High efficiency photocatalytic conversion of CO2 with H2O over Pt/TiO2 nanoparticles

Article abstract of DOI:10.1039/c4ra07457b

The objective of this research was to convert CO₂ to CH₄ using photocatalysis. In this investigation, well-crystallized anatase TiO₂ nanoparticles with diameters of ～18.4 nm and a BET surface area of 108.3 m² g^-1 were synthesized by a solvothermal method and then Pt/TiO₂ nanocomposites with high photocatalytic performance and stability were prepared by photoreduction. The average diameter of the well-dispersed Pt nanoparticles was 1.82 nm. The optimal TiO₂ calcination temperature was 500 °C, while Pt photodeposition time was 1 h. The photoluminescence (PL) spectra demonstrated that the efficiency of photogenerated charge transfer and separation of Pt/TiO₂ was better than that of Pt/P25. For the optimized Pt/TiO₂ photocatalysts, the CH₄ yield reached 60.1 μmol/(g_cat h), together with a H₂ yield of 87.5 μmol/(g_cat h) and an C₂H₆ yield of 2.8 μmol/(g_cat h) at 4 h of irradiation. The photocatalytic CO₂ conversion ability of Pt/TiO₂ was 3.7 times that of Pt/P25. Longer irradiation would result in an increase of H₂ output, but not the output of CH₄ or C₂H₆. With 10.0% CO₂, the CH₄ yield and H₂ yield reached 54.4 μmol/(g_cat h) and 80.2 μmol/(g_cat h), respectively. O₂ was detected and recorded for a systematic analysis of the relationship between CO₂ conversion and water splitting. In addition, the photogenerated electron-hole balance was calculated and an associated reaction mechanism was proposed.

Full text of DOI:10.1039/c4ra07457b

RSC Advances
View Article Online
View Journal | View Issue
PAPER
High efficiency photocatalytic conversion of CO2
with H2O over Pt/TiO2 nanoparticles
Cite this: RSC Adv., 2014, 4, 44442
c
ad
Yan Wang,a Qinghua Lai,a Fan Zhang,ab Xiaodong Shen, Maohong Fan,
*
Yiming Heae and Shenqiang Renf
The objective of this research was to convert CO2 to CH4 using photocatalysis. In this investigation, well-
crystallized anatase TiO2 nanoparticles with diameters of ꢀ18.4 nm and a BET surface area of 108.3 m2 gꢁ1
were synthesized by a solvothermal method and then Pt/TiO2 nanocomposites with high photocatalytic
performance and stability were prepared by photoreduction. The average diameter of the well-dispersed
Pt nanoparticles was 1.82 nm. The optimal TiO2 calcination temperature was 500 ꢂC, while Pt
photodeposition time was 1 h. The photoluminescence (PL) spectra demonstrated that the efficiency of
photogenerated charge transfer and separation of Pt/TiO2 was better than that of Pt/P25. For the
optimized Pt/TiO2 photocatalysts, the CH4 yield reached 60.1 mmol/(gcat h), together with a H2 yield of
87.5 mmol/(gcat h) and an C2H6 yield of 2.8 mmol/(gcat h) at 4 h of irradiation. The photocatalytic CO2
conversion ability of Pt/TiO2 was 3.7 times that of Pt/P25. Longer irradiation would result in an increase
of H2 output, but not the output of CH4 or C2H6. With 10.0% CO2, the CH4 yield and H2 yield reached
54.4 mmol/(gcat h) and 80.2 mmol/(gcat h), respectively. O2 was detected and recorded for a systematic
analysis of the relationship between CO2 conversion and water splitting. In addition, the photogenerated
electron–hole balance was calculated and an associated reaction mechanism was proposed.
Received 22nd July 2014
Accepted 5th September 2014
DOI: 10.1039/c4ra07457b
www.rsc.org/advances
solar energy, it will serve to reduce atmospheric CO2 concen-
trations while providing a renewable energy fuel compatible
with our current energy infrastructure.8
1. Introduction
The rapid increase in anthropogenic CO2 emissions is of great
concern, since fossil fuel consumption is not likely to cease in
the foreseeable future,1–3 and one major CO2 emission source is
the ue gas from power plants.4 Although CO2 capture and
storage (CCS) have been suggested, the ability and environ-
mental consequences of storing several billion tons of CO2 per
year are uncertain and costly.5 Another great challenge that has
attracted high attention is the depletion of fossil fuels and the
subsequent contribution to an energy crisis.6 One promising
strategy for solving both issues is to develop new utilization
methods for converting CO2 into useful chemicals and clean
fuels, which however is energy-intensive since CO2 is very
stable. A potentially cost-effective approach is to harness the
nearly limitless energy available in sunlight by using efficient
photocatalysts to perform these conversions.7 If CO2 can be
converted into useful products like CO, CH4 or CH3OH using
Various attempts have been made at photocatalytic CO2
conversion.9–11 Among the reported photocatalysts, TiO2 has
been investigated most frequently due to its robust reactivity,
commercial availability and chemical stability. However, TiO2
still has several disadvantages including the limit to only UV
light absorption due to its wide band-gap, and low quantum
efficiency resulting from the quick recombination of photo-
generated electron–hole pairs.12–14 To resolve these issues and
increase the photocatalytic efficiency of TiO2, great endeavors
have been made and they can be generally classied into two
kinds. The rst way to enhance the photoactivity of TiO2 is
through TiO2 modication. Numerous modications have been
made, including metal or nonmetal ions doping, dye sensiti-
zation, semiconductor combination and noble metal loading.15
Noble metal loading refers to the loading of Ag, Au, Pt, Pd, etc.
