Modification of SrTiO3 as a photocatalyst for hydrogen evolution from aqueous methanol solution

Article abstract of DOI:10.1016/j.jphotochem.2018.07.010

The photocatalyst SrTiO₃ was modified by three strategies, i.e., the addition of CuO to obtain CuO/SrTiO₃ composites, the introduction of an excess amount of Sr into the SrTiO₃ framework and the introduction of meso- and macropores from template-assisted synthesis. Physico-chemical characterization by elemental analysis (ICP-OES), XRD, N₂-sorption, Hg-intrusion, scanning electron microscopy and UV/Vis-spectroscopy was applied to determine the chemical composition, crystal phase, morphology, textural properties and optical absorption of the modified SrTiO₃ materials. The prepared materials were studied as photocatalysts after loading with 0.5 wt.-% Pt in a slurry reactor for the hydrogen evolution from aqueous methanol solution at ambient temperature under UV light irradiation. In comparison to unmodified SrTiO₃, the highest increase in photocatalytic activity of more than threefold, i.e., from 39 to 130 μmol h^?1, was achieved by combining the addition of 5 mol-% of Sr excess (over Ti) and the introduction of meso-/macropores by using both carbon spheres and polyvinylpyrrolidone as templates.

Full text of DOI:10.1016/j.jphotochem.2018.07.010

JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Journal of Photochemistry & Photobiology A: Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Modification of SrTiO3 as a photocatalyst for hydrogen evolution from
aqueous methanol solution
Sepideh Banakhojasteh, Steffen Beckert, Roger Gläser⁎
Institute of Chemical Technology, Universität Leipzig, Linnéstr. 3, 04103 Leipzig, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords:
The photocatalyst SrTiO3 was modified by three strategies, i.e., the addition of CuO to obtain CuO/SrTiO3
composites, the introduction of an excess amount of Sr into the SrTiO3 framework and the introduction of meso-
and macropores from template-assisted synthesis. Physico-chemical characterization by elemental analysis (ICP-
OES), XRD, N2-sorption, Hg-intrusion, scanning electron microscopy and UV/Vis-spectroscopy was applied to
determine the chemical composition, crystal phase, morphology, textural properties and optical absorption of
the modified SrTiO3 materials. The prepared materials were studied as photocatalysts after loading with 0.5 wt.-
% Pt in a slurry reactor for the hydrogen evolution from aqueous methanol solution at ambient temperature
under UV light irradiation. In comparison to unmodified SrTiO3, the highest increase in photocatalytic activity of
more than threefold, i.e., from 39 to 130 μmol h−1, was achieved by combining the addition of 5 mol-% of Sr
excess (over Ti) and the introduction of meso-/macropores by using both carbon spheres and poly-
vinylpyrrolidone as templates.
Photocatalytic activity
Strontium titanate
Excess amount of Sr
Composite materials
Templating
Hydrothermal synthesis
1. Introduction
templates has received attention because of its low cost [14,15]. Also
other related materials such as TiO2 [16], TiO2/SiO2 [17], or SnO2 [18]
In the past decades, solar water splitting using semiconductor
photocatalysts was intensively studied as a green process for producing
hydrogen [1,2]. Towards this end, many research efforts were devoted
to utilize photocatalytically active materials [3,4]. Among the widely
investigated photocatalysts, SrTiO3 is a promising material owing to its
high chemical stability, suitable band gap and band location, abun-
dance of the constituent elements and environmental friendliness.
SrTiO3 is able to split water and to degrade organic pollutants under
UV-light irradiation [5,6]. Despite these attractive characteristics,
SrTiO3 still suffers from low electron and hole mobility as well as a high
recombination rate of the photogenerated charge carriers [7]. More-
over, its practical application is limited due to its wide band gap
(3.2 eV) [8], resulting in a low absorbance in the UV region of the solar
spectrum. To overcome these drawbacks and to improve their photo-
catalytic activity, SrTiO3 photocatalysts can be modified, e.g., by
doping with cations [9,10], by combination with narrow band gap
semiconductors [11,12] or by generation of oxygen vacancies and/or
defects through impurities or non-stoichiometric synthesis [10,13].
Several studies have focused on the preparation of porous SrTiO3
materials by a variety of synthesis routes. In particular, the combination
of carbon spheres and polyvinylpyrrolidone (PVP) as sacrificial
were prepared with meso- and macropores. The advantage of SrTiO3
photocatalysts with ordered and uniform macropores lies in their large
specific surface area, improved mass transfer and high light absorption
efficiency caused by an enhanced photon harvesting ability [19,20].
Another approach for increasing the photocatalytic activity of
SrTiO3 is the addition of an excess amount of Sr [13]. This Sr excess not
only effectively generates SrO defects in the solid framework of SrTiO3,
but also lowers the band gap and reduces the electron-hole re-
combination rate resulting in a significantly improved photocatalytic
activity under UV/Vis light irradiation. So far, however, the simulta-
neous addition of a Sr excess and the introduction of meso- and mac-
roporosity for increasing the activity of SrTiO3-based photocatalysts has
yet to be investigated.
In the present study, we report on the improvement of the photo-
catalytic activity of SrTiO3 for hydrogen evolution from aqueous me-
thanol solution. Towards that goal, SrTiO3 is modified by three strate-
gies, i.e., the combination with CuO as
a narrow band gap
semiconductor to obtain CuO/SrTiO3 composites, the addition of an
excess amount of Sr, and the introduction of meso- and macroporosity
via template-assisted synthesis. For the first time, the simultaneous
introduction meso- and macropores and the addition of an excess
⁎
Corresponding author.
