The effect of calcination atmosphere upon the photocatalytic performance of Au-La2O3/TiO2 for hydrogen production from formic acid

Article abstract of DOI:10.1016/j.apcata.2017.08.027

A simple modified sol-gel method by calcinating titanium dioxide precursor with gold and lanthanum in heat treatment under H₂/CO₂ atmosphere has been proposed to prepare highly photocatalytic Au-La₂O₃/TiO₂ (ALT) hybrid. Characterization results show that gold nanoparticles with small particle size are embedded in a nearly homogeneous component, which may induce a strong interaction of La and Au co-doping to obtain an appropriately modified bandgap structure and extend the absorption edge to visible light region. The formation of La₂O₂CO₃ on the surface could facilitate to the basicity optimization of the photocatalyst and contribute to the C–H bond cleavage in HCOO- to release H₂ and CO₂ from formic acid (FA). The large surface area and small-sized mesopores structure could provide more active sites for the adsorption of reactant molecules and improve the photocatalytic activity. Benefiting from these positive effects, the ALT photocatalyst calcinating in H₂/CO₂ demonstrated a great potential in FA photocatalytic decomposition under simulated sunlight irradiation and produced an optimal H₂ generation rate of 178.4 μmol g_cat^?1 h^?1 (14.9 mmol g_Au^?1 h^?1), which was approximately 1.3, 1.9 and 2.6 times higher than the same catalysts calcinating in H₂, CO₂, and N₂, respectively.

Full text of DOI:10.1016/j.apcata.2017.08.027

AppliedCatalysisA,General547(2017)96–104
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
The effect of calcination atmosphere upon the photocatalytic performance of
Au-La2O3/TiO2 for hydrogen production from formic acid
Minghui Wu, Meng Zhang, Tao Lv, Minxue Guo, Jianjun Li, Chika A. Okonkwo, Qing Liu,
Lishan Jia⁎
Department of Chemical and Biochemical Engineering, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, Fujian, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Photocatalyst
Hydrogen
Au-La2O3/TiO2
H2/CO2 gas treatment
A simple modified sol-gel method by calcinating titanium dioxide precursor with gold and lanthanum in heat
treatment under H2/CO2 atmosphere has been proposed to prepare highly photocatalytic Au-La2O3/TiO2 (ALT)
hybrid. Characterization results show that gold nanoparticles with small particle size are embedded in a nearly
homogeneous component, which may induce a strong interaction of La and Au co-doping to obtain an appro-
priately modified bandgap structure and extend the absorption edge to visible light region. The formation of
La2O2CO3 on the surface could facilitate to the basicity optimization of the photocatalyst and contribute to the
CeH bond cleavage in HCOO- to release H2 and CO2 from formic acid (FA). The large surface area and small-
sized mesopores structure could provide more active sites for the adsorption of reactant molecules and improve
the photocatalytic activity. Benefiting from these positive effects, the ALT photocatalyst calcinating in H2/CO2
demonstrated a great potential in FA photocatalytic de−c1omposition under simulated sunlight irradiation and
−1 (14.9 mmol
g
h
−1), which was approximately
−1
produced an optimal H2 generation rate of 178.4 μmol gcat
h
Au
1.3, 1.9 and 2.6 times higher than the same catalysts calcinating in H2, CO2, and N2, respectively.
1. Introduction
doping rare earth would minimize the recombination of photo-
generated electrons and holes, furthermore facilitate the absorption of
Hydrogen gas has been recognized as the clean and renewable en-
ergy source to remedy the pollution of fossil fuels for the sustainable
future [1–3]. Despite the useful application, hydrogen gas always suf-
fers from extremely low density, inappropriate handle of the trans-
portation and storage [2]. Considerable attentions have been received
on the liquid phase hydrogen storage using small molecules which are
capable of controllable H2 releasing under mild condition and can mi-
tigate the risk of transportation and storage [4–6]. In particular, formic
light in the visible range [17,18]. The rare earth element, lanthanum,
characterized by an electronic configuration [Xe] 5d1 6s2, might be
effective for photocatalytic processes through an increase of the elec-
tron-hole lifetime [19,20]. In recent years, many conclusions have
shown that the surface plasmon resonance effect of Au nanoparticles
and the interaction between Au and titanium dioxide can significantly
enhance the photocatalytic performance of TiO2-based photocatalysts.
Moreover, it is important to promote the decomposition of formic acid
by dehydrogenation pathway, which generates H2 and CO2 and inhibit
the dehydration pathway generating H2O and CO. The study considers
that the light induced hydroxyl radical benefits to the oxidation of
carbon monoxide adsorbed on metal to carbon dioxide [21]. In parti-
cular, the surface valence states and acid-base properties of the pho-
tocatalysts tend to improve the photocatalytic hydrogen production
from formic acid [21,22].
acid as
a non-toxic and non-flammable matter which contains
4.3 ∼ 4.4% (w/w) of hydrogen, has been considered as a promising
hydrogen source [2,7,8]. The TiO2-based photocatalysts to decompose
formic acid for hydrogen production have caused great concern by
scientific researchers [9,10]. However, the lack of absorbance for
visible light and fast recombination of photogenerated electrons and
holes are the fatal drawbacks which greatly limit its practical applica-
tion [11–13].
