Version of Record: https://www.sciencedirect.

com/science/article/pii/S0360319916002184
Manuscript_1d3601de8585c102325a7c9fb2f325b0
Experimental and First-principles DFT Study on Oxygen Vacancies on Cerium Dioxide and Its Effect on Enhanced
Photocatalytic Hydrogen Production

Ying Huang1,2, Chang-Feng Yan1,2,*, Chang-Qing Guo1, Yan Shi 1

1
Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangdong, 510640, China
2
University of Chinese Academy of Sciences, Beijing, 100039 China
*
E-mail: yancf@ms.giec.ac.cn

Abstract
Experimental and computational (DFT) approaches were carried out to investigate oxygen vacancies on Cerium dioxide.
Computational result indicates oxygen vacancies can shorten the band gap of CeO2 and enhance the absorption for visible
light because Ce (III) created by generation of oxygen vacancies is easier for excitation than Ce (IV) under same irradiation.
The order of calculated band gap for Ce16O31, (111) < (113) < (133), and the absorption data suggests that (111) plane may
be the most ideal position for oxygen vacancies to enhance photocatalytic activity. The result has been tested and verified by
experimental study, including UV-Vis spectra, XPS, XRD and PL. The enhanced hydrogen productions of CeO2 samples,
from 34.8 μmol to 63.0 μmol, are 2.0 to 3.5 times as high as that of commercial CeO2 (17.8 μmol).

Key words: first principle; DFT; CASTEP; cerium dioxide; density of states; photocatalysis

1 Introduction

Converting solar energy into fuel via photocatalytic water splitting to generate hydrogen is an attractive scientific and
technological goal to solve the energy crisis [1]. Single semiconductor only absorbs a narrow range of solar light. In order to
enhance the solar light absorption, at least two catalysts are combined to a Z-scheme [2] or p-n conjugation [3] photocatlyst
system. In this case, selection of catalysts is the key to achieve this goal. Among all reported semiconductor support, cerium
dioxide is a stable, nontoxic and multifunctional catalyst [4-7]. Due to the special f and d electron orbital structure of cerium
atoms, visible light absorption and quantum efficiency of cerium dioxide can be distinctly enhanced by doping some
non-noble semiconductors like TiO2 [4, 5], Bi2O3 [6] and Cu2O [7], even though its wide band gap, low quantum efficiency
and high recombination hinder its further application in photocatalysis. In this case, cerium dioxide can be utilized as a
potential catalyst for solar energy conversion.

Experimental research on CeO2 indicate that defects such as oxygen vacancies and step edges are the most reactive sites on
the surfaces of metal oxides [8], especially for cerium dioxide. Oxygen vacancy defects on CeO2 can be rapidly formed and
eliminated, giving it a high “oxygen storage capacity”, and these defects also start up Ce4+/Ce3+ redox cycles [9] to enhance
the activities of redox reactions. It has been proved that the generation of Ce3+ caused by oxygen vacancy is the main active
oxygen species for oxidation [10]. Many experimental studies have been reported that electrons can be easily trapped in the
oxygen vacancy sites during light excitation [11-15] and each missing oxygen atom may leave two extra electrons at the site
of the vacancy which leads to creation of surface electron centers for the formation of the negatively charged species [11,
17]. It is one of the factors for enhancement of photocatalytic activities. However, how oxygen vacancies on CeO2 enhance
photocatalytic activity is still unclear.

DFT calculation can obtain deeper insight into the catalytic properties of catalysts. It has been reported that existence of
oxygen vacancies on (110) or (111) planes has higher oxidation reactivity than other planes because of the better geometric,
energetic and electronic effects [17]. Multiple oxygen vacancy on the 2 × 1 reconstructed surface leads to dramatically
improve the activity for catalytic oxidation by significantly migrating the lattice oxygen atoms and oxygen vacancies [18] .
The negative U behavior of oxygen vacancies on CeO2 also leads to excellent photocatalytic behavior [19].