on the surface of TiO2. The loaded noble metal has a higher
work function than TiO2, which induces the formation of a
Schottky barrier in the TiO2-metal region. Thus, the photo-
generated electrons can migrate from TiO2 conduction band
(CB) to the metal phase, retarding the recombination of elec-
tron–hole pairs and subsequently increasing the photocatalytic
activity.16,17 Among the noble metal loaded TiO2 photocatalysts,
Pt/TiO2 has attracted much attention due to its much higher
photoactivity than bare TiO2. For example, Zhang et al. prepared
aDepartment of Chemical & Petroleum Engineering, University of Wyoming, Laramie,
WY, 82071, USA. E-mail: mfan@uwyo.edu
bSchool of Chemical and Environmental Engineering, China University of Mining and
Technology, Beijing 100083, China
cCollege of Materials Science and Engineering, Nanjing University of Technology,
Nanjing, 210009, China
dSchool of Energy Resources, University of Wyoming, Laramie, WY, 82071, USA
eDepartment of Materials Physics, Zhejiang Normal University, Jinhua, 321004, China
fDepartment of Chemistry, University of Kansas, Lawrence, KS, 66045, USA
44442 | RSC Adv., 2014, 4, 44442–44451
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Pt/P25 photocatalyst and used it to convert CO2 to CH4 through 0.28 mol CH3COOH (glacial, Fisher-Scientic) and 0.12 mol
photocatalysis, and the CH4 yield was 2.34 mmol/(gcat h).18 Zhai deionized (DI) H2O were mixed to form solution A. Then, 0.35
and coworkers also tested the photoactivity of Pt/P25 to convert mol CH3CH2OH and 0.04 mol titanium n-butoxide (reagent
CO2, and the CH4 yield was 11 mmol/(gcat h), which was 9.2 times grade, 97%, Sigma-Aldrich) were mixed to form solution B.
that of pure P25.19
Next, solution A was dropped into solution B under stirring to
The second way to enhance the photoactivity of TiO2 is to make a homogeneous transparent yellowish sol. The nal sol
improve the crystallinity and surface area of TiO2, so that the was stirred for 1 h before being introduced into a 200 mL
photogenerated charge carriers have better migration and Teon-lined stainless steel autoclave. The autoclave was main-
ꢂ
separation ability. TiO2 nanoparticles prepared with high crys- tained at 180 C for 24 h. TiO2 powder was obtained by centri-
tallinity and large surface areas are widely utilized in organic fuging, washing with ethanol and DI water for several times,
decomposition and H2 generation.20–23 Jiao et al. prepared TiO2 and then drying at 80 C for 24 h. Finally, the materials were
ꢂ
hollow spheres with well-crystallized anatase phase and a large heated with a temperature increase rate of 2 ꢂC minꢁ1 and then
surface area of 104 m2 gꢁ1 to obtain high photocatalytic phenol calcined at 500 C for 2 h.
ꢂ
degradation rates.24 Yu and coworkers prepared Pt/TiO2 nano-
The Pt/TiO2 was prepared by the photoreduction method.
sheets with a surface area of 94 m2 gꢁ1 and used them in pho- First, 0.2 g TiO2 was suspended in a mixture of 30 mL DI H2O, 6
tocatalytic H2 generation. The photocatalytic H2 generation mL 1 mg mLꢁ1 H2PtCl6 (8 wt% chloroplatinic acid solution,
rates of the Pt/TiO2 nanosheets were 1.5 times higher than that Sigma-Aldrich) and 10 mL 0.1 mol Lꢁ1 CH3OH (HPLC grade,
of Pt/P25 (SBET ¼ 41 m2 gꢁ1).25 Nevertheless, the research of Fisher-Scientic). Then, the whole suspension was stirred for
photocatalytic CO2 conversion using TiO2 with a large surface 0.5 h before being irradiated with a 500 W Xe lamp (Newport)
area is still scarce, and most research using Pt/TiO2 for photo- for a certain time. Next, the Pt/TiO2 samples were separated by
catalytic CO2 conversion only focuses on P25 with a surface area ltration, washed repeatedly with DI water, and nally dried at
of around 50 m2 gꢁ1
the photocatalytic ability of Pt/TiO2 systems. In order to further (Degussa) was also prepared by the same procedure.
improve the photocatalytic CO2 conversion ability of Pt/TiO2 UV-vis absorption spectra were obtained on a Cary 4000 UV-
photocatalysts, an effective synthetic method to produce TiO2 vis spectrophotometer (Varian), using MgO as reference (99+%,
with good crystallinity and a large surface area is needed. metals basis, Alfa Aesar). Photoluminescence (PL) spectra were
,
24 which largely limits the improvement of 80 ꢂC in a vacuum oven for 24 h. For comparison, Pt/P25
The solvothermal method has been gaining continuous measured by a Cary Eclipse uorescence spectrophotometer
interest because of its ease in controlling the size, crystal phase, (Varian). BET (Brunauer–Emmett–Teller) surface area was
size distribution and minimizing agglomeration.15 In this measured by a Quantachrome Autosorb IQ automated gas
investigation, anatase TiO2 nanoparticles with large surface sorption analyzer with an outgas temperature of 180 ꢂC for 3 h
areas were synthesized by a novel solvothermal method. Pt and a N2 adsorption–desorption isotherm. Scanning electron
nanoparticles were deposited on the surface of TiO2 through microscopy (SEM) images were obtained by an FEI Quanta 450
photoreduction. The inuence of TiO2 substrate and Pt depo- eld emission scanning electron microscope with a Schottky
sition time were thoroughly studied. Although CH4, CO and eld emission gun at an accelerating voltage of 20 kV. High
CH3OH were widely reported as photocatalytic CO2 conversion resolution scanning transmission electron microscopy (HR-
products, the concentration change with reaction time hasn't STEM) was taken by an FEI Tecnai G2 F20 S-Twin 200 kV
been fully investigated. This doesn't facilitate the under- transmission electron microscope. X-ray diffraction (XRD) data
standing of the reaction mechanism. 10.0% CO2 with the were collected using a Rigaku Smart Lab automated powder
balance gas of N2 was used to mimic the ue gas and to test the diffraction system. Runs were made with a scan range of 20–80ꢂ
photocatalytic CO2 conversion efficiency as well. Although it is at 5ꢂ minꢁ1 with a step increment of 0.02ꢂ. X-ray photoelectron
widely accepted that the photocatalytic CO2 conversion is spectroscopy (XPS) analysis was performed using a Physical
accompanied with water splitting, only reductive products, such Electronics ESCA 5800 spectrometer. High-resolution XPS scans
as CH4, CH3OH, H2, were reported in many investigations.8,26 were performed with a pass energy of 23.5 eV and a step size of
The lack of O2 might raise questions regarding the reliability of 0.1 eV. Adventitious carbon was used to determine the induced
previous experimental data and generate confusion on the offset of the binding energy of the C1s peak (284.8 eV) due to
relationship of CO2 conversion and water splitting. So in this surface charging effects. Inductively coupled plasma optical
work, a detailed analysis on CO2 conversion together with water emission spectroscopy (ICP-OES) analysis was tested on a Per-
splitting was conducted. The balance of generated electrons kinElmer Optima 8300 ICP-OES spectrometer.
and holes was calculated. This might provide some insight into
understanding photocatalytic CO2 conversion mechanisms.