Numerous works have been performed and the results indicate that
the improved bandgap structure and appropriate surface properties of
TiO2 are obviously the main factors for the enhancement of photo-
catalytic activity [14–16]. Interestingly, the modification of TiO2 by
The focus is to propose an efficient method to prepare catalysts with
excellent structure and properties. Seung Hyo Noh et al. [23] in-
vestigated that using CO2 treatment at high temperature can control the
interactions between the encapsulated Cu and the NeC shell. And in
our previous study, we have combined H2 and CO2 as the atmospheric
⁎
Corresponding author.
E-mail address: jials@xmu.edu.cn (L. Jia).
http://dx.doi.org/10.1016/j.apcata.2017.08.027
Received 8 April 2017; Received in revised form 16 August 2017; Accepted 19 August 2017
Availableonline24August2017
0926-860X/©2017ElsevierB.V.Allrightsreserved.

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
Fig. 3. Nitrogen adsorption-desorption isotherms (inset) and BJH pore-size distribution
curves of (a) ALT-H2/CO2, (b) ALT-H2, (c) ALT-CO2, and (d) ALT-N2.
Fig. 1. XRD patterns for samples of (a) ALT-H2 (b) ALT-H2/CO2, (c) ALT-CO2, and (d)
ALT-N2.
2. Experimental
treatment species to adjust the interaction between nickel and
La2O2CO3 and obtained improved photocatalytic activity [24]. It is
considered that atmospheric heat treatment is an effective process to
improve the performance of photocatalysts by enhancing the interac-
tion between the components. However, to the best of our knowledge,
there is no report about influence of various types of atmospheric
treatment to the photocatalytic activities of the TiO2-based photo-
catalysts.
In this paper, we used different atmospheres in heat treatment
process for titanium dioxide precursor sol-gels doped with gold and
lanthanum. The Au-La2O3/TiO2 photocatalysts calcinating in re-
ductive/acidic gas (H2/CO2) could obtain excellent performance for the
decomposition of formic acid. The effects of different atmospheric (H2/
CO2, H2, CO2, and N2) treatment on the crystal structure and surface
properties of Au-La2O3/TiO2 were systematically investigated.
2.1. Fabrication of Au-La2O3/TiO2 photocatalysts
All reagents were analytical grade purchased from Sinopharm
Chemical Reagent Co. Ltd. All gases were purchased from Linde Gas
(Xiamen) Co., Ltd and the purity was 99.99%. Au/La2O3-TiO2 (ALT)
NPs were synthesized by a modified sol-gel method [25]. 25.0 mL
ethanol was slowly added into 35.0 mL tetrabutyl titanate. And 4.0 mL
acetic acid and 1.0 mL hydrochloric acid were added dropwise. A
homogeneous solution called A was obtained by magnetically stirring
for 30 min and sonicating for 20 min. After that, 15.0 mL ethanol and
5.0 mL distilled water were mingled with a certain amount of lan-
thanum nitrate and gold chloride acid to make solution B. Subse-
quently, solution B was added dropwise into solution A. After magnetic
stirring and ultrasonic treatment, a uniform yellow sol was obtained.
Fig. 2. SEM images of (A) ALT-H2 (B) ALT-
H2/CO2, (C) ALT-CO2, and (D) ALT-N2.
97

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
Fig. 4. TEM images of (a) ALT-H2/CO2, (b)
ALT-CO2, (c) ALT-H2, and (d) ALT-N2.
The sol solution was dried in an oven at 80 °C for 24 h to obtain gel and
transferred into a tube furnace and calcinated 3 h at 500 °C in H2/CO2
(73.6% H2, 18.4% CO2, N2 as balance gas), H2, CO2 and N2, respec-
tively. The rate of gas flow was 100 mL min−1 and the heating rate of
the reactor was 5 °C min−1. Finally, Au-La2O3/TiO2eH2/CO2, Au-
La2O3/TiO2eH2, Au-La2O3/TiO2eCO2, and Au-La2O3/TiO2eN2 (20 wt
% La, 1.2 wt% Au, abbreviated as ALT-H2/CO2, ALT-H2, ALT-CO2, and
ALT-N2) photocatalysts were synthesized. In addition, pure TiO2 (PT),
Au/TiO2 (AT) and La2O3/TiO2 (LT) were prepared with the same pro-
cedure (without lanthanum nitrate or gold chloride acid in solution B)
and calcinated in H2/CO2 mix gas for comparison.
Tecnai 30 electron microscope at an accelerating voltage of 300 kV.
TEM images produced under the assistance of SigmaScan Pro software
(SPSS Inc., version 4.01.003). Ultraviolet-visible (UV-vis) diffused re-
flection spectroscopy of the photocatalyst was recorded using a brand
Varian Carry 5000 UV–vis spectrophotometer. X-ray photoelectron
spectroscopy (XPS) was performed with brand PHI Quantum 2000
Scanning ESCA microprobe consisting of a monochromatized micro-
focused Al X-ray source. Photoluminescence spectrum (PL) was mea-
sured in the wavelength range from 400 to 700 nm at room temperature
with a brand Hitachi F-7000 spectrophotometer using a xenon lamp as
the excitation source. The excitation wavelength was 370 nm. Cyclic
voltammetry (CV) and Electrochemical impedance spectra (EIS) were
recorded on an electrochemical workstation (CHI660D) in a standard
three-electrode system with a working electrode, a Pt wire counter
electrode, and a standard Hg/Hg2Cl2 (saturated KCl) reference elec-
trode. 0.1 M Na2S and 0.1 M Na2SO3 solution was used as electrolyte,
and a xenon lamp (300 W) was used as a light source. EIS was tested
over a frequency range of 0.01–100 KHz with an arc amplitude of 5 mV.
