst, we from 50 to 600 °C at a heating rate of 10 °C·min−1. The con-
also prepared CeO2 particles for XPS and catalyst activity centration of CO2 (NH3) in the tail gas was continuously
studies (SI, Activity and Characterization of CeO2 particles monitored by a TCD detector.
prepared by coprecipitation). The IR measurements were performed on a Nicolet 6700
2.3. Material Characterization. The morphology of FTIR spectrometer (Thermo-Nicolet) with a DRIFTS (diffuse
the obtained powder of CeO2, ZrxCe nanorods, and ZrO2 reflectance fourier-transform spectroscopy) reactor by the fol-
was characterized by a JEOL JEM-2100 transmission electron lowing procedures.35−37
microscope operated at an accelerating voltage of 200 kV.
Samples for TEM analyses were prepared by dispersing the (1) Pretreatment: 10 mg of catalyst powder was placed in
powdered products in ethanol and then deposited on a carbon the cell, and the catalyst was heated at 200 °C for 10 min
film coated on a copper grid. in an argon environment. Part of the preadsorbed water
The powder X-ray diffraction patterns of the investigated and CO2 were removed.
catalysts were obtained from an X-ray diffractometer equipped (2) CO2 adsorption: Argon was flowed in the IR cell (flow
with a Cu Kα radiation source (in the 2θ range 10−85° with a of Ar is 40 mL·min−1). After 1 min, the flowing gas was
scanning speed of 8°/min). switched to CO2 with the flow 40 mL·min−1.
10447 DOI: 10.1021/acscatal.8b00415
ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

(3) Methanol adsorption: Argon was flowed in the IR cell.

After 1 min, the Ar flow (40 mL·min−1) was passed
through a methanol saturator and brought the methanol
into the IR cell.
(4) CO2 after methanol adsorption: after methanol adsorp-
tion, the catalyst was purged with Ar flow (40 mL·min−1)
for 5 min to remove nonadsorbed methanol vapor in the
IR cell. CO2 (40 mL·min−1) was then flowed into the cell.
DMC adsorption: Argon was flowed in the IR cell. After
1 min, 8 μL of DMC was injected into the Ar flow and
blown into the cell.
The temperature was maintained at 25 °C when CO2,
methanol, and DMC were introduced into the reaction unit.
Adsorption studies at 25 °C resulted in evolution of the
adsorbed species in a time scale of 1 s, allowing identification
of the surface species involved in the interaction with adsorbed
species.
2.4. Catalytic Performance Evaluation. DMC synthesis
was conducted in a slurry reactor with continuous mechanical
stirring. Typically, 0.2 g of catalyst was placed into an autoclave
reactor with 35 mL of methanol. Before reaction, the reactor
was purged with CO2 several times to remove air and then
pressurized up to 3 MPa with CO2 at room temperature. The
reaction temperature and pressure of the reactor were raised to
140 °C and 6.8 MPa for carrying out the reaction for 2 h and
then cooled down. 1-Propanol (CH3CH2CH2OH) as an internal
standard substance was added into the mixture; products were
analyzed by a gas chromatograph (FID-GC, O Hua 9160).
We note that DMC was the only product in the catalytic reac-
tions without the formation of any other byproducts (except
for a small amount of water). In this work, the DMC yield was
defined as follows.
DMC yield = nDMC(mmol)/mcatalyst (g)

3. RESULTS AND DISCUSSION

3.1. Catalyst Morphologies. The TEM images of the
CeO2, ZrxCe, and ZrO2 nanorods are shown in Figure 1. The
CeO2 nanorods have a mean diameter of ∼13 nm and an
average length of ∼150 nm. The length and diameter of the
ZrxCe nanorods varied with mole ratios of Zr/Ce. The diam-
eters of Zr0.05Ce and Zr0.1Ce nanorods were slightly decreased Figure 1. TEM images of CeO2, ZrxCe nanorods, and pure ZrO2.
to 11 and 9 nm, and the lengths were remarkably shortened
to 63 and 40 nm, respectively. However, the diameters and
lengths were slightly increased when the loading of Zr content
was further increased. For the Zr0.2Ce and Zr0.3Ce, the
measured mean diameters and lengths are 13 and 63 nm and
15 and 72 nm, respectively. Similar results were reported by
Liyanage and Yang et al.28,38 HRTEM images have identified
the crystallographic features of the ZrxCe nanorods. The lattice
fringes of 0.31 and 0.27 nm are attributed to the (111) and
(200) planes of CeO2 (JCPDS 34-0394), respectively. By con-
trast, ZrO2 shows nanoparticle morphology with a mean
particle size of 17 nm. The HRTEM image of ZrO2 sample
shows the lattice fringes of the ZrO2 (the d spacing of 0.29 nm
is due to (011) plane of the ZrO2).
3.2. XRD Analysis. Figure 2 shows the XRD patterns of the Figure 2. XRD patterns of the CeO2, ZrxCe nanorods, and pure ZrO2
prepared catalysts. The CeO2 showed intense and sharp dif- catalysts.
fraction peaks of cubic fluorite structure (JCPDS no. 34-0394).
The ZrxCe (x = 0.05, 0.1, 0.2, 0.3) catalysts exhibited broader peaks at 28.5°, 33.2°, 47.6°, 56.5°, 59.2°, and 69.4° were
peaks which shifted to higher 2θ, suggesting that Zr was suc- assigned to the (111), (200), (220), (311), (222) and (400)
cessfully doped into the ceria lattice structure. The diffraction crystal facets of CeO2. The calculated lattice parameter values for
10448 DOI: 10.1021/acscatal.8b00415
ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

