An investigation of the optical properties and water splitting potential of the coloured metallic perovskites Sr1?xBaxMoO3

Article abstract of DOI:10.1016/j.jssc.2015.12.002

The solid solution Sr_1?xBa_xMoO₃ (x=0.00, 0.025, 0.050, 0.075, 0.100 and 1.00) has been synthesised. Rietveld refinement of X-ray diffraction data shows that all materials crystallise with cubic (Pm-3m) symmetry and that a miscibility gap exists from x=0.1–1.0. The optical properties of the metallic perovskites Sr_1?xBa_xMoO₃ have been investigated by a combination of UV–vis spectroscopy and density functional theory (DFT). Upon increasing x from 0 to 1 in Sr_1?xBa_xMoO₃ there is a reduction in the measured band gap from 2.20?eV to 2.07?eV. The measured band gap is attributed to the electronic transition from the Mo 4d t_2g band to the e_g band. The potential of SrMoO₃ and BaMoO₃ as water-splitting photocatalysts was explored but there was no evidence of hydrogen or oxygen evolution, even with the presence of a Pt co-catalyst.

Full text of DOI:10.1016/j.jssc.2015.12.002

Journal of Solid State Chemistry 234 (2016) 87–92
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
An investigation of the optical properties and water splitting potential
of the coloured metallic perovskites Sr1 Ba MoO
ꢀx
x
3
a
a
a
c
c
b
a
H.A. Hopper , J. Le , J. Cheng , T. Weller , R. Marschall , J.Z. Bloh , D.E. Macphee ,
a
a,
n
A. Folli , A.C. Mclaughlin
Chemistry Department, University of Aberdeen, Meston Walk, Aberdeen AB24 3UE, United Kingdom
DECHEMA Research Institute, Theodor-Heuss-Allee 25, 60486 Frankfurt am Main, Germany
Institute of Physical Chemistry, Justus-Liebig-University Giessen, 35392 Giessen, Germany
a
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
The solid solution Sr1 Ba MoO3 (x¼0.00, 0.025, 0.050, 0.075, 0.100 and 1.00) has been synthesised.
ꢀx
x
Received 27 October 2015
Accepted 1 December 2015
Available online 4 December 2015
Rietveld refinement of X-ray diffraction data shows that all materials crystallise with cubic (Pm-3m)
symmetry and that a miscibility gap exists from x¼0.1–1.0. The optical properties of the metallic per-
x 3
ovskites Sr1ꢀxBa MoO have been investigated by a combination of UV–vis spectroscopy and density
Keywords:
Perovskite
Photocatalysis
Optical properties
functional theory (DFT). Upon increasing x from 0 to 1 in Sr1ꢀxBa
measured band gap from 2.20 eV to 2.07 eV. The measured band gap is attributed to the electronic
transition from the Mo 4d t2g band to the e band. The potential of SrMoO and BaMoO as water-
splitting photocatalysts was explored but there was no evidence of hydrogen or oxygen evolution, even
with the presence of a Pt co-catalyst.
x 3
MoO there is a reduction in the
g
3
3
&
2015 Elsevier Inc. All rights reserved.
1
. Introduction
oxidation potential of water (1.23 V). Often, a sacrificial reagent
such as sodium sulphite or silver nitrate, which will act as hole or
electron scavengers respectively, or a co-catalyst may be required
to complete the process [3–5].
With the rapidly expanding human population, the need for
reliable sources of energy is ever-increasing. Fossil fuel sources are
diminishing, causing costs to rise as it becomes necessary to ex-
ploit harder-to-reach reserves. Photocatalysis is a process which
can be used to produce solar fuels using light energy. The fuel of
interest is hydrogen, which can be used in a hydrogen fuel cell to
produce only water. Hydrogen can be obtained through a number
of routes, two examples of which are steam reforming of natural
gas, and electrolysis of water. Both of these processes require a
high energy input, usually in the form of fossil fuels [1].