E-mail address: roger.glaeser@uni-leipzig.de (R. Gläser).
https://doi.org/10.1016/j.jphotochem.2018.07.010
Received 5 December 2017; Received in revised form 23 May 2018; Accepted 5 July 2018
1010-6030/©2018ElsevierB.V.Allrightsreserved.
Pleasecitethisarticleas:Banakhojasteh,S.,JournalofPhotochemistry&PhotobiologyA:Chemistry(2018),
https://doi.org/10.1016/j.jphotochem.2018.07.010

S. Banakhojasteh et al.
JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxx–xxx
amount of Sr is presented.
dissolved in 50 cm3 deionized water and stirred continuously for 1 h
at room temperature to obtain a clear solution. The obtained mixture
was transferred into a 60 cm3 PTFE-lined autoclave and aged statically
at 453 K for 12 h. The resulting brown powder was filtered off and
purified by repeated washing with a mixture of ethanol and deionized
water (volume ratio 1:1), and dried under air at 353 K for 4 h. The as-
prepared carbon spheres along with PVP were used as templates to
synthesize the porous metal oxides.
2. Material and methods
2.1. Chemicals
Absolute ethanol (C2H6O), dimethylformamide (DMF), methanol,
sodium
hydroxide,
polyvinylpyrrolidone
(PVP,
Mw
=
1,300,000 g mol−1), and tetrabutyl titanate (TBT) were purchased from
Sigma Aldrich (Germany). Glucose monohydrate (C6H12O6·H2O),
strontium nitrate (Sr(NO3)2) and hexachlorplatinic acid (H2PtCl6) were
obtained from Alfa Aesar (Germany). Copper (II) nitrate trihydrate (Cu
(NO3)2·3H2O) was purchased from VWR (Germany). All chemicals were
used as received without further purification. H2 and N2 were supplied
by Air Liquide with a purity of > 99.999 vol.-%. Deionized water was
used in all preparations and photocatalytic experiments.
2.2.3.2. Porous SrTiO3. A series of porous SrTiO3 material was
synthesized as described in Section 2.2.1 through a template-assisted
hydrothermal method. To a solution of TBT and Sr(NO3)2 with a molar
ratio of nTi:nSr = 1:1.05, 1 g PVP was added. Upon PVP addition, the
solution became viscous. When the solution turned transparent, varying
amounts of carbon spheres (50, 100, 150, 200 mg) were added and the
pH was adjusted to 13 by the addition of an aqueous 3 M NaOH
solution. The mixture was added into a PTFE-lined stainless steel
autoclave (85 cm3) and kept at 353 K for 48 h. Finally, the solid
products were removed and calcined at 823 K for 6 h (heating rate of
1 K min−1) under static air. The obtained materials are denoted as
SrTiO3-5%-PC50, SrTiO3-5%-PC100, SrTiO3-5%-150 and SrTiO3-5%-
PC200, respectively, with the last value designating the amount of
carbon spheres added during the synthesis in mg.
2.2. Preparation of the photocatalysts
2.2.1. Synthesis of pure-SrTiO3 and SrTiO3 with excess amount of Sr
SrTiO3 was synthesized by a one-step, template-free hydrothermal
method according to Li et al. [6]. Tetrabutyl titanate (TBT) was added
into 10 cm3 of ethanol (“solution A”). Sr(NO3)2 was dissolved in 30 cm3
of deionized water (“solution B”). The solution A was added dropwise
into solution B under gentle stirring with different molar ratios nTi :nSr
of 1:1.00, 1:1.05, 1:1.07, 1:1.09, and 1:1.12, respectively. The pH was
adjusted to 13 by adding an aqueous 3 M NaOH solution under vigorous
stirring for 20 min. Thereafter, the mixture was transferred into a
polytetrafluoroethylene (PTFE)-lined autoclave (85 cm3) and heated at
453 K for 48 h. After cooling to room temperature, the solid product
was separated by filtration and washed three times with deionized
water and ethanol. Then, the solid was calcined in static air at 823 K for
6 h with a heating rate of 10 K min−1. According to the respective
amount of excess Sr (in mol-% over Ti) added during synthesis, the
obtained materials are denoted as pure-SrTiO3, SrTiO3-5%, SrTiO3-7%,
SrTiO3-9% and SrTiO3-12%.
2.2.3.3. SrTiO3-9%-PC150. Another synthesis was conducted as
described above, but with a molar ratio of nTi:nSr = 1:1.09, 1 g PVP
and 150 mg of carbon spheres.
2.2.3.4. SrTiO3-5%-P and SrTiO3-5%-C50. In addition, mesoporous
materials denoted as SrTiO3-5%-P and SrTiO3-5%-C50 were prepared
for a molar ratio of nTi:nSr = 1:1.05 using only PVP-template (1 g) or
only carbon spheres (50 mg) as template, respectively.
2.3. Characterization
The calcined SrTiO3 with added excess of Sr, the CuO/SrTiO3
composites and the porous SrTiO3 materials were characterized by N2-
sorption, Hg-intrusion, laser diffraction, elemental analysis (by optical
emission spectrometry with inductively coupled plasma, ICP-OES),
powder X-ray diffraction (XRD), UV/Vis spectroscopy, and scanning
electron microscopy (SEM).
2.2.2. Synthesis of CuO and of CuO/SrTiO3 composites
CuO was synthesized according to the method reported by Chen
et al. [21]. In a typical preparation, Cu(NO3)2·3H2O (0.07 mol) was
dissolved in 30 cm3 of a DMF–ethanol mixed solvent (volume ratio 1:2),
stirred for 15 min to form a clear solution, and then transferred into a
50 cm3 PTFE-lined stainless steel autoclave. The autoclave was sealed,
maintained at 403 K for 15 h, and, then, cooled to room temperature.
