Journal of Molecular Structure 994 (2011) 325–331

Contents lists available at ScienceDirect

Journal of Molecular Structure

journal homepage: www.elsevier.com/locate/molstruc

Copper and cerium co-doped titanium dioxide on catalytic photo reduction

of carbon dioxide with water: Experimental and theoretical studies
Dongmei Luo a,⇑, Ye Bi a, Wei Kan a, Ning Zhang b, Sanguo Hong b
a
Department of Chemistry and Chemical Engineering, Qiqihar University, Qiqihar 161006, China
b
Department of Chemistry, Nanchang University, Nanchang 330031, China

a r t i c l e i n f o a b s t r a c t

Article history: The catalytic activities of copper and cerium co-doped titanium dioxide were studied experimentally and
Received 18 February 2011 theoretically in the synthesis of methanol by the photo reduction of carbon dioxide with water firstly.
Received in revised form 17 March 2011 Photo catalysts copper and cerium co-doped titanium dioxide were prepared via the equivalent-volume
Accepted 18 March 2011
incipient wetness impregnation method. The catalysts were characterized by XRD, Raman, BET, and elec-
Available online 13 April 2011
trochemistry analyses. The catalytic properties were determined in the synthesis of methanol from CO2 in
the aqueous solution. The experimental results suggested that Cu/Ce–TiO2 catalysts obviously enhanced
Keywords:
the efficiency of the photocatalytic reduction of CO2. The methanol yield could reach up to 180.3 lmol/g-
Copper and cerium co-doped titanium
dioxide
cat rapidly. The different effects of copper and cerium on the surface of titanium dioxide have been
Carbon dioxide calculated at the Becke’s three-parameter hybrid exchange functional together with the Lee–Yang–Parr
Methanol correlation functional (B3LYP) level. Our results revealed that Ce atoms affect the reaction more
Density functional theory profoundly than Cu atoms do. Ce atoms activated H2O and CO2 molecules, while Cu atoms act as the
channel of photoelectrons in real time and prevent the recombination of electrons and holes.
Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved.

1. Introduction well as Au atoms adsorbed on TiO2 (1 1 0) by Giordano et al. [15].

Most theoretical studies focused on p and d areas, whereas metals
CO2 is one of the largest carbon sources. It is of interest to trans- in f area are rarely involved. In this work, the catalytic activities of
fer CO2 into methanol which is one of the most important indus- Cu-doped TiO2 and Ce-doped TiO2 were studied experimentally
trial chemical materials [1,2]. Therefore, the research on and theoretically in the synthesis of methanol by photoreduction
synthesis of methanol from CO2 and H2O is gaining increasing of CO2 and H2O in order to compare the different effect of dopants
attention in recent years [3–11]. In most of the studies catalysts on the TiO2 surface.
such as Cu/TiO2 have been investigated extensively. Catalysts
doped with lanthanide metal ions [12,13], such as Ce4+, have been
2. Experimental and computational details
shown to increase the photocatalytic efficiency for selected reac-
tions due to its functional f-orbitals. To the best of our knowledge,
2.1. Preparation of catalysts
the research on the auxiliary catalyst Ce is seldom. Cu/Ce–TiO2 cat-
alysts were prepared by equivalent-volume incipient wetness
All reagents were of analytical grade and used without further
impregnation method and further characterized by XRD, Raman,
purification as purchased from SCRC (Sinopharm Chemical Reagent
BET, and electrochemistry analyses for investigating the phase
Co., LtdS, China). Cu/Ce–TiO2 catalysts were prepared via the equiv-
composition, surface structure, chemisorbing ability and photon
alent-volume incipient wetness impregnation method [16,17]. The
absorbing properties. Their catalytic activities were evaluated by
catalysts prepared via co-impregnation method was expressed in
photocatalytic synthesis of methanol from CO2, and this co-doped
(Cu–Ce)/TiO2. It was prepared by impregnating TiO2 slurry [18]
photocatalyst increased the yield of CH3OH rapidly. On the other
with Cu(NO3)2 and Ce(NO3)4 solution for 24 h, being dried at
hand, in the aspect of theoretical research there are many studies
100 °C for 1 h, then calcined at 500 °C for 2 h and reduced at
on the adsorption of metal ions on TiO2 surface and the interaction
300 °C for 7 h under a flowing of 5% (volume ratio) H2/N2. The cat-
between reactant and the dopant ions, for example, Ba adsorption
alysts prepared by incipient impregnation method were expressed
on the TiO2 (1 1 0) surface by San Miguel et al. [14], and Cu, Ag, as
in Ce–Cu/TiO2 and Cu–Ce/TiO2. Ce–Cu/TiO2 catalysts were pre-
pared first by impregnating TiO2 slurry with Cu(NO3)2 solution,
⇑ Corresponding author. Tel.: +86 452 2738216. after being dried and calcined, the powders were impregnated
E-mail address: [email protected] (D. Luo). with Ce(NO3)4 solution [19]. Then reduction step under 5% H2/N2