© 2016 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
In this paper, density of states (DOS), band structure and optical properties of Ce16O32 and Ce16O31 models were calculated
by Materials Studio software in order to analyze the correlation between enhanced electron transfer and oxygen vacancies
on cerium dioxide. Calculation was performed by the CASTEP program within the Perdew-Wang 1991 PW91 generalized
gradient approximation (GGA). In order to demonstrate the accuracy of computational data, cerium dioxide synthesized by
glycine-nitrate combustion method was characterized by UV-Vis spectra, XPS, XRD, and PL. The result indicates that
oxygen vacancies exist on the surface of particles which leads to enhancement of the visible light absorption and reduction
of electron-hole recombination, thus enhance the photocatalytic activity of water splitting for hydrogen production.

2 Methods

2.1 Computational details

First-principles calculations in this paper were performed by the CASTEP program within the Perdew-Wang 1991 (PW91)
[20] generalized gradient approximation (GGA) in materials studio software. Cutoff energy was set as 410 eV. A series of
models were built based on a 2×2×1 cerium dioxide supercell, and k-points [21] was set as 1×1×3. All lattices of CeO2 with
single vacancy were built according to the XPS quantitative analysis. The cerium 4f, 5d, 6s and oxygen 2s, 2p electrons
were treated as valence electrons. Besides, ultrasoft pseudopotentials were used and the relevant representations were
carried out in reciprocal space. The set SCF tolerance was 5.0×10-7.

2.2 Experimental

CeO2 catalyst was prepared by glycine-nitrate combustion method as our preview work [22, 23]. The powder was
characterized after calcined at 300 °C, 400 °C, 500 °C, 600 °C, respectively. X-ray diffraction (XRD) analysis was
performed with the PANalytical X’Pert diffractometer (X’Pert PRO MPD, PW3040/60) within the 2-θ ranged from 5° to 80°
by a speed of 6° per minute with Cu-Kα (λ = 0.154060 nm) radiation (40 kV, 40 mA). The crystallite size was calculated by
the Scherrer equation from the XRD spectra. UV-vis spectroscopy in the 300nm-1000nm was measured with a Lambda 750
Diffuse Reflectance Spectroscopy. X-Ray photoelectron spectroscopy (XPS) analysis were conducted through a thermo
ESCALAB 250XI multifunctional imaging electron spectrometer (Thermo Fisher Scientific Inc.) equipped with an Al K
radiation source. Fluorescence spectra were obtained using a photoluminescence (PL) spectrometer (Perkin Elmer, LS-55).
Photocatalytic reactions for overall water splitting were carried out in a 15 mL quartz cell with a 25 mm× 25 mm window
for light transmission at room temperature under visible light irradiation (λ> 400 nm, ≈0.18 W/cm2). The catalyst powder
(30 mg) was suspended in 10 vol. % methanol aqueous solution (15 mL) by magnetic stirring. In this paper, methanol plays
a role as the sacrifice agent to scavenge photogenerated holes. The evolved gases were determined with a gas chromatograph
equipped with a TCD detector.

3. results and discussion

3.1 Computational study

The models of Ce16O31 lattice with single oxygen vacancy was built according to atomic compositions (%) measured by
XPS. There are three possible sites for oxygen vacancy in this 2×2×1 cerium dioxide supercell. Two layers of oxygen atoms
can be seen in the right side of figure 2, and three sites for oxygen vacancy lie in the second oxygen layer (layer 2). Site A,
B and C lie in (111) plane, (331) plane and (133) plane of supercell lattice, respectively. And each of these three sites
missing an oxygen atom creates a model with single oxygen vacancy.

It is known that electron-hole generation under light irradiation depends on the band structure and the electronic density of
states (DOS), so calculated partial DOS (Figure 2) and the band structures (Table 2) were considered. PDOS of cerium
atoms in Ce16O31 lattices also have a little shift towards negative potential than Ce16O32 lattice, which leads to the negative

2
shift of valence band potential of. And the band gaps of Ce16O31 are all shortened compare to those of Ce16O32. It provide a
favorable condition for catalysts with oxygen vacancies to be excited by visible light.