The solid product was collected by centrifugation and washed thor-
oughly with deionized water and ethanol. The collected dark solid was
dried in a vacuum oven at 333 K for 8 h. Finally, CuO was obtained by
heating in air at 673 K for 4 h (heating rate 10 K min−1).
The CuO/SrTiO3 composites were prepared through an impregna-
tion method. 1 g of the pure-SrTiO3 sample were dispersed in 20 cm3
absolute ethanol at room temperature. Subsequently, 20 cm3 deionized
water containing varying amounts of Cu(NO3)2·3H2O (0.1 and 1.0 wt.-
% with respect to pure-SrTiO3) were added to the SrTiO3 suspension
under stirring for 1 h at room temperature. Then, an aqueous 0.1 M
NaOH solution was added dropwise until the pH of the mixture reached
10. After further stirring for 1 h, the black CuO/SrTiO3 composite
precipitate was collected by centrifugation (2500 min−1, 10 min),
rinsed three times with a mixture of deionized water and absolute
ethanol (volume ratio 1:1) and dried at 353 K in air for 6 h. Finally, the
as-prepared composites were calcined at 673 K for 3 h (heating rate of
10 K min−1).
A micrometrics ASAP2010-Physisorption Analyzer was utilized to
record the N2-sorption isotherms for calcined and spent catalysts. The
samples were evacuated at 523 K under vacuum (3∙10−11 MPa for 6 h)
before measurement and the isotherms were taken at 77 K. The specific
surface area (ABET) was determined by the Brunauer–Emmett–Teller
(BET) model. The specific pore volume (Vp) was estimated from the N2
uptake at a P/P0 value of 0.99. The pore width distribution (Dp) was
deduced from the desorption branch using Barrett-Joyner-Halenda
(BJH) method. The mean macropore width of the meso-/macroporous
series of SrTiO3 was determined by Hg-intrusion in a Quantachrome
instrument (Poremaster 60 GT). A contact angle of 141.3° for Hg was
used for further calculations. The cumulative pore volume at a given
pressure represents the total volume of mercury taken up by the sample
at that pressure. The mean pore width was calculated by applying the
Washburn equation. The particle size distribution for the ground and
sieved fraction was determined based on laser diffraction in aqueous
suspension (Particle size analyzer 1064L, Cilas S. A). The elemental
analysis was carried out using a Perkin Elmer Optima 8000 ICP-OES
instrument with a Scott Cross Flow atomizer. Prior to analysis, the
samples were digested in a PTFE vessel in a mixture of 2.0 cm3 HF
(48%, Roth), 3.0 cm3 HNO3 (69%, Roth) and 3.0 cm3 HCl (35%, Roth)
using an Anton Parr Multiwave 3000 microwave oven (unpulsed mi-
crowave oven (1400 W) with 8 XF100 rotor at 448 K for 70 min). After
digestion, 12.0 cm3 H3BO4 (> 99.9995%, Alfa Aesar) were added and
the mixture was again treated for 10 min at 448 K. Powder XRD
2.2.3. Synthesis of porous SrTiO3 with PVP and/or carbon spheres as
templates
2.2.3.1. Carbon spheres. Carbon spheres were prepared by
a
hydrothermal method [22]. 5 g of glucose monohydrate were
2

S. Banakhojasteh et al.
JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxx–xxx
patterns for all calcined porous and composite SrTiO3 catalysts were
recorded at room temperature using a Siemens D5000 diffractometer.
The diffracted intensity of Cu-Kα radiation (ʎ = 0.154 nm) was mea-
sured in the range of 2ϑ between 4.0° and 90.0° with a step size of 0.02°
and a counting time of 0.2 s for phase identification. UV/Vis absorption
measurements in diffuse reflectance mode were performed on a Lambda
650 S instrument (Perkin Elmer) equipped with a 150 mm integration
ball. The measurements were carried out in a wavelength range of
200–800 nm. The step size was 1 nm with a recording speed of 300 nm
min−1. Scanning electron microscopy (SEM) was conducted using a
LEO GEMINI 1530 SEM instrument (ZEISS Germany). Prior to analysis,
the samples were placed on a sample holder using carbon tape and
covered with an Au film to make them conductive. The accelerating
voltage used was 5–20 kV and the measuring distance 10–15 mm.
2.4. Photocatalytic experiments
The photocatalytic experiments were carried out in batch mode in a
stirred slurry reactor made from glass (VR = 577 cm3) equipped with a
(cooled) middle pressure Hg-lamp (Heraeus, 150 W, emission spectrum
shown in Fig. S1, Supplementary information) held in double-walled
quartz glass tube. Typically, 150 mg of the photocatalyst were added to
a mixture of 38 cm3 methanol and 342 cm3 water. Hexachlorplatinic
acid solution (183 μl, c = 62.6 mmol L−1) was added and reduced in-
situ to elemental Pt by the incident light of the Hg-lamp resulting in
deposition of Pt onto the catalyst surface (corresponding to a catalyst
loading of 0.5 wt.-% Pt). The reactor was connected to a gas rack and
the whole system was flushed with N2. After starting irradiation, the
amount of evolved hydrogen was determined by sampling from the gas
phase above the reaction suspension and analysis by gas chromato-
graphy (GC, HP 6890 Series, Co. Hewlett Packard).
Fig. 1. XRD patterns of CuO, SrTiO3 and the CuO/SrTiO3 composites.