For comparison of CV plots, common scan rate of 100 mV s−1 was
chosen for all the samples.
2.2. Material characterization
X-ray diffraction (XRD) patterns were obtained on a Panalytical pert
spathic powder diffractometer with Cu Kα radiation. The morphology
and distribution of particles were studied by SEM on a ZEISS SIGMA
instrument, using an accelerating voltage of 15 kV. Surface areas of the
samples were measured by using Brunauer-Emmett-Teller (BET)
method and the pore-size distributions were derived from the deso-
rption branches of isotherms based on Barrett–Joyner–Halenda (BJH)
model based on a Micromeritics TriStar 3000 Automatic physical ad-
sorption instrument with N2 as absorbed gas. Energy dispersive spec-
trometer (EDS) analysis was observed with a LEO-1530 incorporated
electron microscope (Germany). The morphology characterization and
element mapping of the particles were carried out in a high resolution
transmission electron microscopy (HRTEM) of a Philip Analytical FEI
2.3. Photocatalytic decomposition of formic acid
The catalytic reaction of FA solution for H2 production was carried
out in a self-made annular quartz reactor. The photocatalyst (0.1 g) was
constantly suspended in aqueous formic acid solution using a magnetic
stirrer. Prior to irradiation, the solution was continuously bubbled with
98

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
Fig. 5. HRTEM images of (a) ALT-H2/CO2,
(b) ALT-CO2, (c) ALT-H2, and (d) ALT-N2.
blank samples without lanthanum doping (in Fig. S1). It could be at-
tributed that lanthanum doping can effectively inhibit the transforma-
tion of anatase to rutile phase [19]. What’s more, although there is a
relatively high doping amount (La: 20 wt%) in the samples, no addi-
tional diffraction peaks corresponding to La and/or its oxides can be
observed. However, La element was identified in EDS (Fig. S2). The fact
indicates that the lanthanum is highly dispersed in the sample due to
heat treatment for homogeneous precursor sol-gels at high temperature
[19,26,27].
The diffraction peaks of ALT-H2 (Fig. 1a) catalyst are narrower with
high intensity than those of ALT-H2/CO2, ALT-CO2, and ALT-N2, sug-
gesting that the crystallinity and the average crystal size increased in
accordance with the particle size statistical distribution measurements
shown in Fig. S3. The peaks at 2θ = 38.2° and 44.4° in ALT-H2 corre-
spond to (111) and (200) planes of Au crystal. In ALT-H2/CO2 (Fig. 1b),
there is only a very weak gold peak, which may be attributed to the heat
treatment process. The carbon dioxide in the H2/CO2 gas may react
with La2O3 on the surface of the sample to form La2O2CO3 and prevent
the reduction of gold ions and the growing up of gold nanoparticles
[28]. The absence of the characteristic gold peaks in ALT-CO2 and ALT-
N2 samples can be attributed to lacking of reducing gas in calcination
(Fig. 1c,d).
Table 1
The band-gap values of the prepared samples.
Catalysts
Band-gap/eV
PT
LT
AT
3.16
3.02
2.97
2.84
2.77
2.66
2.56
ALT-N2
ALT-CO2
ALT-H2/CO2
ALT-H2
N2 at a rate of 100 mL min−1 for 30 min. Then the gas content was
measured to confirm that there is no oxygen present. Irradiation was
conducted by a 300 W xenon lamp. The light intensity at the solution
interface was 99.5 mW cm−2 measured by photoradiometer (PL-
MW2000). The reaction temperature was kept at 50 °C. And the evolved
gases were gathered and analyzed by GC equipped with TCD detectors.
3. Results and discussion
3.1. XRD characterization
Fig. 1 shows the XRD patterns of ALT samples calcinating in dif-
ferent atmospheres at 500 °C for 3 h. Clearly, a series of samples contain
the anatase phase TiO2 (JCPDS File no. 01-0562) without rutile (JCPDS
File no. 21–1276). A small amount of rutile phase was found in the
3.2. SEM and BET characterization
Morphology of the samples was examined by SEM images. A com-
parison of Fig. 2 reveals that ALT-H2/CO2 are well dispersed, small-
99

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
Fig. 6. XPS spectra of O, La, C and Au elements for (A) ALT-CO2, (B) ALT-H2/CO2, (C) ALT-H2, and (D) ALT-N2.
sized and less-densified particles with porous channels, while ALT-H2
are irregularly agglomerations; ALT-CO2 and ALT-N2 particles were
slightly agglomerated resulting in rough surface. Fig. 3 displays that the
samples exhibit the typical type IV N2 adsorption-desorption isotherms
according to the IUPAC classification with H2 hysteresis loop, which
can be observed in the pores with narrow necks and wider bodies (ink-
bottle pores) [29–31]. All samples present the obvious hysteresis loops
at a relative pressure (P/P0) range of 0.4–0.9, suggesting the presence of
mesopores. The hysteresis loop of ALT-H2/CO2 (Fig. 3a, inset) at the
relative pressure of 0.4–0.6 indicated the presence of numberous small-
sized mesopores. The uptakes of N2 on ALT-N2 (Fig. 3d, inset) increase
sharply when pressure > 0.65, implying large-sized mesopores [30].