A photocatalytic process which would utilise visible light en-
ergy for hydrogen-generation would prove to be invaluable, even
more so if the starting feedstock was water. Unfortunately, there
have been very few materials reported which can perform overall
water splitting [2]. Many studies focus on a material’s ability to
2
Titanium dioxide, TiO , is the most widely reported photo-
catalytic material. Its prevalence in photocatalytic studies is tes-
tament to its high stability, especially in aqueous solutions, and its
band gap energy of 3.2 eV in anatase, which is the most photo-
2
catalytically active polymorph [6]. TiO photocatalysts have many
applications, including being used in self-cleaning tiles and for the
purification of water, but they are limited since they only absorb
UV light, which represents only a very small fraction of the
available solar energy, i.e. 0–5%, depending on latitude, period of
the year and time of the day – visible light makes up ∼40%. [7–9] A
photocatalyst which could utilise visible light would therefore be a
much more attractive prospect.
An alternative photocatalytic material is haematite,
α
2 3
-Fe O
which, with its band gap energy of 2.0–2.3 eV is capable of ab-
sorbing visible light. Haematite is also stable in most electrolytes
and can be prepared from cheap and abundant elements. [10]
Haematite has been investigated as a potential candidate for
photoelectrochemical splitting of water and for photocatalytic
oxidation processes but is limited in its abilities by its conduction
band edge of 0.0 V vs SHE [11]. It is also limited by poor con-
ductivity and a high rate of electron-hole pair recombination
[12,13].
perform either side of the reaction; H
necessarily both. The main reason for this is that in order for a
heterogeneous semiconductor material to produce H , the bottom
2 2
or O production but not
2
of the conduction band must be more negative than the reduction
potential of water (∼0.00 V vs SHE at pH¼0). In order to produce
O
2
the top of the valence band must be more positive than the
n
Corresponding author.
E-mail address: a.c.mclaughlin@abdn.ac.uk (A.C. Mclaughlin).
A surprising recent development was the report of the metallic
http://dx.doi.org/10.1016/j.jssc.2015.12.002
022-4596/& 2015 Elsevier Inc. All rights reserved.
0

88
H.A. Hopper et al. / Journal of Solid State Chemistry 234 (2016) 87–92
oxide, Sr
x
NbO
3
as a photocatalytic material [14]. Non-stoichio-
collected on a PANalytical Empyrean diffractometer equipped with
metric phases, 0.75oxo0.9, were prepared, which displayed a
a Johansson monochromator between 5° and 110° 2
size of 0.0065652°.
The microstructure of powder samples of BaMoO and SrMoO
3 3
were analysed by a scanning electron microscope (SEM) on a Hi-
tachi S520 SEM.
Diffuse-reflectance spectra were recorded on an Agilent Cary
WinUV spectrophotometer. The spectra were transformed ac-
cording to the Kubelka–Munk method.
θ, with a step
3
ꢀ1
high electronic conductivity (s¼3 ꢁ 10 S cm
for x¼0.8). The
absorption energy gaps were measured to be ꢂ1.9 eV and the
indirect band gap energy was calculated by DFT to be 2.38 eV for
the stoichiometric phase [14,15]. Sr
conductors so that the band below the conduction band, termed
ꢀ1, has been defined as the highest fully occupied band and the
band above it as the lowest unoccupied band, B [14,15]. Transi-
tions associated with photon absorption would then either be
from Bꢀ1 to CB or CB to B
Initial results suggest Sr
x
NbO
3
(x¼0.8, 0.9) are metallic
B
1
The photocatalytic activity of the samples was evaluated by
monitoring the hydrogen or oxygen evolution in aqueous solutions
for up to 20 h. Hydrogen and oxygen production were measured in
a home-made closed gas system. Pure Argon (6.0) was used as
carrier gas controlled with mass flow controller (Bronckhorst), the
evolved gases under irradiation were analysed with a gas chro-
matograph (Shimadzu GC 2014AT) equipped with thermal con-
ductivity detector and a micropacked ShinCarbon St 100/120 col-
umn. The reactor is made of borosilicate glass, but with a quartz
window for full wavelength transmission. Prior to each measure-
ment, the double-walled glass reactor was flushed with Argon for
one hour to remove all traces of air before starting irradiation with
a Newport 150 W solar simulator (AM 1.5G). The intensity was
1
.
x
3
NbO shows promise as a photo-
catalytic material. Photocatalytic activity was reported using visi-
ble light for both oxidation and reduction water-splitting pro-
cesses, without the need for catalyst additives such as Pt, although
sacrificial reagents were used to facilitate the processes [14,15].