Table 1. Interplanar Spacing, Lattice Parameters, Specific CeO2 lattice. The ID/IF2g values indicate the amounts of defects
Surface Area, and Pore Volume of the Catalysts in the catalysts.28,42 The ID/IF2g values of CeO2 and ZrxCe
catalysts are shown in Figure S1 in the Supporting Information,
interplanar lattice SBET Vpore
2θ spacing (Å) parameterb (m2/g) (cm3/g) which increased in the following order: CeO2 < Zr0.05Ce <
Zr0.1Ce < Zr0.2Ce < Zr0.3Ce, indicating that the amounts of
CeO2 28.64 3.11 5.39 75 0.51
defects in the catalysts increased with the Zr contents. No
Zr0.05Ce 28.76 3.10 5.38 83 0.65
Raman shifts of ZrO2 are observed in ZrxCe, which indicated
Zr0.1Ce 28.79 3.10 5.37 76 0.52
ZrO2 and CeO2 formed a solid solution in line with XRD
Zr0.2Ce 28.88 3.09 5.36 72 0.51
results. For the pure ZrO2, the bands at 146, 266, 319, 461, and
Zr0.3Ce 28.94 3.08 5.35 59 0.45
643 cm−1 could be assigned to tetragonal ZrO2.43,44
ZrO2 30.38 3.07 3.58 46 0.34
a
The introduction of Zr species into CeO2 increases the
Calculated by the (111) plane from XRD. bCalculated for ZrxCe absorbance in the visible light region (see the UV−vis spectra
catalysts by the cubic phase and for ZrO2 by tetragonal.
in Figure S2 in Supporting Information). The adsorption edge
of ZrxCe catalysts slightly shifted to the higher wavenumbers,
CeO2, Zr0.05Ce, Zr0.1Ce, Zr0.2Ce, Zr0.3Ce, and ZrO2, are 5.39, increasing the absorption intensity in the range of 200−500 nm.
5.38, 5.37, 5.36, 5.35, and 3.58 Å, respectively (Table 1). The Thus, the results indicate that the substitution of Ce4+ with Zr
lattice parameter decreased with the Zr content because the species facilitates the formation of Ce3+ ions, leading to a rise
ionic radius of Zr4+ (0.84 Å) is smaller than that of Ce4+ of oxygen vacancy concentration in Zr-doped CeO2 nanorods
(0.97 Å). These results confirmed the incorporation of Zr4+ through distorting the lattice structure.45
into the lattice of CeO2, resulting in the formation of a solid 3.4. Concentration of the Surface Oxygen Vacancies
solution accompanied by the shrinking interplanar spacing, and Chemical States of the Catalysts. Figure 4 shows the
which decreased from 3.11 to3.08 Å.37 The ZrO2 catalyst only XPS spectra of the catalysts. The characters V and U are
showed the tetragonal phase (t-ZrO2), which differs from the related to the spin−orbit splitting of Ce 3d5/2 and Ce 3d3/2.46
structure of pure CeO2 and ZrxCe nanorods.39 In Figure 4a, Ce 3d spectra can be deconvoluted into eight
The XRD patterns in Figure 2 and lattice parameters in peaks: U‴ (∼916.8 eV), U″ (∼907.4 eV), U′ (∼903.4 eV),
Table 1 agree with those disclosed by Liu et al.31 Figure 2 also U (∼900.9 eV), V‴ (∼898.3 eV), V″ (∼888.9 eV),
shows that ZrxCe gave broader peaks than CeO2, suggesting V′ (∼884.9 eV), and V (∼882.4 eV). The four U bands repre-
that Zr doping decreases the crystallite size of nanorods. sent the Ce 3d3/2, and the other V bands represent the Ce
3.3. Raman Spectra. The Raman spectra in Figure 3 3d5/2. The 3d104f 0 state of Ce4+ species is labeled as U‴, U″, U,
further confirmed the formation of ceria−zirconia solid V‴, V″, and V, whereas the 3d104f1 state of Ce3+ species is
labeled as U′ and V′. The V′ and U′ peaks can be used to
quantitatively estimate the concentration ratio of Ce3+ by
calculating the ratio between the area of V′, U′ peaks and the
total area of V, V″, V‴, U, U″, and U‴ peaks based the
following equation
A Ce3+
C[Ce3 +]% = × 100%
A Ce3+ + A Ce4+

A Ce3+ = AV′ + AU′ ...

A Ce4+ = AV + AV ″ + AV ‴ + AU + AU ″ + AU ‴

ACe3+: Photoelectron peaks areas of Ce3+. ACe4+: Photoelectron

peaks areas of Ce4+
Figure 3. Raman spectra of CeO2, ZrxCe, and ZrO2 catalysts. The percentages of Ce3+ to the total Ce (Cetotal) are 17.9%,
20.7%, 21.5%, 19.8%, and 18.7%, for the CeO2, Zr0.05Ce,
solutions. The bands at 462.5 and 596 cm−1 of CeO2 nanorods Zr0.1Ce, Zr0.2Ce, and Zr0.3Ce (Table 2)., respectively. The ratio
correspond to the symmetrical stretching F2g mode of Ce−O of Ce3+ to Cetotal decreased in the order Zr0.1Ce > Zr0.05Ce >
and oxygen vacancies, respectively.40 For the ZrxCe, the F2g Zr0.2Ce > Zr0.3Ce > CeO2. Furthermore, the Ce3+/Cetotal ratio
band broadened with the increase of Zr4+ content and grad- of ZrxCe catalysts is higher than the pure CeO2 sample. Oxygen
ually shifted to 462.6, 462.7, 462.8, and 462.9 cm−1 for Zr0.05Ce, vacancy associated exposed Ce3+ ions on CeO2 are potentially
Zr0.1Ce, Zr0.2Ce, and Zr0.3Ce, respectively. The substitution of potent surface sites in the reaction of DMC synthesis from CO2
Ce4+ with Zr4+ decreased the average length of the Ce−O bond, and methanol. These results implied that the addition of Zr
resulting in a shift to higher energy in the Raman spectra.29,31 promoted the formation of Ce3+ on the surface of the CeO2
This blue shift is due to the variation of interatomic force in catalyst.47 The transformation of Ce4+ (0.97 Å) in CeO2 to a
the ceria crystals, which is influenced by the variation of bond larger Ce3+ (1.10 Å) could compensate for lattice contrac-
length as well as the lattice spacing as evidenced by XRD tion induced by the slightly smaller radius of Zr4+ (0.84 Å).48,49
results. It is noted that the intensity of F2g decreases with the It has been proved that the existence of Ce3+ could generate
increase of x value, revealing structural distortion due to the oxygen vacancies on the catalyst surface.50 Thus, the con-
oxygen vacancies.41 The weak bands at 254 and 596 cm−1 were centration of Ce3+ increased with the concentration of oxygen
assigned to second-order transverse acoustic (2TA) mode and vacancies. The Zr0.1Ce catalyst has the highest Ce3+/Cetotal
detect induced (D), respectively. The band at 596 cm−1 is ratio than other ZrxCe catalysts; hence, it has the highest
related to the oxygen vacancies caused by the Ce3+ ion in the concentration oxygen vacancies.
10449 DOI: 10.1021/acscatal.8b00415
ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

Figure 4. (a) Ce 3d and (b) O 1s XPS spectra of CeO2, ZrxCe nanorods, and ZrO2.

Table 2. Relative Ratios of the Surface Concentration of ZrxCe catalysts exhibit two Zr 3d peaks at 181.8 and 184.2 eV
Ce3+ and Oxygen Vacancy Based on XPS Analysis that could be related to the Zr 3d3/2 and 3d5/2 spin−orbit
splitting. The binding energy of Zr 3d in the ZrxCe catalysts
catalyst CeO2 Zr0.05Ce Zr0.1Ce Zr0.2Ce Zr0.3Ce ZrO2
(182.1 and 184.5 eV) was higher than that in ZrO2.52 These
surface content of 17.9 20.7 21.5 19.8 18.7 results further confirmed that Zr4+ entered into the CeO2
Ce3+
lattice and formed the solid solutions, as shown by the results
surface content of 7.1 9.4 10.4 8.8 7.7 6.0
OV (%) of XRD and TEM.49,52
bulk atomic ratio 0.039 0.13 0.24 0.37 3.5. TPD Analysis of the Catalysts. It has been reported
Zr/Ce (%)a that the surface oxygen vacancy acts as Lewis acid sites and
surface atomic ratio 0.059 0.11 0.27 0.47 plays a key role in the adsorption of CO2.13 Thus, we further
Zr/Ce (%)b
a
studied the nature of the catalyst surface by CO2-TPD. The
Determined by ICP. bDetermined by XPS. results are shown in Figure 5. The amount of CO2 desorbed

