International Journal of Hydrogen Energy 62 (2024) 250–255

Contents lists available at ScienceDirect

International Journal of Hydrogen Energy

journal homepage: www.elsevier.com/locate/he

Single-layer GaInS3: Water-splitting photocatalyst with high solar

conversion efficiency and long carrier lifetime from
first-principles investigation
Zhi Long a, Yi Xiang a, Guo-Zhen Zhang a, b, Xi Qin a, Song Wu a, Wen-Hao Song a, Xing-Ming Liu a,
Jie Cheng a, Li-Li Liu c, Shi-Fa Wang a, Yong Wei a, Lei Hu a, *, Xiang-Kai Deng d, **,
Chun-Ming Yang e, Xing Zou a, ***
a
College of Electronic and Information Engineer, Chongqing Three Gorges University, Chongqing, 404100, China
b
Institute of Intelligent Information Processing, South China Business College of Guangdong University of Foreign Studies, Guangzhou, 510545, Guangdong, China
c
College of Teacher Education, Chongqing Three Gorges University, Chongqing, 404100, China
d
Chongqing College of Mobile Communication, Chongqing, 401520, China
e
College of Chemistry and Chemistry Engineering, Yan’an University, Shaanxi Key Laboratory of Chemical Reaction Engineering, Yan’an, 716000, China

A R T I C L E I N F O A B S T R A C T

Handling Editor: Suleyman I. Allakhverdiev Two-dimensional (2D) materials with a high solar to hydrogen (STH) conversion are in high demand due to the
escalating energy crisis and environmental degradation. The STH efficiency is strongly correlated with the charge
Keywords: carrier separation, which can be improved by the inside electric dipole. Herein, the water-splitting capabilities of
Two-dimensional material single-layer (SL) GaInS3 are explored by employing first-principles calculations. Theoretical outcomes reveal that
Water-splitting photocatalyst
SL GaInS3 exhibits a large intrinsic electric dipole and has a thermally stable structure at 300 K. SL GaInS3
GaInS3
displays a direct bandgap of 1.83 eV, as well as suitable band edges and abundant visible absorption (105 cm− 1)
First-principles calculation
NAMD for solar-driven water-splitting. The electronic structure and carrier mobility calculations demonstrate the high
separation efficiency of charge carriers, which is strongly affirmed by the quite long carrier lifetime of 8.26 ns.
The overpotential analysis of charge carriers suggests that the oxygen generation of SL GaInS3 can be accom­
plished without a cocatalyst, while hydrogen generation can be completed with the aid of cocatalysts. The STH
efficiency of SL GaInS3 reaches up to 16.8%, apparently surpassing the commercial standard (10%). In brief, the
high STH and long carrier lifetime promise the possible application of SL GaInS3 for solar water splitting. What’s
more, this work provides straight evidence that 2D photocatalysts with inside dipoles exhibit a long carrier
lifetime.

1. Introduction separated to take part in superficial reactions [9]. Impressively, the in­
side electric field of 2D photocatalysts is able to enhance the separation
Since graphene was synthesized successfully [1], 2D materials have of charge carriers remarkably [10,11]. What’s more, the charge sepa­
caught widespread attention because of applications in nanotechnology ration efficiency can be quantitatively scaled by the carrier lifetime [12].
[2], photodectors [3,4], photocatalysts [5,6], etc. Among these precious However, the carrier lifetimes of quite a large number of water-splitting
applications, water-splitting photocatalysts have the potential to address photocatalysts with intrinsic dipoles are still unknown [13–16].
the growing problem of energy scarcity and environmental pollution The M2X3 (M = Ga, In; X = O, S, Se, and Te) monolayers are
[7]. Compared with 3D counterparts, 2D photocatalysts provide more renowned for their strong inside electric dipole [17], which induces
reaction sites due to larger specific surface areas [8] and minimize the excellent water-splitting behavior [18,19] and out-of-plane piezoelec­
migration distance so that more charge carriers can be effectively tricity [20,21]. Due to these attractive phenomena, the derivatives of

* Corresponding author.
** Corresponding author.
*** Corresponding author.
E-mail addresses: [email protected] (L. Hu), [email protected] (X.-K. Deng), [email protected] (X. Zou).