The photocatalytic activity of the prepared materials was evaluated
under UV light irradiation at ambient temperature and pressure. It was
determined from the amount of hydrogen formed as a function of re-
action time as the hydrogen evolution rate (HER). For all investigated
catalysts, the HER was constant over at least 60 min. The stability of the
photocatalysts was studied by switching off the Hg-lamp after 120 min
of reaction, flushing the reactor after with N2 and, again, starting ir-
radiation of the suspension and analysis of the gas phase for hydrogen
and determination of HER.
Fig. 2. UV/Vis spectra of CuO, SrTiO3 and the CuO/SrTiO3 composites.
activities can be neglected.
The UV/Vis diffuse reflectance spectra of CuO, SrTiO3 and the CuO/
SrTiO3 composites are presented in Fig. 2. While SrTiO3 absorbs light
below about 392 nm in the UV light region, CuO absorbs in the entire
visible light region due to its narrow band gap. The composites loaded
with 0.1 wt.% and 1.0 wt.% CuO show an enhanced absorbance in the
visible light region. The observed change in the color from the white
SrTiO3 to the black CuO/SrTiO3 composites also supports the ability of
CuO/SrTiO3 for absorbing light in the visible light region. The visible
light absorption range of the composite photocatalysts gradually in-
creases as the CuO loading increases. The band gap value of 0.1 wt.-%
CuO/SrTiO3 and 1.0 wt.-% CuO/SrTiO3 was estimated through plotting
of a Tauc-plot and amounts to 3.0 and 2.6 eV, respectively (see Fig. S3,
Supplementary information).
3. Results and discussion
3.1. CuO/SrTiO3 composites
In Fig. 1 the XRD patterns of CuO, SrTiO3 and the CuO/SrTiO3
composites are shown. The SrTiO3 sample possesses diffraction reflexes
of a cubic perovskite structure (JCPDS Card No. 035-0734), which is in
agreement with results reported by Solanki et al. [23]. The diffraction
reflexes of the CuO sample can attributed to the monoclinic CuO crystal
phase (JCPDS Card No. 045-0937) [24]. No characteristic reflexes were
observed for Cu(OH)2 or Cu2O [25]. For the CuO/SrTiO3 composites,
the diffraction reflexes for SrTiO3 were clearly visible. However, no
typical diffraction reflexes of CuO were observed in the XRD pattern of
the 0.1 wt.-% CuO/SrTiO3 composite, which may be due to the low
content of CuO. Only at the relatively high CuO loading of 1.0 wt.-%,
the diffraction reflexes of both CuO and SrTiO3 are clearly visible
(Fig. 1) [12,23]. Note that at the high CuO loading of 1.0 wt.-%, ad-
ditional reflexes of impurity phases are visible, e.g., at 2 θ = 36, 46, or
52°, the origin of which, was, however, not further analyzed.
To evaluate the photocatalytic properties of the CuO/SrTiO3 com-
posites, the hydrogen generation from aqueous methanol solution
under UV light irradiation was investigated. As illustrated in Fig. 3, the
hydrogen evolution rate of the parent SrTiO3 is 39 μmol h−1, while CuO
alone did not result in any detectable hydrogen generation. Whereas
both of the two CuO/SrTiO3 composite catalysts show significant H2
generation activity, the amount of hydrogen formed over CuO/SrTiO3
with 1.0 wt.-% CuO content is 120 μmol h−1. As expected, this value is
higher than that for CuO/SrTiO3 with the lower CuO content of 0.1 wt.-
% (53 μmol h−1). Furthermore, without Pt co-catalyst, the photo-
The CuO/SrTiO3 composite particles are agglomerated (shown by
SEM images for 1.0 wt.-% CuO content in Fig. S2, Supplementary in-
formation). This agglomeration is most likely due to a sintering during
the high temperatures of calcination during the preparation. Since all
materials were calcined before use in the catalytic experiments, an ef-
fect of this agglomeration on the comparison of the photocatalytic
catalytic activity of the 1.0 wt.-% CuO/SrTiO3 was around 31 μmol h−1
.
Thus, it can be concluded that the CuO/SrTiO3 composites exhibit a
3

S. Banakhojasteh et al.
JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxx–xxx
Fig. 3. Hydrogen evolution rate for photocatalytic conversion of aqueous me-
thanol solution over Pt-containing CuO, SrTiO3 and CuO/SrTiO3 composites.
Fig. 4. XRD patterns of pure-SrTiO3 and SrTiO3-X samples (X: excess amount of
Sr in mol-% over Ti).
significantly higher photocatalytic activity for hydrogen generation
than the unmodified SrTiO3. Moreover, due to their favorable absorp-
tion properties, these catalysts are also attractive candidates for hy-
drogen formation from aqueous methanol solution in the visible light
region.
(a) isotherms according to the IUPAC classification [28], suggesting the
presence of mesopores (see Fig. S4, Supplementary information). The
specific surface areas (ABET) of the materials slightly increases as the
excess amount of Sr (with respect to Ti) increased by up to 9 mol-% and
decreases for the sample containing 12 mol-% Sr excess (see Table 1).
The pore width distributions as calculated from the desorption branch
using the Barrett-Joyner-Halenda (BJH) method [29] for mesoporous
materials indicates slightly larger pore widths in the presence of larger
Sr excess (see Table 1). Both, the largest specific pore volume and
surface area are observed for SrTiO3-9%.
Fig. 5 shows the UV/Vis absorption spectra of SrTiO3 with various
excess amounts of Sr. The pure-SrTiO3 absorbs light in the UV region at
wavelengths below 392 nm. With increasing excess amount of Sr, a
slight shift in the absorption from about 392 nm to 400 nm is observed.