0022-2860/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2011.03.044
326 D. Luo et al. / Journal of Molecular Structure 994 (2011) 325–331

at 300 °C for 7 h was to form Cu0 or Cu1+. Cu–Ce/TiO2 catalysts from the an instructions leaflet was more than or equal to
were synthesized by the almost same method mentioned of 3000 mVml/mg (benzene, H2)).
Ce–Cu/TiO2. The only difference was that the TiO2 slurry was
impregnated with Ce(NO3)4 solution and then impregnated with 2.4. Computational details
Cu(NO3)2 solution.
The non-defective rutile (1 1 0) surface and the anatase (1 0 1)
2.2. Characterizations of catalysts surface of TiO2 are modeled using neutral hydrogen-saturated clus-
ter Ti8O22H12 and neutral stoichiometric cluster Ti12O24, respec-
XRD spectra of catalysts were recorded with an X-ray diffrac- tively. The M-doped clusters (M = Ce, Cu) are rutile (1 1 0)
tometer (type D/max-3B made in Japan) over the range surface and anatase (1 0 1) surface of TiO2 for research the interac-
20° 6 2h 6 80°, operating at 40 kV and 10 mA, using Cu Ka radia- tion between M and TiO2 surface by optimization of M only,
tion (k = 1.5418 Å) combined with the nickel filter. The specific sur- respectively. The geometry optimization and frequencies of
face area of catalysts was measured by multipoint BET of N2 M–3CO2 and M–2H2O–CO2 have been calculated, and all results
adsorption in an ASAP 2020 (Micrometrics). Raman spectra were were obtained with the Restricted open shell Kohn–Sham density
measured on NXR FT-RAMAN with a laser at a wavelength of functional theory (DFT), calculations have been performed using
1064 nm. For all samples, the scanning times were 100 and the ap- the Becke three-parameter hybrid exchange functional together
plied laser beam was 20–50 mW. The photoelectrochemical mea- with the Lee–Yang–Parr correlation functional [20], B3LYP with
surements were performed with an electrochemical set-up different basis sets and among the basis sets LANL2DZ for Ti and
(IM6ex, ZAHNER, Germany) and a three-electrode cell equipped Cu [21], 6-31G for O and C [22], and Stuttgart RSC 1997 ECP EMSL
with a flat quartz window. A platinum flag (1  2 cm) served as a Basis Set for Ce [23] , 6-311G for H [24] were chosen as the optimal
counter electrode and a Ag/AgCl as a reference electrode. The flat- ones according to the data from calculations of the cluster model.
band potential measurements were performed in NaOH (0.02 All calculations were carried out using GAUSSIAN 03 program
mol/L) electrolytes. Electrodes were illuminated by the full light package [25].
of a 125 W UV lamp. The value of photocurrent was taken as a dif-
ference between photocurrent under irradiation and in the dark.
3. Results and discussion