2.2 Photocatalytic CO2 conversion experiments
The experimental apparatus built for photocatalytic CO2
conversion is shown in Fig. 1. It consists of three parts,
including a light source unit, a photocatalytic reaction system,
2. Experimental
2.1 Materials preparation and characterization
All chemicals were purchased and directly used without any and online product measurement equipment. The photo-
further purication. TiO2 was prepared by a solvothermal catalytic CO2 conversion was carried out in a 132 mL stainless
method. First, 0.22 mol CH3CH2OH (200 proof, Decon Labs), steel reactor (4) with an inner Teon liner and a quartz window
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44442–44451 | 44443

View Article Online
RSC Advances
Paper
Fig. 1 Schematic diagram of the photocatalytic CO2 conversion set
up. (1a) CO2 (1b) Ar (1c) H2 (2a) mass flow meter (2b) mass flow
controller (3a) light intensity controller (3b) light source (4) reactor (5)
pressure gauge (6) quartz glass (7) pressure relief valve (8a) thermo-
couple (8b) heating tape (8c) temperature controller (8d) temperature
displayer (9) gas chromatograph (10) computer.
Fig. 2 XRD patterns of TiO2, Pt/TiO2, used Pt/TiO2, P25 and Pt/P25.
˚
where D is the crystal size, l is the X-ray wavelength (1.54056 A),
on the top of the reactor. A 500 W Xe lamp (3b) was used for
illumination. A heating tape (8b) with a temperature controller
(8c) was used to regulate the reaction temperature when heating
was needed. Typically, 0.02 g photocatalyst was placed on a
Teon catalyst holder in the upper region of the reactor and 5
mL DI water was pre-injected into the bottom of the reactor.
Before reaction, CO2 was used to purge the system for 1 h to
completely remove the air inside the reactor. The pressure of
CO2 (laser grade, ꢀ21 mmol in total) inside the reactor was
regulated to 4 bar. Then the Xe lamp was turned on and the
photocatalysts would react with the water vapor and CO2 under
illumination. Aer 4 hours of reaction, the Xe lamp was turned
off and both the gas and liquid components were analyzed. An
SRI 8610 C gas chromatograph with two FID detectors and one
TCD detector was used to detect the gas component. The
columns used were 60 MS-13X, 60 HAYESEP-D and 60M MXT-1.
The liquid component was detected by an Agilent Technologies
1260 Innity high-performance liquid chromatograph (HPLC)
with a Refractive Index Detector and Zorbax carbohydrate
analysis column and Agilent 5975 C gas chromatography mass
spectrometer (GC-MS) with a Triple-Axis Detector and an HP-
5MS column.
b is the full-width at half maximum (FWHM) of the diffraction
peak and q is the diffraction angle.27 The as calculated average
crystal sizes are 13.9 nm for Pt/TiO2, and 22.4 nm for Pt/P25.
3.1.2 SEM and STEM images. The shape and size of the
nanoparticles are shown by SEM and STEM images. Fig. 3(a)
shows a representative SEM image of the TiO2 nanoparticles
ꢂ
calcined at 500 C. The average TiO2 nanoparticle size is 18.4
nm, while that for P25 is 27.5 nm. Fig. 3(b) is a high-angle
annular dark eld (HAADF) STEM image of the Pt/TiO2 nano-
particles. Ultrane Pt with an average size of 1.82 nm (Fig. 3(c))
is evenly distributed on the surface of TiO2. A smaller particle
size usually indicates a larger surface area. The BET surface area
3. Results and discussions
3.1 Characteristics of catalysts
3.1.1 XRD analysis. The structure of photocatalysts was
characterized by XRD. Fig. 2 shows the XRD patterns of TiO2, Pt/
TiO2, used Pt/TiO2, P25 and Pt/P25. It can be observed that the
deposition of Pt did not generate any Pt peaks, which might
result from the low loading amount of Pt (less than 1% of Pt/
TiO2) and the extremely small size of Pt nanoparticles (1–3
nm).18 Only typical anatase TiO2 diffraction peaks of (101), (103),
(004), (112), (200), (105), (211), (204), (116), (220) and (215) were
observed for Pt/TiO2 samples. The average crystal size was
calculated using Debye–Scherrer equation by peak (101), (004)
and (200):
Fig. 3 (a) SEM image of TiO2 (b) STEM image of Pt/TiO2 (c) Pt particle
D ¼ 0.89 ꢃ l/(b cos q)
size distribution for Pt/TiO2.
44444 | RSC Adv., 2014, 4, 44442–44451
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
of Pt/TiO2 (110.4 m2 gꢁ1) is much larger than that of Pt/P25 (59.7
m2 gꢁ1), and this is one of the reasons for the high photoactivity
of Pt/TiO2.
3.1.3 XPS analysis. Compared with XRD, XPS is more
sensitive for detection of surface species with low content.
Hence, Pt/TiO2 photocatalyst was investigated by XPS to verify
the existence and oxidation states of Pt species. From Fig. 4,
only part of the Pt species were fully reduced to metallic Pt. The
most intense peaks at 71.3 eV and 74.4 eV are due to metallic Pt
4f core-level. The second set of doublets at 72.8 eV and 76.2 eV
can be assigned to the photoemission of Pt2+ chemical states
such as PtO and Pt(OH)2. The third doublets of Pt at higher
binding energies of 74.8 eV and 78.0 eV are assigned to Pt4+
such as PtO2 or Pt(OH)4.28–30 The ratio of platinum species, Pt0/
Pt2+/Pt4+, was calculated with the peak areas of Pt 4f7/2
,
.