A pronounced effect of the Sr excess on the UV/Vis absorption was,
however, not observed.
The pure-SrTiO3 and SrTiO3-X with excess of Sr are active for the
photocatalytic hydrogen generation from aqueous methanol solution
under UV light irradiation (Fig. 6). The hydrogen evolution rate in-
creases with increasing excess of Sr reaching a maximum for 9 mol-% of
excess amount of Sr. Thus, the HER increases from 39 μmol h−1 for the
parent SrTiO3 to 92 μmol h−1 for SrTiO3-9%, while for 12 mol-% excess
Sr, it is significantly lower (21 μmol h−1). It has already been demon-
strated using transient spectroscopy that Na-O defects in NaTaO3 as
another perovskite photocatalyst promote the recombination of pho-
togenerated electron-hole pairs [30]. Therefore, the low photocatalytic
activity at the high excess Sr amount of 12 mol-% may also be attrib-
uted to the formation of a too high density of SrO defects, whereas at 9
mol-% of Sr excess, the number of defects is minimized and, therefore,
the recombination of photogenerated electron-hole pairs in SrTiO3 is
suppressed [10]. It may also be considered that composites of SrO and
SrTiO3 form at high Sr excess of 12 mol-% leading to additional defect
centers for electron-hole recombination. Furthermore, and although
low in general, SrTiO3-9% has the highest specific pore volume and
The results presented above clearly indicate that the CuO/SrTiO3
composites are photocatalytically active for hydrogen generation from
aqueous methanol solution. The valence band (VB) and the conduction
band (CB) of CuO are located at 2.07 and 0.55 eV, while those of SrTiO3
are at 2.20 and −1.32 eV, respectively. Since the CB of SrTiO3 is higher
than that of CuO, the electrons can easily transfer from SrTiO3 to CuO.
Under UV light irradiation, the photogenerated electron-hole pairs are
excited from the VB to the CB of SrTiO3, creating holes in the VB. In
addition, the sacrificial electron donor consumes holes in the VB rapidly
leaving electrons in the CB of SrTiO3. Then, the electrons in the CB of
SrTiO3 will migrate to the CuO, which improves the separation of
photogenerated electron-hole pairs, suppressing recombination be-
tween electrons and holes, enhancing charge migration and photo-
catalytic activity for hydrogen generation from aqueous methanol so-
lution.
3.2. SrTiO3 with excess of Sr
XRD patterns of the samples synthesized with various nSr/nTi atomic
ratios by a template-free hydrothermal method show diffraction pat-
terns of a cubic perovskite structure (JCPDS No.79-0176) [26]. No
additional reflexes were observed, even for the highest nSr/nTi atomic
ratio of 1.12 (Fig. 4). Furthermore, the high intensity of the main (110)
reflex indicates that these samples possess high crystallinity [27]. Note
that both the crystallinity and the phase purity of the materials with
different nSr/nTi atomic ratios are comparable. Differences in these
materials are, thus, not due to their long-range ordering. The average
crystallite size of SrTiO3 with excess amount of Sr as calculated by the
Scherrer equation from the XRD data is about 30 nm. From elemental
analysis, a nSr/nTi-ratio of 0.97 in the parent SrTiO3 material was de-
termined. The nSr/nTi atomic ratio lower than one suggests that a
substitution of OH− to O2− may form Sr2+ defects with consideration
of charge compensation, leading to the general formula Sr1-xTiO3-
2x(OH)2x. Calcination of the precursor possessing Sr2+ defects and OH−
substitution should result in the formation of SrTiO3 possessing SrO
defects, i.e., Sr1-XTiO3-X, via dehydration. It is generally accepted that
SrO defects act as recombination centers for photo-induced electrons
and holes decreasing the photocatalytic activity. Therefore, addition of
Sr(NO3)2∙3H2O to the precursor before annealing was attempted to
eliminate the SrO defects.
Table 1
Specific BET surface areas (ABET), specific pore volumes (Vp) and pore widths
(Dp) determined by the BJH method from the desorption branch of the N2
sorption isotherms at 77 K of SrTiO3 with different amounts of excess Sr.
Sample
ABET/m2 g−1
Vp/cm3 g−1
Dp/nm
Pure-SrTiO3
SrTiO3-5%
SrTiO3-7%
SrTiO3-9%
SrTiO3-12%
17
16
19
21
16
0.06
0.05
0.06
0.07
0.06
12.5
14.3
11.9
14.9
16.5
The N2 sorption analysis of the prepared materials revelas type-IV
4

S. Banakhojasteh et al.
JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxx–xxx
Table 2
Specific BET surface areas (ABET), specific pore volumes (Vp), pore widths (Dp),
pore volume (VHg) and median diameter estimated from Hg-intrusion (DHg) for
meso-/macroporous SrTiO3 with 5% Sr excess from template-assisted synthesis.
Samples
ABET/m2 g−1 Vp/cm3 g−1 Dp/nm VHg/cm3 g−1 DHg/nm
SrTiO3 -5%
SrTiO3-5 %-PC50 16
SrTiO3-5%
PC100
SrTiO3-5%-
PC150
16
0.05
0.04
0.07
14.3
11.9
13.4
–
0.30
0.33
–
65.5
72.5
22
25
23
0.10
0.08
16.8
14.5
0.36
0.48
75.5
SrTiO3-5%-
PC200
115.0
SrTiO3-5%-P
SrTiO3-5%-C50
23
18
0.06
0.06
14.9
14.4
–
–
–
–
is a slight change from 30 to 40 nm for the crystal size when template-
assisted synthesis is used for the porous materials. These results are in
agreement with SEM analysis (not shown).