Schematic diagram of electron transfer and partial DOS of Ce (III), O and Ce (IV) are showed in Figure 3. For cerium atoms,
generation of oxygen vacancy created Ce3+ [11] leads to the empty valence orbits being occupied by extra electrons obtained
from removed oxygen ions [8], thus the energy of these orbits shift toward negative potential, especially for 4f orbit and
then shorten the band gaps. The DOS energy ranges of Ce 4f and O 2p are overlapped which leads to electron transfer
between them during excitation. However, DOS of Ce 4f for Ce (III) is larger than that of Ce (IV) in positive energy range,
which indicates Ce (III) is easier for excitation than Ce(IV) under same irradiation. The light absorption of of Ce16O32 and
Ce16O31 lattice with single oxygen vacancy in (111) plane, (113) plane and (133) plane are showed in Figure 4. It shows that
all Ce16O31 lattices have red shift compare to Ce16O32, which suggests oxygen vacancies enhance the visible light absorption.
The result is in line with that of DOS.

In summary, the computational result indicates Ce (III) created by the formation of oxygen vacancies can shorten the band
gap of CeO2 lattice and enhance its absorption for visible light because Ce(III) is easier for excitation than Ce(IV). The order
of calculated band gap for Ce16O31, (111) < (113) < (133), and the absorption data suggests that (111) plane may be the most
ideal position for oxygen vacancies to enhance photocatalytic activity.

3.2 Experimental study

In order to test and verify the accuracy of computational results, experimental study on CeO2 prepared by glycine-nitrate
combustion method was carried out. The UV-vis spectra of the catalysts are shown in Figure 5. All CeO2 samples have
better visible light absorption (λ>400 nm) than that of commercial CeO2. And the order of visible light absorbance is
700 °C > 600 °C ≈ 400 °C > 500 °C > 300 °C > commercial CeO2. The band gaps of catalysts are estimated from the plot of
the (αhν)2 versus photon energy (hν) [24, 25] as shown in Table 3. The band gap of commercial CeO2 was estimated to be
3.1 eV, which is matched with reported band gap [26]. The red shift of the CeO2 samples might be attributed to trivalent
ionic cerium compounds.

Figure 6 shows the XPS spectra for Ce 3d of CeO2 (400 oC). Among all of these ten peaks, v, v’’, v’’’, u, u’’ and u’’’ are
attributed to the present of Ce (IV), and v0, v’, u0 and u’, however, are attributed to the present of Ce (III) [27]. The amounts
of Ce (III) and Ce (IV) were calculated from the components of the Ce 3d core level using the following equations [27, 28]:
Ce (III) = v0 + v’ + u0 + u’ (1)
Ce (IV) = v + v’’ + v’’’ + u + u’’ + u’’’ (2)
% Ce (III) = [Ce (III) / ( Ce(III)+ Ce(IV))]×100 (3)
The calculated results from equations (1)-(3), which has similar atomic composition as those of the computational models as
shown in table 1, indicates that the amount of Ce (III) does exist on the surface of CeO2. The details of measured results are
shown in Table 3. The ratios of O to Ce atoms in CeO2 and Ce2O3 lattice are 2 and 1.5, respectively. The measured results
suggest that the oxygen amount on the surface of particles meets stoichiometric ratio with Ce (IV) and Ce (III) for the
commercial CeO2, however, stoichiometric oxygen amounts of other samples are far larger than their measured oxygen
amounts, which means oxygen vacancies can be significantly formed on the surface of CeO2 crystallite during synthetizing
in glycine-nitrate combustion method. The atomic percentage of Ce (III) increases with the increasing of the relative oxygen
vacancy amount with an order of 700 °C > 600 °C > 500 °C > 300 °C > commercial CeO2, which is in line with the order of
their band gaps. And the generation of each oxygen vacancy in the crystal during calcination may generate a Ce3+ state [29]
Thus, the atomic percentage of Ce (III) is the main factor to the band gap, rather than that of Ce (IV) or ratio of Ce (IV) to
Ce (III). And the increase of oxygen vacancy amount leads to increase of atomic percentage of Ce (III). Among all samples,
CeO2 (400 oC) is a special samples with highest atomic percentage of Ce (III) and the ratio of Ce (IV) to Ce (III) and it will
be discussed later.