Table 2 shows the concentration of Ce3+ and oxygen

vacancy increased with increasing Zr content to a level and
then decreased with further increase in Zr content. This
decrease may be explained by the aggregation of Zr on the
surface of the catalysts.31 Table 2 also lists the surface and bulk
Zr/Ce of the catalysts, determined by XPS and ICP. The
surface Zr/Ce atomic ratios are slightly higher than the bulk
ratio, further indicating aggregation of Zr on the surface of the
catalysts. Aggregation of Zr on the surface of Zr0.3Ce catalysts
is also evidenced by its low surface area as compared with
those of Zr0.05Ce, Zr0.1Ce, and Zr0.2Ce.
The O 1s spectra of the catalysts in Figure 4b showed three
peaks: (i) lattice oxygen (OL) in the CeO2 lattice at ∼529.0 eV),
(ii) the O component associated with the O2− ions in surface Figure 5. CO2-TPD profile of CeO2, ZrxCe nanorods, and ZrO2.
oxygen vacancies (OV) at ∼530.5 eV, and (iii) chemisorbed
oxygen species (OC) at ∼532.5 eV.51 The concentration of from the catalysts is calculated from the TPD peak area and
surface oxygen vacancies of the catalysts can be estimated by listed in Figure 5. The CO2 desorption peaks appeared in the
the integrated peak areas using the following equation range of 50−200, 200−400, and 400−600 °C, which varied
A OV with the Zr/Ce ratios. The amount of CO2 desorbed from
C[O V ]% = × 100% these sites is calculated and displayed in Table S3. In Figure 5
A OL + A OV + A OC we can see the intensity of the peak below 200 °C increases
and then decreases with higher Zr content. The total amount
AOL: Photoelectron peaks areas of lattice oxygen. AOV:
of adsorbed CO2 of the ZrxCe nanorods was higher than those of
photoelectron peaks areas of surface oxygen vacancies. AOC: CeO2 and ZrO2 catalysts. The Zr0.1Ce gave the highest adsorp-
Photoelectron peaks areas of chemisorbed oxygen tion capacity of CO2 at 0.96 mmol/g−1 among these catalysts.
Table 2 shows the total concentration of OV is 7.1%, 9.4%, The amount of adsorbed CO2 on these catalysts showed a
10.4%, 8.8%, 7.7%, and 6.0%, corresponding to pure CeO2, volcano-shaped curve with respect to Zr content and followed
Zr0.05Ce, Zr0.1Ce, Zr0.2Ce, Zr0.3Ce, and ZrO2, respectively. The the order Zr0.1Ce > Zr0.05Ce > Zr0.2Ce > Zr0.3Ce > CeO2 > ZrO2.
concentrations of OV agree well with the results obtained from It is noted that the order of CO2 adsorption ability of the catalysts
the Ce 3d spectra, which indicated that surface oxygen vacan- is consistent with the variations in the concentration of surface
cies were enhanced by doping the Zr into CeO2 nanorods. oxygen vacancies determined by XPS in Table 2. In Table S3 it
Doping Zr into CeO2 lattice promoted the formation of surface can be noted that the weak and moderate adsorption sites of the
oxygen defects/vacancies.31 ZrxCe catalysts increased with increasing Zr content and decreased
The Zr 3d XPS spectra of ZrxCe nanorods and ZrO2 are with higher Zr content. The Zr0.1Ce nanorods possess the largest
shown in Figure S3 in the Supporting Information. ZrO2 and amount of weak and moderate adsorption sites.
10450 DOI: 10.1021/acscatal.8b00415
ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

The surface acid properties of the catalysts were investigated increasing Zr/Ce ratios. In Figure S6 the yields increased from
by NH3-TPD, and the results are exhibited in Figure S5. The 7.2 to 12.1 mmol/g for CeO2 and from 10.2 to 14.9 mmol/g
amount of NH3 of all of the catalysts is calculated from the for Zr0.1Ce with increasing reaction time. It is also noted that
TPD peak area and also displayed in Figure S5. It is observed further increasing the reaction time has not exhibited an
from FigureS5 that the CeO2, ZrxCe nanorods, and ZrO2 obvious increase of the yields for the Zr0.1Ce catalyst. These
catalyst contained the same types of acidic sites: the desorption results indicated that the reaction may reach a plateau state
peaks in the range of 50−200, 200−400, and 400−600 °C can after 6 h reaction time for Zr0.1Ce catalyst because of the
be assigned to weak, moderate, and strong adsorption, respec- accumulation of water.
tively. The amount of NH3 desorbed from these sites is cal- According to the reaction equation of the DMC synthesis
culated and summarized in Table S3. It is noted that the acid from carbon dioxide and methanol, the thermodynamic infor-
properties of the catalysts were strongly affected after doping mation about this reaction was calculated by the thermody-
Zr into CeO2. The Zr0.1Ce has the highest acidic sites density namic data of various substances.55 The ΔGΘ298K and ΔHΘ298K
compared to other samples, and the adsorption quantity of values are 26.21 and −27.90 kJ/mol, respectively. The ΔG of
NH3 is 0.54 mmol/g−1, which indicated that Zr0.1Ce sample this reaction in our reaction conditions is about 32.95 kJ/mol.
has the largest number of acidic sites. The amount of adsorp- It can be seen that the Gibbs free energy of this reaction at
tion NH3 for all catalysts showed a volcano-shaped curve with ambient temperature is higher than 0, which indicated that this
respect to Zr content and followed the order Zr0.1Ce > reaction does not occur spontaneously at 413 K. However, we
Zr0.05Ce > Zr0.2Ce > CeO2 > Zr0.3Ce > ZrO2. It has to be can see from the estimated result that the ΔG (413 k, 6.8 MPa)
emphasized that the total acidity of ZrxCe nanorods catalysts is smaller than ΔGΘ413K, which indicates that an increase in the
was higher than pure CeO2 and ZrO2 catalysts (except Zr0.3Ce), pressure of the reaction is in favor of the formation of the DMC.
which may be ascribed to a synergic effect between ZrO2 and We also found that ΔG (413 k, 6.8 MPa) is still larger than
CeO2. At the same time, the findings also showed that the ΔGΘ298K; this is due to ΔHΘ413K < 0; hence, the higher temper-
acidity of Zr0.3Ce nanorods was lower than the acidity of CeO2 ature is not in favor of the formation of the DMC.
nanorods. It may be due to the BET surface and pore volume The stability is one of the most important concerns of the
of Zr0.3Ce nanorods (59 m2/g) being lower than pure CeO2 heterogeneous catalyst. The reusability of the catalysts in the
nanorods (74 m2/g) and exposed less acidic sites. Similar to DMC synthesis was tested with six reaction cycles under identical
CO2-TPD results, for all ZrxCe catalysts, weak and moderate reaction as mentioned in the Experimental Section. The results
acidic sites increased first and then decreased with higher Zr are shown in Figure S7 in the Supporting Information. The
content. It can be found that the Zr0.1Ce nanorods possess the activity from these catalysts decreased slightly after six cycles.
largest amount of acidic sites. By doping oxides the surface However, the activity of Zr0.1Ce remained the highest among all
acid−base sites of CeO2 will be varied and will bring about the samples. The activities of the catalysts were reduced by 21%, 24%,
performance differences; similar results were obtained by pre- 22%, 26%, 19%, and 68% after six cycle times corresponding to
vious reports.53,54 CeO2, Zr0.05Ce, Zr0.1Ce, Zr0.2Ce, Zr0.3Ce, and ZrO2, respectively.
3.6. Catalytic Activity. Figure 6 shows the catalyst per- 3.7. Characterization of the Used Catalyst. To deter-
formance over the catalysts. The ZrxCe nanorods gave higher mine the deactivation mechanism, we compared the TEM,
XRD, and XPS results of the fresh and used Zr0.1Ce nanorods.
The TEM results and the XRD patterns of the fresh Zr0.1Ce
and used Zr0.1Ce nanorods are shown in Figure S8 in the
Supporting Information. The TEM images of the morpholo-
gies of Zr0.1Ce nanorods showed that the fresh and used
Zr0.1Ce have the same morphologies after six cycles. However,
the used Zr0.1Ce nanorods had a mean diameter of ∼10 nm
and an average length of ∼46 nm; it appears that the used
Zr0.1Ce nanorods become a little wider and longer than fresh
Zr0.1Ce nanorods (∼9 and ∼40 nm). After six cycles, the used
Zr0.1Ce catalysts still display the characteristic diffraction peaks
corresponding to the cubic fluorite structure, which is fully
consistent with the TEM results.
Figure 7a and 7b compares the deconvoluted Ce 3d and
Figure 6. Catalytic performance of the CeO2, ZrxCe nanorods and O 1s XPS spectra of fresh and used Zr0.1Ce. The relative
ZrO2 catalysts. Methanol conversion = 2nDMC/nCH3OH × 100%. Error abundances of the Ce3+ and Ov of the samples, listed in Figure 7,
bars represent one standard deviation (n = 5). showed that the content of Ce3+ and the surface oxygen
vacancy decreased from 21.5% to 17.4% and from 10.4% to
activity than CeO2 and ZrO2. The yield reached a maximum 7.1%, respectively. This decrease could be attributed to partial
value of 14.2 mmol/g at Zr/Ce = 0.1 and then decreased with filling of surface oxygen vacancy with the O atom of adsorbed
further increasing Zr/Ce ratios. The activity of Zr0.1Ce is 1.3 CO2.56 A number of DFT studies have indicated that the sur-
and 6.7 times higher than that of the pure CeO2 and ZrO2, face oxygen vacancy is thermodynamically unstable and highly
respectively. The activity of the catalysts decreased in the order reactive; experimental studies also showed that the surface
Zr0.1Ce > Zr0.05Ce > Zr0.2Ce > Zr0.3Ce > CeO2 > ZrO2, which oxygen vacancy can interact with O from adsorbed CO2.57,58
followed the trend of the surface oxygen vacancy concen- 3.8. Effect of the Surface Vacancy Concentrations on
trations and the capacity of CO2 adsorption. Meanwhile, it can the CO2 Adsorption and Activity of Catalysts. The
be observed that the methanol conversion reached a maximum surface oxygen vacancy on the catalyst surface could serve as
value of 0.65% at Zr/Ce = 0.1 and then decreased with further Lewis acid sites that promote CO2 adsorption.13,33 Thus, we
10451 DOI: 10.1021/acscatal.8b00415
ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

