Sustainable

Energy & Fuels

View Article Online
PERSPECTIVE View Journal | View Issue
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

Photocatalytic water splitting: advantages and

Cite this: Sustainable Energy Fuels,
challenges
2021, 5, 4560
Katherine Villa, *a José Ramón Galán-Mascarós, ab
Núria López a

and Emilio Palomares ab

Over the last decades, an increasing interest has been devoted to the generation of hydrogen from solar-
powered water splitting devices as a direct approach to generate a clean and sustainable energy vector.
Among the different alternatives, a direct photocatalytic process is the most straightforward, where
a light absorber is able to produce hydrogen and oxygen under light irradiation without an external bias.
Herein, we discuss the main pros and cons of photocatalytic water splitting for hydrogen production,
aiming to offer a clear vision of the critical challenges that need to be overcome for these reactions to
become a major player in the generation of solar fuels. The main parameters limiting solar-to-hydrogen
yields and efficiencies, and the requirements for the descriptors of materials and components are
Received 27th May 2021
Accepted 12th August 2021
introduced. We also describe the most promising semiconductors and material composites, highlighting
their main features, along with future trends to improve photocatalytic stability, efficiency, and kinetics.
DOI: 10.1039/d1se00808k
Thus, our critical perspective grows from the state-of-the-art to focus on the future and viability of
rsc.li/sustainable-energy photocatalytic solar hydrogen.

and solid solutions based on the combination of GaN and

1. Introduction ZnO.9–13 Over the last years, the deposition of protective layers,
The realization of sustainable energy generation from renew- such as Al2O3 and TiO2, onto unstable semiconductors has led
able sources, i.e., water and solar light, is a major challenge to to a renewed interest in short-band gap photocatalysts by
global sustainability goals. Hydrogen, as a clean energy vector increasing their stability over longer reaction times.14,15
with high mass-energy density, is the strongest candidate for Since photocatalytic materials act as photon absorbers and
the development of solar to chemical fuel products. Among the are the main mediators of the water-splitting reaction (WSR), we
emerging technologies that may enable this conversion, pho- can identify herein eight main challenges associated with the
tocatalysis provides great advantages, including simple setup,
mild reaction conditions, and potential scalability.1,2
An approximate solar to hydrogen (STH) conversion effi-
ciency close to 10% is a commonly accepted target for the
process to become economically viable.3 Yet, current photo-
catalytic efficiencies of water splitting based on suspended
particles are still far from such industrial requirements.4 It
should be noted that competitive STH efficiencies of about 20%
have been achieved by monolithic photoelectrochemical
systems using tandem designs.5–7 A more detailed explanation
about these systems and current limitations regarding stability
can be found in ref. 8.
Several efforts have been devoted to the development of new
photocatalytic materials and/or their combination with suitable
redox co-catalysts. Some of the most representative examples
include modied NaTaO3, SrTiO3, Ga2O3, C3N4 photocatalysts,

Fig. 1 Main challenges associated with the use of photocatalytic

a
Institute of Chemical Research of Catalonia (ICIQ), The Barcelona Institute of Science materials for water splitting: (1) redox potential; (2) wavelength light
and Technology (BIST), Av. Paı̈sos Catalans, 16, Tarragona E-43007, Spain. E-mail: absorption; (3) solar to hydrogen efficiency; (4) (photo-)chemical
kvilla@iciq.es stability; (5) toxicity; (6) cost/scarcity; (7) material scalability; and (8)
b
Institució Catalana de Recerca i Estudis Avançats (ICREA), Passeig de Lluı́s Companys fabrication versatility.
23, 08010 Barcelona, Spain

4560 | Sustainable Energy Fuels, 2021, 5, 4560–4569 This journal is © The Royal Society of Chemistry 2021
 View Article Online

Perspective Sustainable Energy & Fuels

type of photocatalysts based on their bandgap values, which are (8) Fabrication versatility: the possibility of shaping a pho-
represented in Fig. 1 and described as follows: tocatalyst with the desired morphology is a very attractive
(1) Redox potential: an ideal photocatalyst should have feature for enhancing light harvesting properties, tuning their
a suitable Conduction Band (CB) for reducing the photo- surface area, and favoring charge transfer processes. Although
generated protons to H2 (0.000 V vs. NHE at pH 0) and a Valence this aspect is not certainly a challenge, it plays a role in
Band (VB) more positive than the potential required to oxidize enhancing the overall water splitting efficiencies. Therefore,
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