Compared to used Pt/TiO2, the Pt/TiO2 and Pt/P25 peaks have a
positive shi. Such a positive shi is postulated for highly
dispersed and small size Pt.31 The negative shi from used Pt/
TiO2 may not only because of the Pt agglomeration, but also due
to the fact that electrons transfer from TiO2 to Pt because of a
strong metal-support interaction (SMSI), which is based on the
electronic and structural effects involving both the active metal
and the support, and it may cause the rich electron states of
Pt.31,32 The negative Pt would inhibit formation of Schottky
barriers between P25 and metallic Pt, which increases the
recombination of photo-induced electrons and holes and ulti-
mately reduces the efficiency of CO2 conversion.33,34 This is
consistent with the fact that Pt nanoparticles grow bigger and
partly explains the low efficiency of CO2 conversion rate by used
Pt/TiO2.35
3.1.4 UV-vis spectroscopy. The UV-vis light absorption is
given in Fig. 5(a). The signicant UV absorption below 390 nm is
assigned to the intrinsic bandgap absorption of TiO2 (ꢀ3.15 eV).
Fig. 5 (a) UV-vis absorption spectra of TiO2, Pt/TiO2, used Pt/TiO2,
P25 and Pt/P25 (b) Corresponding plot of transformed Kubelka–Munk
function versus the energy of the light.
The deposition of Pt on TiO2 or P25 increases its visible light
response ability.25 Secondly, PtO/Pt(OH)2 and PtO2/Pt(OH)4 on
the surface also contribute to the visible light absorption.36
The corresponding diffuse-reectance UV-vis spectra in
Fig. 5(b) were obtained according to the Kubelka–Munk func-
tion: F(R) ¼ (1 ꢁ R)2/2R, in which R is the reectance (%). The
Kubelka–Munk function was then applied to the Tauc equation
(hvF(R))1/n ¼ A(hv ꢁ Eg), where hv is the photo energy, Eg is the
bandgap and A is a constant. For indirect allowed transition, n
¼ 2, since the fundamental absorption of TiO2 possesses an
Fig. 4 XPS spectra of Pt 4f level of Pt/TiO2, used Pt/TiO2 and Pt/P25.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44442–44451 | 44445

View Article Online
RSC Advances
Paper
indirect transition between bands. If (hvF(R))1/2 is plotted
against hv and a tangent line is drawn from the inection point
on the curve, the intersection of the tangent line with the
horizontal axis is the bandgap.37 As calculated, the bandgap of
the Pt/TiO2 is 3.07 eV.
3.1.6 Characteristics of Pt/TiO2 with TiO2 calcined at
different temperatures. With the exception of the representative
photocatalysts, the Pt/TiO2 photocatalysts with TiO2 calcined at
different temperatures were also characterized. The results are
shown in Table 1. It is obvious that with the increasing of
calcination temperature, the TiO2 particle size is increasing
together with the decreasing of BET surface area. Meanwhile,
with the same deposition time, the surface Pt atomic percentage
rst increased and then decreased with a further increasing
calcination temperature ꢂfrom 500 ꢂC. No metallic Pt was
observed on Pt/TiO2-650 C, which agrees with Fig. 8 that the
photocatalytic activity decreased rapidly over this sample. While
at 650 ꢂC, the bandgap still decreased to 2.89 eV; this is probably
due to the growing amount of PtO/Pt(OH)2 and PtO2/Pt(OH)4 on
the surface of TiO2. Aer reaction, the surface Pt amount
decreased, together with an increase in average Pt size (ꢀ1.98
nm), and this agrees with the experiment results of Bai et al.41
3.1.5 PL spectroscopy. The PL emission results from the
recombination of photogenerated electrons and holes, and a
lower PL emission intensity usually indicates a decrease in
radiative recombination and better separation of photo-
generated electrons and holes.38 To understand the efficiency
of photogenerated charge transfer and separation of the Pt/
TiO2 photocatalyst, PL tests were carried out. Fig. 6 presents a
comparison of the PL spectra of several representative
samples. For all the samples, two main emission peaks appear
around 391 nm and 453 nm, which are equivalent to 3.17 and
2.74 eV, respectively. The 391 nm peak is due to the bandgap
transition emission with the energy of emission light approx-
imately equal to the bandgap energy of TiO2.34 The PL signals
at about 430–500 nm mainly result from excitonic PL, due to
surface oxygen vacancies and defects of the TiO2 samples. The
PL peak at 453 nm is attributed to band edge free excitons.39
3.2 Factors affecting the performances of photocatalysts
3.2.1 Preparation conditions of photocatalysts. TiO2
Two small peaks at 472 nm and 486 nm result from bound calcined under different temperatures were used to synthesize
excitons.40 As shown in Fig. 6, the TiO2 bandgap emission is Pt/TiO2 and then used for photocatalytic CO2 conversion. As
lower than P25, indicating the electron–hole recombination shown in Fig. 7, with the increase in calcination temperature
rate through radiative combination is lower than that of P25. from 400–500 ꢂC, the photocatalytic CO2 conversion ability kept
ꢂ
The photodeposition of Pt nanoparticles generates consider- increasing, and reached a maximum value at 500 C. The CH4
able uorescence quenching, which indicates the transfer of yield was 277.2 ppm [60.1 mmol/(gcat h)] aer 4 h of illumina-
electrons from TiO2 to Pt and the favorable contact between tion, together with a H2 concentration of 389.2 ppm [87.5 mmol/
them. In addition, it should be noted that the PL peak of Pt/ (gcat h)] and an C2H6 concentration of 12.4 ppm [2.8 mmol/(gcat
TiO2 is weaker than that of Pt/P25, suggesting its lower elec- h)]. The increase in photoactivity is mainly due to the improved
tron–hole recombination rate through radiative combination. crystallization of anatase TiO2, an appropriate surface Pt
Under UV light irradiation, the electrons from the valence percentage (shown in Table 1), and an ultrane Pt size (shown
band of TiO2 will be excited to the conduction band and then in Fig. 3). When the calcination temperature was higher than
ꢂ
migrate rapidly to Pt nanoparticles, decreasing the direct 550 C, a sharp decrease in photoactivity was observed, which
recombination of charge carriers and nally contributing to could be ascribed to bigger and agglomerated TiO2,42 a low Pt
the high photocatalytic performance.25
content and a decrease of Pt0 state (shown in Table 1).
In addition to the effect of TiO2 calcination temperature, the
inuence of Pt deposition time was investigated. Fig. 8 shows
the photocatalytic activity of P25, Pt/P25 and Pt/TiO2 samples.