The photocatalytic activity was found to be influenced by the Sr
content, x, and the surface area of the material. A Sr-rich layer was
observed on the surface, and was suggested to be providing pro-
tection against photocatalytic degradation and reaction with water
at ambient temperatures.
It has been reported that strongly coloured metals could sub-
stitute the use of semiconductors as photocatalysts since the
electron emissions between bands after irradiation could still
achieve the necessary electron and hole production, the separation
of which could be favoured by increased mobility [14]. Two per-
ovskite compounds which have received little attention in the
2
measured to be 1000 W/m . The temperature of the suspension
was controlled with a Lauda ProLine RP 845 cryostat.
For hydrogen generation, 0.2 g of powder was suspended in
200 mL oxalic acid solution (0.025 M). For oxygen evolution, 0.1 g
of powder was suspended in 200 mL silver nitrate solution
literature are SrMoO
3 3 3 3
and BaMoO . SrMoO and BaMoO are
strongly red and purple coloured respectively, are good metallic
(
0.005 M). Before each measurement, the powders were sus-
pended 10 minutes in the respective solution by ultrasonication.
wt% platinum co-catalyst was deposited on SrMoO by in-situ
photodeposition, adding H PtCl to the suspension before starting
to irradiate. For comparison, impregnation with H PtCl followed
by drying at 80 °C and 2 h calcination in air at 180 °C was also
performed for Pt decoration of 1 wt% on SrMoO
4
ꢀ1
3
ꢀ1
conductors (s∼10 S cm
BaMoO respectively at 290 K) and reported to be cubic per-
ovskites with space group Pm-3m [16,17]. SrMoO has recently
been reported to undergo crystallographic phase transitions at
66 K (cubic Pm-3m to tetragonal I4/mcm) and 152 K (tetragonal
I4/mcm to orthorhombic Imma) [18]. Given that both SrMoO and
BaMoO are strongly coloured metals suggests that they may also
and ∼10 S cm
3
for SrMoO and
3
1
3
3
2
6
2
6
2
3
3
.
3
The DFT calculations were carried out using the freely available
package, CP2K/Quickstep [19]. The density functional im-
plementation in Quickstep is based on a hybrid Gaussian plane
wave scheme. Orbitals are described by an atom centred Gaussian-
type basis set while an auxiliary plane wave basis set is used to re-
expand the electron density in the reciprocal space. The core
electrons are represented by analytic Goedecker–Teter–Hutter
(GTH) pseudopotentials [20,21]. The 2s, 2p electrons of the O
atoms, the 4s, 4p, 5s, 4d electrons of Mo, the 4s, 4p, 5s electrons of
Sr and the 5s, 5p, 6s electrons of Ba were treated as valence
have photocatalytic potential.
3 3
The optical properties of both SrMoO and BaMoO have not
previously been reported. We have synthesised the materials
Sr1ꢀxBa MoO
x
3
(x¼0.00, 0.025, 0.050, 0.075, 0.100 and 1.00) and
have evaluated the materials' potentials for water splitting. We
show that it is possible to tune the optical band gap in
Sr1ꢀxBa
in Sr1ꢀxBa
from 2.20 eV to 2.07 eV. The measured band gap is attributed to
the electronic transition from the Mo 4d t2g band to the e band.
Finally we show that a miscibility gap exists in Sr1ꢀxBa MoO for
x
MoO
3
solid solutions so that upon increasing x from 0 to
1
x
MoO
3
there is a reduction in the measured band gap
g
x
3
electrons. The basis sets were double- basis functions with one
ζ
0
.1oxo1.0.
set of polarisation functions (DZVP) [22]. The plane wave cutoff
used for the density was 500 Ry. The wave function optimisation
was achieved using the traditional matrix diagonalization proce-
dure as the more efficient orbital transformation (OT) minimiser
implemented in CP2K is not suitable for metallic systems. Fermi
smearing was used to facilitate the self-consistent field (SCF)
convergence with the electronic temperature of 300 K. The density
functional used was the Perdew–Burke–Ernzerhof (PBE) func-
tional, [23] plus the Gimmer's dispersion correction (D3) [24]. Si-
milar settings were used in previous studies [25–27].
