Perovskite BaTaO2N: From Materials Synthesis to Solar Water Splitting [original]

REVIEW

www.advancedscience.com

Perovskite BaTaO2N: From Materials Synthesis to Solar

Water Splitting

Mirabbos Hojamberdiev,* Ronald Vargas, Fuxiang Zhang, Katsuya Teshima,

and Martin Lerch

Barium tantalum oxynitride (BaTaO2N), as a member of an emerging class of

perovskite oxynitrides, is regarded as a promising inorganic material for solar

water splitting because of its small band gap, visible light absorption, and

suitable band edge potentials for overall water splitting in the absence of an

external bias. However, BaTaO2N still exhibits poor water-splitting

performance that is susceptible to its synthetic history, surface states,

recombination process, and instability. This review provides a comprehensive

summary of previous progress, current advances, existing challenges, and

future perspectives of BaTaO2N for solar water splitting. A particular

emphasis is given to highlighting the principles of photoelectrochemical (PEC)

water splitting, classic and emerging photocatalysts for oxygen evolution

reactions, and the crystal and electronic structures, dielectric, ferroelectric,

and piezoelectric properties, synthesis routes, and thin-ﬁlm fabrication of

BaTaO2N. Various strategies to achieve enhanced water-splitting performance

of BaTaO2N, such as reducing the surface and bulk defect density, engineering

the crystal facets, tailoring the particle morphology, size, and porosity, cation

doping, creating the solid solutions, forming the heterostructures and

heterojunctions, designing the photoelectrochemical cells, and loading

suitable cocatalysts are discussed. Also, the avenues for further investigation

and the prospects of using BaTaO2N in solar water splitting are presented.

M. Hojamberdiev, M. Lerch

Institut für Chemie

Technische Universität Berlin

Straße des 17. Juni 135, 10623 Berlin, Germany

E-mail: [email protected]

R.Vargas

InstitutoTecnológicodeChascomús(INTECH)–ConsejoNacionalde

InvestigacionesCientíﬁcasyTécnicas(CONICET)

UniversidadNacionaldeSanMartín(UNSAM)

AvenidaIntendenteMarino,Km8,2,(B7130IWA),Chascomús,Provincia

deBuenosAiresArgentina

The ORCID identiﬁcation number(s) for the author(s) of this article

can be found under https://doi.org/10.1002/advs.202305179

This is an open access article under the terms of the Creative Commons

Attribution License, which permits use, distribution and reproduction in

any medium, provided the original work is properly cited.

DOI: 10.1002/advs.202305179

1. Photoelectrochemical (PEC)

Water Splitting

To tackle the rising global energy demand

and to reduce greenhouse gas emissions

from the combustion of diminishing fos-

sil fuels, the development of renewable en-

ergy is indispensable. Hydrogen is regarded

as a potential zero-emission energy car-

rier with the highest gravimetric energy

density (120 MJ kg−1) despite its low vol-

umetric energy density (8 MJ L−1)[1] and

it can play a vital role in the successful

implementation of the Paris Agreement,[2]

which is essential for the achievement of

the 17 United Nations Sustainable De-

velopment Goals. However, the current

global hydrogen production still heavily

relies on fossil fuels (>95%), which is

the cheapest option in most parts of the

world.[3,4] Solar-driven water splitting is one

of the most environmentally benign chem-

ical processes to convert abundant solar en-

ergy into storable and transportable green

hydrogen.[5–8] Solar-driven water splitting

proceeds over cocatalyst-assisted semicon-

ductor according to three consecutive steps

(Figure 1a): i) the absorption of pho-

tons with higher energy values to excite

R. Vargas

Escuela de Bio y Nanotecnologías

Universidad Nacional de San Martín (UNSAM)

Avenida Intendente Marino, Km 8,2, (B7130IWA), Chascomús, Provincia

de Buenos Aires Argentina

F. Zhang

State Key Laboratory of Catalysis

iChEM

Dalian Institute of Chemical Physics

Chinese Academy of Sciences

Dalian National Laboratory for Clean Energy

Dalian 116023, P.R. China

K. Teshima

Department of Materials Chemistry

Shinshu University

4-17-1 Wakasato, Nagano 3808553, Japan

K. Teshima

Research Initiative for Supra-Materials

Shinshu University

4-17-1 Wakasato, Nagano 3808553, Japan

2305179 (1 of 47)

www.advancedsciencenews.com www.advancedscience.com

Figure 1. a) Working principle of the photoelectrochemical cell for water splitting using a photoanode and a photocathode. b) OER mechanism for acid