Pure TiO2 had a better photocatalytic activity than that of P25,
with a CH4 production of 12.4 ppm and a H2 production of 68.3
ppm, while P25 could only produce 3.8 ppm CH4, 11 ppm CO
and 23.7 ppm H2. The homemade TiO2 preferentially produced
CH4, rather than CO, and had a higher photocatalytic water
splitting ability. The Pt loading greatly promoted the photo-
catalytic activity of both TiO2 and P25. As can be seen in Fig. 8,
with the increase of Pt deposition time, the photocatalytic
performance rst sharply increased and then decreased slowly.
The optimal photodeposition time was 1 h under the given
experimental conditions. Considering that Pt was deposited on
TiO2 by photoreduction, the variation of the photocatalytic
performance of Pt/TiO2 might be due to different Pt nano-
particle content. With a short deposition time, not enough Pt
was reduced. From ICP-OES test results, with a deposition time
of 0.25 h, Pt counts for only 4.48& of the catalyst weight. It is
accepted that Pt has a good ability to trap photogenerated
electrons, but if not enough Pt was deposited, insufficient sites
Fig. 6 PL spectra of TiO2, Pt/TiO2, used Pt/TiO2, P25 and Pt/P25.
44446 | RSC Adv., 2014, 4, 44442–44451
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Table 1 Characteristics of Pt/TiO2 with TiO2 calcined at different temperatures
Samples
Structurea Bandgapb (eV) TiO2 particle sizec Pt particle sizec Pt weight percentaged Pt0 : Pt2+ : Pt4+ ratiod BET (m2 gꢁ1
)
Pt/TiO2-400 ꢂ
Pt/TiO2-450 ꢂ
Pt/TiO2-500 ꢂ
Pt/TiO2-550 ꢂ
Pt/TiO2-600 ꢂ
Pt/TiO2-650 ꢂ
C
C
C
C
C
C
Anatase
Anatase
Anatase
Anatase
Anatase
Anatase
Anatase
3.10
3.09
3.07
3.03
2.97
2.89
2.83
14.8 ꢄ 4.3
15.5 ꢄ 5.6
18.4 ꢄ 7.1
19.9 ꢄ 7.7
24.5 ꢄ 8.9
37.5 ꢄ 11.8
18.4 ꢄ 7.1
1.6 ꢄ 0.5
1.7 ꢄ 0.6
1.8 ꢄ 0.5
2.2 ꢄ 0.8
4.6 ꢄ 2.0
N/A
5.38&
8.02&
10.87&
8.87&
6.15&
3.02&
10.63&
18 : 56 : 26
19 : 52 : 29
21 : 40 : 39
21 : 49 : 30
15 : 40 : 45
N/A
128.5
124.6
110.4
49.7
37.7
12.1
Used Pt/TiO2-500 ꢂ
C
1.9 ꢄ 0.7
26 : 43 : 31
108.9
Obtained by XRD. Calculated from UV-vis spectra. c Measured from TEM images. d Analyzed from ICP-OES.
a
b
Fig. 8 Comparison of photocatalytic performances of P25, Pt/P25
Fig. 7 Effect of TiO2 calcination temperature on the concentrations of and Pt/TiO2 with different Pt deposition time (catalyst weight: 20 mg;
products resulting from photocatalytic CO2 conversion over Pt/TiO2 water volume: 5 mL; CO2 pressure: 4 bar; temperature: 80 ꢂC).
(catalyst weight: 20 mg; water volume: 5 mL; CO2 pressure: 4 bar;
temperature: 80 ꢂC).
reaction time, reaction temperature and CO2 concentration on
would not trap enough electrons for CO2 conversion.43 The
the photocatalytic performance of the optimized Pt/TiO2
increase in photodeposition time improved the Pt content,
samples were investigated. Fig. 9 is a plot of the product
promoting the photocatalytic activity. At the optimum Pt
deposition time of 1 h, Pt weight percentage was 10.87& of the
catalyst weight. At a Pt deposition time of 2 h, the Pt weight
percentage was 13.92& of the catalyst weight. When the depo-
concentration under different reaction time. The products
include CH4, CO, H2, C2H6 and O2. As can be seen in Fig. 9, for
the rst 4 h, CH4 concentration kept increasing and reached a
peak of 277.2 ppm. Under longer irradiation, the concentration
sition time was beyond the optimum, too much Pt was depos-
of CH4 did not grow anymore, which might be due to the
ited on the surface of TiO2, which might block the light to the
deterioration of the photocatalyst, resulting from the saturation
active sites on TiO2 and retard the conversion of CO2.25,44 For
of adsorption sites on the Pt/TiO2 surface with intermediate
products.45 This can be proved by repeated use of Pt/TiO2 pho-
tocatalysts. Aer removing the intermediate products, Pt/TiO2
photocatalyst still had 88.6% of its original activity. In the rst 3
h CO was detected with a peak concentration of 25.9 ppm, and
then it disappeared. Aer longer reaction time, CO diminished
when more photogenerated electrons were created; this
phenomenon can be attributed to the further conversion of CO
to CH4, since CH4 formation is an 8-electron photoreduction
process, while the formation of CO only needs 2 electrons.46 The
diminishment of CO can also be explained by the electro-
chemical potential, ꢁ0.53 V per NHE (normal hydrogen
comparison, Pt/P25 was also prepared by the photodeposition
method for 1 h under the same conditions. Aer photo-
deposition, the Pt weight percent was 11.21& of the catalyst
weight, similar to that of Pt/TiO2. The selectivity of P25 was
changed aer photodeposition. CO could not be detected. The
production of CH4 was 74 ppm [16.6 mmol/(gcat h)] while H2 was
151.4 ppm [34.0 mmol/(gcat h)], much smaller than that of Pt/
TiO2.
3.2.2 Effects of reaction conditions. In addition to the
photocatalyst itself, the reaction conditions also show great
effect on the photocatalytic reaction. Hence, the inuence of
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44442–44451 | 44447

View Article Online
RSC Advances
Paper
Fig. 10 shows the effect of reaction temperature on the
photocatalytic performance of Pt/TiO2. It is known that the
photocatalytic efficiency mainly depends on the photogenerated
electron–hole pair separation and reaction, but a higher
temperature can increase the diffusion rate and collision
frequency.4 As shown in Fig. 10, the concentration of C2H6
increased slightly with an increase in temperature, while CH4
output increased 8.7%, with a concentration change from 258.4
ꢂ
ꢂ
ppm at 30 C to 280.9 ppm at 175 C. Output of H2 increased
19.0%, with a concentration change from 358.4 ppm at 30 ꢂC to
426.5 ppm at 175 ꢂC. The increased temperature speeded up the
desorption process during the heterogeneous photocatalytic
reaction, resulting in a higher reaction rate.49 Hence, the output
of products was increased. Meanwhile, it can be noticed that the
reaction temperature had a stronger inuence on H2 produc-
tion than on CH4. This phenomenon may be attributed to an
improved concentration of water vapor with the increased
temperature, which is benecial for water splitting.