https://doi.org/10.1016/j.ijhydene.2024.03.017
Received 8 December 2023; Received in revised form 25 January 2024; Accepted 2 March 2024
Available online 12 March 2024
0360-3199/© 2024 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Z. Long et al. International Journal of Hydrogen Energy 62 (2024) 250–255

M2X3 monolayers come into the sights of researchers. Specifically, SL ensemble. After simulation, no broken chemical bonds appeared, and
GaInO3 was predicted to hold strain-tunable electronic and optical only slight atomic position shifts could be observed (Fig. S2). Hence, SL
properties that are suitable for use in optoelectronic applications [22, GaInS3 maintains good thermal stability at 300 K.
23]. The layer-dependent piezoelectric property was theoretically
proved in 2D GaInS3 nanosheets [24]. The In2X2Y (X, Y = S, Se, Te) 2.2. Electronic property and band edge
monolayers exhibit remarkable photocatalytic properties and high STH
efficiency [15]. Very notably, SL GaInX3 (X = S, Se, Te) is characterized For solar water-splitting, appropriate electronic structures such as
by asymmetric structures and high carrier mobility [25], promising moderate bandgaps are necessary. Herein, the band structure is first
excellent water-splitting behavior [26]. Among GaInX3 (X = S, Se, Te), investigated at the HSE06 level. The valence band maximum (VBM) and
SL GaInS3 holds the most significant inside dipole and the widest the conduction band minimum (CBM) of SL GaInS3 occur at the Γ point,
bandgap, meaning the lowest possibility for charge carrier recombina­ forming a direct bandgap of 1.83 eV. The S-3p and Ga-4s orbitals mainly
tion during water-splitting. Up to now, the water-splitting characteris­ contribute to the CBM, while the VBM is primarily contributed by the S-
tics of SL GaInS3 have not been explored. 3p and In-5p orbitals (Fig. 2a). Fig. 2b suggests that the partial charge
This work systematically investigates the photocatalytic perfor­ density directly reveals that the CBM is mainly distributed in the bottom
mance of SL GaInS3 by first-principles calculations coupled with ab Ga–S layer (the bottom surface), and the VBM is located at the top sulfur
initio nonadiabatic molecular dynamics (NAMD). Theoretical results layer (the top surface). Thus, the CBM and VBM of SL GaInS3 are
show that SL GaInS3 possesses appropriate band edges and sufficient spatially separated, which is further confirmed by Fig. S3 and facilitates
visible absorption, making it feasible for solar water splitting. The STH the rapid separation of photogenerated holes and electrons. Very
efficiency of SL GaInS3 reaches as high as ~17% and far surpasses the Notably, the separated CBM and VBM enable 2D materials to exhibit
economic value (10%). The carrier lifetime of SL GaInS3 is calculated to glorious prospects for solar water splitting [30–36]. In addition, the
be 8.26 ns. Hence, SL GaInS3 is a powerful candidate for water splitting spin-orbital coupling (SOC) does not need to be considered for SL GaInS3
and will contribute to solving the increasing energy shortage problem. (Fig. S4).
For 2D materials with vacuum level differences between the top
2. Result and discussion
(001) and bottom (001) surfaces, the CBM and VBM levels relative to the
vacuum level of the (001) surface (Ebot CBM ) can be calculated via [35,
2.1. Structure and stability
37–39]:
The optimized in-plane lattice parameters of SL GaInS3 are a = b = bot
ECBM = E′CBM − Evacuum
bot
Eq. (1)
3.769 Å, which approaches the previously calculated (3.74 Å) [25]. SL
GaInS3 is characterized by a five-atom thickness, with the indium (In) bot
EVBM = E′VBM − Evacuum
bot
Eq. (2)
and gallium (Ga) atomic layers sandwiched by the sulfur (S) layers
(Fig. 1a), producing an effective thickness of 0.828 nm (Fig. S1 of the Ebot
vacuum is the vacuum level of the (001) surface of SL GaInS3. The
Supplementary Material). A theoretical electric dipole moment of 0.511 calculated values of Ebot bot
CBM and EVBM for SL GaInS3 are − 4.05 and − 5.91
Debye exists in SL GaInS3 between the top (001) and bottom (001) eV, respectively. Similarly, the CBM and VBM levels relative to the
surfaces. For comparison, the theoretical intrinsic dipole moments of SL vacuum level of the (001) surface (Etop
VBM ) can be obtained using Eq. (3)
MXY (M = Mo, W; X, Y=S, Se, Te) range from 0.173 Debye (WSSe) to and Eq. (4).
0.375 Debye (MoSTe) [27]. Please note 1 e⋅Å = 4.807 Debye.
Commonly, more prominent electric dipole moments mean stronger
top
ECBM = E′CBM − Evacuum
top
Eq. (3)
electric forces to separate the generated holes and electrons [28,29]. As
a result of the intrinsic dipole, an electrostatic potential difference ΔΦ =
top
EVBM = E′VBM − Evacuum
top
Eq. (4)
1.79 eV in vacuum appears for SL GaInS3 (Fig. 1c), with the vacuum
level of the top surface higher than that of the bottom surface. Etop
vacuum is the vacuum level of the (001) surface. The calculated values of
Previous calculations have demonstrated the mechanical and Etop top
CBM and EVBM are − 5.87 and − 7.65 eV, respectively. Taking into ac­
dynamical stability of SL GaInS3 [25]. Herein, the thermal stability is count that the CBM is predominantly located at the (001) surface, Eq. (1)
studied by ab initio molecular dynamics simulations within the NVT is employed. Similarly, considering the VBM is located at the (001)