Fig. 5. UV/Vis spectra of pure-SrTiO3 and SrTiO3-X with different amounts X of
excess Sr.
N2-sorption isotherms of porous SrTiO3, SrTiO3-C50 and SrTiO3-P
materials correspond to type-IV(a) (see Figs. S7 and S8, Supplementary
information), according to the IUPAC classification [28]. This suggests
the presence of a mesopore system in these materials. However, the
porous SrTiO3 exhibits a type-II isotherm with a type-IV(a) hysteresis
loop at high relative pressure (P/P0) range of 0.8–1.0, indicating the
presence of macropores of ca. 75 nm in width (DHg). It is worth men-
tioning that the meso- and macroporous materials (prepared with both
carbon spheres and PVP as templates) exhibit specific surface areas
(ABET) slightly higher than those of the mesoporous series (from
synthesis with either carbon spheres or PVP as template), and a larger
specific volume of meso- and macropores (Table 2). The largest specific
pore volume of 0.10 cm3 g−1 was observed for the SrTiO3-5%-PC150
sample. Note that the overall specific surface area and pore volumes as
well as the differences between the materials prepared with different
templates as detected by N2 sorption are rather low. The major differ-
ence between the materials is the presence of macropores as determined
by Hg-intrusion, with increasing specific pore volume (from 0.30 to
0.48 cm3 g−1) and increasing pore width (from 65.5 to 115.0 nm) with
increasing amount of carbon spheres added during synthesis (see
Table 2).
Moreover, the pore width distributions calculated from the deso-
rption branch using the Barrett-Joyner-Halenda (BJH) method for both
meso-/macroporous and mesoporous materials (see Fig. S9,
Supplementary information) is similar for all materials ranging for
mesopores from 15 to 50 nm with a maximum at about 30 to 40 nm and
for meso-/macropores from 10 to 100 nm with a maximum about 40 to
70 nm. The porous SrTiO3-PC150 sample exhibits two types of pores,
i.e., small mesopores at 10–25 nm and large meso-/macropores cen-
tered at 40–90 nm, while the relative intensities of two main peaks
varies. In addition, the specific surface area of porous SrTiO3 with
meso-/macropores increases as the amount of carbon spheres added
during synthesis increases.
Fig. 6. Hydrogen evolution rate as a function of the excess Sr amount X in the
photocatalytic conversion of aqueous methanol solution over Pt-containing
SrTiO3-X.
surface area of the investigated materials, which might contribute to
facilitated transport of reactants and products to the active sites. In
conclusion, the introduction of an excess amount of 9 mol-% of Sr re-
sults in an almost doubled photocatalytic activity with respect to the
unmodified SrTiO3.
3.3. Meso-/macroporous SrTiO3
The XRD patterns of porous SrTiO3 materials with 5 mol-% excess Sr
show diffraction reflexes of a cubic perovskite structure (JCPDS No.79-
0176) [26]. Evidently, the introduction of additional porosity does not
affect the crystal structure of the formed SrTiO3 phase (see Fig. S6,
Supplementary information). The diffractograms of SrTiO3-5%-C50 and
SrTiO3-5%-P also confirm the presence of the perovskite structure (see
Fig. S6, Supplementary information) [26]. The average particle size of
prepared materials are summarized in Tab. S1, Supplementary in-
formation. Particle sizes of the porous SrTiO3-5% with meso- and
macropores increase from 21 to 27 μm as the amount of carbon spheres
used during synthesis increases. The particle size of the mesoporous
materials SrTiO3-5%-C50 and SrTO3-5%-P prepared with only carbons
spheres or PVP are in the same range, i.e., 28 and 29 μm, respectively.
The average of crystallite size as determined using the Scherrer equa-
tion on the cubic (2 theta = 32.25°) diffraction reflexes show that there
Mercury porosimetry was carried out on the four SrTiO3-5% pow-
ders from synthesis with both carbon spheres and PVP as templates.
Pore width distributions for the materials are presented in Fig. S10,
Supplementary information. The largest pore width (DHg) of 115 nm
was observed for the porous SrTiO3-5%-PC200.
The UV/Vis absorption spectra of porous SrTiO3 samples are shown
in Fig. 7. All porous samples absorb light below about 392 nm similar to
SrTiO3, corresponding to a band gap of about 3.1–3.2 eV. With in-
creasing amount of carbon spheres added during synthesis, the light
absorption increases in the whole wavelength region above 392 nm.
With increasing amount of carbon spheres added during synthesis
up to 150 mg, the hydrogen evolution rate over the SrTiO3 photo-
catalysts more than doubles from about 60 μmol h−1 to 130 μmol h−1
while it decreases to about 85 μmol h−1 for 200 mg (see Fig. 8). The
,
5

S. Banakhojasteh et al.
JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxx–xxx
SrTiO3 with meso- and macropores shows significantly higher photo-
activity than the CuO/SrTiO3 composites and SrTiO3 with an excess of
Sr. This can be attributed to the unique meso-/macroporous structure.
Thus, the meso-/macropore system within the SrTiO3 photocatalyst
provides a high density of active sites and may allow for faster diffusion
of reactants. Moreover, the interconnected macroporous network can
have the advantage of improving charge carrier mobility and assist the
separation of electron-hole pairs at the catalyst surface. Therefore, the
recombination process of the electron-hole pairs is effectively inhibited
resulting in an obvious improvement of hydrogen formation for the
porous SrTiO3 photocatalysts.