water to O2 (1.23 V vs. NHE at pH 0). Only a few wide- and semiconductors that offer versatility in terms of fabrication
intermediate bandgap semiconductors satisfy both thermody- enable the design of tailored-made photocatalytic systems for
namic requirements, which makes this parameter a bottleneck improving photon absorption, charge transfer mechanisms,
of WSR. adsorption of reactants, etc.
(2) Wavelength light absorption: solar light energy is mainly In this perspective, we discuss the capabilities of short,
composed of visible and NIR wavelengths, where the UV region intermediate, and wide bandgap semiconductors for decom-
only corresponds to 5%. Unfortunately, most of the visible-light posing water under light irradiation with a focus on their main
responsive semiconductors are unstable, leading to their deac- limitations. Moreover, we describe the advantages and current
tivation over time. In the case of NIR-activated photocatalytic challenges needed to overcome for the full realization of solar
systems, the examples are very limited, and the enhancement fuel generation by photocatalytic approaches.
provided respect to the reaction yields is not as satisfactory as
expected. 2. Semiconductors for water splitting
(3) Solar to hydrogen efficiency: current photocatalytic STH
efficiencies are approaching 1%,16 which is one order of The photocatalytic generation of solar fuels by suitable photo-
magnitude below the target efficiency required to compete with catalytic materials suspended in water is certainly one of the
the well-established energy economy based on fossil fuels. most attractive and challenging pathways for decarbonizing
Therefore, the development of efficient photocatalytic systems current energy systems. From the outside, it seems as a simple
is of utmost importance to minimize the costs associated with approach combining water, solar light, and a semiconductor.
the direct hydrogen generation from solar light and water However, each of these components plays a key role in the
without electrical inputs. overall efficiency of the reaction, and mostly, many factors
(4) (Photo-)chemical stability: another important require- including thermodynamic, kinetic, and other requirements
ment for practical applications is the stability of the semi- hold back this technology from its maximum potential. For
conductors against oxidative/reductive pathways that instance, photocatalytic materials, which usually consist of
simultaneously take place under light irradiation.4 Ideally, metal oxide semiconductors, should efficiently absorb visible
photocatalysts should exhibit a lifetime of at least 10 years.4 light irradiation up to a l < 520 nm to reach about 10% STH
Therefore, their stability over multiple reusability tests should efficiency.4,18 Moreover, as described before, such semi-
be assured. Moreover, attention should be paid to the structural conductors should fulll the thermodynamic requirements for
and morphological changes that may take place upon reaction, driving the decomposition of water into H2 and O2. Some
which further affect their photocatalytic performance.17 research directions involve the addition of organic compounds
(5) Toxicity: to avoid environmental issues, photocatalysts that act as sacricial agents, which results in higher H2 gener-
should also exhibit null toxicity and a high degree of biocom- ation yields than pure water. However, this approach is not
patibility. However, this property is oen underestimated or considered as a WSR since the electrons are originated solely
even not assessed at all in most reports. from organic molecules, and thus the organic molecule is
(6) Cost/scarcity: common photocatalytic heterostructures consumed during the process, that is, not sustainable.19
involve the use of noble metals, such as Pt or/and RuO2 as co- The photocatalytic reaction is initiated by the activation of
catalysts. Despite their high efficiency for enhancing electron– a semiconductor with light wavelengths higher or equal than
hole pair separation, these materials are not cost-effective for the bandgap energy. Upon illumination, the electrons present
large-scale applications and should be replaced by earth- in the VB are promoted to the CB, leaving positive holes in the
abundant counterparts. former. Then, the photogenerated electron–hole pairs can
(7) Material scalability: the development of efficient light- either perform a redox reaction or recombine in the bulk.
harvesting photocatalytic systems usually entails the combina- Depending on the type of bandgap, i.e., direct or indirect, such
tion of costly fabrication techniques and multi-step synthesis electron–hole transitions occur differently. Therefore, their
approaches that are either not scalable or add an important bandgap and band structures strongly inuence the resulting
scalability limitation. On the other hand, the conguration of photocatalytic efficiencies. As shown in Fig. 2A, in a direct
the photocatalyst also affects their recovery and reusability, bandgap semiconductor, the bottom of the conduction band
which inuences the overall process scalability. For instance, lies directly above the top of the valence band in a reciprocal k-
the use of powders in large liquid volumes entails several issues space.13 Therefore, the excitation of the photocatalyst with light
associated with their dispersion and recovery for further reuse. energies $ bandgap results in a rapid promotion of electrons
Therefore, major efforts should be devoted to the fabrication of from the VB to the CB. On the contrary, when the bottom of the
scalable supported photocatalysts that can facilitate their real- CB and the top of the VB of a semiconductor are not aligned, it
world application under several photocatalytic cycles. is considered an indirect bandgap semiconductor.20 In this

This journal is © The Royal Society of Chemistry 2021 Sustainable Energy Fuels, 2021, 5, 4560–4569 | 4561
 View Article Online

Sustainable Energy & Fuels Perspective

band gaps, electronic levels are then ltered to particular

properties. In several cases, new materials have been proposed
as adsorbents although there are some limitations to these
procedures: (i) most of the computed properties are bulk ones,
with little attention to surfaces or defects (quite common in
these materials);29 (ii) the particularities of interfaces with
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

water, for instance, are typically not considered.30

The most representative examples of wide, intermediate, and
short bandgap semiconductors with diverse band structure
properties are discussed below, correlating their main proper-
ties with the main challenges described in the introduction.

2.1. Wide-band gap semiconductors $ 3 eV

Wide band-gap semiconductors based on metal oxides offer the
advantage of a high (photo-)chemical stability, which is a crucial
condition for long-term reactions and reusability in multiple
reaction cycles. However, large bandgap values limit their light
absorption to the UV wavelength range (100–400 nm) that
Fig. 2 Band structure of semiconductors. (A) Schematic illustration of corresponds only to 5% of the solar spectrum. The most widely
the direct and indirect bandgap semiconductors and (B) schematic studied UV-semiconductors for water splitting are described
illustration of n-type and p-type semiconductors. below.
Titanium dioxide. TiO2 is a reference material for diverse
applications in the photocatalytic eld, ranging from water
case, the excitation involves not only photon absorption but also cleaning, antifogging, to energy applications. Since the pio-
phonon absorption or emission.21 Thus, the electron transitions neering work by Fujishima and Honda on the photo-
occur at a lower probability and slower rates than direct photon electrochemical water splitting involving photoelectrodes of
absorption.4 As a result, the photogenerated charge pair carriers TiO2/Pt in 1972,31 a large number of studies dealing with the use
in indirect bandgap semiconductors exhibit a longer lifetime, of TiO2 for photocatalytic hydrogen generation from water have
which is benecial for the overall photocatalytic performance by been reported, including extensive modeling.32 Despite the
decreasing electron–hole recombination rates.22 However, these attractive properties of TiO2, i.e. non-toxicity, corrosion resis-
types of semiconductors are not as good light absorbers as tance, and low cost, its photocatalytic efficiency for water
direct bandgap semiconductors, requiring larger thickness to splitting is still low. In fact, the maximum theoretical STH
efficiently absorb the incident light. A factor that not only efficiencies under UV light irradiation for the anatase and rutile
affects the costs associated with material synthesis but also the crystalline phases are 1.3 and 2.2%, respectively.33 Fig. 3 shows
resulting carrier extraction efficiencies.23 some strategies to improve the efficiency of wide-bandgap
Moreover, semiconductors are classied into intrinsic or photocatalysts, including semiconductor doping to extend its
extrinsic. Intrinsic semiconductors are in a pure form while light absorption to the visible (Fig. 3A),4,34 However, many
extrinsic semiconductors are materials containing impurities doping strategies oen end up in localized (trapped) states that
that can be intentionally added or intrinsically present.24 Based enhance charge pair recombination, due to the created inter-
on the type of impurity, extrinsic semiconductors can be clas- mediate states, or introduce stability issues.35 Other approaches
sied as n- or p-type semiconductors (Fig. 2B). The introduction also involve size and shape optimization through dimension-
of impurities into pure semiconductors, also known as doping, ality for improving light harvesting and charge mobility
leads to the creation of additional electronic levels. The donor (Fig. 3B),36 and couplings to conjugated polymers, and/or
impurities lead to the shiing of the Fermi level to higher formation of dual semiconductor heterostructures to mini-
energy close to the CB, resulting in a negative, n-type, semi- mize electron–hole recombination.
conductor with increased electron density. On the contrary, the Zinc oxide. ZnO possesses similar redox potential and
addition of acceptor impurities to intrinsic semiconductors bandgap to TiO2, including biocompatibility. An interesting
results in a positive, p-type, semiconductor, in which the Fermi feature of this material is its versatility in terms of fabrication
level is shied to lower energy and the higher proportion of with diverse shapes and lower cost.37 Despite its large bandgap,
carriers are holes.24 the presence of oxygen vacancies in ZnO can lead to visible-light
Over the last years, extensive materials surveys with response, which is desirable for solar-driven applications.
computed properties obtained from atomistic simulations have Nevertheless, ZnO is pH-sensitive and usually, it exhibits
been employed to identify new families of materials.25–27 Data- stability problems under light irradiation, which is a factor that
bases are then generated as described in ref. 28. Typically in should not be underestimated.38 A study comparing the pho-
these approaches the electronic structure calculations for bulk tocatalytic performance of TiO2 and ZnO for water splitting by
systems are presented at a high accuracy level and the reported simulated sunlight irradiation showed that TiO2 nanostructures