(blue line) and alkaline (red line) conditions. The black and green lines indicate that the oxygen evolution involves the formation of a peroxide (M–OOH)

SocietyofChemistry.

electrons from the valence band to the conduction band, ii) the

separation and transfer of photo-excited charge carriers (elec-

trons and holes) from the bulk to the surface of the semiconduc-

tor, and iii) the initiation of water redox reactions by the involve-

ment of photo-excited charge carriers[9]:

Hydrogen evolution reaction (HER):

2H++2e−→H2(acidic media)(1)

2H2O+2e−→H2+2OH−(alkaline media)(2)

Oxygen evolution reaction (OER):

2H2O→O2+4e−+4H+(acidic)(3)

4OH−→O2+4e−+2H2O(alkaline)(4)

Particularly, photoelectrochemical (PEC) water splitting is a

powerful yet complex process, where the redox potentials for the

decomposition of water determine the required band gap energy

and band-edge potentials for semiconductors to be used as pho-

toanodes and photocathodes. However, to eﬃciently and sustain-

ably split water into H2and O2, several key criteria must be met

simultaneously: i) suﬃcient voltage must be generated upon irra-

diation to split water, ii) the bulk band gap must be small enough

to absorb a signiﬁcant portion of the solar spectrum, iii) for the

unbiased operation of a PEC cell, the band edge potentials at

the surfaces must straddle the hydrogen and oxygen redox po-

tentials (the conduction band minimum must be more negative

than E0(H+/H2)=0 V versus RHE at pH 0 and the valence band

maximummustbemorepositivethanE0(O2/H2O) =1.23 V ver-

sus RHE at pH 0), iv) the system must be stable for a long period

of reaction time, v) charge transfer from the semiconductor sur-

face to the electrolyte must be facile to minimize energy losses,

and vi) low-cost and earth-abundant elements must be used.[10]

In addition to these important requirements, the photoelec-

trode materials must exhibit at least >10% solar-to-hydrogen

(STH) conversion eﬃciency for their commercial viability.[11] The

U.S. Department of Energy estimated the cost of green hydrogen

produced by the PEC process to be US$5.7 kg−1in 2020 (with a

20% STH eﬃciency) and to lower to US$2.1 kg−1(with a 25%

STH eﬃciency) in the more distant future.[12] It has also been

suggested that the STH eﬃciency of 25% and the photoelectrode

lifetime of 10 years are required for the PEC systems to be eco-

nomically consistent with fossil fuel-based hydrogen production

processes.[13] Recently, monolithic systems and integrated PEC

modules under light concentration reached STH eﬃciencies as

high as 17.12%[14] and 19%.[15] However, the high capital and

operational costs signiﬁcantly hamper their economic viability.

To date, no cost-eﬀective and highly eﬃcient PEC systems have

been developed to satisfy all those key criteria despite signiﬁcant

progress in this ﬁeld. Thus, further studies are necessary to dis-

cover novel materials with unprecedented physicochemical and

optoelectronic properties that can simultaneously meet those key

criteria set for the design and application of the PEC system for

green hydrogen production.

As one of the two key reactions of overall water splitting,

the water oxidation reaction proceeds via the following gen-

eral steps (Figure 1b)[16]: i) water dissociation and formation of

surface-bonded OHads, ii) the further oxidation of OHads to Oads,

and iii) the formation of OOHads, which is a precursor for O2

evolution[17,18]:

H2O⇄OHads +H++e−(5)

OHads ⇄Oads +H++e−(6)

H2O+Oads ⇄OOHads +H++e−(7)

OOHads ⇄O2+H++e−(8)

Using ﬁrst-principles periodic density functional theory (DFT)

calculations, Man et al.[19] proposed that the binding energies of

reaction intermediates (e.g., HO*, O*, and HOO*) are responsi-

ble for the origin of OER activity over select oxide surfaces. When

the binding energy of oxygen is quite high, the OER overpoten-

tial is limited by the formation of HOO* species, and otherwise,

the formation of HO* species is dominant. Therefore, various

2305179 (2 of 47)