Figure 7. XPS spectra of the fresh and used Zr0.1Ce catalyst.

further investigated the relationship between the surface because the surface of CeO2 nanorods is more defective.
oxygen vacancy, CO2 adsorption capacity, and catalyst activity In CO2 methanation,15,61 some authors proposed that the sur-
in Figure 8. The adsorption capacity of CO2 of the catalysts face oxygen vacancy played an important role in the reduction
of CO2 to CO and proved that the surface oxygen vacancy cata-
lyzed the rate-determining step with a much lower activation
temperature. According to the previous reports and the obtained
results in this work, doped Zr into CeO2 lattice resulted in more
oxygen vacancies on the CeO2 surface serving as the active sites,
which can promote the interaction between CO2 and ZrxCe
nanorods.
3.9. In Situ FTIR Study. We have further employed in situ
FTIR to characterize the structure of adsorbed species and the
nature of the active sites. Figure 9a shows that CO2 adsorption
on CeO2 produced carbonate and bicarbonate.64,65
Zr doping decreased the IR intensity of bidentate carbonate.
Bidentate carbonate and bidentate bicarbonate can be assigned
to CO2 adsorbed on the surface oxygen vacancy sites. CO2 may
adsorb on the surface oxygen vacancy site through inserting
Figure 8. Relationship between catalyst activity, CO2 adsorption one O atom in the vacancy site on the surface of the catalysts.
amount, and concentration of the surface oxygen vacancy. The C atoms bind the O atom of the catalyst surface to form
the carbonate. A number DFT studies have demonstrated that
increased with the surface oxygen vacancy concentrations, the surface oxygen vacancy assists CO2 activation and adsorp-
indicating that surface oxygen vacancies promote CO2 adsorp- tion.57,59,66 This proposition is further supported by our XPS
tion. Oxygen vacancies can be associated with Ce3+. XRD, and CO2-TPD results.
Raman, and XPS results in Figures 2, 3, and 4 showed that Zr Figure 9b shows that the adsorption of methanol on CeO2
doping led to the formation of Ce3+ in the fluorite structure of produced terminal methoxy (1106 cm−1) and bridged methoxy
CeO2. The Lewis acid nature of oxygen vacancy allowed it (1052 cm−1) on Ce4+. Zr doping produced a methoxy
to interact with nonbonding electrons in the O atom of CO2 (1156 cm−1) on Zr4+. The formation of these positive methoxy
through Lewis acid−base interactions.13 A number of theo- bands is accompanied by an increase in the H2O band in the
retical and experimental studies suggest that the surface oxygen 3300−3400 cm−1 region and the negative OH band at 3674 cm−1.
vacancy allows inserting one of the O atoms of CO2 into the This observation has also been reported by former studies on
oxygen vacancy, forming a bent CO2− intermediate.59,60 Marco the interaction of alcohol with the surface hydroxyl groups of
Fronzi et al.60 suggested that a higher energy is needed to form metal oxides.67,68 This interaction allowed the transfer of the
an oxygen vacancy reflecting a higher reactivity of a reduced H atom from methanol’s OH to a coordinately unsaturated
nanocluster toward CO2. Our results showed ZrxCe nanorods O2− center on the catalyst surface, producing a bridging OH
with increasing surface oxygen vacancy concentration exhibited group. The H in methanol’s OH could also react the surface
a propensity to increase CO2 adsorption. OH group to produce surface water (Figure S9).69 Accordingly,
Figure 8 also shows that the catalytic activity of ZrxCe the whole process is summarized in Scheme S1.
catalysts increased with the surface oxygen vacancy concen- Because of its high activity, we have further studied the
tration. Among these catalysts, the Zr0.1Ce nanorods which interaction of CO2 with adsorbed methoxy species on Zr0.1Ce,
exhibited the highest activity among the investigated samples shown in Figure 10a. Exposure of adsorbed methoxy species
possessed the highest concentration of surface oxygen vacancy. to CO2 resulted in a rapid decrease in a terminal methoxy at
Oxygen vacancies promote the direct DMC synthesis from 1106 cm−1 and the formation of monomethyl carbonate. The
CO2/methanol. On the basis of the previous reports,61,62 CO2 IR spectra of monomethyl carbonate were further verified by
could be adsorbed and activated on the oxygen vacancy sites the adsorption of DMC on the catalysts in Figure 10b. Mono-
over the defect surface of metal oxide. In DFT studies, the methyl carbonate has been considered to be the reaction inter-
CO bond cleavage of CO2 is achieved by using one of the mediate of direct synthesis of DMC from CO2 and methanol.
O atoms to fill the oxygen vacancy with a small barrier, which Adsorption of DMC on Zr0.1Ce produced not only monomethyl
is easier than that on the metal surface.60 Wu et al.63 studied carbonate species but also adsorbed methoxy species. Adsorbed
the CeO2 nanocrystals with different morphology (rod, cube, DMC was not detected because the low-pressure condition used
octahedral) and found CeO2 nanorods bind CO2 strongly favors its conversion to monomethyl carbonate.69
10452 DOI: 10.1021/acscatal.8b00415
ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