2
. Experimental
Sr1ꢀxBa
were prepared by typical solid-state synthesis. Stoichiometric
quantities of SrCO (Z99.9%), BaCO (Z99.99%) and MoO (99.5%)
were ground, pressed into pellets and heated at 900 °C in air for
0 h. The resultant Sr1ꢀxBa MoO powders were reground, re-
pelleted and heated at 1100 °C under flowing 5% H /N for 20 h,
x
MoO
3
(x¼0.00, 0.025, 0.050, 0.075, 0.100 and 1.00)
3
3
3
1
x
4
2
2
Sr1ꢀxBa
x
MoO
3
(x¼0.00, 0.025, 0.050, 0.075, 0.100 and 1.00)
with a heating ramp of 5 °C per minute and a cooling rate of 2 °C
per minute. This step was repeated until examination via XRD
indicated that the materials were phase pure.
perovskite structures were modelled by periodic 4 ꢁ 4 ꢁ 5 super-
cells containing a total of 400 atoms. The theoretical lattice para-
meters were determined by locating the most stable geometries
when varying the cell lengths. Larger 6 ꢁ 6 ꢁ 6 supercells consist-
The crystal structures of the Sr1ꢀxBa
amined by X-ray powder diffraction at room temperature on a
PANalytical X'Pert3 Powder diffractometer using a Cu K
1 source
¼1.54059 Å) between 5° and 110° 2 , with a step size of
.0131303°. X-ray diffraction patterns for Rietveld refinement were
x 3
MoO phases were ex-
α
ing of 1080 atoms of SrMoO
tion of the density of states (DOS). Only the
in all of the calculations due to the very large cells.
3
and BaMoO
3
were used for calcula-
(λ
θ
Γ
point was employed
0

H.A. Hopper et al. / Journal of Solid State Chemistry 234 (2016) 87–92
89
3
. Results and discussion
for samples with 0.1oxo1.0 which can be modelled by including
∼
4 perovskite phases with different lattice parameters.
3.1. Structural characterisation
Miscibility gaps have been reported for other perovskite sys-
3 3 x 3
tems. For example the perovskite system (CH NH )Pb(I Cl1ꢀx) has
The X-ray diffraction patterns for the Sr1ꢀxBa
lutions are shown in Fig. 1. The structural parameters are sum-
marised in Table 1. All patterns could be indexed on the cubic Pm-
x
MoO
3
solid so-
a miscibility gap of a similar magnitude (0.5oxo0.95), which was
cited as a possible reason for difficulties encountered during co-
evaporation as assembly of perovskite solar cells [32]. Another
3
m space group as previously reported for BaMoO
3
[16] or SrMoO
3
study of the perovskite system, Sr
2
Ru2ꢀxCo
x
O
6ꢀ
δ
(0.5rxr1.5),
[16,17]. The cubic perovskite structure is shown in the inset of
found a significantly smaller miscibility gap just at x¼0.67, al-
lowing for a separation between two different crystal phases: Ru-
rich Pnma and Co-rich I2/c [33].
Fig. 2. All materials are single phase, and changes in the peak
positions indicate solid solution formation (Fig. 1 (inset)). Peaks for
the scheelite phases, AMoO
full reduction of the material.
Rietveld refinement [29] was performed on the high quality
X-ray diffraction patterns of all Sr1ꢀxBa MoO phases. The GSAS
4
[28] were not observed, indicating
The SEM images of BaMoO
The particle size is ∼12 mm for BaMoO
particles are irregular in morphology and appear similar in shape
3
and SrMoO
3
are shown in Fig. 4.
3
and ∼6 mm for SrMoO
3
. The
x 3
and size to that previously reported for Sr NbO [14,15].
x
3
programme with EXPGUI interface was used for Rietveld refine-
ment of the data [30]. The Rietveld refinement fits to the high
3.2. Optical and photocatalytic properties
3 3
resolution XRD patterns of BaMoO and SrMoO phases shown in
The absorbance spectra (from Kubelka–Munk (KM) transfor-
mations) for SrMoO and BaMoO are shown in Fig. 5. The corre-
sponding Tauc plots show that the optical absorption profile
Fig. 2. There was no evidence of peak splitting or superstructure
peaks to suggest a change in symmetry across the series. The
background was fitted with the Chebyschev polynomial function
and peak shapes were modelled using a pseudo-Voigt function.