2.3. Photocatalytic reaction 3.1. Physicochemical properties

The pohotocatalytic reduction of CO2 was conducted with cata- 3.1.1. Structural and textural properties
lysts in a hermetic high-pressure reaction still equipped with Initially, we obtained that the optimal Ce loading amount is
heating system. Fig. 1 schematically illustrates the reactor system. approximately 0.5 wt.% for Cu/Ce–TiO2 catalysts, which was con-
The system was illuminated by an UV lamp (@125 W) with a pri- sistent with the result of Ce/TiO2 catalysts. On the basis of that,
mary emission at the wavelength of 365 nm. For typical batches, the Ce amount of Cu/Ce–TiO2 catalysts was fixed on 0.5 wt.%. The
150 mg of catalysts powder was suspended in 150 ml of 0.2
mol/L NaOH aqueous solution. Supercritical-fluid grade CO2 was
bubbled through the reactor for at least 30 min to purge air and Table 1
to saturate the solution. The reactor was tightly closed during The specific surface area of catalysts.

the reaction and a magnetic stirrer agitated the catalysts-sus- Catalysts (Cu–Ce)/TiO2 Ce–Cu/TiO2 Cu–Ce/TiO2
pended solution at the bottom to prevent sedimentation of the cat- 2
Surface area/(m g 1
) 43.002 44.932 52.963
alysts and irradiated 0–16 h maintaining at 50 °C, 70 °C and 90 °C,
respectively. After the photocatalytic reaction finished, the liquid (Cu–Ce)/TiO2 catalysts were impregnated with Ce(NO3)4 solution and Cu(NO3)2
solution at the same time. Ce–Cu/TiO2 catalysts were prepared first by impreg-
sample was collected in vials wrapped in aluminum foil. After nating TiO2 slurry with Cu(NO3)2 solution, after being dried and calcined, the
the catalysts sedimentation, The solution sample and the con- powders were impregnated with Ce(NO3)4 solution. Cu–Ce/TiO2 catalysts were
densed sample were analyzed in the SP-6800 GC/TCD equipped impregnated with Ce(NO3)4 solution and then impregnated with Cu(NO3)2 solution.
with a 4 m long Porapak Q column (the sensitivity of the analysis

Fig. 1. Schematic of photocatalytic reaction. Fig. 2. XRD patterns of the samples.

 D. Luo et al. / Journal of Molecular Structure 994 (2011) 325–331 327

specific surface area of three catalysts was listed in Table 1. The

specific surface area of the samples increased in the following or-
der: Cu–Ce/TiO2 > Ce–Cu/TiO2 > (Cu–Ce)/TiO2 > Cu/TiO2. Appar-
ently, the slight change in specific surface area has little
contribution to the reaction activity.
The XRD diffraction patterns of Cu–Ce/TiO2, Ce–Cu/TiO2 and
(Cu–Ce)/TiO2 were shown in Fig. 2. Cu diffraction peaks (2h = 43.3°
and 2h = 52°) were observed on Cu–Ce/TiO2 and (Cu–Ce)/TiO2,
which revealed that Cu were well dispersed on the surface of
catalysts. No CuO diffraction peak was found near 2h = 35.6°, which
indicated that CuO has been successfully reduced. Whereas, no Cu
diffraction peak was observed on Ce–Cu/TiO2, which perhaps due
to the process of Ce–Cu/TiO2 catalysts impregnated by Ce(NO3)4
solution lastly, and Ce enveloped the surface of Ce–Cu/TiO2. In addi-
tion, the appearance of peak at around 2h = 28° in the Ce–Cu/TiO2
and (Cu–Ce)/TiO2 samples suggested the presence of a CeO2 phase.
These results implied that the sequence of impregnation affected
the structure of catalysts and phase on the surface of the samples.
Fig. 3. Raman spectra of the samples.