Furthermore, a template-assisted hydrothermal synthesis was ap-
plied to obtain the SrTiO3-9%-PC150 photocatalyst. The photocatalytic
activity for hydrogen generation over this catalyst amounts to
97 μmol h−1 and is, thus, higher than for both SrTiO3-5% and SrTiO3-
9% (see Section 3.2). However, SrTiO3-9%-PC150 shows a lower pho-
tocatalytic activity for hydrogen generation from aqueous methanol
solution than SrTiO3-5%-PC150 (see Fig. 8).
Moreover, the stability of the meso-/macroporous photocatalyst
SrTiO3-5%-PC150 was studied (see Fig. S11, Supplementary informa-
tion). After three runs, the catalyst is still highly active. In addition, the
photogenerated efficiency reached 383% within 120 min of UV illu-
mination, which is higher than that of the first run.
Fig. 7. UV/Vis spectra of porous SrTiO3-5%-PC (with varying amount of carbon
spheres and constant amount of PVP as templates), SrTiO3-5% (synthesized
without template), SrTiO3-5%-P (synthesized with only PVP as template),
SrTiO3-5%-C50 (synthesized with only carbon spheres as template).
The effect of calcination temperature on the photocatalytic activity
of the synthesized meso-/macroporous SrTiO3-5%-PC150 catalyst is
shown in Fig. 9. As the calcination temperature increases from 823 K to
1023 K, the photocatalytic activity for hydrogen generation from aqu-
eous methanol solution decreases. This can be explained by the lower
specific surface area, but higher crystallinity of the catalyst present
after calcination at higher temperature (Fig. S12, Supplementary in-
formation) [35]. Consequently, a proper synthesis procedure is needed
to provide a balance between crystallinity and specific surface area of
the photocatalyst.
4. Conclusions
Modification of SrTiO3 by addition of CuO, an excess amount of Sr
and the introduction of additional meso-/macroporosity efficiently in-
creases the photocatalytic activity for hydrogen evolution from aqueous
methanol solution. Depending on the type of modification, the origin of
the improved photocatalytic activity may, however, be different. It may
be due to a modification of the band gap (for CuO/SrTiO3 composites),
the elimination of SrO defects in the SrTiO3 framework (for an excess of
Sr) or due to an increase in specific surface area and pore volume (for
Fig. 8. Hydrogen evolution rate as a function of mass of carbons spheres added
during synthesis of meso-/macroporous Pt-containing SrTiO3 (black cubes) and
for SrTiO3-9%-PC150 (blue cube) in the photocatalytic conversion of aqueous
methanol solution. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of this article).
maximum conversion is observed for the porous SrTiO3-5%-PC150
photocatalyst which also exhibits the highest specific surface area of
25 m2 g−1 (compare Table 2). A high surface area provides a high
number of active sites for the photocatalytic conversion and promotes
light harvesting [31]. Yin et al. [32] synthesized ordered macroporous
BiOBr with controllable dual porosity via a modified colloidal crystal
templating method and verified that light harvesting as well as charge
migration in periodically ordered macroporous architectures has a
strong dependence on the surface area. The specific surface areas of the
ordered macroporous BiOBr was found to be 17 m2 g−1. Compared to
BiOBr, the specific surface areas and pore volumes of the pure BiPO4
nanorods were smaller (3.648 m2 g−1 and 0.0087 cm3 g−1) due to the
dense and smooth morphology. Recently, McDonald et al. [33] men-
tioned that one of the main reasons for the charge recombination in
BiVO4 is the long diffusion path length of the photo-induced electrons.
Modifying porous BiVO4, especially by ordered pore systems, can
shorten this diffusion path length and thus facilitate charge migration,
readily providing access to and sorption of reactants on the active sites
[33,34]. Thus, the introduction of additional porosity to SrTiO3 in-
creases the photocatalytic activity for hydrogen generation. Note that
Fig. 9. Hydrogen evolution rate as a function of calcination temperature in the
photocatalytic conversion of aqueous methanol solution over Pt-containing
meso-/macroporous SrTiO3-5%-PC150.
6

S. Banakhojasteh et al.
JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxx–xxx
meso-/macroporous SrTiO3), respectively. The inhibition of electron-
hole recombination is, of course, an accompanying effect that is likely
superimposed and also plays an important role. The highest increase in
hydrogen evolution rate with respect to the unmodified SrTiO3 was
achieved by combining the addition of Sr excess (5 mol-% with respect
to Ti) and the introduction of meso-/macropores by using both carbon
spheres and polyvinylpyrrolidone as templates. With this catalyst, the
hydrogen evolution rate amounts to 130 μmol h−1 which is more than
threefold higher than that of the unmodified SrTiO3. The above men-
tioned approaches may, therefore, als be applied for other inroganic
semiconductor photocatalysts for activity improvement.
It should be noted, however, that both for excess Sr and for added
meso-/macroporosity, a maximum in photocatalytic activity was ob-
served, i.e., the activity decreases, if a larger Sr excess or more porosity
was present. Therefore, a subtle balance between the effects of band
gap modification, defect site density and porosity are important for
further improving the activity of semiconductor photocatalysts for hy-
drogen generation from aqueous solutions.
[4] H. Jeong, T. Kim, D. Kim, K. Kim, Int. J. Hydrogen Energy 31 (2006) 1142–1146.
[5] G. Yang, W. Yan, J. Wang, Q. Zhang, H. Yang, J. Sol-Gel Sci. Technol. 71 (2014)
159–167.