21983844, 2023, 33, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/advs.202305179 by Technische Universitaet Berlin, Wiley Online Library on [06/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

www.advancedsciencenews.com www.advancedscience.com

Figure 2. a) Crystal and band structures of TiO2and IPCE spectra of reduced TiOxnanotube arrays before and after Ar/H2treatment at diﬀerent

photocurrent-potential plots for a 2-μm-thick WO3electrode illuminated with AM 1.5G simulated sunlight: in 1 M aq. HClO4(curve A) and after the

structures of CuWO4and linear potential sweep curves of the doped CuWO4nanoﬂake photoanodes with diﬀerent Mo doping concentrations. Re-

Ag/Fe2O3,andCo-Pi/Ag/Fe

2O3photoelectrodes under chopped light illumination in 1 M NaOH (pH 13.6). Reproduced with permission.[47] Copyright

2016, WILEY-VCH Verlag GmbH & Co. KGaA. e) Crystal and band structures of ZnFe2O4and linear scanning J–V curves of ZnFe2O4photoanodes

Co. KGaA. f) Crystal and band structures of BiVO4and I-V characteristics of the optimized WO3-nanorods (green), WO3-nanorods/BiVO4(blue) and

strategies to stabilize HOO* species compared with HO* species

were developed.[20] The suitable photoanodes must satisfy the

expected conditions: ΔG(OHads)=CO=1.23 eV, ΔG(Oads)=2×

CO=2.46 eV, and ΔG(OOHads)=3×Co =3.69 eV.[18] The water

oxidation reaction requires the transfer of four electrons and four

protons, which needs high overpotential, with the simultaneous

formation of the O–O bond in comparison to the two-electron-

transfer water reduction reaction.[21] Therefore, water oxidation

is thermodynamically and kinetically more challenging than

water reduction. As shown in Figure 1b, a suﬃcient amount of

energy must be supplied in each step to drive the water oxidation

reaction. Thus, it leads to high energy barriers and slow kinetics

in addition to the uphill reaction thermodynamics.[22] A sluggish

water oxidation reaction is the rate-determining step that gov-

erns the reaction rate of water splitting.[23] The water oxidation

reaction is generally enhanced by modifying the photoanodes

with oxygen evolution cocatalysts that promote the eﬃcient

charge separation of photo-excited electron-hole pairs.[24]

According to the three basic stages of water splitting, the eﬃ-

ciency of the water oxidation reaction is mainly determined by the

light absorption, the separation eﬃciency of photo-excited charge

carriers, and the surface catalytic reaction[9]:

𝜂=𝜂absorption ×𝜂separation ×𝜂reaction (9)

which can be controlled by modulating the physicochemical, op-

toelectronic, and surface properties of n-type semiconductors.

Also, the valence band edge potentials of photocatalysts must be

more positively positioned than that of E0(O2/H2O) and the water

oxidation active sites must be suﬃcient on the photocatalyst sur-

face to hinder the recombination of photo-excited electrons and

holes.[25]

2. Oxide-Based Photocatalysts for PEC OER

Since the ﬁrst successful demonstration of solar-induced unas-

sisted water splitting over a TiO2photoanode by Honda and

Fujishima,[26] a large number of studies have been conducted

to maximize the STH eﬃciency of various heterogeneous oxide-

based semiconductors but few have shown relatively outstanding

water oxidation performance and stability. The overview of some

n-type oxide semiconductors for PEC OER is shown in Figure 2

and Table 1.[27]

TiO2is one of the most widely studied oxide semiconductors

because of its low cost, chemical stability, earth abundance,

non-corrosiveness, and non-toxicity. TiO2typically exists in

three crystalline structures: anatase (tetragonal, I41/amd), rutile

(tetragonal, P42/mnm), and brookite (orthorhombic, Pbca). In

addition to other factors, the STH eﬃciency of TiO2depends

2305179 (3 of 47)

www.advancedsciencenews.com www.advancedscience.com

Table 1. Overview of some n-type oxide semiconductors for PEC OER.