4562 | Sustainable Energy Fuels, 2021, 5, 4560–4569 This journal is © The Royal Society of Chemistry 2021
 View Article Online

Perspective Sustainable Energy & Fuels

Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

Fig. 3Strategies to improve the applicability of wide bandgap semiconductors. (A) Schematic illustration of semiconductor doping. Reproduced
with permission.4 Copyright 2020, American Chemical Society. (B) Effects of dimensionality of photocatalyst properties. Reproduced with
permission.36 Copyright 2012, Elsevier. (C) Solid solution engineering. Reproduced with permission.48 Copyright 2018, Wiley-VCH GmbH.

have superior performance in comparison to ZnO.37 Moreover, also the conduction band potentials to favor the reduction
recent modeling investigations pointed out the competition pathway.48
between OER and H2O2 synthesis by this material.39 Strontium titanate. SrTiO3 is a material that has many
Zinc sulde. ZnS has a bandgap of about 3.7 eV,40 entailing advantages over the semiconductors described before. It not
the same disadvantages regarding UV-light activation as dis- only exhibits good chemical and photochemical stability but
cussed for the previous materials. Extensively reviewed also has a higher conduction band potential, which favors
computational and experimentally ZnS,41 which has a negative hydrogen formation. Moreover, this material exhibits an ABO3-
conduction band potential ( 1.1 eV, NHE), is a very attractive type perovskite structure, offering a versatile and exible
material for H2 generation, even in the absence of a co-cata- approach to modify its structure and composition by replacing
lyst.42 According to theoretical studies, ZnS also presents the Sr and Ti cations or O anions.49,50 SrTiO3 presents two
a higher photo-carrier generation, resulting from its direct bandgaps; indirect (3.2 eV) and direct (3.75 eV), thus, its pho-
bandgap, in comparison with other materials such as TiO2.43 toactivation is limited to UV light irradiation. Domen et al.
Moreover, its photocatalytic efficiency has been remarkably achieved a maximum quantum efficiency of 30% at 360 nm and
improved in the presence of some co-catalysts, e.g., Au, Pt, CuS, 56% at 365 nm for a modied SrTiO3:Al sample.10 Owing to its
etc.43,44 Although, sulde-based catalysts suffer self-photo- great potential, the performance of this material was also
corrosion, different approaches involving the addition of examined at a larger scale, reaching a 0.4% STH efficiency,
sacricial agents, coupling with other semiconductors, and which is slightly lower than the one obtained at the lab scale.51
formation of solid solutions (Fig. 3C) may overcome such DFT simulations have shown potential doping strategies for this
a stability issue.45–47 This strategy is considered as one of the material.52 A common issue associated with the doping of wide-
more efficient strategies to not only tune the bandgap values but bandgap photocatalysts is that it affects negatively their

This journal is © The Royal Society of Chemistry 2021 Sustainable Energy Fuels, 2021, 5, 4560–4569 | 4563
 View Article Online

Sustainable Energy & Fuels Perspective

quantum efficiencies in the UV region, which is usually where

the highest efficiencies are reached.53

2.2. Intermediate-band gap semiconductors 2.2 $ 3.0 eV

Semiconductors with an intermediate band-gap exhibit visible-
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

light absorption. Thus, they are ideal materials for performing

water splitting under solar light. However, only a few of them
have a favorable redox potential for carrying out the whole WSR.
Here, we will describe classical and new photocatalytic mate-
rials that have been explored over the past years.
Cadmium sulde. CdS presents a bandgap of 2.4 eV and has
been widely investigated for hydrogen generation from water
and organic pollutants.47,54 This photocatalyst can be prepared
in multiple shapes or combined with others, which is benecial
for improving its photoactivity.55 Modeling studies comparing
the performance of a one-dimensional CdS nanotube and its
bulk counterpart showed that dimensionality has important
effects on the resulting photocatalytic activities and charge
transfer processes.55 A facetted CdS nanotube showed an
optimal morphology and a stabilized valence band edge,
decreasing the electron–hole recombination rates. Likewise,
crystalline phases also inuence the H2 generation rates. CdS
exhibit two crystalline phases, i.e., cubic and hexagonal. The Fig. 4 Intermediate band gap semiconductors. (A) Release of Cd ions
latter is considered the most photoactive crystal phase for H2 by the CdS and modified-CdS over different consecutive tests under
visible light irradiation. Reproduced with permission.14 Copyright 2017,
evolution, due to the presence of (0001) crystal facets.56,57 The
Elsevier. (B) Band structures of tantalum (oxy-)nitrides. Reproduced
maximum STH efficiency obtained so far for a CdS-modied with permission.77 Copyright 2014, the Royal Society of Chemistry.
material under simulated solar light is 1.63%.58 As a result of
the photogenerated holes during the light irradiation, CdS
suffers self-photocorrosion.59 Moreover, it has been found that chemical stability and visible-light response.64 All these features
the presence of O2 in the liquid media also contributes to its are very attractive for the realization of large-scale photo-
corrosion by reacting with the photogenerated electrons to catalytic reactions. Since the pioneering work by Wang et al. in
produce superoxide radicals that accelerate the process.60 Main 2009 that reported a conjugated polymer semiconductor (g-
strategies to avoid the photocorrosion of CdS involve removal of C3N4) for this application,65 different strategies have been
oxygen from the media,61 protection layer to capture the holes,14 investigated to overcome the low quantum efficiencies, e.g.,
coupling with other semiconductors,60 and addition of sacri- doping, heterostructuring,66 and structure modication.67
cial agents (Na2S/Na2SO3). For instance, by combining the rst Extensive simulations on the impact of the structure on the
two strategies, in which the surface of CdS was modied with an electronic structure have been reported.68 This photocatalyst
inert Al2O3 layer, the photostability of CdS was notoriously has suitable valence and conduction band positions of 1.1 eV
enhanced, as illustrated in Fig. 4A. The leaking of Cd2+ ions was and 1.6 eV vs. NHE, respectively.69 Computational studies
hampered in the rst two cycles in comparison to the unmod- seemed to suggest that carbon nitride structures could split
ied CdS.14 Simulations have analysed in detail the electronic water according to thermodynamic criteria. However, these
structure of these materials.62 Overall, the coating of unstable results are very demanding in terms of accuracy and thus the
semiconductors with protective metal oxide layers and co- choice of the particular functional could affect the robustness of
catalysts have led to a remarkable stability progress of visible- the predictions. Indeed, experimental studies have shown
light responsive semiconductors.7,63 For instance, the deposi- kinetic limitations associated with the electron–hole recombi-
tion of a thin TiO2 layer onto a silicon photoanode followed by nation that decreases the number of holes available for water
an additional iridium layer led to a remarkable improvement of oxidation.69
the performance of Si that was maintained for hours in Silicon carbide. Among the families of metal-free semi-
comparison to the bare one.15 conductors, SiC also offers interesting properties for water
Carbon nitride. The quest for suitable photocatalysts that splitting and again the topology of the nanostructure affects the
efficiently reduce water to hydrogen has promoted the devel- position of the active photocatalytic bands.70 It has proper redox
opment of alternative materials to the classical inorganic band potentials for water splitting and exhibits high chemical
semiconductors. Among them, graphitic carbon nitride is stability and visible-light absorption (bandgap 2.3–3.3 eV).71
a promising n-type semiconductor for solar fuel generation. It is However, the H2 evolution rates have been reported to decay
biocompatible, easy to fabricate from organic precursors, and over time for SiC-based photocatalytic systems.72,73 Moreover,
consists of a completely metal-free structure with good