Figure 10. DRIFTS spectra of adsorbed species on Zr0.1Ce nanorods:

(a) Interaction of CO2 with adsorbed methoxy and (b) adsorbed
DMC.
Figure 9. DRIFTS IR spectra of adsorbed species on CeO2 and ZrxCe
nanorods: (a) adsorbed CO2 and (b) adsorbed methanol. Scheme 2. Proposed Mechanism of DMC Synthesis on the
ZrxCe Nanorods
3.10. Proposed Mechanism for DMC Formation over
the Surfaces of ZrxCe Nanorods. Scheme 2 illustrates the
proposed mechanism for the DMC synthesis which begins
with the activation of adsorbed CO2 by an oxygen vacancy site
through a Lewis acid−base interaction. It should be noted that
the linear OH in the vicinity of oxygen vacancy site is a
hypothetic species which will be further investigated by in situ
IR study. DFT studies have suggested that the oxygen vacancy
site could locate between Ce and Zr atoms.70 Zr0.3Ce produced
a high intensity of terminal methoxy, as shown in Figure 9b for
the insertion into adsorbed methanol. The abundance of
terminal methoxy could be attributed to the high specific
activity (i.e., catalyst activity on the basis of oxygen vacancy
sites) of Zr0.3Ce when compared with that of Zr0.1Ce. Thus,
the DMC activity is governed by the abundance of adsorbed
active species and surface oxygen vacancy sites. acid−base pair sites plays a key role in the direct synthesis of
The surface oxygen vacancy promoted CO2 adsorption and DMC.3,69,72 Activated CO2 could react with the terminal
activation of a number of CO2-related reactions.15,59,71 The CH3O−M/CH3−M bond in metal oxides to produce methyl
XPS and CO2-TPD results show that doping Zr into the CeO2 carbonate species, as evidenced by a decrease in terminal
nanorods promotes the generation of the surface oxygen CH3O−M at 1106 cm−1 and an increase in methyl carbonate
vacancy. The oxygen vacancy sites allow CO2 to adsorb as in Figure 10a. Subsequent reaction of methyl carbonate with
bidentate carbonate as evidenced by the presence of IR bands CH3 from adsorbed methoxy should lead to DMC with regen-
at 1018 and 1288 cm−1 (Figure 9a). At a neighboring acidic site, eration of oxygen vacancy.
methanol may adsorb as methyl and methoxy species, respec- According to the refs 12 and 72 ,elementary mechanistic
tively. A number of studies have suggested that the existence of steps for the formation of DMC from CO2 and methanol were
10453 DOI: 10.1021/acscatal.8b00415
ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

proposed by the Langmuir−Hinshelwood (LH) and the Eley− ORCID

Rideal (ER) mechanism. Some studies revealed that the CO2 Steven S. C. Chuang: 0000-0002-6279-0584
and methanol adsorbed on the surface of the CeO2 nanorods Zhong Li: 0000-0001-6087-6854
in two separate steps, which are consistent with the Langmuir−
Hinshelwood mechanism; meanwhile, the adsorption and the Notes
activation of CO2 is the rate-determining step.53,73Determining The authors declare no competing financial interest.
the dynamic behavior of CO2 in our reaction needs further
work in the future, such as the study of the initial formation
rate of DMC with different pressures of CO2 and the effect of
■ ACKNOWLEDGMENTS
This work was supported by the National Natural Science
the amount of the methanol on the DMC yield. In the present
Foundation of China (U1510203).

■
work, based on the XPS, CO2-TPD, and FTIR results, it can be
found that CO2 and methanol interact with the surface oxygen
vacancy and the surface hydroxyl, respectively; this is con- REFERENCES
sistent with the LH mechanism. Moreover, there existed a (1) Keller, N.; Rebmann, G.; Keller, V. Catalysts, Mechanisms and
linear relationship between the DMC yield of the catalysts and Industrial Processes for the Dimethyl Carbonate Synthesis. J. Mol.
the adsorption capacity of the CO2. Hence, we considered that Catal. A: Chem. 2010, 317, 1−18.
the steps of the adsorption and activation of the CO2 may be (2) Tamboli, A. H.; Chaugule, A. A.; Kim, H. Catalytic Develop-
rate-determining steps of the direct synthesis of the DMC from ments in the Direct Dimethyl Carbonate Synthesis from Carbon
CO2 and methanol. Dioxide and Methanol. Chem. Eng. J. 2017, 323, 530−544.
Alternative proposed mechanisms for DMC formation also (3) Tomishige, K.; Sakaihori, T.; Ikeda, Y.; Fujimoto, K. A Novel
Method of Direct Synthesis of Dimethyl Carbonate from Methanol
involved the surface oxygen vacancy over the Ce0.1Ti0.9O2 and
and Carbon Dioxide Catalyzed by Zirconia. Catal. Lett. 1999, 58,
H3PW12O40/Ce0.1Ti0.9O2 catalysts.32,33 The surface oxygen 225−229.
vacancies could act as Lewis acid sites to interact with the (4) Zhang, Z. F.; Liu, Z. W.; Liu, Z.-W.; Lu, J. DMC Formation over
O atom of CO2. Methanol could adsorb at adjacent oxygen Ce0.5Zr0.5O2 Prepared by Complex-Decomposition Method. Catal.
vacancy to produce the intermediate which on further reaction Lett. 2009, 129, 428−436.
adsorbed CO2 to form DMC and then regenerates oxygen (5) La, K. W.; Jung, J. C.; Kim, H.; Baeck, S. H.; Song, I. K. Effect of
vacancy Acid-Base Properties of H3PW12O40/CexTi1‑xO2 Catalysts on the
It is important to note that methyl carbonate which has been Direct Synthesis of Dimethyl Carbonate from Methanol and Carbon
produced from methanol and CO2 can be also produced from Dioxide: A TPD Study of H3PW12O40/CexTi1‑xO2 Catalysts. J. Mol.
DMC, as shown in Figure 10b. This observation further Catal. A: Chem. 2007, 269, 41−45.
confirms the microreversibility of the steps involved in the (6) Zhou, Y. J.; Wang, S. J.; Xiao, M. Novel Cu-Fe Bimetal Catalyst
proposed mechanism in Scheme 2. for the Formation of Dimethyl Carbonate from Carbon Dioxide and
Methanol. RSC Adv. 2012, 2, 6831−6837.
4. CONCLUSION (7) La, K. W.; Youn, M. H.; Chung, J. S.; Baeck, S. H.; Song, I. K.
Synthesis of Dimethyl Carbonate from Methanol and Carbon Dioxide
In summary, a series of Zr-doped CeO2 nanorods was by Heteropolyacid/Metal Oxide Catalysts. Solid State Phenom. 2007,
synthesized via a hydrothermal method, and the effect of the 119, 287−290.
doping content of Zr on the lattice structure, microstructure, (8) Bansode, A.; Urakawa, A. Continuous DMC Synthesis from CO2
especially the amount of oxygen vacancy, as well as the cata- and Methanol over a CeO2 Catalyst in a Fixed Bed Reactor in the
lytic activity of DMC synthesis from CO2 and methanol were Presence of a Dehydrating Agent. ACS Catal. 2014, 4, 3877−3880.
studied in detail. Zr0.1Ce nanorods which gave the highest (9) Saada, R.; Kellici, S.; Heil, T.; Morgan, D.; Saha, B. Greener
amount of oxygen vacancies exhibited the highest activity for Synthesis of Dimethyl Carbonate using a Novel Ceria-Zirconia
DMC synthesis. The DMC synthesis activities were found Oxide/Graphene Nanocomposite Catalyst. Appl. Catal., B 2015,
to be correlated with the concentration of the surface oxygen 168−169, 353−362.
(10) Kumar, P.; With, P.; Srivastava, V. C.; Gläser, R.; Mishra, I. M.
vacancy. FTIR results suggest that bidentate carbonate and
Efficient Ceria-Zirconium Oxide Catalyst for Carbon Dioxide
terminal methoxy species are actively involved in DMC syn- Conversions: Characterization, Catalytic Activity and Thermody-
thesis. The results of this study provide insight into the effect namic Study. J. Alloys Compd. 2017, 696, 718−726.
of Zr doping on DMC synthesis on CeO2 nanorods catalysts (11) Honda, M.; Tamura, M.; Nakagawa, Y.; Nakao, K.; Suzuki, K.;
and provide a technical basis for devising a new strategy to Tomishige, K. Organic Carbonate Synthesis from CO2 and Alcohol
design high-efficiency CeO2-based catalysts.