The corresponding BJH pore-size distribution curves demonstrated that
the mesopore sizes of ALT-H2/CO2, ALT-H2, ALT-CO2 and ALT-N2 are
2–5 nm, 2–7 nm, 2–12 nm and 2–15 nm, respectively. The corre-
of Au NPs. The (101) crystal face of TiO2 support corresponds to the d-
value of 0.35 nm. Meanwhile, d = 0.30 nm and 0.29 nm are assigned to
the (002) plane of La2O3 and (103) plane of .La2O2CO3, respectively. As
presented in Fig. 5a and b, La element is in the form of La2O3 and
La2O2CO3 in the ALT-H2/CO2 and ALT-CO2 composites. It is the pri-
marily evidence of the formation of La2O2CO3, which will be further
explained in XPS as follows. In Fig. 5a, Au NPs (5 ∼ 6 nm) are em-
bedded in both closely adjacent TiO2 and La2O3 which implies that Au
NPs have strong interaction with TiO2 and La2O3. The gold nano-
particles on the surface of ALT-H2 (Fig. 5c) are mainly hemispherical
with bigger particle size of 7 ∼ 13 nm. However, no lattice fringes
corresponding to Au NPs were found in ALT-CO2 and ALT-N2 (Fig. 5b,
d). Furthermore, the element distributions in ALT-H2/CO2 were carried
out by means of elemental mapping (Fig. S4). The elements of O, La and
Ti in the images are in good agreement with the selected area, sub-
stantiating that they are well distributed and of high homogeneity in
the chemical composition of the nanoframes. Combining with the cor-
responding TEM image (Fig. 4a), it is speculated that the specific fried-
egg-like ALT-H2/CO2 sample is a homogeneous multilayer entirety
constituting by Au NPs, La2O3 and TiO2. And this particular mor-
phology and structure may contribute to improve light absorption
ability and photocatalytic performance of the material.
sponding specific surface areas are 186 m2 g−1
,
159 m2 g−1
,
96 m2 g−1, and 84 m2 g−1 (Table S1). ALT-H2/CO2 has the much larger
surface area compared with pure TiO2 (84.5 m2 g−1). The large surface
area possibly relates to the photocatalytic activity due to more surface
active sites on the surface of the sample [32,33]. These results indicate
that the H2/CO2 atmosphere can effectively regulate the morphology
and structure of the sample.
TEM images are shown in Fig. 4. From Fig. 4a, it can be observed
that ALT-H2/CO2 sample is in a peculiar fried-egg-like shape with ob-
vious regular boundary. This phenomenon hasn’t been discovered in
ALT-CO2, ALT-H2 and ALT-N2 which are consisted of tanglesome na-
noparticles (Fig. 4b,c,d). HRTEM images are disclosed in Fig. 5. The
lattice fringe with d-value of 0.22 nm is ascribed to (111) crystal plane
3.3. UV–vis characterization
The UV–vis diffuse reflectance spectra (DRS) were recorded for ALT
samples calcinating in different atmospheres and blank samples of AT,
LT and PT calcinating in H2/CO2 for comparison shown in Fig. S5.
100

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
Fig. 7. (A) PL spectra of (a) ALT-N2, (b) ALT-CO2, (c) ALT-H2, and (d) ALT-H2/CO2. (B) CV tests on (a) ALT-H2/CO2, (b) ALT-H2, (c) ALT-CO2, and (d) ALT-N2. (C), (D) EIS curves of ALT
samples in dark and under the xenon lamp irradiation, respectively.
Strong absorption peaks in the range of 500 nm ∼ 650 nm caused by
the excellent surface plasmon resonance effect on Au NPs can be ob-
served from the DRS curves of ALT samples calcinating in different
atmospheres. The intensity of the absorption peak is significantly dif-
ferent, which is closely related to the process of atmospheric treatment.
It can be observed that the LT sample harvests markedly improved
absorbance than PT, which can be elucidated that La 5d orbit and Ti 3d
orbit together constitute a conduction band making the forbidden band
width narrowed and broadening the photocatalytic light response range
[34,35]. The bandgap values of samples were calculated by the formula
Eg = 1240/λ [36], where λ is the absorption edge; the results are
shown in Table 1. It can be clearly seen that the bandgaps of LT
(3.02 eV) and AT (2.97 eV) are lower than PT (3.16 eV). After co-
doping of Au and lanthanum, the band gaps of ALT composites calci-
nating in different atmospheres are remarkably reduced with the se-
quence of ALT-H2 (2.56 eV) < ALT-H2/CO2 (2.66 eV) < ALT-CO2
(2.77 eV) < ALT-N2 (2.84 eV). The fact that the introduction of both
Au and La makes the bandgaps of ALT samples become narrower may
attribute to the strong synergistic effect between gold and lanthanum in
atmospheric heat treatment process.