3
Figure 7 shows the the X-ray diffraction patterns of CeO2 samples prepared by glucine-nitride combustion method along
with that of commercial CeO2. Eight main reflections belong to a typical fcc fluorite structure of CeO2 (Reference code:
03-065-2975), corresponding respectively to the [111], [200], [220], [311], [222], [400], [331] and [420] planes. Among all
the CeO2 samples, the diffraction peaks are getting stronger and sharper with increasing of calcination temperature, which
indicates the improvement of the crystallization structure of catalysts. And the CeO2 crystallite sizes were estimated by
using the Scherrer equation as shown in Table 2. CeO2 crystallite size is smaller under calcination temperature of 400 °C
than under 300 °C, which is attributed to the insufficient crystallization under 300 °C. However, the enlarged size of CeO2
crystallite under calcination temperature of 500 °C, 600 °C and 700 °C are caused by sintering of particles. No peaks for
trivalent ionic cerium compounds were found on the patterns, which suggests the Ce3+ may be highly separated on the
surface of particles and the existence of Ce3+ may be possibly caused by generation of oxygen vacancies.

To ensure the existence of oxygen vacancy, Photoluminescence spectra (Figure 7) of CeO2 samples and commercial CeO2
were obtained at excitation wavelengths of 325 nm and 350 nm. Under 325 nm excitation, peak 1 at 530 nm is assigned to
the oxygen vacancy with one trapped electron [30, 31]. Strong emission at around 460 nm to 480 nm is related to the present
of dislocations [32]. And under 350 nm excitation, the emission from around 400 to 500 nm is associated with charge
transfer from the Ce 4f energy level to the O 2p level (valance band) [33]. Peak 2 at around 430 nm is associated with the
charge transfer between Ce 4f energy states [33, 34], and peak 3 at around 470 nm is attributed to oxygen defects [33, 35].
Weaker optical emission of CeO2 samples than that of commercial CeO2 indicates less electron-hole recombination on the
surface of particles.

No inactivation happened after a 2000-minutes testing. The hydrogen yields increase stably and the final yields were in line
with the measured band gaps. The better light absorption, the higher hydrogen yield. The CeO2 sample under calcination
temperature of 400 oC has the highest hydrogen production of 63 μmol after 2040 min irradiation is attributed to its
crystallite size and shorten band gap caused by generation of oxygen vacancies.

Effective electron transfer between cerium and oxygen atoms and rich of oxygen vacancies lead to lower probability of
electron-hole recombination, and thus enhancing the photocatalytic performance for hydrogen production, which is in line
with the computational study.

4 Conclusion
Computational result indicates oxygen vacancies can shorten the band gap of CeO2 and enhance the absorption for visible
light because Ce (III) created by generation of oxygen vacancies is easier for excitation than Ce (IV) under same irradiation.
The order of calculated band gap for Ce16O31, (111) > (113) > (133), and the absorption data suggests that (111) plane may
be the most ideal position for oxygen vacancies to enhance photocatalytic activity. Experimental study illustrates that cerium
dioxide synthesized by glycine-nitrate combustion method owns its narrower bandgap (2.76 eV ~ 2.95 eV) than the reported
CeO2 (3.3eV). Oxygen vacancies were found on the surface of CeO2 by PL spectra and quantitative analysis of XPS and
were proved to be well separated by XRD. It also indicates that oxygen vacancies generation leads to higher visible light
absorption and less electron-hole recombination on the surface of particles, which is qualitatively matched with the
computational result.

Acknowledgement
The authors gratefully acknowledge the financial support of CAS Renewable Energy Key Lab., National Science
Foundation of China (51576201), Natural Science Foundation of Guangdong Province (2015A030312007), Guangdong
Science and Technology Project (2013B050800007) and Guangzhou Science and Technology Project (2013J4500027).