Data was successfully fitted with the Pm-3m space group for all x.
The structural and goodness of fit parameters are summarised in
Table 1.
3
3
corresponds to band gaps of 2.20 eV and 2.07 eV respectively. The
materials are very strongly coloured: purple for BaMoO
for the SrMoO material. The red or purple colour of Sr1ꢀxBa
is justified by the absorption tail extending to nearly 700 nm.
Sr1ꢀxBa MoO
3
and red
3
x
MoO
3
x
3
materials (x¼0–0.1) are also red in colour and the
A linear relationship between the a cell parameter and value of
x was observed (Fig. 3). As the value of x, and therefore the size of
the A-site cation increased, a was shown to increase in accordance
band gaps are reported in Table 1. There is no discernible change
in the optical band gap for x¼0–0.10, but this is most likely to be a
result of the limited precision of the KM approach over such a
small doping range.
2
þ
with Vegard's law (the ionic radius is 1.61 Å for Ba
in compar-
ison to 1.44 Å Sr2 [31]). The Mo–O bond length increases linearly
þ
The photocatalytic activity of the samples was evaluated by
monitoring the hydrogen or oxygen evolution in aqueous solutions
with x from 1.9875(1) Å for x¼0.00 to 2.0204(1) Å for x¼1.00
(Table 1; Fig. 3). For all phases the Mo and O fractional occupancies
(with oxalic acid or silver nitrate as the sacrificial reagent). How-
refined to within 71% of the full occupancy and were fixed at 1.0.
The Ba and Sr occupancies were refined as shown in Table 1.
Attempts to synthesise phase pure samples for 0.1oxo1.0
were unsuccessful, contrary to what has been reported previously
ever there was no evidence of hydrogen or oxygen evolution even
with the presence of Pt co-catalyst. The detection limit is
∼
0.5 mmol/h.
[
17]. We propose that this discrepancy is a result of the better
resolution of current X-ray diffractometers. The X-ray diffraction
patterns for Sr1ꢀxBa MoO (x¼0–0.1, 1.0) contain sharp symmetric
peaks. Significant broadening of the Bragg reflections is observed
3
.3. DFT calculations
x
3
x 3
The calculated lattice parameters of Sr1ꢀxBa MoO are in good
agreement with the experimental values (Table 1; Fig. 3), falling
well within the 2% uncertainty margin of standard DFT calcula-
tions. More importantly, the DFT calculations accurately reproduce
the increase in a upon changing x from 0–1, therefore confirming
the experimental findings.
The electronic structures of SrMoO
3 3
and BaMoO were in-
vestigated by plotting the calculated projected DOS (Fig. 6). Their
conduction bands (CBs) are mainly made up of Mo 4d orbitals, and
2
are partially occupied by non-bonding t2g electrons. The un-
g
occupied e bands are higher in energy than the t2g bands, with
parts overlapping according to the PBE functional. The valence
bands (VBs) are fully occupied, mainly having O 2p character. The
measured optical band gaps, 2.20 eV for SrMoO
BaMoO , can be attributed to either the electronic transition from
the O 2p band to the Mo 4d t2g band or from the Mo 4d t2g band to
the e band (denoted as Bꢀ1 to CB and CB to B in Ref. [14], re-
spectively). Similar to what is argued for SrNbO by Xu et al. [14],
the former assignment seems unlikely given that the DFT gaps of
ꢀ1 to CB for SrMoO and BaMoO are 3.11 eV and 2.71 eV, re-
3
and 2.07 eV for
3
g
1
3
B
3
3
spectively, both being significantly larger than the experimental
gaps. Note that the PBE functional tends to underestimate the
band gaps due to the well-known delocalisation error [34], in-
dicating that the real gaps could be even larger. This seems to
suggest that the experimentally measured gaps are due to the
Fig. 1. X-ray powder diffraction patterns of Sr1ꢀxBa
.075, 0.100 and 1.000). The inset shows the (211) peak evidencing the change in
cell parameters with increasing x.
x 3
MoO (x¼0.000, 0.025, 0.050,
electronic transition from Mo 4d t2g to e
g
.