In addition to the effects of reaction time and temperature,
the inuence of CO2 concentration on the photoactivity was also
investigated. For these tests, 1.0% and 10.0% concentration of
CO2 diluted with N2 were used. From Fig. 11, the CH4 produc-
tion of 1.0% CO2 was 167.9 ppm, accounting for 60.8% of the
CH4 yield of pure CO2. For the 10.0% CO2, the CH4 production
was 251 ppm, accounting for 90.5% of the CH4 yield of pure CO2
and demonstrating the potential ability to photocatalytically
convert ue gas into CH4. As can be seen, the H2 production
with diluted CO2 was even higher than the H2 production with
pure CO2, which may be because there was more cH bonded to
form H2, instead of reacting with cCH3, due to the decreased
CO2 content.50 No C2H6 was observed with diluted CO2 since
there was enough cH and all cCH3 bonded with cH instead of
cCH3.
Fig. 9 Effect of reaction time on the concentrations of products
resulting from photocatalytic CO2 conversion over Pt/TiO2 (catalyst
weight: 20 mg; water volume: 5 mL; CO2 pressure: 4 bar; temperature:
80 ꢂC).
electrode) for CO and ꢁ0.24 V per NHE for CH4, which indicates
that CH4 is more stable than CO.26 In addition to CH4 and CO,
C2H6 was detected and reached its highest concentration (12.3
ppm) at 4 h. This value remained unchanged at longer irradi-
ation. The generation of C2H6 occurs when the system has an
abundance of photogenerated electrons, and can't provide
enough cH for cCH3, since cH has participated in H2 produc-
tion.47 Output of H2 kept increasing during 8 h of irradiation,
reaching 548.8 ppm. This demonstrated the photocatalytic
water splitting ability of Pt/TiO2 to form cH, which is in accor-
dance with the report of Nada, in which H2O is an electron
acceptor and competes with CO2 to accept electrons.45 The high
H2 yield may be partly due to enough water coverage on the
catalyst surface, which contributes to the generation of cH and
the transport of cH to CO2.48 The different increases of CH4 and
H2 demonstrate a difference in the intermediate balance
concentration and further different reaction active sites.42 Along
with the reduction products, O2 was detected in our system.
Aer reaction, the liquid was carefully analyzed by Agilent 1260
Innity HPLC and Agilent 5975 C GC-MS, but potential products
such as CH3OH, CH2OH and HCOOH were not detected.
Control experiments were carried out to rule out possibili-
ties of carbon contamination during the preparation process,
such as incomplete reaction of titanium n-butoxide, or from
CH3OH during Pt photodeposition. When He (5.0 UHP,
99.999%) was used as the carrier gas instead of introducing
CO2 into the reactor, while all the other condition remained
the same, no CH4/C2H6 was detected, suggesting that all the
CH4 and C2H6 came from CO2 conversion. Also, without light
illumination, no product was detected, demonstrating light
illumination is the driving force and CO2 conversion is a
photocatalytic process.
Fig. 10 Effect of temperature on the concentrations of products
resulting from photocatalytic CO2 conversion over Pt/TiO2 (catalyst
weight: 20 mg; water volume: 5 mL; CO2 pressure: 4 bar).
44448 | RSC Adv., 2014, 4, 44442–44451
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
the H2 output decreased from 389.2 ppm to 332.1 ppm.
Subsequently, Pt/TiO2 gradually became stable and the CH4
evolution didn't change aer ve rounds of reaction. Aer three
rounds of reaction, there was no C2H6 produced, due to the
deterioration of photocatalysts. Aer 10 rounds of reaction, the
CH4 concentration was 207.9 ppm and the H2 was 257.8 ppm,
demonstrating 75.8% of the original CH4 output; this was still
2.8 times that of Pt/P25. In comparison, the CH4 concentration
over Pt/P25 dropped to 60.2 ppm only aer one cycle,
accounting for 81.4% of its original photocatalytic ability and
was lower than that of Pt/TiO2.
3.4 Reaction mechanism
The generation of electron–hole pairs upon UV irradiation and
the migration of the charges on Pt/TiO2 photocatalyst are
obvious. The photoexcited electrons generated by TiO2 are
trapped by the Pt nanoparticles and the surface reduction
process takes place on Pt nanoparticles, while the holes stay on
TiO2 to oxidize H2O to O2. However, the photocatalytic CO2
conversion process is not clear. Based on previous literature and
the data obtained in this research, photocatalytic CO2 conver-
sion might occur according to the following process: CO2 is rst
reduced to CO at the Pt sites, while at the same time cH is
formed from water splitting. Then the as-formed CO reacts with
cH to form CH4 and C2H6.46,51 Formation of O2 was detected. For
example, over the Pt/TiO2 catalyst and a reaction time of 4 h, the
O2 concentration reached 664.4 ppm, the CH4 concentration
reached 277.2 ppm, together with a H2 concentration of 389.2
ppm and an C2H6 concentration of 12.4 ppm. From the
concentration of the reductive products, the stoichiometric
formation of O2 was calculated by assuming that the oxidation
of water to form O2 is the only source to provide electrons. The
O2 calculated concentration (O2,c), based on the quantities of
measured H2, CH4, and C2H6, can be estimated as follows:
Fig. 11 Effect of CO2 concentration on the concentrations of prod-
ucts resulting from photocatalytic CO2 conversion over Pt/TiO2
(catalyst weight: 20 mg; water volume: 5 mL; total pressure: 4 bar;
temperature: 80 ꢂC).