REFERENCES
4
1. Tran, P.D., et al., Recent advances in hybrid photocatalysts for solar fuel production. Energy & Environmental Science,
2012. 5(3): p. 5902-5918.
2. Kudo, A., Recent progress in the development of visible light-driven powdered photocatalysts for water splitting.
International Journal of Hydrogen Energy, 2007. 32(14): p. 2673-2678.
3. Xie, S., et al., NiO decorated Mo:BiVO4 photoanode with enhanced visible-light photoelectrochemical activity.
International Journal of Hydrogen Energy, 2014. 39(10): p. 4820-4827.
4. Tian, M., et al., Visible light driven nanocrystal anatase TiO2 doped by Ce from sol-gel method and its
photoelectrochemical water splitting properties. International Journal of Hydrogen Energy, 2014. 39(25): p.
13448-13453.
5. Wang, Y., et al., Ordered mesoporous CeO2-TiO2 composites: Highly efficient photocatalysts for the reduction of CO2
with H2O under simulated solar irradiation. Applied Catalysis B-Environmental, 2013. 130: p. 277-284.
6. Li, L. and B. Yan, CeO2-Bi2O3 nanocomposite: Two step synthesis, microstructure and photocatalytic activity. Journal of
Non-Crystalline Solids, 2009. 355(13): p. 776-779.
7. Hu, S., et al., Preparation of Cu2O/CeO2 heterojunction photocatalyst for the degradation of Acid Orange 7 under
visible light irradiation. Catalysis Communications, 2011. 12(9): p. 794-797.
8. Campbell, C.T. and C.H.F. Peden, Chemistry - Oxygen vacancies and catalysis on ceria surfaces. Science, 2005.
309(5735): p. 713-714.
9. Ansari, S.A., et al., Band gap engineering of CeO2 nanostructure using an electrochemically active biofilm for visible
light applications. Rsc Advances, 2014. 4(32): p. 16782-16791.
10. Shan, W.J., et al., Syngas production from partial oxidation of methane over Ce1-XNiXOY catalysts prepared by
complexation-combustion method. Applied Catalysis a-General, 2006. 311: p. 24-33.
11. Tu, W.G., Y. Zhou, and Z.G. Zou, Photocatalytic Conversion of CO2 into Renewable Hydrocarbon Fuels:
State-of-the-Art Accomplishment, Challenges, and Prospects. Advanced Materials, 2014. 26(27): p. 4607-4626.
12. Iqbal, J., et al., Oxygen-vacancy-induced green emission and room-temperature ferromagnetism in Ni-doped ZnO
nanorods. New Journal of Physics, 2009. 11.
13. Mahammadunnisa, S., et al., NiO/Ce1-xNixO2-delta as an alternative to noble metal catalysts for CO oxidation.
Catalysis Science & Technology, 2013. 3(3): p. 730-736.
14. Kumari, L., et al., Controlled Hydrothermal Synthesis of Zirconium Oxide Nanostructures and Their Optical Properties.
Crystal Growth & Design, 2009. 9(9): p. 3874-3880.
15. Wang, Z., Z. Quan, and J. Lin, Remarkable changes in the optical properties of CeO2 nanocrystals induced by lanthanide
ions doping. Inorganic Chemistry, 2007. 46(13): p. 5237-5242.
16. Koppenol, W.H. and J.D. Rush, Reduction Potential of the Co2/Co2.- Couple - a Comparison with Other C1 Radicals.
Journal of Physical Chemistry, 1987. 91(16): p. 4429-4430.
17. Song, Y.-L., et al., A DFI' plus U study of CO oxidation at CeO2(110) and (111) surfaces with oxygen vacancies. Surface
Science, 2013. 618: p. 140-147.
18. Zhang, J., X.-Q. Gong, and G. Lu, Catalytic activities of CeO2(110)-2 x 1 reconstructed surface. Surface Science, 2015.
632: p. 164-173.