Compound Structure type and

symmetry

Band gap Theoretical STH eﬃciency

and photocurrent density

Performance/eﬃciency Limitations Reference

TiO2Tetragonal anatase,

I41/amd

3.2 eV 1.3%,

1.1 mA cm−2

75% IPCE fast recombination rate, rapid backward

reaction, and a large overpotential for

HER, and bandgap limits the STH

eﬃciency to ≈1%

[29–31]

WO3Monoclinic, P21/n2.6 eV 4.8%,

3.9 mA cm−2

75% IPCE,

2.4 mA cm−2at 1.23 V vs RHE

rapid recombination rate, slow charge

transfer at semiconductor/electrolyte,

a restricted light absorption up to

450 nm, instability at pH>5, and

limited theoretical STH eﬃciency

[31, 36]

CuWO4Triclinic distorted

wolframite, P

2.2 eV ≈13%,

10.7 mA cm−2

>20% IPCE,

≈0.15-0.16 mA cm−2,

and 0.62 mA cm−2at 1.23 V vs

RHE

low light absorption coeﬃcient and

high bulk charge transfer resistance

[37–42]

𝛼-Fe2O3Trigonal corundum, R

3c2.0-2.2 eV 16.8%,

12.6 mA cm−2

≈18% IPCE,

≈6mAcm

−2at 1.23 V vs RHE,

≈40% IPCE, and ≈0.55% STH

a low absorption coeﬃcient, short

excited-state lifetime, short

hole-diﬀusion length, low charge

carrier mobility, poor electric

conductivity, and poor OER kinetics

[44–47]

ZnFe2O4Cubic spinel, Fd

3m2.0 eV 20%,

≈11 mA cm−2

25% IPCE

1.0 mA cm−2at 1.23 V vs RHE

rapid surface and bulk recombination

rate and poor minority career

[49, 51]

BiVO4Monoclinic scheelite,

C2/c

2.4 eV 9.1%,

7.4 mA cm−2

6.72 mA cm−2at 1.23 V vs RHE,

≈90% IPCE, 8.1% STH, and

1.75% ABPE at 0.6 V vs RHE

high recombination rate, poor charge

transport properties, low carrier

collection eﬃciency, inadequate water

oxidation kinetics, and limited

theoretical STH eﬃciency

[31, 56, 58]

on its polymorphs. The photocatalytic overall water splitting

reaction can take place on rutile but hardly on anatase and

brookite and becomes feasible for anatase and brookite only

under prolonged UV light irradiation.[28] Recently, compact

anatase TiO2layers fabricated on the FTO by introducing a Ti

interlayer and suboxide TiO2nanotubes exhibited the highest

incident photon-to-current eﬃciency (IPCE) value of 75% at

300 nm[29] and 340 nm (Figure 2a),[30] respectively. Although

the anatase-TiO2polymorph has better electron mobility and

conductivity in comparison to the rutile-TiO2and brookite-TiO2

polymorphs, its theoretical STH eﬃciency and photocurrent

density can only reach the maximum of 1.3% and 1.1 mA cm−2,

respectively, because of wide optical bandgap energy of 3.2 eV.[31]

In addition to the low STH eﬃciency, a fast recombination

rate of photo-excited charge carriers, a rapid backward reaction

(recombination of H2and O2), and a large overpotential for the

HER also hamper the practical application of TiO2in solar water

splitting despite considerable progress.[32,33]

The monoclinic phase with space group P21/nis the most sta-

ble polymorph of WO3at room temperature and has a perovskite-

like structure without an A-site ion. Theoretical studies on the

correlation between the crystal structure and the band gap of

WO3revealed that the bandgap energy values of diﬀerent WO3

polymorphs decrease in the following order: monoclinic >or-

thorhombic >triclinic >tetragonal ≥cubic.[34] Also, the struc-

tural modiﬁcation by introducing the nitride ions into the lattice

of the monoclinic WO3phase led to the reduction of the bandgap

energy from 2.6 eV to 1.9 eV.[35] Highly transparent nanoporous

WO3ﬁlms fabricated by layer-by-layer deposition of a colloidal so-

lution of tungstic acid and annealing exhibited a maximum IPCE

of 75% and a photocurrent density of 2.4 mA cm−2at 1.23 V ver-

sus RHE under simulated solar irradiation (Figure 2b).[36] How-

ever, the rapid recombination of photo-excited charge carriers,

slow charge transfer at the WO3/electrolyte, restricted light ab-

sorption up to 450 nm, instability at pH >5, and a theoreti-

cal STH eﬃciency of 4.8% and a photocurrent density of 3.9

mA cm−2still limit its application.[31]