4564 | Sustainable Energy Fuels, 2021, 5, 4560–4569 This journal is © The Royal Society of Chemistry 2021
 View Article Online

Perspective Sustainable Energy & Fuels

this material usually presents a low surface area that limits its evolution.33 These materials can be synthesized in two-
wide applicability in catalysis,74 leading also to low H2 genera- dimensional (2D) structures (e.g., thin sheets), providing
tion rates. Therefore, different synthetic methods aiming at a versatile way to modify their band gaps to meet the thermo-
increasing its surface area have been developed in the last years, dynamic requirements for water oxidation and reduction.57 In
including sol–gel, carbon-based template, and polycarbosilane comparison to metal oxides, these disuldes show higher
pyrolysis.74 For instance, Hao et al. developed modied SiC photosensitivity and longer lifetimes.90 Fig. 5A illustrates the
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

nanowires by carbothermal reduction.71 The as-synthesized importance of the structure of metal suldes in their photo-
material showed a maintained H2 generation over 30 h of catalytic activity as co-catalysts. The bulk form of MoS2
visible light irradiation, demonstrating the importance of composed of compact layers does not contain active sites for
morphology and surface optimization towards stable water splitting, while the single-layer exhibits active edge and
photocatalysts. surface sites that favor H2 generation. When combined with
Tantalum (oxy-)nitrides. In the early 2000s, new TaON and CdS, it results in boosted photocatalytic yields.91 A novel
Ta3N5 materials were introduced as efficient visible-light approach involves the combination of 2D in-plane hetero-
responsive photocatalysts for H2 and O2 evolution.75 The elec- structures, containing WS2/MoS2.92 As shown in Fig. 5B, such
tronic structure of different compounds in this family has been WS2/MoS2 heterostructures were easily prepared by a combina-
recently reviewed.76 Fig. 4B shows the band structures of TaON tion of hydrothermal and ultrasonication methods. These 2D
and Ta3N5 with bandgaps of 2.4 and 2.1 eV, respectively.77 These photocatalytic systems showed an improved charge pair sepa-
emerging photocatalysts show similar properties to classical ration, resulting in remarkably high H2 generation rates. Thus,
metal oxide-based photocatalysts, such as corrosion resis- this work opens the door for alternative material congurations
tance,78 but with the advantage of a smaller bandgap and more based on pure 2D in-plane heterostructures.
negative valence band potentials. This is attributed to their
hybridized valence band that is composed of N(2p) and O(2p)
orbitals and the conduction band of empty d orbitals of the
3. Advantages vs. challenges
respective metal ions. For instance, in the case of TaON, this It has been almost 50 years since the rst study dealing with
material presents conduction and valence band levels of 0.3 light-assisted water splitting.31 Along the way, a myriad of works
and 2.2. V vs. NHE at pH 0, respectively. However, overall one- aiming at improving photocatalytic efficiencies has been re-
step water splitting by oxynitrides (d0-type) has not been ach- ported. Most common strategies include doping of the photo-
ieved yet, due to the fast charge recombination.79 Thus, for catalysts, optimization of their morphology to maximize light
achieving higher charge separation, these materials are usually harvesting, coupling to co-catalysts to enhance redox reactions,
coupled with other materials or included in a Z-scheme the combination of dual semiconductors to improve electron–
system.53,80 Another approach to improve the reduction capa- hole separation, introducing protective layers for stability, and
bilities of oxynitrides includes surface modication. As development of new materials and reactors to increase the
example, the decoration of TaON with ZrO2 nanoparticles photocatalytic yields.26,93,94 Although maximum STH efficiencies
results in the suppression of surface defects, due to the of purely photocatalytic systems are still around 1%, there is
formation of Ta–O–Zr bonds.81 still a considerable amount of published works on this topic, as
illustrated in Fig. 6A. Such continuous interest is due to the

2.3. Short-band gap semiconductors 1.3 $ 2.2

Cuprous oxide. Cu2O is a p-type photocatalyst with a band
gap between 2.0–2.2 eV.82 This material is non-toxic and fullls
the requirements for the whole photocatalytic WSR, involving
hydrogen generation and water oxidation, under visible light
irradiation.83 More importantly, theoretically it could be
possible to achieve an overall 18% STH conversion efficiency
(AM 1.5 spectrum) with this material.84 Unfortunately, the
photocatalyst is unstable as it undergoes oxidation to CuO and/
or reduction to metallic Cu under light irradiation,85 limiting its
performance in part as the surface modication under these
conditions reduces the band gap further.86 Strategies to mitigate
this problem include tandem systems with other photo-
catalysts,87,88 decoration with a protective layer to suppress the
self-photodegradation,89 or facet and morphology
engineering.85
Fig. 5 2D Transition metal disulfides. (A) Structure of bulk and a single
Transition metal disuldes. From this category, WS2 and
layer of MoS2 and its combination with a photocatalyst. Adapted with
MoS2 with indirect band gaps between 1.3–1.8 eV are attractive permission.91 Copyright 2017, The Royal Society of Chemistry. (B)
semiconductors for replacing highly cost precious metals, due Fabrication steps of WS2/MoS2 heterostructures. Reproduced with
to their unique conduction properties that favor hydrogen permission.92 Copyright 2021, Elsevier.