19. Huang, B., R. Gillen, and J. Robertson, Study of CeO2 and Its Native Defects by Density Functional Theory with
Repulsive Potential. Journal of Physical Chemistry C, 2014. 118(42): p. 24248-24256.
20. Perdew, J.P., K. Burke, and M. Ernzerhof, Generalized gradient approximation made simple. Physical Review Letters,
1996. 77(18): p. 3865-3868.
21. Monkhorst, H.J. and J.D. Pack, Special Points for Brillouin-Zone Integrations. Physical Review B, 1976. 13(12): p.
5188-5192.
22. Yan, C.F., et al., Synthesis of mesoporous Co-Ce oxides catalysts by glycine-nitrate combustion approach for CO
preferential oxidation reaction in excess H-2. International Journal of Hydrogen Energy, 2014. 39(32): p. 18695-18701.
23. Huang, Y., et al., Enhanced Photoreduction Activity of Carbon Dioxide over Co3O4/CeO2 Catalysts under Visible Light
Irradiation. International Journal of Photoenergy, 2015: p. 30808-30808.
5
24. Li, X., et al., Adsorption of CO2 on heterostructure CdS(Bi2S3)/TiO2 nanotube photocatalysts and their photocatalytic
activities in the reduction of CO2 to methanol under visible light irradiation. Chemical Engineering Journal, 2012. 180: p.
151-158.
25. Li, H.L., et al., Photocatalytic reduction of carbon dioxide to methanol by Cu2O/SiC nanocrystallite under visible light
irradiation. Journal of Natural Gas Chemistry, 2011. 20(2): p. 145-150.
26. Matejova, L., et al., Preparation, characterization and photocatalytic properties of cerium doped TiO2: On the effect of
Ce loading on the photocatalytic reduction of carbon dioxide. Applied Catalysis B-Environmental, 2014. 152: p. 172-183.
27. Beche, E., et al., An XPS investigation of (La2O3)1-x (CeO2)2x (ZrO2)2 compounds. Surface and Interface Analysis,
2012. 44(8): p. 1045-1050.
28. Abiaad, E., et al., Preparation and Characterization of Ceo2 under an Oxidizing Atmosphere - Thermal-Analysis, Xps,
and Epr Study. Chemistry of Materials, 1993. 5(6): p. 793-797.
29. Krishnan, A., et al., One-pot synthesis of ultra-small cerium oxide nanodots exhibiting multi-colored fluorescence.
Journal of Colloid and Interface Science, 2013. 389: p. 16-22.
30. Zhang, R., et al., Promotional effect of fluorine on the selective catalytic reduction of NO with NH3 over CeO2-TiO2
catalyst at low temperature. Applied Surface Science, 2014. 289: p. 237-244.
31. Li, D., et al., Visible-light-driven photocatalysis on fluorine-doped TIO2 powders by the creation of surface oxygen
vacancies. Chemical Physics Letters, 2005. 401(4-6): p. 579-584.
32. Suresh, R., V. Ponnuswamy, and R. Mariappan, The role of oxidizing agents in the structural and morphological
properties of CeO2 nanoparticles. Materials Science in Semiconductor Processing, 2014. 21: p. 45-51.
33. Saranya, J., et al., Cobalt-doped cerium oxide nanoparticles: Enhanced photocatalytic activity under UV and visible light
irradiation. Materials Science in Semiconductor Processing, 2014. 26: p. 218-224.
34. Palard, M., et al., Effect of hydrothermal ripening on the photoluminescence properties of pure and doped cerium oxide
nanoparticles. Materials Chemistry and Physics, 2010. 120(1): p. 79-88.
35. Wang, G.F., et al., Synthesis, characterization and photoluminescence of CeO2 nanoparticles by a facile method at room
temperature. Journal of Alloys and Compounds, 2010. 493(1-2): p. 202-207.