■
over CeO2 with 2-cyanopyridine: Scope and Mechanistic Studies. J.
Catal. 2014, 318, 95−107.
ASSOCIATED CONTENT (12) Santos, B. A. V.; Pereira, C. S. M.; Silva, V.; Loureiro, J. M.;
*
S Supporting Information Rodrigues, A. E. Kinetic Study for the Direct Synthesis of Dimethyl
The Supporting Information is available free of charge on the Carbonate from Methanol and CO2 over CeO2 at High Pressure
Conditions. Appl. Catal., A 2013, 455, 219−226.
ACS Publications website at DOI: 10.1021/acscatal.8b00415.
(13) Xin, C.; Hu, M.; Wang, K.; Wang, X. Significant Enhancement
Supporting figures and tables related to the additional of Photocatalytic Reduction of CO2 with H2O over ZnO by the
UV−vis spectra, XPS, NH3-TPD, TEM, and XRD data Formation of Basic Zinc Carbonate. Langmuir 2017, 33, 6667−6676.
(PDF) (14) Zhou, G.; Liu, H.; Cui, K.; Jia, A.; Hu, G.; Jiao, Z.; Liu, Y.;

■
Zhang, X. Role of Surface Ni and Ce Species of Ni/CeO2 Catalyst in
CO2 Methanation. Appl. Surf. Sci. 2016, 383, 248−252.
AUTHOR INFORMATION (15) Wang, F.; He, S.; Chen, H.; Wang, B.; Zheng, L.; Wei, M.;
Corresponding Authors Evans, D. G.; Duan, X. Active Site Dependent Reaction Mechanism
*E-mail: [email protected]. over Ru/CeO2 Catalyst toward CO2 Methanation. J. Am. Chem. Soc.
*E-mail: [email protected]. 2016, 138, 6298−6305.