Fig. 1. (a) In the side view, E represents the intrinsic electric field; (b) In the top view, the rhombic dashed line represents the primitive unit cell, and a rectangular
dashed line denotes the orthogonal supercell; (c) The electrostatic potential of SL GaInS3.

251
Z. Long et al. International Journal of Hydrogen Energy 62 (2024) 250–255

Fig. 2. (a) The band structure and partial density of states; (b) The partial charge density of CBM and VBM, where the isosurface is set to 0.01 e/Å3; (c) The band
alignment of SL GaInS3, where the symbol E is the electric field due to the intrinsic dipole.

surface, Eq. (4) is adopted. That is, the physically meaningful CBM and noteworthy the maximum absorption reaches up to 1.3 × 105 cm − 1. On
VBM levels are − 4.05 and − 7.65 eV, respectively. The basic qualifica­ the other hand, the maximum optical transition from VBM to CBM of SL
tion of solar water-splitting is that the aligned CBM level is higher than GaInS3 takes place near the Γ point, corresponding to a high electronic
the H+/H2 redox potential (EH+ /H2 ) and the aligned VBM level is lower state density. The maximum optical transition from VBM-1 to CBM oc­
than the O2/H2O oxidation potential (EO2 /H2 O ). At pH = 0, the EH+ /H2 and curs at the k-point near the Γ point (Fig. 3b). In brief, the allowed
EO2 /H2 O are − 4.44 and − 5.67 eV, respectively. Hence, SL GaInS3 satisfies transition further affirms the strong optical absorption in SL GaInS3.
the energetic requirement for solar water-splitting (Fig. 2c). Considering The theoretical exciton binding energy Eb is 0.82 eV by considering
EH+ /H2 = − 4.44 eV + pH × 0.059 eV and EO2 /H2 O = − 5.67 eV + pH × the energy difference between the quasiparticle bandgap (2.46 eV) in
0.059 eV, the EH+ /H2 and EO2 /H2 O become − 4.26 and − 5.49 eV at pH = 3, Fig. S5 and the optical bandgap (1.64 eV) in Fig. 3a. The calculated Eb of
respectively. At this condition, the band edges of SL GaInS3 are also SL GaInS3 is comparable with that of recently predicted 2D photo­
favorable for water splitting (Fig. 2c). catalysts such as SL WSSe (0.83 eV) [43] and SL In2S3 (0.82 eV) [18].
The external potential of photoexcited electrons Ue during hydrogen The relatively weak exciton binding for SL GaInS3 can be attributed to
generation is the energy difference between the EH+ /H2 level and the the intrinsic electric field and the spatially separated CBM and VBM.
After light absorption, electrons jump from the VBM (VBM-1) to the
aligned CBM. The external potential of photoexcited holes Uh during
CBM, and hence, photoexcited electrons and holes are produced.
oxygen generation is the difference between the aligned VBM and the
EH+ /H2 level [40]. The theoretical Uh and Ue values for SLGaInS3 are 3.21
and 0.39 V at pH = 0, respectively. At pH = 3, the calculated Uh and Ue
2.4. Carrier mobility and lifetime
are 3.39 and 0.21 eV, respectively. As indicated by the high Uh and Ue ,
the photogenerated electrons and holes of SL GaInS3 preferentially react
The photocatalytic performance is essentially determined by the
with water rather than corrode itself.
separation efficiency of photo-excited electrons and holes. Herein, using
the deformation potential (DP) theory (SM-1), the carrier mobility is
2.3. Optical absorption and exciton binding calculated to reveal the separation efficiency of photoexcited carriers:

eℏ3 C2D
Abundant solar energy is required to split water, and hence, the μ2D = Eq. (5)
kB Tm∗ md Ed2
optical absorption coefficient of SL GaInS3 is calculated using α(ω) =
√̅̅̅
2ω[(ε1 (ω)2 + ε2 (ω)2 )1/2 − ε1 (ω)]1/2 [39] and shown in Fig. 3a, where Deduced from Eq. (5), the carrier mobility can be significantly
ε1 (ω) and ε2 (ω) are the real and imaginary parts of the complex modified by the Ed value. For photoexcited electrons, the Ed value is the
dielectric function at the G0W0+BSE level [41,42]. The light absorption slope of the linear fitting between the (ECBM − Φbottom ) and the strain ε, as
of SL GaInS3 starts in the infrared region, and three conspicuous ab­ the CBM charge is distributed on the (001) surface. However, the CBM
sorption peaks are at 2.16, 2.56, and 3.20 eV, respectively. It is very distribution is not previously considered in the Ed calculations of SL
GaInS3 [25], though it is imperatively needed [21,44]. As shown in
Table 1, the Ed value is different from that in the previous calculation
[25]. The C2D and m∗ possess isotropic characters. Due to the fast energy
change near the CBM, the m∗ value is relatively small (0.21 me ). As a
result, SL GaInS3 exhibits superior or comparable electron mobility
(~1500 cm2⋅V− 1⋅s− 1) as typical photocatalysts such as InSe (1000–1100
cm2 ⋅V− 1⋅s− 1) [45–47] and MoS2 (~200 cm2⋅V− 1⋅s− 1) [48].

Table 1
The effective mass m* of electrons and holes, elastic constant C2D, deformation
potential Ed, and carrier mobility μ of SL GaInS3 along x and y directions. me is
the electron mass.
Species direction m*(me ) C2D (N⋅m− 1) Ed (eV) μ (cm2⋅V− 1⋅s− 1)
Electron x 0.21 91.97 5.43 1506 (202.87) [25]
0.21 91.97
y 0.21 91.97 5.30 1583 (203.08) [25]
Hole x 0.58 91.97 2.35 1030 (3.20) [25]
0.58 91.97
Fig. 3. (a) The absorption coefficients at the G0W0+BSE level; (b) The transi­
y 0.58 91.97 2.58 849 (3.13) [25]
tion dipole moment from VBM (VBM-1) to CBM.