XPS spectra of La element. For La 3d spectra in Fig. 6, two groups of
doublet peaks are indicative of two bonding states. The spin-orbit
splitting between 3d3/2 and 3d5/2 levels is 16.9 eV while the separation
between the satellite and main peak is approximately 3.5 eV, which
agree with reported values for La3+ compounds [40]. The 3d5/2 peaks
of ALT-CO2, ALT-H2/CO2 and ALT-H2, and are positively shifted by 0.9,
0.7, 0.5 eV compared with ALT-N2. This may be caused by the forma-
tion of new groups or the changes of chemical states on the surface of
the samples in the atmosphere treatment process. Specifically, the ob-
tained La 3d5/2 spectrum for ALT-N2 is composed by a doublet with
834.3/837.9 eV, which is consistent with that of the recorded for pure
La2O3. The positive shift of the doublet to 835.2/838.7 eV for ALT-CO2
attributes to the formation of La2O2CO3 in the hybrid [39], which is in
agreement with O spectrum analysis. It is also consistent with the
HRTEM results shown in Fig. 5. Moreover, C 1 s spectra were performed
for the samples. As observed in Fig. 6, the recorded C 1 s peaks of ALT-
CO2 and ALT-H2/CO2 were fitted into three peaks, corresponding to
2−
CeC, CeO, and CO3
groups. Whereas, only two peaks ascribed to
CeC and CeO are found in the sample of ATL-H2 and ATL-N2 with no
2−
2−
CO3
group observed. It reveals that the CO3
groups should origi-
nate from the formed La2O2CO3 phase exactly and the acidic CO2 in the
atmosphere contributes to the formation of La2O2CO3.
3.4. XPS measurements
The formation of La2O2CO3 has positive influence for the samples
calcinating in H2/CO2 gas. La2O3 is able to react with CO2 and convert
to La2O2CO3. And La2O2CO3 can easily decompose to La2O3 revertively.
This dynamic formation and decomposition of the complex during the
reaction may release highly active Au NPs, preventing the sintering of
the photocatalyst, and to a certain extent, promoting the spontaneous
dispersion of Au NPs [19,41], which can be seen in Fig. S6. The particle
size of Au NPs in ALT-H2/CO2 is much smaller (3.2 nm) than ALT-H2
(13.8 nm). Compared with the Au 4f spectra, ALT-H2 (Fig. 6C) has the
strongest intensity of peaks while ALT-N2 (Fig. 6D) has the lowest. The
result can be explained that reducing gas H2 is lacking in the heat
As shown in Fig. 6, the XPS fitting patterns belong to the composites
calcining in CO2, H2/CO2, H2 and N2. The spectra of O 1 s are wide and
asymmetric, indicating that at least two kinds of oxygen species are
present on the surface. The peaks located at about 529 and 530 eV are
assigned to crystal lattice oxygen in the metal oxides and adsorbed
oxygen (oxygen vacancies), respectively [37,38]. A new peak at higher
binding energy approximately 532 eV can be identified in ALT-CO2 and
2−
ALT-H2/CO2 which is attributed to CO3
[39]. It can be speculated
that the formation of La2O2CO3 took place when ALT hybrids were
calcinated in CO2 and H2/CO2 atmospheres. We further analyzed the
101

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
Additionally, the results reveal that different atmospheric heat treat-
ments have different influences on the amount and ratio of Au0 and
positive Au ions. ALT-H2/CO2 has the appropriate intensity and ratio of
the Au 4f peaks. And the proper state on the surface of the samples is
closely related to the photocatalytic ability according to the literature
[39,42].
3.5. PL, CV and EIS
Fig. 7A shows the photoluminescence (PL) spectra of the ALT
samples calcinating in different atmospheres in the wavelength range
from 400 to 650 nm. The emission at 416 nm is generally exhibited due
to the transition of free electrons from low donor level of the oxygen
vacancies in the valance band. Moreover, the PL bands of anatase TiO2
nanoparticles located at around 471, 485 and 588 nm are attributed to
the oxygen vacancies and defects [42–44]. It is consistent with the XPS
result. PL is to analyze the charge separation efficiency and the fate of
electron–hole pairs in semiconductor particles [42]. All the catalysts
have semblable curves and the spectra intensity follows the order of
ATL-N2 > ATL-CO2 > ATL-H2 > ATL-H2/CO2. The weakest ex-
citonic PL spectrum of ALT-H2/CO2 implies the highest separation rate
of photo-induced charge carriers and the prolonged lifetime of photo-
generated electron-hole pairs [42,45–47]. For the purpose of making
further study on the charge transfer process and the separation effi-
ciency of the photogenerated charge carriers, cyclic voltammetry (CV)
and electrochemical impedance spectroscopy (EIS) were performed.
From Fig. 7B, it can be observed that CVs don’t have any obvious peaks,
suggesting that the behavior is free from redox reactions or is purely
based on the electrostatic mechanism [48–50]. The ALT-H2/CO2 con-
tains broader plot implying more capacitance than the same samples
calcinating in H2, CO2 or N2 atmosphere which endow less narrow plots
indicating that the capacitances are less. The CV results agree with the
EIS in Fig. 7C and D. Generally, the smaller radius of the arc sub-
stantiates the higher efficiency of charge transfer [50,51]. On the basis
of this charge transfer resistance value, the separation efficiency of
electrons-holes can be illuminated clearly [48]. EIS of ALT samples
were performed under both dark and visible light conditions. The dia-
meters under the xenon lamp illumination were smaller than that
without light irradiation. This phenomenon confirms that visible light
irradiation can efficiently accelerate the charges mobility [52]. Ad-
ditionally, ALT-H2/CO2 has the smallest impedance value compared
with the other samples, which conveys that photocatalyst calcinating in
H2/CO2 mix gas has the highest charge transfer rate.
Fig. 8. (A) H2 evolution rate over ALT samples calcinated in N2, CO2, H2, and H2/CO2. (B)
Recycling performance of ALT-H2/CO2 towards photocatalytic H2 generation.