3.1.2. Raman spectroscopy properties

FT-Raman spectroscopy is more surface-sensitive than X-ray
diffraction because the exciting line of the former technique falls
in the near IR region and is less penetrating than X-rays [26].
Therefore, a comparison of the results of Raman spectroscopy with
those from X-ray diffraction can be used to estimate the difference
between the surface and bulk compositions of the catalysts. Raman
spectra of three catalysts at room temperature are shown in Fig. 3.
The absence of 446 cm1 and 609 cm1 peaks, characteristic of the
rutile crystal phase, suggests that the powders are composed of the
anatase phase only. Raman modes at 148.44 cm1, 199.01 cm1,
397.58 cm1, 515.97 cm1, and 639.81 cm1 are assigned as Eg,
Eg, B1g, A1g (or B1g) and Eg modes in anatase phase, respectively
[27]. A strong sharp band at 148.44 cm1 (Eg) can be observed on
Cu–Ce/TiO2 sample, whereas the bands are weak for Ce–Cu/TiO2
and (Cu–Ce)/TiO2, which implied that the particle size of Cu–Ce/
TiO2 sample are large crystallites [28]. As shown in Fig. 3, for
Cu–Ce/TiO2 sample, the 148.44 cm1 peak shifts to the red and
shrinks asymmetrically which implied that the particle size of
sample are large. It summarizes the main experimental observa-
Fig. 4. Time development of photocurrent transients for the samples. tions that are associated with changes in the particle size: with

Fig. 5. Catalytic activity of the three samples.

328 D. Luo et al. / Journal of Molecular Structure 994 (2011) 325–331

Table 2
(a) Optimized parameters of M/TiO2 clusters.

M/Ti8O22H12 (rutile(110)) M/Ti12O24 (anatase(101))

RCe–O 2.197 RCu–O 2.031 RCe–O 2.000 RCu–O 2.020
Covalent radius/Å Ce 1.650 Cu 1.170 Ti 1.320 O 0.730
Ionic radius/Å Ce 1.034 Cu 0.730 Ti 0.650 O 1.400

‘‘M’’ refers to the metal atoms loaded on the surface of titanium dioxide, RCe–O refers
to the bond length of Ce atom between O atom of titanium dioxide surface, and RCu–
O refers to the bond length of Cu atom between O atom of titanium dioxide surface.

Cu/Ti 8O22H12
an increase in the particle size the peak shifts to the red, shrinks
asymmetrically and increases in intensity.

3.1.3. Photovoltaic properties

The time variations of photocurrent generated under 125 W UV
illumination of samples were shown in Fig. 4. It was evident that
all samples investigated generated similar action spectra apart
from their photocurrents and the photocurrents of modified Cu/
TiO2 with Ce were higher than that of Cu/TiO2. After the initial rise
of photocurrent immediately upon switching-on the light, a rapid
Ce/Ti 8O22H12 decay was observed after the light switched off. Such a shape of
photocurrent transient is a typical fingerprint of intense surface
recombination processes [29,30]. These states acted as electron–
(b) hole recombination centers, as clearly manifested by the instant
photocurrent decay after switching-on the light [31]. Furthermore,
I–t characteristic clearly revealed that the photocurrent of Cu–Ce/
TiO2 catalysts is highest, suggesting the catalysts were advanta-
geous for the production of photoelectrons, by acting as light-
harvesting units [32].

3.2. Catalytic performance

The catalytic activities of the samples were determined in the

synthesis of methanol from CO2. As shown in Fig. 5, Cu–Ce/TiO2
is the most active catalyst comparing the CH3OH yield with the
Ce/Ti12O24 other catalysts, which due to the procedure of catalysts prepara-
tion. Ce and Cu were well-proportioned on the surface of carrier
TiO2 for Cu–Ce/TiO2, which were stochastic for (Cu–Ce)/TiO2 and
layered for Ce–Cu/TiO2. The methanol yield could reach up to
180.3 lmol/g-catal rapidly, which is higher than the work reported
by Tseng et al. [6]. Therefore, the modified Cu/TiO2 by Ce could im-
prove the efficiency of producing methanol.