252
Z. Long et al. International Journal of Hydrogen Energy 62 (2024) 250–255

Though a relatively large effective mass (0.58 me) of photoexcited

OOH∗ → ∗ + O2 + H+ + e−
holes is acquired, the reduced Ed value leads to the comparatively large
hole mobility (~1000 cm2⋅V− 1⋅s− 1). Though the carrier mobility may be Where * indicates the adsorbed material, i.e., SL GaInS3. Fig. 5a illus­
overestimated due to the intrinsic defect of the DP theory to some extent trates the relaxed structures of the OH*, O*, and OOH* intermediates.
[49], the high mobility for photoexcited electrons and holes helps them The Gibbs free energy change (ΔG) of each step is expressed as:
promptly arrive at the reaction sites. Very notably, the photoexcited
electrons with high speed will move towards the (001) surface with the ΔG = ΔEads + ΔEZPE − TΔS + ΔGu + ΔGpH Eq. (6)
intrinsic positive charges, where the HER is expected. Conversely,
where ΔEads , ΔEZPE and ΔS denote the intermediate adsorption energy,
photoexcited holes with high speed will move towards the (001) surface
the difference in zero-point energy, and the entropy change. T is the
with intrinsic negative charges, where the OER is expected. In brief, the
room temperature (300 K). ΔGu = − eU, and U is the electrode poten­
high-speed carriers will be separated effectively with the aid of the in­
tial. ΔGpH denotes the free energy correction by hydrion concentration.
side electric field.
As clearly seen in Fig. 5a, all the formations of OH*, O*, and OOH*
The carry mobility of SL GaInS3 suggests the low recombination of
intermediates are endothermic in the absence of solar irradiation. The
charge carriers, i.e., the long carrier life. In this section, the lifetime of
limiting step (Ul = 3.26 V) emerges during the transformation from O*
carriers is calculated using the density functional theory combined with
to OOH*. On the other hand, though the maximum external potential
TD-DFT implemented in the Hefei NAMD [50–52]. After photon ab­
provided by photoexcited holes is Uh = 3.21 V at pH = 0, it can be used
sorption, the photoexcited electron (hole) translates to the CBM (VBM).
to overcome the reaction barrier. The reason is that the exothermic
Therefore, the carrier lifetime can be regarded as the relaxation time
energy (− 1.89 eV) in the second step is remarkably higher than the
during which the photoexcited electron on the CBM decays to the VBM
exothermic energy (0.05 eV) in the third step. Thus, SL GaInS3 can
[53,54]. The electron-hole recombination process of photoexcited car­
competently perform the OER without a cocatalyst. At pH = 3, under the
riers in SL GaInS3 is depicted in Fig. 4. Assuming the decay is fully
eternal potential of holes (3.39V), the OER behaves as a downhill
exponential [55], the carrier lifetime τ of SL GaInS3 is calculated to be
exothermic process (Fig. S8a). Notably, SL PdSeO3 is a water-splitting
8.26 ns through fitting f(t) = e(− t/τ) ,where f(t) denotes the population of
photocatalyst receiving wide attention as no cocatalyst is required [40].
remaining carriers as a function of time t. As comparisons, the carrier
The HER can be decomposed into two steps:
lifetime of SL GaInS3 is longer than those of most 2D optoelectronic
materials, such as MoS2 (0.39 ns) [56], MoSSe (0.44 ns), MoSTe (1.31 ∗ + H + + e− →H ∗
ns) [57], and LiAlTe2 (1.69 ns) [58]. The long lifetime in SL GaInS3
primarily results from the electronic properties, wherein the intrinsic H ∗ + H + + e− → ∗ +H2
electric field segregates electrons and holes into different domains.
The HER is expected to occur on the Ga atoms where the CBM is
Moreover, the long carrier lifetime of SL GaInS3 also provides straight
present, and its ΔG is plotted in Fig. 5b. In the absence of any external
evidence that M2X3 [18] and GaInX3 (M = Ga, In; X = S, Se, Te) [25]
voltage, the HER barrier is 2.04 eV. Under the maximum external po­
photocatalysts with inside dipoles also exhibit enough carrier lifetime.
tential of 0.39 V provided by the photogenerated electrons at pH = 0,
the HER still exhibits a barrier of 1.65 eV. Consequently, SL GaInS3 re­
2.5. HER and OER
quires a cocatalyst to enhance its hydrogen generation ability. As
demonstrated before, the HER barrier of g-C3N4 was decreased by a
The OER is expected to occur on the (001) sulfur atomic layers,
cocatalyst [59]. At pH = 3, under the eternal potential of electrons (0.
where the VBM resides on. The OER can be divided into four procedures:
21V), the HER completion also requires cocatalysts (Fig. S8b).
∗ + H2 O → OH∗ + H+ + e−
2.6. STH efficiency
OH∗ → O∗ + H+ + e−
The STH efficiency (ηSTH ) is the critical factor in measuring the
O∗ + H2 O → OOH∗ + H+ + e− photocatalytic properties and can be obtained by Eq. (7):
ηSTH = ηabs × ηcu Eq. (7)

where ηabs is the light absorption efficiency and ηcu is the carrier utili­
zation efficiency. Firstly, the ηabs value can be calculated by Eq. (8):
∫∞
P(ℏw)d(ℏw)
Eq. (8)
E
ηabs = ∫ ∞g
0
P(ℏw)d(ℏw)