This journal is © The Royal Society of Chemistry 2021 Sustainable Energy Fuels, 2021, 5, 4560–4569 | 4565
 View Article Online

Sustainable Energy & Fuels Perspective

Solar to chemical energy. This technology enables the

conversion of solar energy into storable chemical molecules
that can be easily transported and used as needed. Such a way of
storage provides a promising alternative to counter the inter-
mittent character of sunlight, contributing to the development
of green technologies.
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

3.2. Disadvantages
Product separation. The photocatalytic generation of oxygen
and hydrogen from water occurs simultaneously in the same
reactor, which complicates the separation of the products and
may lead to the undesired recombination of the O2 and H2
molecules to generate water over again. This inuences the
overall stoichiometry of the WSR with the adverse effect of
increasing the pressure in the reactor and decreasing the pho-
toactivity of the materials.4 As a potential solution to counter the
drawback reaction, the coupling of two semiconductors able to
perform reduction and oxidation reaction separately, known as
Z-scheme, has been also studied. This conguration consists of
two compartments, containing each photocatalyst separated by
a membrane and interconnected by a redox couple. Thus, the
protons and electrons generated resulting from water oxidation
are transferred to the second compartment that eventually leads
Fig. 6 Impact of photocatalytic water splitting. (A) Bar chart of the to hydrogen generation by reductive pathways. However, the Z-
number of articles published per year from 2000–2020 on photo- scheme approach involves a two-photon absorption and the
catalytic water splitting. The data were obtained by searching the addition of extra compounds that affect negatively the
keywords ‘‘photocatalysis + water splitting’’ in Scopus as TOPIC. (B)
economic viability of photocatalytic water splitting. Other
Schematic illustration showing the main advantages and challenges of
photocatalytic water splitting. photocatalytic membrane reactors based on composite polymer
membranes or panel reactors coupled to ceramic membranes
that consist of a single-side illumination with no need for
cost-effective advantages that entail the generation of fuels from additional use of sacricial agents are quite promising for direct
widely available renewable sources; water and solar energy. photocatalytic water splitting.4,96 Further research on this
However, the feasibility of this technology is still limited by direction is encouraged to improve the overall performance.
important constraints regarding result reproducibility, opera- Low quantum efficiencies. One of the most limiting steps of
tion factors, and process scalability (Fig. 6B). photocatalytic reactions is the fast recombination rates.
Unfortunately, the required time of the photogenerated elec-
tron–hole pairs to react (ms) is usually longer than the
3.1. Advantages recombination rates (usually ns). Therefore, the latter domi-
Simple set-up. It only requires water, a suitable photo- nates the overall photocatalytic performances of the materials,
catalyst, and a light source without the need for external circuits resulting in low quantum efficiencies.97 Common strategies to
or electrolytes.1 The photocatalyst is usually in a powder form extend the lifetime of the photogenerated charge carriers
that can be synthesized by easy scalable wet chemical methods, include the addition of co-catalysts that promote H2 and O2
i.e., hydrothermal, sol–gel, precipitation, among others. More- evolution, which is reinforced by their spatial separation onto
over, the reaction proceeds at room temperature and pressure the photocatalytic structures.4 Other approaches involve
conditions, which is benecial in comparison to other energy- a decrease of the particle size of the photocatalysts or the
intensive technologies. development of single-layer materials, which shorten the
Particulate photocatalysts with suitable redox potentials. migration trajectory of the electron–hole pairs from the bulk to
The availability of single photocatalysts that are able to carry out the surface.98 In this sense, computational studies are key to
the whole WSR, such as CdS (bandgap 2.4 eV), is very attractive nd suitable photocatalytic systems without involving tedious
for minimizing the complexity of the entire system. experiments. For instance, a screening of over 50 000 inorganic
Clean solar fuel generation. The photocatalytic reaction does compounds yielded 36 types of stable two-dimensional (2D)
not involve the use of fuel-based energy sources nor involves the materials with proper band edge positions and their possible
generation of toxic by-products. Therefore, this emerging tech- combinations for efficient photocatalytic performance, such as
nology stands as an environmentally friendly approach MoS2/MoSe2, WS2/WSe2, MoSe2/WSe2, and MoS2/WS2.99.
compared with other reactions, e.g., thermocatalysis and high- However, these studies are limited by: (i) the errors in the esti-
temperature electrolysis.95 mation of the band gaps typical of most common GGA

4566 | Sustainable Energy Fuels, 2021, 5, 4560–4569 This journal is © The Royal Society of Chemistry 2021
 View Article Online

Perspective Sustainable Energy & Fuels

approaches, (ii) the system idealization, for instance the pres- reaction mechanisms are not fully well-understood. As a result,
ence of vacancies and impurities are neglected and (iii) the lack research on this topic is still limited to lab-scale experiments,
of detailed understanding and characterization of interfaces with few scaling-up initiatives. Ideal photocatalysts should
when more than one material is considered. absorb visible-light irradiation, exhibit high (photo-)chemical
Reproducibility. An increasing concern about photocatalytic stability, show minimal electron–hole pair recombination and
reactions is the reproducibility of the reported rates by different present suitable redox potentials for decomposing water,
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