3.6. Photocatalytic activity
The photocatalytic process was optimized and determined by
comparing the calcinating time and temperature of samples and the
concentration of FA solution (Fig. S7). The results showed that the
sample obtained the maximum output of H2 production when the cal-
cination temperature was 500 °C; the H2 production gradually in-
creased when the calcination time was 1, 2, and 3 h, then remained
almost unchanged beyond 3 h; the amount of H2 was increased initially
then decreased with the concentration of formic acid. In brief, the op-
timum calcinating condition for the samples was at 500 °C for 3 h, and
the concentration of FA solution was 6 mol L−1 for suitable hydrogen
production. As shown in Fig. 8A, ALT-H2/CO2 exhibits a remarkable
Fig. 9. Schematic illustration for the reaction mechanism of photocatalytic decomposi-
tion over ALT-H2/CO2 photocatalyst.
photocatalyti−c1 activity to obtain
a hydrogen production rate of
178.4 μmol gcat h−1 (14.9 mmol g−A1u h−1), which is 1.3, 1.9 and 2.6
times higher than the same catalysts calcinating in H2, CO2, and N2. The
hydrogen production rate by unit surface area is shown in Table S2, and
ALT-H2/CO2 produced the highest H2 evolution rate (0.96 μmol
treatment atmosphere source for ALT-CO2 and ALT-N2. And it is cal-
culated that there are 62.6% Au0 (86.8/83.1 eV) and 37.4% positive Au
ions (88.2/84.5 eV) in ALT-CO2 while 83.3% Au0 and 16.7% positive
Au ions in ALT-H2/CO2; 89.2% Au0 and 10.8% positive Au ions in ALT-
H2; 54.4% Au0 and 45.6% positive Au ions in ALT-N2. The Au0 on the
surface of ALT-CO2 and ALT-N2 may be due to the fact that a small
amount of gold hydroxide is self-decomposed at high temperature.
h
−1 m−2). It indicates that the stronger interaction between the com-
ponents may improve the photocatalytic performance. The stability of
the photocatalyst was investigated by repetitive time-circle photo-
catalytic hydrogen generation for consecutive five cycles (10 h) in
102

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
h−1
,
−1
Fig. 8B. There was no distinguishable decline observed in catalytic ef-
fect after five periods under visible light illumination, suggesting the
splendid stability and reusability of as-prepared ALT-H2/CO2 sample
with the prolonged reaction duration. The photocatalytic stability of
ALT-H2/CO2 was also measured shown in Fig. S8A. Within five hours,
the photocatalyst could stably promote the decomposition of formic
acid and produce hydrogen continuously, which implied the good po-
tential for application though slight amount of H2 was consumed in the
pretreatment process for prepared catalysts. When the xenon lamp is
off, there is approximately no hydrogen production generates (Fig.
S8B). And there was no CO detected in the decomposition of FA (Fig.
S9). Although the series of ALT photocatalysts have a certain activity
for H2 liberation, the photocatalyst treated by H2/CO2 atmosphere ex-
hibits an obviously preferable performance, due to H2/CO2 gas treat-
ment plays a crucial role to form special morphologies and enhance the
interactions between the components. Additionally, to compare this
result with H2 production by photocatalysis, the representative results
using doped TiO2 are presented in Table S3, which implies H2/CO2 gas
treatment catalyst indeed has good catalytic performance for photo-
catalytic hydrogen production.
catalyst achieved a sufficient H2 generation rate of 178.4 μmol
g
cat
which was much higher than the same photocatalysts calcinating in H2,
CO2 or N2. The mixed atmosphere (H2/CO2) treatment can effectively
control the morphology and surface states of the photocatalyst, and
enhance the interaction between the components, which are beneficial
to the photocatalytic decomposition of FA for hydrogen production.
Acknowledgments
This work was supported by the NSFC project (No.21536010) and
the general program of the National Natural Science Foundation of
China (Grant No.21176203). The authors thank Analysis and Testing
Center of Xiamen University for the analysis and observation work in
this study.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.apcata.2017.08.027.
References
3.7. Reaction mechanism
[1] C. Chang, M. Chen, W. Du, J. Gliniak, J. Lin, H. Wu, H. Chan, J. Yu, T. Wu, Chem 21
(2015) 6617–6622.
Combining with the analyses of the characterizations and photo-
catalytic performance above, a possible mechanism for the formic acid
decomposition process over ALT-H2/CO2 photocatalyst is illustrated in
Fig. 9. The Au, La doped titanium dioxide precursor was synthesized
with sol-gel method and effectively controlled by the process of atmo-
sphere heat treatment to make the catalysts absorb light in the visible
region. ALT composites calcinating in H2/CO2 forms a smaller meso-
porous structure and larger surface area. And the nano-scale Au NPs
embedded into the support had strong interation with TiO2 and La2O3,
which is profitable for the separation of photo-generated electrons and
holes. The formed La2O2CO3 may play the role in regulating the basicity
of the surface, which contributes to the cleavage of CeH band in HCOO-
and the desorption of CO2 on the surface of the catalyst [19,53–55],
resulting in effectively improved photocatalytic performance. In the
schematic diagram, La-doped TiO2 NPs absorb photons under visible
light irradiation, which excite photogenerated electrons to the con-
duction band (CB) and simultaneously produce photogenerated holes in
the valence band (VB). In the first step, the electrons are readily
transferred to Au NPs on the surface of La2O3-TiO2 support to promote
the decomposition of HCOOH intoHCOO- and H+ [56]. The proton can
be reduced by an electron to release H2. The formate then reacts with a
photogenerated hole in TiO2 to form a free radical of HCOO·. In the
second step, the formate radical is oxidized by holes and releases a
proton to form CO2. In this procedure, the La2O2CO3 on the surface of
the La2O3-TiO2 support can promote the decomposition of FA and the
favorable desorption of surface-adsorbed CO2, resulting in speeding up
the photocatalytic reaction rate [21,53,54]. Meanwhile, the protons
released from HCOO· are reduced by the electrons and liberate hy-
drogen efficiently from the formic acid solution [56].