6
Figure 1. Ce16O31 lattice and the positions of oxygen vacancy in (111) plane A, (113) plane B and (133) plane C. Red
balls are oxygen atoms and white balls are cerium atoms.

Figure 2. Partial DOS for Cerium atoms and oxygen atoms of 2×2×1 CeO2 supercell without oxygen vacancy, and
Ce16O31 lattice with single oxygen vacancy in (111) plane, (113) plane and (133) plane.

Figure 3. Schematic diagram of electron transfer (a) and partial DOS of Ce3+ (b), O (c) and Ce4+ (d).

Figure 4. Calculated light absorption of Ce16O32 and Ce16O31 lattice with single oxygen vacancy in (111) plane, (113)
plane and (133) plane.

Figure 5. UV-vis absorption spectra of CeO2 samples and commercial CeO2.

Figure 6. XPS spectra of CeO2 samples and commercial CeO2. Yan C.F. et al. [20]

Figure 7. X-ray diffraction (XRD) patterns of CeO2 samples and commercial CeO2.

Figure 8. Photoluminescence spectra of CeO2 samples and commercial CeO2 at excitation wavelengths of 325 nm (a)
and 350 nm (b)

Figure 9. Photocatalytic activities of CeO2 samples and commercial CeO2.

Figure 1. Ce16O31 lattice and the positions of oxygen vacancy in (111) plane A, (113) plane B and (133) plane C. Red
balls are oxygen atoms and white balls are cerium atoms.
Figure 2. Partial DOS for Cerium atoms and oxygen atoms of 2×2×1 CeO2 supercell without oxygen vacancy, and
Ce16O31 lattice with single oxygen vacancy in (111) plane, (113) plane and (133) plane.
Figure 3. Schematic diagram of electron transfer (a) and partial DOS of Ce3+ (b), O (c) and Ce4+ (d).
Figure 4. Calculated light absorption of Ce16O32 and Ce16O31 lattice with single oxygen vacancy in (111) plane, (113)
plane and (133) plane.
Figure 5. UV-vis absorption spectra of CeO2 samples and commercial CeO2.
Figure 6. XPS spectra of CeO2 samples and commercial CeO2. Yan C.F. et al. [20]
Figure 7. X-ray diffraction (XRD) patterns of CeO2 samples and commercial CeO2.
Figure 8. Photoluminescence spectra of CeO2 samples and commercial CeO2 at excitation wavelengths of 325 nm (a)
and 350 nm (b).
Figure 9. Photocatalytic activities of CeO2 samples and commercial CeO2.
Table 1-Atomic compositions (%) measured from the O 1s, Ce 3d and Co 3s spectra of CeO2
along with the models for calculation.

Table 2- calculated bandgap of Ce16O32 and Ce16O31 lattice with single oxygen vacancy in (111)
plane, (311) plane and (331) plane.

Table 3- Estimated band gaps and crystallite size of commercial CeO2 and other CeO2 samples

Table 4- Atomic compositions (%) measured from the O 1s and Ce 3d spectra commercial CeO2
and other CeO2 samples
 Table 1-Atomic compositions (%) measured from the O 1s, Ce 3d and Co 3s spectra of CeO2
along with the models for calculation.
Atomic compositions (%)
Ce
Samples/models
O
Ce(III) Ce(IV)

CeO2 (measured) 65.8 6.7 27.5

Ce16O31 (111) A 66.0 3.5 30.5
Ce16O31 (113) B 66.0 6.4 27.6
Ce16O31 (133) C 66.0 4.8 29.2

Table 2- calculated bandgap of Ce16O32 and Ce16O31 lattice with single oxygen vacancy in (111)
plane, (311) plane and (331) plane.
lattices Bandgap (eV) Valence band potential Conduction band
(eV) potential (eV)
Ce16O32 2.42 1.65 -0.77
Ce16O31 (111) A 2.37 1.60 -0.77
Ce16O31 (113) B 2.39 1.61 -0.78
Ce16O31 (133) C 2.40 1.62 -0.78

Table 3- Estimated band gaps and crystallite size of commercial CeO2 and other CeO2 samples
Estimated band gaps The crystallite size of
Samples
(eV) CeO2 (nm)
Commercial CeO2 3.10 10.5
300 oC 2.95 15.1
400 oC 2.86 9.7
500 oC 2.90 13.5
600 oC 2.87 18.6
700 oC 2.76 30.3

Table 4- Atomic compositions (%) measured from the O 1s and Ce 3d spectra commercial CeO2
and other CeO2 samples
Atomic compositions (%) Stoichiometric Relative oxygen
Catalysts Ce (IV)/Ce (III) oxygen vacancy amount
O Ce (III) Ce (IV)
amount (%)1 (%)2
Commercial CeO2 65.8 6.7 27.5 4.1 65.1 -0.7
300 oC 63.8 9.1 27.1 3.0 67.8 4.0
400 oC 62.9 11.9 25.2 2.1 68.2 5.3
500 oC 62.2 9.7 28.1 2.9 70.7 8.5
600 oC 61.8 10.3 27.9 2.7 71.3 9.5
700 oC 56.3 10.4 33.4 3.2 82.3 26.0
1 Stoichiometric oxygen amount (%) = 1.5× percentage of Ce (III) (%) +2× percentage of Ce
(IV) (%)
2 Relative oxygen vacancy amount (%) = Stoichiometric oxygen amount (%) - Atomic percentage
of O (%)