research groups, especially if the photocatalysts are modied without representing a risk for the environment. Despite
with co-catalysts. There are a lot of variables that may inuence multiple efforts in this direction, efficient photocatalytic
such inconsistencies, ranging from photocatalyst synthesis systems that fulll these requirements have not been achieved
methods, reagent brands, inherent impurities, to reaction yet. Towards major progress in this eld, it is thus necessary to
conditions.97 Many works tend to normalize the photocatalytic link the resulting performances with the photocatalytic prop-
activities based on the product rates per total mass of the erties through the integration of experimental and theoretical
material. This approach does not involve other factors such as studies and the development of robust protocols for the
light reection and scattering that are equally important to the appropriate benchmarking of testing protocols to facilitate
whole photocatalytic system. Thus, the catalyst-mass- solid reproducibility through open data strategies. From the
normalization is not as accurate as apparent quantum yields simulations point of view, access to more accurate methods to
(AQY), which are independent of the reactor and light source.18 estimate band gaps and level alignment, addressing the mate-
AQY estimation takes into account the amount of products rial gap by considering impurities and interfaces, obtaining
formed per total rate of incident photons and is considered parameters to understand long-term stability, and the access to
a standard measure of the photoactivity of a material. Since excited state dynamic simulations that could provide kinetic
apparent quantum yields are usually measured using only parameters, are key developments required to fully understand
monochromatic light, a recent study proposed an alternative the photocatalytic performance. This will provide fundamental
method to determine AQY under simulated solar light, which knowledge about mechanistic insights that will promote further
provides a fair comparison for heterostructured photocatalytic developments in photocatalyst design and improve the lters
systems.100 Therefore, certication, open raw data, and digita- employed in the theoretical simulations to investigate new
lization could be mitigating strategies to improve the repro- families of materials. Likewise, immobilized photocatalysts
ducibility of the whole synthetic, characterization, and testing should become a priority over powder systems to facilitate the
protocols. transition to large-scale testing.
Scalability of semiconductor fabrication. Major efforts are
undergone to develop novel and efficient photocatalytic systems
needed to meet the STH efficiency requirements. To direct the Conflicts of interest
photocatalytic water splitting towards real-world applications, it There are no conicts to declare.
is still necessary to achieve major milestones on the scalability
of the photocatalytic materials, keeping in mind the ease of
large-scale fabrication and recovery. The most simple way of Acknowledgements
using photocatalytic materials is as a powder. However, the
agitation and recovery of powder suspensions in large volumes EP thanks ICINN for the project PID2019-109389RB-I00, NL
of water is challenging, especially when the particle size of the thanks RTI2018-101394-B-I00, and the AGAUR for the project
materials is in a nanometer range. Thus, immobilized photo- SGR-2017SGR00978. ICIQ and ICREA are also acknowledged.
catalysts are usually more desirable for large-scale applications.
To overcome mass-transfer limitations on immobilized mate- References
rials, an inner-motile photocatalyst lm has been recently re-
ported.101 It involves the combination of photocatalyst with 1 X. Yang and D. Wang, ACS Appl. Energy Mater., 2018, 1,
magnetic response in a cilia-based morphology that enhances 6657–6693.
light harvesting. Upon external actuation, the lm improves the 2 S. Lin, H. Huang, T. Ma and Y. Zhang, Adv. Sci., 2021, 8,
ow and mixing at the microscale, enhancing mass transfer and 2002458.
desorption of products. Another strategy that holds great 3 T. Setoyama, T. Takewaki, K. Domen and T. Tatsumi,
promise for the manufacturing of photocatalysts is emerging Faraday Discuss., 2017, 198, 509–527.
three-dimensional (3D) printing technologies that enable the 4 Q. Wang and K. Domen, Chem. Rev., 2020, 120, 919–985.
fabrication of hierarchical structures with interconnected 5 W.-H. Cheng, M. H. Richter, M. M. May, J. Ohlmann,
porous morphology and potential scalability.102 D. Lackner, F. Dimroth, T. Hannappel, H. A. Atwater and
H.-J. Lewerenz, ACS Energy Lett., 2018, 3, 1795–1800.
4. Conclusions and outlook 6 J. L. Young, M. A. Steiner, H. Döscher, R. M. France,
J. A. Turner and T. G. Deutsch, Nat. Energy, 2017, 2, 1–8.
Photocatalytic water splitting is a key technology to move 7 E. Verlage, S. Hu, R. Liu, R. J. R. Jones, K. Sun, C. Xiang,
towards more environmentally friendly routes for solar fuel N. S. Lewis and H. A. Atwater, Energy Environ. Sci., 2015, 8,
generation. The reaction is operationally simple but the 3166–3172.

This journal is © The Royal Society of Chemistry 2021 Sustainable Energy Fuels, 2021, 5, 4560–4569 | 4567
 View Article Online

Sustainable Energy & Fuels Perspective

8 C. Ros, T. Andreu and J. R. Morante, J. Mater. Chem. A, 2020, 33 J. Li and N. Wu, Catal. Sci. Technol., 2015, 5, 1360–1384.
8, 10625–10669. 34 M. A. Aziz Aljar, M. Zulqarnain, A. Shah, M. S. Akhter and
9 H. Kato, K. Asakura and A. Kudo, J. Am. Chem. Soc., 2003, F. J. Iikhar, AIP Advances, 2020, 10, 070701.
125, 3082–3089. 35 K. Villa, A. Black, X. Domènech and J. Peral, Solar Energy,
10 Y. Ham, T. Hisatomi, Y. Goto, Y. Moriya, Y. Sakata, 2012, 86, 558–566.
A. Yamakata, J. Kubota and K. Domen, J. Mater. Chem. A, 36 K. Nakata and A. Fujishima, J. Photochem. Photobiol. C:
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

2016, 4, 3027–3033. Photochem. Rev., 2012, 13, 169–189.