10454 DOI: 10.1021/acscatal.8b00415

ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

(16) Wang, F.; Li, C.; Zhang, X.; Wei, M.; Evans, D. G.; Duan, X. (34) Mai, H. X.; Sun, L. D.; Zhang, Y. W. Shape-Selective Synthesis
Catalytic Behavior of Supported Ru Nanoparticles on the {100}, and Oxygen Storage Behavior of Ceria Nanopolyhedra, Nanorods. J.
{110}, and {111} Facet of CeO2. J. Catal. 2015, 329, 177−186. Phys. Chem. B 2005, 109, 24380−24385.
(17) Grabowski, R.; Słoczyński, J.; Sliwa, M.; Mucha, D. Influence of (35) Yu, J.; Chuang, S. S. C. The Structure of Adsorbed Species on
Polymorphic ZrO2 Phases and the Silver Electronic State on the Immobilized Amines in CO2 Capture: An in Situ IR Study. Energy
Activity of Ag/ZrO2 Catalysts in the Hydrogenation of CO2 to Fuels 2016, 30, 7579−7587.
Methanol. ACS Catal. 2011, 1, 266−278. (36) Isenberg, M.; Chuang, S. S. C. The Nature of Adsorbed CO2
(18) Samson, K.; Śliwa, M.; Socha, R. P.; Góra-Marek, K.; Mucha, and Amine Sites on the Immobilized Amine Sorbents Regenerated by
D.; Rutkowska-Zbik, D.; Paul, J. F.; Ruggiero-Mikołajczyk, M.; Industrial Boiler Steam. Ind. Eng. Chem. Res. 2013, 52, 12530−12539.
Grabowski, R.; Słoczyński, J. Influence of ZrO2 Structure and Copper (37) Yu, J.; Chuang, S. S. C. The Role of Water in CO2 Capture by
Electronic State on Activity of Cu/ZrO2 Catalysts in Methanol Amine. Ind. Eng. Chem. Res. 2017, 56, 6337−6347.
Synthesis from CO2. ACS Catal. 2014, 4, 3730−3741. (38) Yang, D.; Wang, L.; Sun, Y. Z.; Zhou, K. Synthesis of One-
(19) Huygh, S.; Bogaerts, A.; Neyts, E. C. How Oxygen Vacancies Dimensional Ce1‑xYxO2‑x/2 (0 < x < 1) Solid Solutions and Their
Activate CO2 Dissociation on TiO2 Anatase (001). J. Phys. Chem. C Catalytic Properties: The Role of Oxygen Vacancies. J. Phys. Chem. C
2016, 120, 21659−21669. 2010, 114, 8926−8932.
(20) Liu, B.; Li, C. M.; Zhang, G. Q.; Yan, L. F.; Li, Z. Direct (39) Lara-García, H. A.; Romero-Ibarra, I. C.; Pfeiffer, H.
Synthesis of Dimethyl Carbonate from CO2 and Methanol over CaO- Hierarchical Na-Doped Cubic ZrO2 Synthesis by a Simple Hydro-
CeO2 Catalysts: The Role of Acid-Base Properties and Surface thermal Route and its Application in Biodiesel Production. J. Solid
Oxygen Vacancies. New J. Chem. 2017, 41, 12231−12240. State Chem. 2014, 218, 213−220.
(21) Imagawa, H.; Suda, A.; Yamamura, K.; Sun, S. Monodisperse (40) Wang, R. G.; Mutinda, S. I.; Fang, M. H. One-pot hydrothermal
CeO2 Nanoparticles and Their Oxygen Storage and Release synthesis and high temperature thermal stability of CexZr1‑xO2
Properties. J. Phys. Chem. C 2011, 115, 1740−1745. nanocrystals. RSC Adv. 2013, 3, 19508−19514.
(22) Liu, L.; Cao, Y.; Sun, W.; Yao, Z.; Liu, B.; Gao, F.; Dong, L. (41) Li, X.; Ni, C.; Yao, C.; Chen, Z. Development of Attapulgite/
Morphology and Nanosize Effects of Ceria from Different Precursors Ce1‑xZrxO2 Nanocomposite as Catalyst for the Degradation of
on the Activity for NO Reduction. Catal. Today 2011, 175, 48−54. Methylene Blue. Appl. Catal., B 2012, 117−118, 118−124.
(23) Mann, A. K. P.; Wu, Z.; Calaza, F. C.; Overbury, S. H. (42) Gao, R.; Zhang, D.; Maitarad, P.; Shi, L.; Rungrotmongkol, T.;
Adsorption and Reaction of Acetaldehyde on Shape-Controlled CeO2 Li, H.; Zhang, J.; Cao, W. Morphology-Dependent Properties of
Nanocrystals: Elucidation of Structure-Function Relationships. ACS MnOx/ZrO2-CeO2 Nanostructures for the Selective Catalytic
Catal. 2014, 4, 2437−2448. Reduction of NO with NH3. J. Phys. Chem. C 2013, 117, 10502−
(24) Zhou, F.; Zhao, X. M.; Xu, H.; Yuan, C. G. CeO2 Spherical 10511.
Crystallites: Synthesis, Formation Mechanism, Size Control, and (43) Si, R.; Zhang, Y. W.; Li, S.-J.; Lin, B.-X.; Yan, C. H. Urea-Based
Electrochemical Property Study. J. Phys. Chem. C 2007, 111, 1651− Hydrothermally Derived Homogeneous Nanostructured Ce1‑xZrxO2
1657. (x = 0−0.8) Solid Solutions: A Strong Correlation between Oxygen
(25) Wu, Z.; Li, M.; Howe, J.; Meyer, H. M.; Overbury, S. H. Storage Capacity and Lattice Strain. J. Phys. Chem. B 2004, 108,
Probing Defect Sites on CeO2 Nanocrystals with Well-defined Surface 12481−12488.
Planes by Raman Spectroscopy and O2 Adsorption. Langmuir 2010, (44) Fernández López, E.; Sánchez Escribano, V.; Panizza, M.;
26, 16595−16606. Carnasciali, M. M.; Busca, G. Vibrational and Electronic Spectro-
(26) Wu, Z.; Li, M.; Overbury, S. H. On the Structure Dependence scopic Properties of Zirconia Powders. J. Mater. Chem. 2001, 11,
of CO Oxidation over CeO2 Nanocrystals with Well-defined Surface 1891−1897.
Planes. J. Catal. 2012, 285, 61−73. (45) Wang, Y.; Li, B.; Zhang, C.; Cui, L.; Kang, S.; Li, X.; Zhou, L.
(27) Wang, S. P.; Zhou, J. J.; Zhao, S. Y.; Zhao, Y. J.; Ma, X. B. Ordered Mesoporous CeO2-TiO2 Composites: Highly Efficient
Enhancements of Dimethyl Carbonate Synthesis from Methanol and Photocatalysts for the Reduction of CO2 with H2O under Simulated
Carbon Dioxide: The in Situ Hydrolysis of 2-Cyanopyridine and Solar Irradiation. Appl. Catal., B 2013, 130−131, 277−284.
Crystal Face Effect of Ceria. Chin. Chem. Lett. 2015, 26, 1096−1100. (46) Liao, X.; Chu, W.; Dai, X.; Pitchon, V. Bimetallic Au-Cu
(28) Liyanage, A. D.; Perera, S. D.; Tan, K.; Chabal, Y.; Balkus, K. J. Supported on Ceria for PROX Reaction: Effects of Cu/Au Atomic
Synthesis, Characterization, and Photocatalytic Activity of Y-Doped Ratios and Thermal Pretreatments. Appl. Catal., B 2013, 142−143,
CeO2 Nanorods. ACS Catal. 2014, 4, 577−584. 25−37.
(29) Fei, Z.; Xie, X.; Dai, Y.; Liu, H.; Chen, X.; Tang, J.; Cui, M.; (47) Silva, L. P. C.; Terra, L. E.; Coutinho, A. C. S. L. S.; Passos, F.
Qiao, X. HCl Oxidation for Sustainable Cl2 Recycle over the B. Sour Water-Gas Shift Reaction over Pt/CeZrO2 Catalysts. J. Catal.
CexZr1−xO2Catalysts: Effects of Ce/Zr Ratio on Activity and Stability. 2016, 341, 1−12.
Ind. Eng. Chem. Res. 2014, 53, 19438−19445. (48) Nagai, Y.; Yamamoto, T.; Tanaka, T.; Yoshida, S.; Nonaka, T.;
(30) Chen, G.; Xu, Q.; Yang, Y.; Li, C.; Huang, T.; Sun, G.; Zhang, Okamoto, T.; Suda, A.; Sugiura, M. X-ray Absorption Fine Structure
S.; Ma, D.; Li, X. A Facile and Mild Strategy to Construct Analysis of Local Structure of CeO2-ZrO2 Mixed Oxides with the
Mesoporous CeO2-CuO Nanorods with Enhanced Catalytic Activity Same Composition Ratio (Ce-Zr = 1). Catal. Today 2002, 74, 225−
toward CO Oxidation. ACS Appl. Mater. Interfaces 2015, 7, 23538− 234.
23544. (49) Cai, W.; Zhong, Q.; Zhao, W. Solvent Effects on Formation of
(31) Liu, X.; Ding, J.; Lin, X.; Gao, R. H.; Li, Z.; Dai, W. L. Zr-doped Cr-Doped Ce0.2Zr0.8O2 Synthesized with Cinnamic Acid and their
CeO2 Nanorods as Versatile Catalyst in the Epoxidation of Styrene Catalysis in Oxidation of NO. Chem. Eng. J. 2014, 246, 328−336.
with Tert-butyl Hydroperoxide as the Oxidant. Appl. Catal., A 2015, (50) Zhang, Y.; Yuwono, A. H.; Wang, J.; Li, J. Enhanced
503, 117−123. Photocatalysis by Doping Cerium into Mesoporous Titania Thin
(32) Wada, S.; Oka, K.; Watanabe, K.; Izumi, Y. Catalytic Films. J. Phys. Chem. C 2009, 113, 21406−21412.
Conversion of Carbon Dioxide into Dimethyl Carbonate using (51) Wang, X.; Jiang, Z.; Zheng, B.; Xie, Z.; Zheng, L. Synthesis and
Reduced Copper-Cerium Oxide Catalysts as Low as 353 K and 1.3 Shape-dependent Catalytic Properties of CeO2 Nanocubes and
MPa and the Reaction Mechanism. Front. Chem. 2013, 1, 1−8. Truncated Octahedra. CrystEngComm 2012, 14, 7579−7582.
(33) Chiang, C. L.; Lin, K. S.; Yu, S. H.; Lin, Y. G. Synthesis and (52) Postole, G.; Chowdhury, B.; Karmakar, B.; Pinki, K.; Banerji, J.;
Characterization of H 3PW12O40 /Ce 0.1Ti0.9O2 for Dimethyl Auroux, A. Knoevenagel Condensation Reaction over Acid-Base
Carbonate formation via Methanol Carbonation. Int. J. Hydrogen Bifunctional Nanocrystalline CexZr1‑xO2 Solid Solutions. J. Catal.
Energy 2017, 42, 22108−22122. 2010, 269, 110−121.