3.3 Photocatalyst stability
Although Pt/TiO2 photocatalyst demonstrates high photo-
catalytic performance in CO2 conversion, its stability is still very
important for practical applications. Repeated uses of Pt/TiO2
on photocatalytic CO2 conversion were conducted to explore the
Pt/TiO2 stability. These experiments were tested every 4 h for ten
times. Each time, aer one round, the reacted gas was evacu-
ated and new CO2 and DI water were introduced. As can be seen
in Fig. 12, aer one cycle, the CH4 output dropped to 245.6 ppm,
accounting for 88.6% of its original photocatalytic ability, while
O2,c ¼ 0.5H2,m + 0.5COm + 2CH4,m + 3.5C2H6,m
where H2,m, COm, CH4,m and C2H6,m refer to the measured
concentration of H2, CO, CH4 and C2H6, respectively. The
calculated O2 concentration and the measured O2 formation
concentration (O2,m) under different reaction time are listed in
Table 2. The errors between them are listed. Considering the
actual O2 formation is similar to the calculated stoichiometric
Table 2 Calculated O2 concentration vs. measured O2 concentration
Time (h)
O2,m (ppm)
O2,c (ppm)
Errora (%)
1
2
3
4
5
6
7
8
221.2
479.7
589.1
664.4
699.6
725.7
761.4
785.3
252.8
12.50f
9.45
529.75
656.35
783.4
798.15
815.95
837.15
859.2
10.25
15.19
12.35
11.06
9.05
Fig. 12 Time course of photocatalytic CO2 conversion over Pt/TiO2
(catalyst weight: 20 mg; water volume: 5 mL; CO2 pressure: 4 bar;
temperature: 80 ꢂC).
8.60
a
Error was calculated by (O2,c ꢁ O2,m)/O2,c
.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44442–44451 | 44449

View Article Online
RSC Advances
Paper
O2 formation value, the following reaction pathways at pH ¼ 7 Department of Energy in the US, as well as Nanjing University of
are proposed:19,26
Technology in China.
TiO2 + hv / TiO*2(eꢁ + h+)
Notes and references
1 S. Cui, W. Cheng, X. Shen, M. Fan, A. T. Russell, Z. Wu and
X. Yi, Energy Environ. Sci., 2011, 4, 2070.
Valance band (photooxidation)
2 B. Dutcher, M. Fan, B. Leonard, M. D. Dyar, J. Tang,
E. A. Speicher, P. Liu and Y. Zhang, J. Phys. Chem. C, 2011,
115, 15532.
2H2O + 4h+ / 4H+ + O2; Eroedox ¼ 0.82 V vs. NHE
3 B. T. Zhang, M. Fan and A. E. Bland, Energy Fuels, 2011, 25,
1919.
Conduction band (electrons transfer)
TiO2 (eꢁ, CB) + Pt / TiO2 (CB) + Pt (eꢁ)
4 P. Usubharatana, D. McMartin, A. Veawab and
P. Tontiwachwuthikul, Ind. Eng. Chem. Res., 2006, 45, 2558.
5 L. Cao, S. Sahu, P. Anilkumar, C. E. Bunker, J. Xu,
K. S. Fernando, P. Wang, E. A. Guliants, K. N. Tackett and
Y. P. Sun, J. Am. Chem. Soc., 2011, 133, 4754.
6 D. Ravelli, D. Dondi, M. Fagnoni and A. Albini, Chem. Soc.
Rev., 2009, 38, 1999.
7 S. Kwon, M. Fan, A. T. Cooper and H. Yang, Crit. Rev. Environ.
Sci. Technol., 2008, 38, 197.
8 O. K. Varghese, M. Paulose, T. J. LaTempa and C. A. Grimes,
Nano Lett., 2009, 9, 731.
Pt surface (photoreduction)
2H+ + 2eꢁ / H2; Eroedox ¼ ꢁ0.42 V vs. NHE
CO2 + 2H+ + 2eꢁ / CO + H2O; Eroedox ¼ ꢁ0.53 V vs. NHE
CO2 + 8H+ + 8eꢁ / CH4 + 2H2O; Eroedox ¼ ꢁ0.24 V vs. NHE
2CO2 + 14H+ + 14eꢁ / C2H6 + 4H2O; Eoredox ¼ ꢁ0.27 V vs. NHE
9 M. Halmann, Nature, 1978, 275, 115.
10 W. Fan, Q. Zhang and Y. Wang, Phys. Chem. Chem. Phys.,
2013, 15, 2632.
4. Conclusions
11 Q. Liu, Y. Zhou, J. Kou, X. Chen, Z. Tian, J. Gao, S. Yan and
Z. Zou, J. Am. Chem. Soc., 2010, 132, 14385.
12 H. Huang, D. Li, Q. Lin, W. Zhang, Y. Shao, Y. Chen, M. Sun
and X. Fu, Environ. Sci. Technol., 2009, 43, 4164.
13 W. Tu, Y. Zhou, Q. Liu, Z. Tian, J. Gao, X. Chen, H. Zhang,
J. Liu and Z. Zou, Adv. Funct. Mater., 2012, 22, 1215.
14 L. Luo, A. T. Cooper and M. Fan, J. Hazard. Mater., 2009, 161,
175.
15 X. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891.
16 H. Tada, T. Kiyonaga and S. I. Naya, Chem. Soc. Rev., 2009,
38, 1849.
17 Y. Xie, K. Ding, Z. Liu, R. Tao, Z. Sun, H. Zhang and G. An, J.
Am. Chem. Soc., 2009, 131, 6648.
18 Q. H. Zhang, W. D. Han, Y. J. Hong and J. G. Yu, Catal. Today,
2009, 148, 335.
19 Q. Zhai, S. Xie, W. Fan, Q. Zhang, Y. Wang, W. Deng and
Y. Wang, Angew. Chem., 2013, 125, 5888.
20 J. Liao, L. Shi, S. Yuan, Y. Zhao and J. Fang, J. Phys. Chem. C,
2009, 113, 18778.
21 C. Hu, X. Zhang, W. Li, Y. Yan, G. Xi, H. Yang, J. Li and
H. Bai, J. Mater. Chem. A, 2014, 2, 2040.
In summary, well-crystallized anatase TiO2 nanoparticles with a
high surface area and a diameter of ꢀ18.4 nm were prepared by
a solvothermal method at 180 ꢂC. Pt/TiO2 nanocomposites with
an average Pt nanoparticle size of 1.82 nm were then prepared
by photoreduction and used for photocatalytic CO2 conversion.