11 Y. Sakata, T. Hayashi, R. Yasunaga, N. Yanaga and 37 S. Hernández, D. Hidalgo, A. Sacco, A. Chiodoni,
H. Imamura, Chem. Commun., 2015, 51, 12935–12938. A. Lamberti, V. Cauda, E. Tresso and G. Saracco, Phys.
12 K. Maeda, T. Takata, M. Hara, N. Saito, Y. Inoue, Chem. Chem. Phys., 2015, 17, 7775–7786.
H. Kobayashi and K. Domen, J. Am. Chem. Soc., 2005, 127, 38 P. D. Nguyen, T. M. Duong and P. D. Tran, J. Sci. Adv. Mater.
8286–8287. Devices, 2017, 2, 399–417.
13 J. Liu, Y. Liu, N. Liu, Y. Han, X. Zhang, H. Huang, Y. Lifshitz, 39 S. R. Kelly, X. Shi, S. Back, L. Vallez, S. Y. Park,
S.-T. Lee, J. Zhong and Z. Kang, Science, 2015, 347, 970–974. S. Siahrostami, X. Zheng and J. K. Nørskov, ACS Catal.,
14 X. Ning, W. Zhen, Y. Wu and G. Lu, Appl. Catal., B, 2018, 2019, 9, 4593–4599.
226, 373–383. 40 G.-J. Lee and J. J. Wu, Powder Technol., 2017, 318, 8–22.
15 Y. W. Chen, J. D. Prange, S. Dühnen, Y. Park, M. Gunji, 41 S. Hamad, S. M. Woodley and C. R. A. Catlow, Mol. Simul.,
C. E. D. Chidsey and P. C. McIntyre, Nat. Mater., 2011, 10, 2009, 35, 1015–1032.
539–544. 42 N. Zeug, J. Buecheler and H. Kisch, J. Am. Chem. Soc., 1985,
16 Q. Wang, T. Hisatomi, Q. Jia, H. Tokudome, M. Zhong, 107, 1459–1465.
C. Wang, Z. Pan, T. Takata, M. Nakabayashi, N. Shibata, 43 Y. Hong, J. Zhang, F. Huang, J. Zhang, X. Wang, Z. Wu,
Y. Li, I. D. Sharp, A. Kudo, T. Yamada and K. Domen, Nat. Z. Lin and J. Yu, J. Mater. Chem. A, 2015, 3, 13913–13919.
Mater., 2016, 15, 611–615. 44 J. Zhang, Y. Wang, J. Zhang, Z. Lin, F. Huang and J. Yu, ACS
17 L. Liu and A. Corma, Nat. Rev. Chem., 2021, 5, 256–276. Appl. Mater. Interfaces, 2013, 5, 1031–1037.
18 T. Hisatomi, K. Takanabe and K. Domen, Catal. Lett., 2015, 45 Z. Lei, W. You, M. Liu, G. Zhou, T. Takata, M. Hara,
145, 95–108. K. Domen and C. Li, Chem. Commun., 2003, 2142–2143.
19 H. Idriss, Catal. Sci. Technol., 2020, 10, 304–310. 46 C. Xing, Y. Zhang, W. Yan and L. Guo, Int. J. Hydrogen
20 D.-K. Seo and R. Hoffmann, Theor. Chem. Acc., 1999, 102, Energy, 2006, 31, 2018–2024.
23–32. 47 K. Villa, X. Domènech, U. M. Garcı́a-Pérez and J. Peral, RSC
21 K. W. Böer and U. W. Pohl, in Semiconductor Physics, ed. Adv., 2016, 6, 36681–36688.
K. W. Böer and U. Pohl, Springer International 48 M. Dan, A. Prakash, Q. Cai, J. Xiang, Y. Ye, Y. Li, S. Yu, Y. Lin
Publishing, Cham, 2014, pp. 1–29. and Y. Zhou, Solar RRL, 2019, 3, 1800237.
22 J. Zhang, P. Zhou, J. Liu and J. Yu, Phys. Chem. Chem. Phys., 49 B. L. Phoon, C. W. Lai, J. C. Juan, P.-L. Show and G.-T. Pan,
2014, 16, 20382–20386. Int. J. Hydrogen Energy, 2019, 44, 14316–14340.
23 A. Walsh, Y. Yan, M. N. Huda, M. M. Al-Jassim and 50 P. Jana, C. Mata Montero, P. Pizarro, J. M. Coronado,
S.-H. Wei, Chem. Mater., 2009, 21, 547–551. D. P. Serrano and V. A. de la Peña O'Shea, Int. J. Hydrogen
24 Design of Advanced Photocatalytic Materials for Energy and Energy, 2014, 39, 5283–5290.
Environmental Applications, ed. J. Coronado, F. Fresno, M. 51 Y. Goto, T. Hisatomi, Q. Wang, T. Higashi, K. Ishikiriyama,
D. Hernández-Alonso and R. Portela, Springer-Verlag, T. Maeda, Y. Sakata, S. Okunaka, H. Tokudome,
London, 2013. M. Katayama, S. Akiyama, H. Nishiyama, Y. Inoue,
25 I. E. Castelli, D. D. Landis, K. S. Thygesen, S. Dahl, T. Takewaki, T. Setoyama, T. Minegishi, T. Takata,
I. Chorkendorff, T. F. Jaramillo and K. W. Jacobsen, T. Yamada and K. Domen, Joule, 2018, 2, 509–520.
Energy Environ. Sci., 2012, 5, 9034–9043. 52 M. Kumar, P. Basera, S. Saini and S. Bhattacharya, Sci. Rep.,
26 I. E. Castelli, F. Hüser, M. Pandey, H. Li, K. S. Thygesen, 2020, 10, 15372.
B. Seger, A. Jain, K. A. Persson, G. Ceder and 53 P. Yohan, H. Takashi and D. Kazunari, in Current
K. W. Jacobsen, Adv. Energy Mater., 2015, 5, 1400915. Developments in Photocatalysis and Photocatalytic Materials,
27 D. W. Davies, K. T. Butler and A. Walsh, Chem. Mater., 2019, Elsevier, 2020, pp. 141–157.
31, 7221–7230. 54 K. Villa, X. Domènech, U. M. Garcı́a-Pérez and J. Peral,
28 J. Pan and Q. Yan, J. Semicond., 2018, 39, 071001. Catal. Lett., 2016, 146, 100–108.
29 N. Daelman, F. S. Hegner, M. Rellán-Piñeiro, M. Capdevila- 55 P. Garg, P. Bhauriyal, A. Mahata, K. S. Rawat and B. Pathak,
Cortada, R. Garcı́a-Muelas and N. López, J. Chem. Phys., ChemPhysChem, 2019, 20, 383–391.
2020, 152, 050901. 56 D. Jing and L. Guo, J. Phys. Chem. B, 2006, 110, 11139–
30 F. Ambrosio, G. Miceli and A. Pasquarello, J. Chem. Phys., 11145.
2015, 143, 244508. 57 S. Kahng, H. Yoo and J. H. Kim, Adv. Powder Technol., 2020,
31 A. Fujishima and K. Honda, Nature, 1972, 238, 37–38. 31, 11–28.
32 H. Eidsvåg, S. Bentouba, P. Vajeeston, S. Yohi and
D. Velauthapillai, Molecules, 2021, 26, 1687.

4568 | Sustainable Energy Fuels, 2021, 5, 4560–4569 This journal is © The Royal Society of Chemistry 2021
 View Article Online

Perspective Sustainable Energy & Fuels

58 F. Jiang, Gunawan, T. Harada, Y. Kuang, T. Minegishi, 81 K. Maeda, H. Terashima, K. Kase, M. Higashi, M. Tabata
K. Domen and S. Ikeda, J. Am. Chem. Soc., 2015, 137, and K. Domen, Bull. Chem. Soc. Jpn., 2008, 81, 927–937.
13691–13697. 82 S. Rej, M. Bisetto, A. Naldoni and P. Fornasiero, J. Mater.
59 X. Chen and W. Shangguan, Front. Energy, 2013, 7, 111–118. Chem. A, 2021, 9, 5915–5951.
60 X. Ning and G. Lu, Nanoscale, 2020, 12, 1213–1223. 83 M. Hara, T. Kondo, M. Komoda, S. Ikeda, J. N. Kondo,
61 X. Ning, J. Li, B. Yang, W. Zhen, Z. Li, B. Tian and G. Lu, K. Domen, M. Hara, K. Shinohara and A. Tanaka, Chem.
Published on 12 August 2021. Downloaded by CHULALONGKORN UNIVERSITY on 12/4/2023 9:16:15 AM.