[2] M.F. Kuehnel, D.W. Wakerley, K.L. Orchard, E. Reisner, Angew. Chem. 54 (2015)
9627–9631.
[3] Y. Zhu, L. Li, L. Fu, Y. Wu, Int. J. Hydrogen Eerg. 40 (2015) 14290–14296.
[4] D. Jantke, L. Pardatscher, M. Drees, M. Cokoja, W.A. Herrmann, F.E. Kuhn,
Chemsuschem 9 (2016) 2849–2854.
[5] K. Koh, M. Jeon, D.M. Chevrier, P. Zhang, C.W. Yoon, T. Asefa, Appl. Catal. B:
Environ. 203 (2017) 820–828.
[6] L. Yang, X. Hua, J. Su, W. Luo, S. Chen, G. Cheng, Appl. Catal. B: Environ. 168–169
(2015) 423–428.
[7] Y. Karatas, A. Bulut, M. Yurderi, I.E. Ertas, O. Alal, M. Gulcan, M. Celebi, H. Kivrak,
M. Kaya, M. Zahmakiran, Appl. Catal. B: Environ. 180 (2016) 586–595.
[8] S.M. Barrett, S.A. Slattery, A.J.M. Miller, ACS Catal. 5 (2015) 6320–6327.
[9] L. Clarizia, I. Di Somma, R. Marotta, P. Minutolo, R. Villamaina, R. Andreozzi, Appl.
Catal. A 518 (2016) 181–188.
[10] K. Tedsree, T. Li, S. Jones, C.W. Chan, K.M. Yu, P.A. Bagot, E.A. Marquis,
G.D. Smith, S.C. Tsang, Nature Nanotech. 6 (2011) 302–307.
[11] Z. Zhao, X. Zhu, M. Zuo, J. Xu, Y. Wang, CrystEngComm 18 (2016) 1636–1644.
[12] D. Wang, H. Shen, L. Guo, C. Wang, F. Fu, Y. Liang, RSC Adv. 6 (2016)
71052–71060.
[13] H. Liu, T. Lv, Z. Zhu, J. Mol. Catal. A: Chem. 404–405 (2015) 178–185.
[14] H. Lee, M. Shin, M. Lee, Y.J. Hwang, Appl. Catal. B: Environ. 165 (2015) 20–26.
[15] F. Liu, X. Yan, X. Chen, L. Tian, Q. Xia, X. Chen, Catal. Today 264 (2016) 243–249.
[16] Q. Liu, X. Yang, Y. Huang, S. Xu, X. Su, X. Pan, J. Xu, A. Wang, C. Liang, X. Wang,
T. Zhang, Energ. Environ. Sci. 8 (2015) 3204–3207.
[17] V. Gomez, A.M. Balu, J.C. Serrano Ruiz, S. Irusta, D.D. Dionysiou, R. Luque,
J. Santamaría, Appl. Catal. A:Gen. 441–442 (2012) 47–53.
[18] T. Luitel, K. Fernando, B.S. Tatum, B.W. Alphenaar, F.P. Zamborini, Electrochim.
Acta 211 (2016) 918–925.
[19] M. Meksi, G. Berhault, C. Guillard, H. Kochkar, Catal. Commun. 61 (2015) 107–111.
[20] Z. Xu, C. Li, W. Lu, J. Synthetic Crystals 44 (2015) 542–545.
[21] Z. Zhang, S. Cao, Y. Liao, C. Xue, Appl.Catal.B: Environ. 162 (2015) 204–209.
[22] Q. Bi, J. Lin, Y. Liu, H. He, F. Huang, Y. Cao, J. Power Sources 328 (2016) 463–471.
[23] S.H. Noh, M.H. Seo, X. Ye, Y. Makinose, T. Okajima, N. Matsushita, B. Han,
T. Ohsaka, J. Mater. Chem. A 3 (2015) 22031–22034.
[24] L. Jia, J. Li, W. Fang, J. Alloy. Compd. 489 (2010) L13–L16.
[25] Y.Q. Zeng, T.X. Wang, S.L. Zhang, Y.N. Wang, Q. Zhong, Appl. Surf. Sci. 411 (2017)
227–234.
[26] X. Zhou, X. Zhang, X. Feng, J. Zhou, S. Zhou, Dyes Pigm. 125 (2016) 375–383.
[27] Y. Liu, S. Zhou, J. Li, Y. Wang, G. Jiang, Z. Zhao, B. Liu, X. Gong, A. Duan, J. Liu,
Y. Wei, L. Zhang, Appl. Catal. B: Environ. 168-169 (2015) 125–131.
[28] Z. Zhang, X.E. Verykios, Appl. Catal. A:Gen. 138 (1996) 109–133.
[29] G. Wang, J. Mol. Catal. A: Chem. 274 (2007) 185–191.
[30] Y. Zhang, M. Jia, J. Yu, J. Fan, L. Wang, Y. Zou, Y. Zhao, J. Solid State Electr. 20
(2016) 733–741.