10455 DOI: 10.1021/acscatal.8b00415

ACS Catal. 2018, 8, 10446−10456
ACS Catalysis Research Article

(53) Fu, Z. W.; Zhong, Y. Y.; Yu, Y. H.; Long, L. Z.; Xiao, M.; Han, Synthesis of Dimethyl Carbonate from Carbon Dioxide over Ceria
D. M.; Wang, S. J.; Meng, Y. Z. TiO2-Doped CeO2 Nanorod Catalyst Nanorod Catalysts. J. Catal. 2016, 340, 295−301.
for Direct Conversion of CO2 and CH3OH to Dimethyl Carbonate:
Catalytic Performance and Kinetic Study. ACS Omega 2018, 3, 198−
207.
(54) Li, A. X.; Pu, Y. F.; Li, F.; Luo, J.; Zhao, N.; Xiao, F. K.
■ NOTE ADDED AFTER ASAP PUBLICATION
This paper published ASAP on October 18, 2018 with errors in
the Supporting Information file. The corrected paper reposted
Synthesis of Dimethyl Carbonate from Methanol and CO2 over Fe-Zr
Mixed Oxides. J. CO2 Util. 2017, 19, 33−39. to the Web on October 24, 2018.
(55) Cai, Q.; Lu, B.; Guo, L.; Shan, Y. Studies on Synthesis of
Dimethyl Carbonate from Methanol and Carbon Dioxide. Catal.
Commun. 2009, 10, 605−609.
(56) An, B.; Zhang, J.; Cheng, K.; Ji, P.; Wang, C.; Lin, W.
Confinement of Ultrasmall Cu/ZnOx Nanoparticles in Metal-Organic
Frameworks for Selective Methanol Synthesis from Catalytic
Hydrogenation of CO2. J. Am. Chem. Soc. 2017, 139, 3834−3840.
(57) Lee, J.; Sorescu, D. C.; Deng, X. Electron-induced Dissociation
of CO2 on TiO2 (110). J. Am. Chem. Soc. 2011, 133, 10066−10069.
(58) Zhang, M. H.; Dou, M. B.; Yu, Y. Z. DFT Study of CO2
Conversion on InZr3 (110) Surface. Phys. Chem. Chem. Phys. 2017,
19, 28917−28927.
(59) Ye, J.; Liu, C.; Mei, D.; Ge, Q. Active Oxygen Vacancy Site for
Methanol Synthesis from CO2 Hydrogenation on In2O3 (110): A
DFT Study. ACS Catal. 2013, 3, 1296−1306.
(60) Fronzi, M.; Daly, W.; Nolan, M. Reactivity of Metal Oxide
Nanocluster Modified Rutile and Anatase TiO2: Oxygen Vacancy
Formation and CO2 Interaction. Appl. Catal., A 2016, 521, 240−249.
(61) Hamid, M. Y. S.; Firmansyah, M. L.; Triwahyono, S.; Jalil, A.
A.; Mukti, R. R.; Febriyanti, E.; Suendo, V.; Setiabudi, H. D.;
Mohamed, M.; Nabgan, W. Oxygen Vacancy-Rich Mesoporous Silica
KCC-1 for CO2 Methanation. Appl. Catal., A 2017, 532, 86−94.
(62) Rui, N.; Wang, Z.; Sun, K.; Ye, J.; Ge, Q.; Liu, C. J. CO2
Hydrogenation to Methanol over Pd/In2O3: Effects of Pd and
Oxygen Vacancy. Appl. Catal., B 2017, 218, 488−497.
(63) Wu, Z.; Mann, A. K. P.; Li, M. J.; Overbury, S. H. Spectroscopic
Investigation of Surface-Dependent Acid-Base Property of Ceria
Nanoshapes. J. Phys. Chem. C 2015, 119, 7340−7350.
(64) Turek, A. M.; Wachs, I. E.; DeCanio, E. Acidic Properties of
Alumina-Supported Metal Oxide Catalysts: An Infrared Spectroscopy
Study. J. Phys. Chem. 1992, 96, 5000−5007.
(65) Di Cosimo, J. I.; Dıez, V. K.; Xu, M.; Iglesia, E.; Apestegúıa, C.
R. Structure and Surface and Catalytic Properties of Mg-Al Basic
Oxides. J. Catal. 1998, 178, 499−510.
(66) Huygh, S.; Bogaerts, A.; Neyts, E. C. How Oxygen Vacancies
Activate CO2 Dissociation on TiO2 Anatase (001). J. Phys. Chem. C
2016, 120, 21659−21669.
(67) Wu, Z.; Li, M.; Mullins, D. R.; Overbury, S. H. Probing the
Surface Sites of CeO2 Nanocrystals with Well-Defined Surface Planes
via Methanol Adsorption and Desorption. ACS Catal. 2012, 2, 2224−
2234.
(68) Yu, Z. Q.; Chuang, S. In Situ IR Study of Adsorbed Species and
Photogenerated Electrons During Photocatalytic Oxidation of
Ethanol on TiO2. J. Catal. 2007, 246, 118−126.
(69) Jung, K. T.; Bell, A. T. An in Situ Infrared Study of Dimethyl
Carbonate Synthesis from Carbon Dioxide and Methanol over
Zirconia. J. Catal. 2001, 204, 339−347.
(70) Wang, H. F.; Gong, X. Q.; Guo, Y. L.; Guo, Y.; Lu, G. Z.; Hu,
P. A Model to Understand the Oxygen Vacancy Formation in Zr-
Doped CeO2 Electrostatic Interaction and Structural Relaxation. J.
Phys. Chem. C 2009, 113, 10229−10232.
(71) Liu, X.; Wang, M.; Zhou, C.; Zhou, W.; Cheng, K.; Kang, J.;
Zhang, Q.; Deng, W.; Wang, Y. Selective Transformation of Carbon
Dioxide into Lower Olefins with a Bifunctional Catalyst Composed of
ZnGa2O4 and SAPO-34. Chem. Commun. 2018, 54, 140−143.
(72) Tomishige, K.; Ikeda, Y.; Sakaihori, T.; Fujimoto, K. Catalytic
Properties and Structure of Zirconia Catalysts for Direct Synthesis of
Dimethyl Carbonate from Methanol and Carbon Dioxide. J. Catal.
2000, 192, 355−362.
(73) Marin, C. M.; Li, L.; Bhalkikar, A.; Doyle, J. E.; Zeng, X. C.;
Cheung, C. L. Kinetic and Mechanistic Investigations of the Direct

10456 DOI: 10.1021/acscatal.8b00415

ACS Catal. 2018, 8, 10446−10456