To circumvent the drawbacks of WO3, triclinic CuWO4with

a distorted wolframite crystal structure (space group P

1) oﬀers

a smaller bandgap energy and limits the formation of soluble

tungstates due to the strong covalency in the metal oxo bonds,

yielding higher photocurrent in neutral pH under visible light

irradiation.[37] The CuWO4photoanodes exhibited photocurrent

densities of ≈0.15–0.16 mA cm−2and 0.62 mA cm−2at the ther-

modynamic potential for water oxidation (1.23 V versus RHE) un-

der simulated solar irradiation (Figure 2c)[37–39] and an IPCE eﬃ-

ciency of >20%.[40] Despite its theoretical STH eﬃciency of ≈13%

and photocurrent density of 10.7 mA cm−2, the PEC performance

of CuWO4is signiﬁcantly hindered by its low light absorption co-

eﬃcient and high bulk charge transfer resistance.[41,42]

𝛼-Fe2O3with a trigonal structure (space group R

3c) has been

intensively explored as a promising photoanode material for PEC

water splitting due to its small bandgap energy, suitable valence

band edge potential to thermodynamically drive the water oxida-

tion reaction, excellent stability under alkaline conditions, earth

abundance, and environmentally friendliness.[43,44] With the

reported bandgap energy of 2.0-2.2 eV, 𝛼-Fe2O3has the potential

to reach a theoretical maximum STH eﬃciency of 16.8% and a

2305179 (4 of 47)

www.advancedsciencenews.com www.advancedscience.com

photocurrent density of 12.6 mA cm−2,[45] which manifestly

exceeds the STH benchmark eﬃciency of 10% required for

practical applications. The 𝛼-Fe2O3nanowires-based photo-

electrode, which was fabricated by a chemical bath deposition

method followed by hydrogen treatment and then loaded with

ultrathin TiO2overlayer and CoPi cocatalyst, exhibited a stable

photocurrent density of ≈6mAcm

−2at 1.23 V versus RHE

over 100 h under AM 1.5G simulated sunlight and an IPCE

value of ≈18% at 420 nm owing to its excellent light absorption

and bulk charge separation capacity, the enhanced electrical

conductivity, and the decreased surface recombination.[46] A

hematite nanosheets-based electrode modiﬁed by Ag and CoPi

nanoparticles showed an IPCE value of ≈40% at 420 nm and an

STH eﬃciency of ≈0.55% due to the improved light harvesting

and the facilitated charge transfer by Ag nanoparticles and the

reduction of surface/bulk recombination and the stabilization

of the photoelectrode surface by CoPi cocatalyst (Figure 2d).[47]

However, several limitations, such as low absorption coeﬃcient

due to an indirect band gap, short excited-state lifetime (≈10−12

s), short hole-diﬀusion length (2-4 nm), low charge carrier mo-

bility (≈10–2 to ≈10–1 cm2V−1s−1), poor electric conductivity, and

poor OER kinetics still hinder achieving the practical maximum

STH eﬃciency of 𝛼-Fe2O3.[44,48]

Cubic spinel ZnFe2O4(space group Fd

3m) also received much

interest due to its narrow bandgap energy (Eg=1.9 eV), high en-

ergy level conduction band minimum, outstanding photochem-

ical stability, low cost, and magnetic recyclability.[49,50] Interest-

ingly, ZnFe2O4with a relatively poor crystallinity but a higher

spinel inversion degree (due to cation disorder) shows a superior

eﬃciency in photo-excited charge carrier separation and an im-

proved majority charge carrier transport compared to ZnFe2O4

with higher crystallinity but a lower inversion degree.[51] The op-

timization of these factors and the addition of a nickel-iron cocat-

alyst overlayer resulted in a new benchmark photocurrent den-

sity of 1.0 mA cm−2at 1.23 V versus RHE, and an IPCE value

of about 25% was reached for ZnFe2O4nanorod photoanode

(Figure 2e).[51] At potentials between 0.8 and 1.3 V versus RHE,

ZnFe2O4can exhibit a considerably higher charge transfer eﬃ-

ciency due to a slower surface charge recombination rate.[52] De-

spite its maximum theoretical STH eﬃciency of about 20% and

a photocurrent density of ≈11 mA cm−2,[49,51] the eﬃciency of

ZnFe2O4is substantially limited by a rapid surface and bulk re-

combination rate of photo-excited charge carriers and poor mi-

nority career.