Appl. Catal., B, 2017, 212, 129–139. Commun., 1998, 357–358.

62 J. A. Nasir, Z. ur. Rehman, S. N. A. Shah, A. Khan, I. S. Butler 84 J. Luo, L. Steier, M.-K. Son, M. Schreier, M. T. Mayer and
and C. R. A. Catlow, J. Mater. Chem. A, 2020, 8, 20752–20780. M. Grätzel, Nano Lett., 2016, 16, 1848–1857.
63 A. G. Scheuermann, J. P. Lawrence, K. W. Kemp, T. Ito, 85 Y. Kwon, A. Soon, H. Han and H. Lee, J. Mater. Chem. A,
A. Walsh, C. E. D. Chidsey, P. K. Hurley and 2014, 3, 156–162.
P. C. McIntyre, Nat. Mater., 2016, 15, 99–105. 86 F. Dattila, R. Garcı́a-Muelas and N. López, ACS Energy Lett.,
64 M. S. Nasir, G. Yang, I. Ayub, S. Wang, L. Wang, X. Wang, 2020, 5, 3176–3184.
W. Yan, S. Peng and S. Ramakarishna, Appl. Catal. B: 87 J.-N. Nian, C.-C. Hu and H. Teng, Int. J. Hyd. Energy, 2008,
Environ., 2019, 257, 117855. 33, 2897–2903.
65 X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, 88 W. Siripala, A. Ivanovskaya, T. F. Jaramillo, S.-H. Baeck and
J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., E. W. McFarland, Sol. Energy Mater. Sol. Cells, 2003, 77, 229–
2009, 8, 76–80. 237.
66 J. Fu, J. Yu, C. Jiang and B. Cheng, Adv. Energy Mater., 2018, 89 A. Paracchino, V. Laporte, K. Sivula, M. Grätzel and
8, 1701503. E. Thimsen, Nature Mater, 2011, 10, 456–461.
67 A. Mishra, A. Mehta, S. Basu, N. P. Shetti, K. R. Reddy and 90 S. Chandrasekaran, L. Yao, L. Deng, C. Bowen, Y. Zhang,
T. M. Aminabhavi, Carbon, 2019, 149, 693–721. S. Chen, Z. Lin, F. Peng and P. Zhang, Chem. Soc. Rev.,
68 B. Zhu, B. Cheng, L. Zhang and J. Yu, Carbon Energy, 2019, 2019, 48, 4178–4280.
1, 32–56. 91 Y. Li, Y.-L. Li, B. Sa and R. Ahuja, Catal. Sci. Technol., 2017,
69 C. Butchosa, P. Guiglion and M. A. Zwijnenburg, J. Phys. 7, 545–559.
Chem. C, 2014, 118, 24833–24842. 92 G.-J. Lai, L.-M. Lyu, Y.-S. Huang, G.-C. Lee, M.-P. Lu,
70 R. Q. Zhang, X. M. Liu, Z. Wen and Q. Jiang, J. Phys. Chem. T.-P. Perng, M.-Y. Lu and L.-J. Chen, Nano Energy, 2021,
C, 2011, 115, 3425–3428. 81, 105608.
71 J.-Y. Hao, Y.-Y. Wang, X.-L. Tong, G.-Q. Jin and X.-Y. Guo, 93 K. Maeda, J. Photochem. Photobiol. C, 2011, 12, 237–268.
Int. J. Hydrogen Energy, 2012, 37, 15038–15044. 94 N. Fajrina and M. Tahir, Int. J. Hydrogen Energy, 2019, 44,
72 M. Aslam, M. T. Qamar, I. Ahmed, A. U. Rehman, S. Ali, 540–577.
I. M. I. Ismail and A. Hameed, Appl. Nanosci., 2018, 8, 95 M. Raque, R. Mubashar, M. Irshad, S. S. A. Gillani,
987–999. M. B. Tahir, N. R. Khalid, A. Yasmin and M. A. Shehzad, J.
73 Y. Zhang, T. Xia, P. Wallenmeyer, C. X. Harris, Inorg. Organomet. Polym., 2020, 30, 3837–3861.
A. A. Peterson, G. A. Corsiglia, J. Murowchick and 96 R. Molinari, T. Marino and P. Argurio, Int. J. Hydrogen
X. Chen, Energy Technol., 2014, 2, 183–187. Energy, 2014, 39, 7247–7261.
74 X.-N. Guo, X.-L. Tong and X.-Y. Guo, in Novel Carbon 97 K. Takanabe, ACS Catal., 2017, 7, 8006–8022.
Materials and Composites, John Wiley & Sons, Ltd, 98 X. Zhang, X. Zhao, D. Wu, Y. Jing and Z. Zhou, Adv. Sci.,
Chichester, UK, 2019, pp. 35–72. 2016, 3, 1600062.
75 M. Hara, G. Hitoki, T. Takata, J. N. Kondo, H. Kobayashi 99 X. Zhang, Z. Zhang, D. Wu, X. Zhang, X. Zhao and Z. Zhou,
and K. Domen, Catal. Today, 2003, 78, 555–560. Small Methods, 2018, 2, 1700359.
76 E. Nurlaela, A. Ziani and K. Takanabe, Mater. Renew. 100 L. Y. Kunz, B. T. Diroll, C. J. Wrasman, A. R. Riscoe,
Sustain. Energy, 2016, 5, 18. A. Majumdar and M. Cargnello, Energy Environ. Sci., 2019,
77 T. Hisatomi, J. Kubota and K. Domen, Chem. Soc. Rev., 2014, 12, 1657–1667.
43, 7520–7535. 101 D. Zhang, W. Wang, F. Peng, J. Kou, Y. Ni, C. Lu and Z. Xu,
78 M. Ahmed and G. Xinxin, Inorg. Chem. Front., 2016, 3, 578– Nanoscale, 2014, 6, 5516–5525.
590. 102 M. J. Martı́n de Vidales, A. Nieto-Márquez, D. Morcuende,
79 K. Maeda, D. Lu and K. Domen, Chem. –Eur. J., 2013, 19, E. Atanes, F. Blaya, E. Soriano and F. Fernández-Martı́nez,
4986–4991. Catal. Today, 2019, 328, 157–163.
80 F. Fresno, O. Iglesias, E. Alfonso-González, M. J. Rivero,
I. Ortiz and V. A. de la Peña O'Shea, Catal. Today, 2021,
362, 22–27.

This journal is © The Royal Society of Chemistry 2021 Sustainable Energy Fuels, 2021, 5, 4560–4569 | 4569