where Eg is the bandgap of the martial and P(ℏw) is the AM 1.5G solar
energy flux at photon energy ℏw. The ηabs reaches 44.3% due to the small
band gap. Next, the ηcu is determined by Eq. (9):
∫ ∞ P(ℏw)
ΔG d(ℏw)
ηcu = ∫ ∞ Emin ℏw Eq. (9)
Eg
P(ℏw)d(ℏw)

where ΔG is the potential difference of photocatalytic water splitting

and taken as 1.23 eV. The minimum energy of the photons (Emin ) utilized
by the redox reactions is given by Eq. (10):
⎧
⎪
⎪ Eg , (Ue ≥ 0.2, Uh ≥ 1.83)
⎨
Eg + 0.2 − Ue , (Ue < 0.2, Uh ≥ 1.83)
Emin = Eq. (10)
⎪
⎪ Eg + 1.83 − Uh , (Ue ≥ 0.2, Uh < 1.83)
Fig. 4. Electron-hole recombination dynamics at 300 K in SL GaInS3. The ⎩
Eg + 2.03 − Ue − Uh , (Ue < 0.2, Uh < 1.83)
population of photoexcited carriers is normalized to 1.

253
Z. Long et al. International Journal of Hydrogen Energy 62 (2024) 250–255

Fig. 5. (a) The Gibbs free energy change (ΔG) of the OER on the upper S atomic layer at pH = 0; (b) The ΔG of the HER on the Ga atomic layers at pH = 0.