3.3. Quantum calculation results

3.3.1. Cu and Ce doped TiO2 clusters of rutile and anatase phase

Cu and Ce doped TiO2 photocatalysts prepared in this work
were used to reduce CO2 in aqueous medium for the synthesis of
methanol. Cu and Ce affect the reaction differently, and what kind
of effect did they reveal during the reaction was the very interest-
CuTi12O24 ing question to discuss, but it is difficult to catch the effect of Cu
and Ce in catalysts during the reaction process under the irradia-
Fig. 6. Geometry structure of M/TiO2. tion of the UV lamp. There is necessary to study the different

Table 3
Part of charge distribution of the clusters.

M/e Ti O Ti O Ti O Ti O M O
Ce–R 1.435 0.685 1.492 0.918 1.400 0.638 1.312 0.660 1.710 0.910
Ce–A 1.491 0.935 1.527 0.892 1.475 0.793 1.486 0.876 1.782 0.903
Cu–R 1.438 0.809 1.505 0.828 1.463 0.682 1.106 0.918 0.497 0.684
Cu–A 1.474 0.912 1.570 0.823 1.476 0.716 1.483 0.795 0.417 0.889

The unit of charge is e, ‘‘M’’ refers to the metal atoms loaded on the surface of titanium dioxide, and ‘‘R’’ refers to rutile and ‘‘A’’ refers to anatase.
 D. Luo et al. / Journal of Molecular Structure 994 (2011) 325–331 329

effects of Cu and Ce on the surface of the catalysts with quantum DETi8 O22 H12 Ce ¼ ETi12 O22 H12 Ce  ETi12 O22 H12  ECe ; DETi8 O22 H12 Ce was
computation. The results of all calculation in this work have indi-  897:3393 kJ=mol:
cated that Ce on TiO2 surface affected the reaction stronger than
Cu stated at the clearance of the crystal and the external surface of
that of Cu. The Cu atoms on the surface of catalysts are channels
the catalysts, and no matter 1 or 0 valence of the Cu atoms that they
of photoelectrons in real time, which prevented recombination of
act as the transports of the electrons from the catalysts surface, and
the electrons and holes.
they prevent the recombination of the electrons and holes
The optimized parameters of M/TiO2 clusters as shown in Fig. 6
accordingly.
were presented in Table 2 both the bond length of the two metal
Time development of photocurrent transients for the samples
ions and the O atoms of the anatase TiO2 were shorter than that
were presented in Fig. 4, and the photocurrent of the Cu–Ce/TiO2
of the rutile TiO2, and the combination of M–O is tighter in the ana-
catalysts was highest, and it indicates that the catalysts prevent
tase phase than that of which in the rutile phase. The anatase TiO2
the photodecomposition of the electrons and holes, by acting as
was the active crystal which was consistent with the characteriza-
light-harvesting units, which was in good consistency with the
tion of Raman spectrum.
analysis of the geometry structure parameters of the M doped
As shown in Table 2 the radius of Cu is much smaller than Ce,
TiO2 above.
while the bond length Ce–O was almost equal to the bond length
The charge distribution of the clusters was list in the Table 3.
Cu–O, wherefore Ce combined with TiO2 surface was more firmly
There were higher charges on the anatase phase surface than that
than Cu, which explained why the loading sequence of Ce and Cu
of the rutile phase as a whole, which revealed the more activate of
affected the activity of the photocatalysts, and Ce entered the crys-
the anatase surface than that of the rutile surface, and it was con-
tal lattice of TiO2 to bring lattice distortion, which be advantageous
gruous with the Raman characterization and the photocurrent
for the production of the photoelectrons and holes, and the bond-
transients. The absolute value of the charge reduced from M was
ing energy DE followed were consistent with optimized parame-
very interesting, and the charge on M atoms minimized about 2
ters of M/TiO2 clusters.
times, and the absolute value of the charge on the Ce atom was
DETi12 O24 Ce ¼ ETi12 O24 Ce  ETi12 O24  ECe ; DETi12 O24 Ce was much lower than that on Ce4+, it revealed that the delocalization
of the positive charge to the surroundings around the Ce atom
 1004:2042 kJ=mol;
would increased the opportunity for the induction of the photo-
electrons and holes.
DETi12 O24 Cu ¼ ETi12 O24 Cu  ETi12 O24  ECu ; DETi12 O24 Cu was
 146:3353 kJ=mol; 3.3.2. Interaction of Cu or Ce with H2O and CO2
Interaction between reactants (H2O and CO2) and Cu as well as
DETi8 O22 H12 Cu ¼ ETi12 O22 H12 Cu  ETi12 O22 H12  ECu ; DETi8 O22 H12 Cu was Ce on the surface of catalysts have been calculated also. Two
 49:4478 kJ=mol; models M–3CO2 and M–2H2O–CO2 (M = Ce and Cu) were selected