XPS analysis indicated that only part of the Pt species were
reduced to metallic Pt. The Pt/TiO2 nanocomposites had a
bandgap of 3.07 eV, and the photogenerated electron–hole
recombination rate was lower than that of Pt/P25. Pt photo-
deposition time and TiO2 calcination temperature had a
signicant impact on the photoactivity of Pt/TiO2, and the
optimal TiO2 cꢂalcination temperature and Pt photodeposition
time were 500 C and 1 h, respectively. The optimized Pt/TiO2
presented a photoactivity 3.7 times higher than that of Pt/P25
and 18.7 times higher than that of P25. Reaction temperature
had a slight inuence on the photocatalytic CO2 conversion
process. By using 10.0% CO2 diluted in N2, the CH4 yield was
90.5% that of pure CO2. Moreover, The Pt/TiO2 nanocomposites
were more stable than Pt/P25 and could be used repeatedly. In
addition, O2 was detected and recorded for a systematic analysis
of the relationship between photocatalytic CO2 conversion and
water splitting, and the balance between the photogenerated
electrons and holes were calculated. It is proposed that CO and
cH were formed at rst and then the as-formed CO reacted with
cH to form CH4 and C2H6.
22 G. Li, S. Ciston, Z. V. Saponjic, L. Chen, N. M. Dimitrijevic,
T. Rajh and K. A. Gray, J. Catal., 2008, 253, 105.
23 S. Lee, I. S. Cho, J. H. Lee, D. H. Kim, D. W. Kim, J. Y. Kim,
H. Shin, J. K. Lee, H. S. Jung and N. G. Park, Chem. Mater.,
2010, 22, 1958.
24 Y. Jiao, C. Peng, F. Guo, Z. Bao, J. Yang, L. Schmidt-Mende,
R. Dunbar, Y. Qin and Z. Deng, J. Phys. Chem. C, 2011, 115,
6405.
Acknowledgements
The authors greatly appreciate the supports from the Clean Coal 25 J. Yu, L. Qi and M. Jaroniec, J. Phys. Chem. C, 2010, 114,
Program and School of Energy Resources in Wyoming, and the
13118.
44450 | RSC Adv., 2014, 4, 44442–44451
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
26 W. N. Wang, W. J. An, B. Ramalingam, S. Mukherjee, 40 M. Zhou, J. Yu, S. Liu, P. Zhai and L. Jiang, J. Hazard. Mater.,
D. M. Niedzwiedzki, S. Gangopadhyay and P. Biswas, J. Am.
Chem. Soc., 2012, 134, 11276.
27 C. Chen, P. Liu and C. Lu, Chem. Eng. J., 2008, 144, 509.
2008, 154, 1141.
41 Y. Bai, W. Li, C. Liu, Z. Yang, X. Feng, X. Lu and K. Y. Chan, J.
Mater. Chem., 2009, 19, 7055.
28 J. Wang, G. Yin, Y. Shao, S. Zhang, Z. Wang and Y. Gao, J. 42 G. Li, N. M. Dimitrijevic, L. Chen, J. M. Nichols, T. Rajh and
Power Sources, 2007, 171, 331. K. A. Gray, J. Am. Chem. Soc., 2008, 130, 5402.
29 R. E. Rettew, N. K. Allam and F. M. Alamgir, ACS Appl. Mater. 43 T. Chen, Z. Feng, G. Wu, J. Shi, G. Ma, P. Ying and C. Li, J.
Interfaces, 2011, 3, 147.
Phys. Chem. C, 2007, 111, 8005.
30 J. Lee and W. Choi, J. Phys. Chem. B, 2005, 109, 7399.
31 O. S. Alexeev, S. Y. Chin, M. H. Engelhard, L. Ortiz-Soto and
M. D. Amiridis, J. Phys. Chem. B, 2005, 109, 23430.
32 A. Lewera, L. Timperman, A. Roguska and N. Alonso-Vante, J.
Phys. Chem. C, 2011, 115, 20153.
44 M. Huang, C. Xu, Z. Wu, Y. Huang, J. Lin and J. Wu, Dyes
Pigm., 2008, 77, 327.
45 N. M. Dimitrijevic, B. K. Vijayan, O. G. Poluektov, T. Rajh,
K. A. Gray, H. He and P. Zapol, J. Am. Chem. Soc., 2011,
133, 3964.
33 F. Zhao, Y. Ikushima, M. Shirai, T. Ebina and M. Arai, J. Mol. 46 Y. Li, W. N. Wang, Z. Zhan, M. H. Woo, C. Y. Wu and
Catal. A: Chem., 2002, 180, 259. P. Biswas, Appl. Catal., B, 2010, 100, 386.
34 L. Nie, J. Yu, X. Li, B. Cheng, G. Liu and M. Jaroniec, Environ. 47 W. Tu, Y. Zhou, Q. Liu, S. Yan, S. Bao, X. Wang, M. Xiao and
Sci. Technol., 2013, 47, 2777. Z. Zou, Adv. Funct. Mater., 2013, 23, 1743.
35 W. Kong, B. Tian, J. Zhang, D. He and M. Anpo, Res. Chem. 48 G. Ketteler, S. Yamamoto, H. Bluhm, K. Andersson,
Intermed., 2013, 39, 1701.
D. E. Starr, D. F. Ogletree, H. Ogasawara, A. Nilsson and
36 F. Li and X. Li, Chemosphere, 2002, 48, 1103.
M. Salmeron, J. Phys. Chem. C, 2007, 111, 8278.
37 W. Fan, Q. Lai, Q. Zhang and Y. Wang, J. Phys. Chem. C, 2011, 49 M. Tahir and N. S. Amin, Appl. Catal., B, 2013, 142, 512.
115, 10694. 50 G. Dey, J. Nat. Gas Chem., 2007, 16, 217.
38 J. Shi, J. Chen, Z. Feng, T. Chen, Y. Lian, X. Wang and C. Li, J. 51 X. Feng, J. D. Sloppy, T. J. LaTempa, M. Paulose,
Phys. Chem. C, 2007, 111, 693.
S. Komarneni, N. Bao and C. A. Grimes, J. Mater. Chem.,
2011, 21, 13429.
39 N. Wu, J. Wang, D. N. Tafen, H. Wang, J. G. Zheng,
J. P. Lewis, X. Liu, S. S. Leonard and A. Manivannan, J. Am.
Chem. Soc., 2010, 132, 6679.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 44442–44451 | 44451