where Ue and Uh are the external potentials driving the HER and OER, Appendix A. Supplementary data
respectively [34]. The ηcu reaches 51.73%. According to Eq. (6), the ηSTH
of SL GaInS3 is 22.9%. Supplementary data to this article can be found online at https://doi.
org/10.1016/j.ijhydene.2024.03.017.
The corrected STH efficiency (η′STH ) is calculated using Eq. (7) by
considering the in-built electric field:
References
∫∞
P(ℏw)d(ℏw)
η′STH = ηSTH × ∫ ∞ 0
∫∞ Eq. (11) [1] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, et al. Electric
0
P(ℏw)d(ℏw) + ΔΦ Eg P(ℏw)
ℏw
d(ℏw) field effect in atomically thin carbon films. Science 2004;306:666–9.
[2] Gooran-Shoorakchaly A, Ahmadchally AA, Soleimani-Amiri S, Gholipour M. Design
of a low-power short-channel electrostatically doped silicene nanoribbon FET. IEEE
where ΔΦ is the vacuum level difference between the (001) and (001) Trans Electron Dev 2021;68:1956–61.
surfaces. The η′STH of SL GaInS3 is 16.8%. The high STH efficiency of SL [3] Rudi SG, Soleimani-Amiri S. Bilayer armchair graphene nanoribbon photodetector
with Stone-Wales defect: a computational study. Mater Sci Semicond Process 2022;
GaInS3 can be attributed to the high generation efficiency of charge
150:106918.
carriers facilitated by strong light absorption, the high separation effi­ [4] Ma Q, Ren G, Xu K, Ou JZ. Tunable optical properties of 2D materials and their
ciency of electrons and holes due to their different positions, and a high applications. Adv Opt Mater 2021;9:2001313.
ηcu resulting from the prolonged carrier lifetime. For comparison, the [5] Dholam R, Patel N, Adami M, Miotello A. Hydrogen production by photocatalytic
water-splitting using Cr-or Fe-doped TiO2 composite thin films photocatalyst. Int J
recently reported η′STH values are 11.68% for SL WSSe [43] and 17% for Hydrogen Energy 2009;34:5337–46.
α-BC2N [60]. [6] Prasad C, Tang H, Liu Q, Bahadur I, Karlapudi S, Jiang Y. A latest overview on
photocatalytic application of g-C3N4 based nanostructured materials for hydrogen
production. Int J Hydrogen Energy 2020;45:337–79.
3. Conclusion [7] Carley S, Konisky DM. The justice and equity implications of the clean energy
transition. Nat Energy 2020;5:569–77.
[8] Cao S, Yu J. Carbon-based H2-production photocatalytic materials. J Photochem
Herein, the water-splitting capabilities of SL GaInS3 have been Photobiol, C 2016;27:72–99.
explored. First of all, Theoretical findings reveal that SL GaInS3 exhibits [9] Singh AK, Mathew K, Zhuang HL, Hennig RG. Computational screening of 2D
materials for photocatalysis. J Phys Chem Lett 2015;6:1087–98.
appropriate band edges and sufficient solar absorption for water split­
[10] Fan Y, Song X, Qi S, Ma X, Zhao M. Li-III-VI bilayers for efficient photocatalytic
ting. Secondly, the electronic structure and carrier mobility property overall water splitting: the role of intrinsic electric field. J Mater Chem A 2019;7:
boost the charge carriers’ separation efficiency of SL GaInS3. Thirdly, the 26123–30.
oxygen generation of SL GaInS3 can be accomplished without a cocat­ [11] Fan Y, Ma X, Wang J, Song X, Wang A, Liu H, et al. Highly-efficient overall water
splitting in 2D Janus group-III chalcogenide multilayers: the roles of intrinsic
alyst, while its hydrogen generation can be completed with the aid of electric filed and vacancy defects. Sci Bull 2020;65:27–34.
cocatalysts. Fourthly, the solar efficiency of SL GaInS3 reaches as high as [12] Cheng C, Long R. Charge-compensated doping extends carrier lifetimes in SrTiO3
16.8%, indicating economic value. Very Impressively, the carrier life­ by passivating oxygen vacancy defects. J Phys Chem Lett 2021;12:12040–7.
[13] Liu X, Cheng P, Zhang X, Shen T, Liu J, Ren J-C, et al. Enhanced solar-to-hydrogen
time of SL GaInS3 achieves a rarely reported value of 8.26 ns. In brief, efficiency for photocatalytic water splitting based on a polarized heterostructure:
this work not only defines SL GaInS3 as a water-splitting photocatalyst the role of intrinsic dipoles in heterostructures. J Mater Chem A 2021;9:14515–23.
but also provides straight evidence that 2D photocatalysts with inside [14] Bao J, Zhu B, Zhang F, Chen X, Guo H, Qiu J, et al. Sc2CF2/Janus MoSSe
heterostructure: a potential Z-scheme photocatalyst with ultra-high solar-to-
dipoles hold a long carrier lifetime. hydrogen efficiency. Int J Hydrogen Energy 2021;46:39830–43.
[15] Wang P, Liu H, Zong Y, Wen H, Xia JB, Wu HB. Two-dimensional In2X2X’(X and
X’= S, Se, and Te) monolayers with an intrinsic electric field for high-performance
photocatalytic and piezoelectric applications. ACS Appl Mater Interfaces 2021;13:
Declaration of competing interest 34178–87.
[16] Yuan J-H, Xue K-H, Miao X. Two-dimensional ABC3 (A= Sc, Y; B= Al, Ga, In; C= S,
The authors declare that they have no known competing financial Se, Te) with intrinsic electric field for photocatalytic water splitting. Int J
Hydrogen Energy 2023;48:5929–39.
interests or personal relationships that could have appeared to influence [17] Ding W, Zhu J, Wang Z, Gao Y, Xiao D, Gu Y, et al. Prediction of intrinsic two-
the work reported in this paper. dimensional ferroelectrics in In2Se3 and other III2-VI3 van der Waals materials. Nat
Commun 2017;8:14956.
[18] Fu CF, Sun J, Luo Q, Li X, Hu W, Yang J. Intrinsic electric fields in two-dimensional
Acknowledgment materials boost the solar-to-hydrogen efficiency for photocatalytic water splitting.
Nano Lett 2018;18:6312–7.
[19] Zhao P, Ma Y, Lv X, Li M, Huang B, Dai Y. Two-dimensional III2-VI3 materials:
This work is supported by the Natural Science Foundation of
promising photocatalysts for overall water splitting under infrared light spectrum.
Chongqing (cstc2021jcyj-msxmX0524), the Natural Science Foundation Nano Energy 2018;51:533–8.
of China (U2030116), and the Reform Program Chongqing Municipal [20] Hu L, Huang XR. Peculiar electronic, strong in-plane and out-of-plane second
Education Commission (223310). harmonic generation and piezoelectric properties of atom-thick α-M2X3 (M = Ga,