[6] T.-Y. Ma, H. Li, T.-Z. Ren, Z.-Y. Yuan, RSC Adv. 2 (2012) 2790–2796.
[7] B. Modak, K. Srinivasu, S.K. Ghosh, RSC Adv. 4 (2014) 45703–45709.
[8] D. Hou, X. Hu, W. Ho, P. Hu, Y. Huang, J. Mater. Chem. A 3 (2015) 3935–3943.
[9] H.-C. Chen, C.-W. Huang, J.C.S. Wu, S.-T. Lin, J. Phys. Chem. C 116 (2012)
7897–7903.
[10] H. Kato, Y. Sasaki, N. Shirakura, A. Kudo, J. Mater. Chem. A 1 (2013)
12327–12333.
[11] P. Jing, J. Du, J. Wang, W. Lan, L. Pan, J. Li, J. Wei, D. Cao, X. Zhang, C. Zhao,
Q. Liu, Nanoscale 7 (2015) 14738–14746.
[12] J. Liu, L. Zhang, N. Li, Q. Tian, J. Zhou, Y. Sun, J. Mater. Chem. A 3 (2015)
706–712.
[13] U. Sulaeman, S. Yin, T. Sato, IOP Conf. Ser.: Mater. Sci. Eng. 18 (2011)
32018–32022.
[14] Q. Kuang, S. Yang, ACS Appl. Mater. Interfaces 5 (2013) 3683–3690.
[15] K.S. Krishna, S. Maity, K.K.R. Datta, J. Nanosci. Nanotechnol. 13 (2013)
3121–3126.
[16] T. Wang, X. Yan, S. Zhao, B. Lin, C. Xue, G. Yang, S. Ding, B. Yang, C. Ma, G. Yang,
G. Yang, J. Mater. Chem. A 2 (2014) 15611–15619.
[17] Z. Yan, J. He, L. Guo, Y. Li, D. Duan, Y. Chen, J. Li, F. Yuan, J. Wang, Catalysts 7
(2017) 82–94.
[18] H. Wang, A.L. Rogach, Chem. Mater. 26 (2014) 123–133.
[19] D. Zhang, G. Li, H. Li, Y. Lu, Chem. Asian J. 8 (2013) 26–40.
[20] J.H. Pan, H. Dou, Z. Xiong, C. Xu, J. Ma, X.S. Zhao, J. Mater. Chem. 20 (2010)
4512–4528.
Acknowledgements
[21] L. Zhao, H. Chen, Y. Wang, H. Che, P. Gunawan, Z. Zhong, H. Li, F. Su, Chem.
S.B. acknowledges generous support through a Research Grant for
Doctoral Candidates from the German Academic Exchange Service
(DAAD, ref. no. 91525602-57048249). Thanks are due to Prof. W.-D.
Einicke and Dipl.-Ing. Heike Rudzik (both from the Institute of
Chemical Technology, Universität Leipzig) for assistance with the ni-
trogen sorption analysis and for carrying out elemental analysis, re-
spectively.
Mater. 24 (2012) 1136–1142.
[22] K.S. Krishna, S. Maity, K.K.R. Datta, J. Nanosci. Nanotechnol. 13 (2013)
3121–3126.
[23] S. Choudhary, A. Solanki, S. Upadhyay, N. Singh, V.R. Satsangi, R. Shrivastav,
S. Dass, J. Solid State Electrochem. 17 (2013) 2531–2538.
[24] T. Kimura, Y. Sekio, H. Nakamura, T. Siegrist, A.P. Ramirez, Nat. Mater. 7 (2008)
291–294.
[25] G. Du, G. van Tendeloo, Chem. Phys. Lett. 393 (2004) 64–69.
[26] G. Sreedhar, A. Sivanantham, S. Venkateshwaran, S.K. Panda, M. Eashwar, J. Mater.
Chem. A 3 (2015) 13476–13482.
[27] Y. Ham, T. Hisatomi, Y. Goto, Y. Moriya, Y. Sakata, A. Yamakata, J. Kubota,
K. Domen, J. Mater. Chem. A 4 (2016) 3027–3033.
Appendix A. Supplementary data
[28] M. Thommes, K. Kaneko, A.V. Neimark, J.P. Olivier, F. Rodriguez-Reinoso,
J. Rouquerol, K.S. Sing, Pure Appl. Chem. 87 (2015) 1051–1069.
[29] E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951) 373–380.
[30] A. Yamakata, T.-a. Ishibashi, H. Kato, A. Kudo, H. Onishi, J. Phys. Chem. B 107
(2003) 14383–14387.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.jphotochem.2018.07.
010.
[31] T. Kimura, APL Mater. 2 (2014) 113301–113309.
[32] J. Xia, S. Yin, H. Li, H. Xu, L. Xu, Y. Xu, Dalton Trans. 40 (2011) 5249–5258.
[33] Y. Park, K.J. McDonald, K.-S. Choi, Chem. Soc. Rev. 42 (2013) 2321–2337.
[34] K.J. McDonald, K.-S. Choi, Energy Environ. Sci. 5 (2012) 8553–8557.
[35] T. Puangpetch, T. Sreethawong, S. Yoshikawa, S. Chavadej, J. Mol. Catal. A: Chem.
287 (2008) 70–79.
References
[1] A. Kudo, Y. Miseki, Chem. Soc. Rev. 38 (2009) 253–278.
[2] T. Jafari, E. Moharreri, A.S. Amin, R. Miao, W. Song, S.L. Suib, Molecules 21 (2016)
1–29.
[3] X. Chen, S. Shen, L. Guo, S.S. Mao, Chem. Rev. 110 (2010) 6503–6570.
7