0
From the DOS plots (Fig. 6), we notice that the possible

90
H.A. Hopper et al. / Journal of Solid State Chemistry 234 (2016) 87–92
Table 1
Refined cell parameters, agreement factors, atomic parameters and bond lengths for Sr1ꢀxBa
, 1/2, 1/2), z), Mo at 1a (0, 0, 0) and O at 3d (1/2, 0, 0). The band gaps estimated from Kubelka–Munk transformation of the absorption curves are also displayed.
x
3
MoO from Rietveld fits against powder X-ray diffraction data. Sr/Ba is at 1b (1/
2
x
1.0
0.100
0.075
0.050
0.025
0.000
3
Mo Uiso (Å )
0.0024(2)
0.0024(2)
0.005(1)
1.002(1)
0.00
4.06300
4.04077(5)
1.497
0.0013(2)
0.0013(2)
0.0098(7)
0.097(1)
0.900(2)
4.00100
3.9817(12)
2.721
0.0017(2)
0.0017(2)
0.0099(8)
0.071(1)
0.922(2)
3.99930
3.98026(9)
2.301
0.0016(2)
0.0016(2)
0.0091(6)
0.046(1)
0.947(1)
3.99740
3.97836(8)
2.554
0.0008(1)
0.0008(1)
0.0087(6)
0.015(1)
0.975(1)
3.99550
3.97679(7)
2.494
0.0015(2)
0.0015(2)
0.0115(7)
0.00
0.991(1)
3.99370
3.97499(3)
2.749
3
A Uiso (Å )
3
O Uiso (Å )
Ba occupancy
Sr occupancy
a (Å) (calculated)
a (Å) (experimental)
2
χ
R
R
wp (%)
(%)
13.30
8.53
14.65
9.75
13.32
7.89
13.79
7.82
13.73
7.73
14.13
8.17
p
Mo–O Bond length (Å)
A–O Bond length (Å)
Band gap (eV)
2.0204(1)
2.8573(1)
2.07
1.9908(1)
2.8155(1)
2.20
1.9901(1)
2.8145(1)
2.20
1.9892(1)
2.8131(1)
2.19
1.9884(1)
2.8120(1)
2.20
1.9875(1)
2.8108(1)
2.20
Fig. 3. The variation of the calculated and experimental cell parameters, a, with x
in the Sr1ꢀxBa
x
MoO
3
solid solutions.
n
g
and anti-bonding e
(having mainly Mo d character) orbitals. If the
n
t2g to e
g
transition is responsible for the optical band gap, as ar-
oct for BaMoO would suggest a smaller
gued above, a smaller
Δ
3
band gap, which is indeed what is found both experimentally and
theoretically. It is also worth mentioning that the same argument
would still apply even if the optical gap was owing to the transi-
tion from filled bonding e
g
orbitals, having mainly O 2p character,
to the partially filled t2g band.
Fig. 2. Rietveld refinement fit to the X-ray diffraction powder patterns of
a) SrMoO and (b) BaMoO . The data were refined in the Pm-3m space group for
(
3
3
both phases. The inset shows the cubic perovskite structure.
4
. Conclusions
electronic bands that are near the Fermi level and involved in
2
ꢀ
optical absorption are all due to the MoO
3
group. Upon in-
The solid solution Sr1 Ba MoO (x¼0.00, 0.025, 0.050, 0.075,
ꢀx
x
3
creasing x from 0 to 1 in Sr1ꢀxBa MoO there is a reduction in the
x
3
0.100 and 1.00) has been successfully synthesised. Single phase
solid solution is only found for xr0.1 and x¼1. It has been re-
ported that a structural transition to tetragonal symmetry occurs
for SrMoO3 below 266 K to alleviate compressive stress [18].