4. Conclusions
ALT-H2/CO2 nanophotocatalyst was fabricated via a simple mod-
ified sol-gel method by calcinating titanium dioxide precursor with gold
and lanthanum in H2/CO2 atmosphere. In this photocatalyst, H2/CO2
treatment provided a preferential surficial basicity and morphology of
the TiO2-based photocatalysts. The co-doping of Au and La obtained a
suitable bandgap structure and enhanced the visible light absorption
capacity. The Au NPs and La2O2CO3 on the surface of the photocatalyst
play important roles to facilitate the decomposition of FA. Meanwhile,
the large surface area and small-sized mesopores could provide more
active sites for the adsorption of reactant molecules and improve the
photocatalytic activity. Benefiting from these positive effects, the
[31] B. Li, J. Zhao, J. Liu, X. Shen, S. Mo, H. Tong, RSC Adv. 5 (2015) 15572–15578.
[32] J. Yu, J. Ran, Energ. Environ. Sci. 4 (2011) 1364.
[33] A. Hui, J. Ma, J. Liu, Y. Bao, J. Zhang, J. Alloy. Compd. 696 (2017) 639–647.
[34] S. Na Phattalung, S. Limpijumnong, J. Yu, Appl. Catal. B: Environ. 200 (2017) 1–9.
[35] T. Ao, Q. Hou, C. Ying, Titanium Indus. Prog. 29 (2012) 13–18.
[36] Z. Yang, J. Lu, W. Ye, C. Yu, Y. Chang, Appl`. Surf. Sci. 392 (2017) 472–480.
[37] J.Q. Jiao, Y.C. Wei, Z. Zhao, W.J. Zhong, J. Liu, J.M. Li, A.J. Duan, G.Y. Jiang, Catal.
Today 258 (2015) 319–326.
[38] L.Q. Jing, X.J. Sun, B.F. Xin, B.Q. Wang, W.M. Cai, H.G. Fu, J. Solid State Chem. 177
(2004) 3375–3382.
[39] W. Gu, J. Liu, M. Hu, F. Wang, Y. Song, A.C.S. Appl. Mater. Inter. 7 (2015)
103

M. Wu et al.
AppliedCatalysisA,General547(2017)96–104
26914–26922.
[49] R. Farma, M. Deraman, Awitdrus, I.A. Talib, R. Omar, J.G. Manjunatha, M.M. Ishak,
N.H. Basri, B.N.M. Dolah, Int. J. Electrochem. Sc. 8 (2013) 257–273.
[50] A. Massa, S. Hernández, A. Lamberti, C. Galletti, N. Russo, D. Fino, Appl. Catal. B:
Environ. 203 (2017) 270–281.
[40] P. Zhang, J. Guo, P. Zhao, B. Zhu, W. Huang, S. Zhang, RSC Adv. 5 (2015)
11989–11995.
[41] A.N. Fatsikostas, D.I. Kondarides, X.E. Verykios, Catal. Today 75 (2002) 145–155.
[42] E. Grabowska, M. Marchelek, T. Klimczuk, G. Trykowski, A. Zaleska-Medynska, J.
Mol. Catal. A: Chem. 423 (2016) 191–206.
[51] L. Jing, Y. Xu, S. Huang, M. Xie, M. He, H. Xu, H. Li, Q. Zhang, Appl. Catal. B:
Environ. 199 (2016) 11–22.
[43] K. Gopinath, S. Kumaraguru, K. Bhakyaraj, S. Thirumal, A. Arumugam, Superlattice
[52] C. Zhai, M. Zhu, F. Pang, D. Bin, C. Lu, P. Yang, Y. Du, A.C.S. Appl. Mater. Inter. 8
Microst. 92 (2016) 100–110.
(2016) 5972–5980.
[44] D.D. Liu, Z.S. Wu, F. Tian, B.C. Ye, Y.B. Tong, J. Alloy Compd. 676 (2016) 489–498.
[45] G. He, Y. Zhong, M. Chen, X. Li, Y. Fang, Y. Xu, J. Mol. Catal. A: Chem. 423 (2016)
70–76.
[53] M. Sridhar, D. Ferri, M. Elsener, J.A. van Bokhoven, O. Kröcher, ACS Catal. 5 (2015)
4772–4782.
[54] H. Li, D. Gao, P. Gao, F. Wang, N. Zhao, F. Xiao, W. Wei, Y. Sun, Catal. Sci. Technol.
[46] Y. Yin, E. Liu, H. Li, J. Wan, J. Fan, X. Hu, J. Li, C. Tang, C. Pu, Cer. Int. 42 (2016)
3 (2013) 2801–2809.
9387–9395.
[55] W. Gu, J. Liu, M. Hu, F. Wang, Y. Song, A.C.S. Appl. Mat. Inter. 7 (2015)
26914–26922.
[56] T. Chen, G. Wu, Z. Feng, G. Hu, W. Su, P. Ying, Chinese J. Catal. 29 (29) (2008)
105–107.
[47] Q. Chen, S. Wu, Y. Xin, Chem. Eng. J. 302 (2016) 377–387.
[48] K. Gurushantha, K.S. Anantharaju, S.C. Sharma, H.P. Nagaswarupa,
S.C. Prashantha, K.R.V. Mahesh, L. Renuka, Y.S. Vidya, H. Nagabhushana, J. Alloy.
Compd. 685 (2016) 761–773.
104