Ce–2H2O–CO2 Cu–2H2O–CO2

Cu–3CO 2 Ce–3CO2

Fig. 7. Geometry Structure of Cu–3CO2, Cu–2H2O–CO2, Ce–3CO2 and Ce–2H2O–CO2.

330 D. Luo et al. / Journal of Molecular Structure 994 (2011) 325–331

Fig. 8. IR spectrum of M–3CO2 and M–2H2O–CO2.

to study the interaction of Ce and Cu with CO2 and H2O molecule. The charge on O atoms of H2O in the Ce–2H2O–CO2 system was
Optimized geometry parameters of M–3CO2 and M–2H2O–CO2 are 0.782 electron which was little smaller than that of H2O mole-
presented in Fig. 7. No imaginary frequency mode was found, cule, and the charge of H atoms was 0.608 and 0.607 which was
which is equivalent to classifying all the species as stable com- much higher than that of H2O molecule. The charge of O atom of
pounds, being at the minimum of their potential energy surface CO2 near to Ce atom was 0.530 electron and the charge of O atom
[33]. Our results reveal that the bond lengths of CO2 and H2O elon- of CO2 far from the Ce atom was 0.196 e. In a word, it was the result
gate significantly. The bond lengths of C–O in Cu–3CO2 are 1.173 Å of the activation from the Ce atom the system, which induce the
and 1.197 Å, whereas it is 1.604 Å in CO2. The distance between Cu potential of CO and the H–O bond lengthened.
and O is 2.053 Å. In the case of Cu–2H2O–CO2, the bond length of IR spectrums of M–3CO2 and M–2H2O–CO2 were shown in
C–O is 1.177 Å and 1.193 Å. The bond lengths of O–H and Cu–O Fig. 8, and IR Spectrums of the systems were more complicated
are 0.965 Å and 1.889 Å, respectively. It is interesting that CO2 is than that of the H2O and CO2, and it was the result from the effect
bound to one H2O. The distance of H–O is 1.749 Å, which is approx- of Ce and Cu atoms to the systems. C–O bond stretching vibration
imate to the hydrogen bond. In a word, Cu activates the CO2 and of CO2 molecule was in 2400–2600 cm1, and C–O bond bending
H2O not much compared to Ce. The main effect of Cu on the surface vibration was in 600–800 cm1. H–O bond stretching vibration of
of the catalysts is to conduct photoelectrons. H2O was in 3500–4000 cm1, and H–O bond rocking vibration
As shown in Fig. 7, the length of the Ce–O bond is 2.211 Å, and was in 1500–2000 cm1. The infrared absorption spectra of
the lengths of the C–O bonds are 1.258 Å and 1.150 Å. It reveals Ce–2H2O–CO2 system was made redshift to long wave entirely,
that Ce activates the CO2 more significantly than Cu. In the and Ce–O–C bond stretching vibration was in 1000–1250 cm1,
Ce–2H2O–CO2 system, the bond lengths of C–O are 1.262 Å and and H–O bond rocking vibration wagging vibration was in 750–
1.151 Å, respectively. The bond length of H–O is 1.009 Å, which in- 1000 cm1. The IR spectrum of Cu–2H2O–CO2 system was redshift
duce the dissociate trend of the H2O, Therefore, Ce activates H2O not so much contrasted to that of Ce–2H2O–CO2 system. H–O bond
and CO2 molecule so severely that they be decomposed. Ce on stretching vibration of Cu–2H2O–CO2 was about the same to that of
the surface of the catalysts helped to produce photoelectrons and H2O molecule, and the O–H–O (Fig. 7 O6–H10–O2) stretching
activate the reactants. vibration was about in 3500 cm1. The IR absorption of Cu–3CO2
Mulliken Atomic charges on the atoms of the two models was almost the same to that of CO2 molecule, and that the
M–3CO2 and M–2H2O–CO2 were labeled in the Fig. 7, and the Cu–3CO2 system increased a new absorption Ce–O stretching
charge values of CO2 (C 0.497, O 0.248, O 0.248) and H2O vibration in about 1000–1250 cm1, and the C–O bond bending
(O 0.797, H 0.399, H 0.399) were far from that of M–3CO2 and vibration was in about 500 cm1.
M–2H2O–CO2. Charges on the O atoms near to the Cu atom of
the Cu–3CO2 system were more negative and the charges on C
atoms were more positive, and the same trend to C for O atoms 4. Conclusions
far from Cu atom, and all electron cloud flow to the O atoms near
to the Cu atom to delocalize the positive charge on the Cu atom Photocatalysts Cu/Ce–TiO2 prepared via equivalent-volume
consequently, and the effect of the Cu atom was to conduct incipient wetness impregnation method can lead to major differ-
electrons according to the geometry structural parameters of ences in the catalytic properties. The optimal reaction temperature
Cu–3CO2 system. The same charge distribution tendency to the is 50 °C and the optimal amount of copper loading is 2 wt.% in the
Cu–3CO2 system of the Ce–3CO2 system except the activation of present catalyst system. Cu–Ce/TiO2 catalyst was highly Raman
Ce to the CO2 molecule. and photovoltaic responses, and the calculation results revealed
 D. Luo et al. / Journal of Molecular Structure 994 (2011) 325–331 331