BaMoO has a tolerance factor of 1.04 (assuming the ionic radii for
6-coordinate Mo
band gap from 2.20 eV to 2.07 eV. This could be explained by
considering the ligand field splitting energy of the octahedral
complex structure [35]. The extent of hybridisation between the
2
2
2
Mo dxꢀy and d
adapted linear combinations of the O 2p orbitals, has an important
impact on the magnitude of the ligand field splitting energy
the stronger the hybridisation, the greater the ligand field splitting
energy. Because Ba has a larger ionic radius than Sr, BaMoO has a
larger lattice parameter, a, so that the Mo–O bonds in BaMoO are
longer than that in SrMoO (Table 1). This would lead to weaker
mixing between the Mo 4d and O 2p orbitals in BaMoO , i.e.
smaller ligand field splitting energy between the non-bonding t2g
z
orbitals having e
g
symmetry and the symmetry-
3
4
þ
2þ
is 0.650 Å, 12-coordinate Ba
is 1.40 Å [31]) and therefore should exhibit a
is 1.61 Å and
2
ꢀ
Δ
oct
;
6-coordinate O
hexagonal perovskite structure. Hence the crystal structures of
both end members of the solid solution are strained. It is most
likely that the size mismatch between Sr
augments this crystal strain so that phase segregation occurs for
0.1oxo1.0.
3
2
þ
2þ
3
and Ba
further
3
3
Previously it has been thought that metals cannot act as

H.A. Hopper et al. / Journal of Solid State Chemistry 234 (2016) 87–92
91
Fig. 5. Ultraviolet–visible absorbance spectra (converted from diffuse reflectance
spectra) for SrMoO and BaMoO . (For interpretation of the references to colour in
this figure, the reader is referred to the web version of this article.)
3
3
3 3
Fig. 6. Partial density of states of SrMoO (top) and BaMoO (bottom) in the cubic
perovskite structure. (For interpretation of the references to colour in this figure,
3 3
Fig. 4. SEM images of (a) BaMoO and (b) SrMoO .
the reader is referred to the web version of this article.)
photocatalysts because of the photogenerated carrier recombina-
tion without an electric field. However, recently photocatalytic
water splitting has been achieved in metallic Sr NbO [14] and it has
x 3
been suggested that strongly coloured metals could substitute the
use of semiconductors as photocatalysts. There are clear similarities
for any of the Sr1ꢀxBa
that the effective mass of the photogenerated charge carriers is
small for Sr NbO (∼0.4m where m is the free electron mass) [36].
In contrast, the effective mass of the charge carriers in SrMoO has
been reported to be ∼2.1m [37]. It is most likely that a small ef-
x 3
MoO materials. A recent study has reported
x
3
e
e
3
between Sr
x
NbO
3
[14] and Sr1ꢀxBa
x
MoO
3
materials. Both niobate
e
and molybdate perovskite phases are metallic and strongly coloured
with similar bandgaps. Furthermore the size and morphology of the
particles is also comparable. However there is no evidence of hy-
drogen or oxygen evolution even with the presence of Pt co-catalyst
fective mass of the photogenerated charge carriers is essential for
observing photocatalytic activity in a strongly coloured metallic
photocatalyst since a high mobility of the electrons is necessary to
prevent recombination of the charge carriers.

92
H.A. Hopper et al. / Journal of Solid State Chemistry 234 (2016) 87–92
Acknowledgements
[14] X. Xu, C. Randorn, P. Efstathiou, J.T.S. Irvine, Nat. Mater. 11 (2012) 595–597.
[
15] P. Efstathiou, X. Xu, H. Ménard, J.T.S. Irvine, Dalton Trans. 42 (2013)
880–7887.
16] V. Nassif, R.E. Carbinio, J.A. Alonso, J. Solid State Chem. 146 (1999) 266–270.
7
We acknowledge the University of Aberdeen for provision of a
studentship for Harriet Hopper. J. Le is supported by a CoPS stu-
dentship at the University of Aberdeen. J. Le and J. C. thank the
Maxwell supercomputing cluster at the University of Aberdeen for
computing time.
[
[17] L.H. Brixner, J. Inorg. Nucl. Chem. 14 (1960) 225–230.
[18] R.B. Macquart, B.J. Kennedy, M. Avdeev, J. Solid State Chem. 183 (2010)
249–254.
[19] J. VandeVondele, M. Krack, F. Mohamed, M. Parrinello, T. Chassaing, J. Hutter,
Comput. Phys. Commun. 167 (2005) 103–128.
[
[
[
[
[
[
20] S. Goedecker, M. Teter, J. Hutter, Phys. Rev. B 54 (1996) 1703–1710.
21] C. Hartwigsen, S. Goedecker, J. Hutter, Phys. Rev. B 58 (1998) 3641–3662.