BiVO4crystallizes in three diﬀerent polymorphs: monoclinic

scheelite, tetragonal scheelite, and tetragonal zircon. The mon-

oclinic scheelite BiVO4(space group C2/c)isann-type semi-

conductor with a direct bandgap energy of 2.4 eV and a valence

band edge potential of ≈2.4 V versus RHE, which is suﬃciently

positive than E0(O2/H2O) =1.23 V versus RHE.[53,54] Therefore,

the monoclinic scheelite BiVO4shows the highest photocatalytic

water oxidation activity among the three polymorphs.[55] Apho-

tocurrent density of 6.72 mA cm−2at 1.23 V versus RHE and

an IPCE of ≈90% were achieved by the combination of BiVO4

with more conductive WO3nanorods in the form of core-shell

heterojunction (Figure 2f).[56] The high recombination rate of

photo-excited charge carriers was signiﬁcantly reduced by creat-

ing a thin absorber layer of BiVO4, which was thinner than the

carrier diﬀusion length. The tandem device constructed with a

GaAs/InGaAsP solar cell exhibited an STH eﬃciency of 8.1%.[56]

The charge separation eﬃciency of 97.1% and charge trans-

fer eﬃciency of 90.1% at 1.23 V versus RHE were obtained by

engineering the hierarchical nanoporosity of BiVO4,[57] and an

applied bias photon-to-current conversion eﬃciency (ABPE) of

1.75% was found at a potential as low as 0.6 V versus RHE for

nanoporous BiVO4photoanodes.[58] However, the water oxida-

tion eﬃciency of BiVO4is still limited by a high electron-hole

recombination rate, poor charge transport properties, a low car-

rier collection eﬃciency, inadequate water oxidation kinetics, and

the bandgap energy limiting an AM 1.5G solar photocurrent den-

sity to 7.4 mA cm−2and an STH eﬃciency to 9.1%.[31]

Despite their encouraging and relatively outstanding perfor-

mance achieved so far, TiO2,WO

3,CuWO

4,Fe

2O3,ZnFe

2O4,and

BiVO4alone are unlikely to satisfy the criteria set for ensuring the

practical application in solar to chemical energy conversion. This

paved the way for the development of novel materials, including

mixed-anion compounds, for PEC applications.

3. Crystal and Electronic Structures of Perovskite

BaTaO2N

Mixed-anion compounds, containing more than one anionic

species in a single phase, are an emerging class of advanced

materials with the potential to contribute to solar fuel produc-

tion in the future.[59] Unlike single-anion compounds, mixed-

anion compounds exhibit diverse structures, chemical and phys-

ical properties, and new functionalities because of diﬀerent an-

ionic characteristics, such as ionic radii, valency, electronegativ-

ity, and polarizability.[60] Also, the combination of hetero-anions

can realize the structures of compounds that cannot be gener-

ally stabilized by homo-anions. Especially, to develop photocat-

alytic materials that can eﬃciently function under visible light,

anions less electronegative than oxygen can be simultaneously

introduced. Having similar chemical, structural, and electronic

properties, oxygen and nitrogen substitute each other in the an-

ion site to form oxynitrides.[61] In oxynitrides, the nitride anions

(N3−) having atomic orbitals with potential energy higher than O

2p atomic orbitals of the oxide anions (O2−) shift the valence band

maximum upward without aﬀecting the conduction band mini-

mum, leading to the increased covalency of metal-anion bonds,

the improved absorption property, and the decreased optical band

gap (Figure 3a).[62,63] Also, most d0transition metal oxynitrides

can absorb photons with absorption band edges in the range of

500–760 nm and have theoretical STH conversion eﬃciencies in

the range of 8–32% and suitable conduction and valence band

edge potentials straddling the proton reduction and water oxida-

tion reaction potentials, respectively, and can drive the overall wa-

ter splitting reaction (Figure 3b).[64]

As a typical representative of the AB(O,N)3perovskites,

BaTaO2N is regarded as one of the promising photocatalysts

for solar water splitting due to its absorption of visible light up

to 660 nm, small bandgap energy (Eg=1.9 eV), good stability

under light irradiation in concentrated alkaline solutions, and

nontoxicity.[65] Moreover, the conduction band minimum and

valence band maximum of BaTaO2N are located at –0.4 V and

1.5 V versus NHE at pH 0, respectively, which should theoret-

ically drive the water-splitting reaction in the absence of an ex-

ternal bias.[66] BaTaO2N can generate a photocurrent density of

2305179 (5 of 47)

Loading more pages...