that Ce atom enter the crystal lattice of TiO2 to bring lattice distor- [14] M.A. San Miguel, J. Oviedo, J.F. Sanz, Structure: THEOCHEM 237 (2006) 769.
[15] L. Giordano, G. Pacchioni, T. Bredow, J.F. Sanz, Surf. Sci. 471 (2001) 21.
tion, which is advantageous for the production of photoelectrons
[16] A.J. Duan, R.L. Li, G.Y. Jiang, J.S. Gao, Z. Zhao, G.F. Wan, D.Q. Zhang, W.Q. Huang,
and holes and possibly lead to the catalytic activity. The results K.H. Chung, Catal. Today 4 (2009) 188.
of all calculation in this work indicated that Ce on TiO2 surface af- [17] V.V. Narkhede, A.D. Toni, V.S. Narkhede, M. Reichinger, A. Birkner, J.W. (Hans)
fected the reaction stronger than that of Cu for higher bonding en- Niemantsverdriet, W. Grunert, H. Gies, J. Mater. Sci. 44 (2009) 6702.
[18] L. Kaluza, M. Zdrazil, Z. Vit, (Eng) React. Kinet. Catal. Lett. 97 (2) (2009) 307.
ergy and shorter bond length with O atom of TiO2 surface. Thus [19] R. Prasad, G. Rattan, Bull. Chem. React. Eng. Catal. 5 (2010) 8.
under the experimental conditions of this work, the methanol yield [20] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 785 (1988) 37.
could reach up to 180.3 l mol/g-cat for Cu–Ce/TiO2 catalyst rap- [21] P.J. Hay, W.R. Wadt, J. Chem. Phys. 270 (1985) 82;
P.J. Hay, W.R. Wadt, J. Chem. Phys. 284 (1985) 82;
idly. The action of Cu on the surface of catalysts is the channels P.J. Hay, W.R. Wadt, J. Chem. Phys. 299 (1985) 82.
of photoelectrons in the real time, which prevented recombination [22] J.D. Dill, J.A. Pople, J. Chem. Phys. 2921 (1975) 62.
of the electrons and holes, The action of Cu on the surface of cata- [23] M. Dolg, H. Stoll, H. Preuss, R.M. Pitzer, J. Phys. Chem. 5852 (1993) 97.
[24] R. Krishnan, J.S. Binkley, R. Seeger, J.A. Pople, J. Chem. Phys. 650 (1980) 72.
lysts is the channels of photoelectrons in the real time, which pre- [25] Gaussian 03, Revision E.01, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria,
vented recombination of the electrons and holes, Ce atoms activate M.A. Robb, J.R. Cheeseman, J.A. Montgomery, Jr., T. Vreven, K.N. Kudin, J.C.
the CO2 and H2O at the ground state from the geometry structural Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G.
Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
parameters of Ce–2H2O–CO2, while Cu atom effect them so weak Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
that it acts as conductor for photoelectrons because of the effect Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J.
of Cu atom to the yield of CH3OH. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J.
Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas,
References D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G.
Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
[1] H. Harada, Ultrason. Sonochem. 73 (1998) 5. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
[2] M. Anpo, Tae H. Kim, M. Matsuoka, Catal. Today 114 (2009) 124. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong,
[3] S.Z. Chen, S.H. Zhong, X.F. Xiao, Chineses J. Catal. 24 (1) (2003) 67. C. Gonzalez, J.A. Pople, Gaussian, Inc., Wallingford CT, 2004.
[4] S.Z. Chen, S.H. Zhong, Acta Phys. – Chim. Sin. 18 (12) (2002) 1099. [26] K. Porkodi, S. Daisy Arokiamary, Mater. Charact. 495 (2007) 6.
[5] S. Hosna, W. Nasution, E. Purnama, et al., Catal. Commun. 313 (2005) 6. [27] K.R. Zhu, M.S. Zhang, Q. Chen, et al., Phys. Lett. A 222 (2005) 340.
[6] I.-H. Tseng, W.-C. Chang, et al., Appl. Catal. B: Environ. 37 (2002) 38. [28] S. Kelly, F.H. Pollak, M. Tomkiewicz, J. Phys. Chem. B 2731 (2007) 101.
[7] H. Yamashita, Y. Fujii, Y-C. Ichihashi, et al., Catal. Today 221 (1998) 45. [29] L.M. Abrantes, L.M. Peter, J. Electroanal. Chem. Interf. Electrochem. 593 (1983)
[8] H. Yamashita, K. Ikeue, et al., Topics Catal. 95 (2002) 18. 150.
[9] M. Anpo, S.G. Zhang, H. Yamashita, Stud. Surf. Sci. Catal. 1587 (1997) 105. [30] L.M. Peter, Chem. Rev. 753 (1990) 90.
[10] K. Ikeue, H. Yamashita, M. Anpo, Chem. Lett. 28 (11) (1999) 1135. [31] R. Beranek, H. Kisch, Electrochem. Commun. 763 (2007) 4.
[11] K. Hirano, K. Inoue, T. Yatsu, J. Photochem. Photobiol. A: Chem. 255 (1992) 64. [32] P.M. Jayaweera, T. Sanjeewa, K. Tennakone, Solar Energy Mater. Solar Cells 91
[12] Y.H. Zhang, H.X. Zhang, Y.X. Xu, Y.G. Wang, J. Solid State Chem. 3490 (2004) (10) (2007) 944.
177. [33] C. Jarosław, J. Chojnacki/Polyhedron 969 (2008) 27.
[13] W. Li, S.I. Shah, C.P. Hunag, O. Jung, C. Ni, Mater. Sci. Eng. B247 (2002) 96.