22] J. VandeVondele, J. Hutter, J. Chem. Phys. 127 (2007) 114105.
23] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868.
24] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 132 (2010) 154104.
25] J. Cheng, M. Sulpizi, J. VandeVondele, M. Sprik, ChemCatChem 4 (2012)
References
[
[
1] S.D. Angeli, F.G. Pilitsis, A.A. Lemonidou, Catal. Today 242 (2015) 119–128.
2] S. Protti, A. Albini, N. Serpone, Phys. Chem. Chem. Phys. 16 (2014)
636–640.
[
[
26] J. Cheng, J. VandeVondele, M. Sprik, J. Phys. Chem. C 118 (2014) 5437–5444.
27] J. Cheng, X. Liu, J.A. Kattirtzi, J. VandeVondele, M. Sprik, Angew. Chem. Int. Ed.
19790–19827.
[
[
[
3] R. Abe, J. Photochem. Photobiol. C 11 (2010) 179–209.
4] Y. Sasaki, H. Nemoto, K. Saito, A. Kudo, J. Phys. Chem. 113 (2009) 17536–17542.
5] B. Wang, C. Li, D. Hirabayashi, K. Suzuki, Int. J. Hydrog. Energy 35 (2010)
5
3 (2014) 12046–12050.
28] K. Yu, E. Tismenko, M.A. Simonov, N.V. Belov, Kristallografiya 12 (1967)
11–512.
[
5
3306–3312.
[
[
29] H.M. Rietveld, Acta Crystallogr. 22 (1967) 151–152.
30] A.C. Larson, R.B. Von Dreele, General Structure Analysis System (GSAS), Los
[
[
[
6] A. Galinska, J. Walendziewski, Energy Fuels 19 (2005) 1143–1147.
7] K. Hashimoto, H. Irie, A. Fujishima, Jpn. J. Appl. Phys. 44 (2005) 8269–8285.
8] L. Kavan, M. Grätzel, S.E. Gilbert, C. Klemenz, H.J. Scheel, Electrochemical and
photoelectrochemical investigation of single-crystal anatase, J. Am. Chem. Soc.
Alamos National Laboratory, United States, 2004, unpublished.
[31] R.D. Shannon, Acta Cryst. 32 (1976) 751–767.
[32] P. Pistor, J. Borchert, W. Fränzel, R. Csuk, R. Scheer, J. Phys. Chem. Lett. 5 (2014)
118 (1996) 6716–6723.
[9] J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo, D.
3308–3312.
W. Bahnemann, Chem. Rev. 114 (2014) 9919–9986.
[33] A.D. Lozano-Gorrín, J.E. Greedan, P. Núñez, C. González-Silgo, G.A. Botton,
G. Radtke, J. Solid State Chem. 180 (2007) 1209–1217.
[34] J. Cheng, M. Sprik, Phys. Chem. Chem. Phys. 14 (2012) 11245–11267.
[35] D.F. Shriver, P.W. Atkins, Inorganic Chemistry, 3rd edn, Oxford University
Press, Oxford (1999), p. p236.
[36] Y. Zhu, Y. Dai, K. Lai, Z. Li, B. Huang, J. Phys. Chem. C 117 (2013) 5593–5598.
[37] H. Mizoguchi, N. Kitamura, K. Fukumi, T. Mihara, J. Nishii, M. Nakamura,
N. Kikuchi, H. Hosono, H. Kawazoe, J. Appl. Phys. 87 (2000) 4617–4619.
[
[
[
[
10] N.J. Cherepy, D.B. Liston, J.A. Lovejoy, H. Deng, J.Z. Zhang, J. Phys. Chem. B 102
1998) 770–776.
11] B.C. Faust, M.R. Hoffman, D.W. Bahnemann, J. Phys. Chem. 93 (1989)
371–6381.
12] C. Baumanis, J.Z. Bloh, R. Dillert, D.W. Bahnemann, J. Phys. Chem. C 115 (2011)
5442–25450.
(
6
2
13] Y.-S. Hu, A. Kleiman-Shwarsctein, A.J. Forman, D. Hazen, J.-N. Park, E.
W. McFarland, Chem. Mater. 20 (2008) 3803–3805.