254
Z. Long et al. International Journal of Hydrogen Energy 62 (2024) 250–255

In; X = S, Se): role of spontaneous electric dipole orientations. RSC Adv 2017;7: [40] Qiao M, Liu J, Wang Y, Li Y, Chen Z. PdSeO3 monolayer: promising inorganic 2D
55034–43. photocatalyst for direct overall water splitting without using sacrificial reagents
[21] Li W, Yan X, Zou X, Liu L, Wang S, Wei Y, et al. α-In2O3 monolayer: a promising and cocatalysts. J Am Chem Soc 2018;140:12256–62.
material as field-effect phototransistor and out-of-plane piezoelectric device. Appl [41] Albrecht S, Reining L, Del Sole R, Onida G. Ab initio calculation of excitonic effects
Surf Sci 2023;614:156198. in the optical spectra of semiconductors. Phys Rev Lett 1998;80:4510.
[22] Du H, Liu S, Li G, Li L, Liu X, Liu B. Versatile GaInO3-sheet with strain-tunable [42] Benedict LX, Shirley EL, Bohn RB. Optical absorption of insulators and the electron-
electronic structure, excellent mechanical flexibility, and an ideal gap for hole interaction: an ab initio calculation. Phys Rev Lett 1998;80:4514.
photovoltaics. Chin Phys B 2019;28:016105. [43] Ju L, Bie M, Tang X, Shang J, Kou L. Janus WSSe monolayer: an excellent
[23] Hu L, Sun Y-F, Cheng J, Qin X, Yang X-Y, Wu S, et al. Single-layer GaInO3: photocatalyst for overall water splitting. ACS Appl Mater Interfaces 2020;12:
promising material for optoelectronic and out-of-plane piezoelectric devices. 29335–43.
Results Phys 2023;52:106847. [44] Liao Y, Zhang Z, Gao Z, Qian Q, Hua M. Tunable properties of novel Ga2O3
[24] Chen W, Yin H, Jiang S, Liu S, Liu C, Wang B, et al. Anomalous layer-dependent monolayer for electronic and optoelectronic applications. ACS Appl Mater
electronic and piezoelectric properties of 2D GaInS3 nanosheets. Appl Phys Lett Interfaces 2020;12:30659–69.
2021;118:213103. [45] Bandurin DA, Tyurnina AV, Yu GL, Mishchenko A, Zolyomi V, Morozov SV, et al.
[25] Vu TV, Hieu NN, Lavrentyev AA, Khyzhun OY, Lanh CV, Kartamyshev AI, et al. High electron mobility, quantum Hall effect and anomalous optical response in
Novel Janus GaInX3 (X = S, Se, Te) single-layers: first-principles prediction on atomically thin InSe. Nat Nanotechnol 2017;12:223–7.
structural, electronic, and transport properties. RSC Adv 2022;12:7973–9. [46] Shi LB, Cao S, Yang M, You Q, Zhang KC, Bao Y, et al. Theoretical prediction of
[26] Zhang X, Chen A, Zhang Z, Jiao M, Zhou Z. Rational design of C2N-based type-II intrinsic electron mobility of monolayer InSe: first-principles calculation. J Phys
heterojunctions for overall photocatalytic water splitting. Nanoscale Adv 2019;1: Condens Matter : Inst Phys J. 2019;32:065306.
154–61. [47] Sen R, Jatkar K. Modulation of electronic and transport properties of bilayer
[27] Chen YQ, Zhang HH, Wen B, Li XB, Wei XL, Yin WJ, et al. The role of permanent heterostructures: InSe/MoS2 and InSe/h-BN as the prototype. Phys Rev B 2020;
and induced electrostatic dipole moments for Schottky barriers in janus MXY/ 101:235425.
graphene heterostructures: a first-principles study. Dalton Trans 2022;51:9905–14. [48] Cai Y, Zhang G. Polarity-reversed robust carrier mobility in monolayer MoS2
[28] Yin W-J, Wen B, Nie G-Z, Wei X-L, Liu L-M. Tunable dipole and carrier mobility for nanoribbons. J Am Chem Soc 2014;136:6269–75.
a few layer janus MoSSe structure. J Mater Chem C 2018;6:1693–700. [49] Mir SH, Yadav VK. Recent advances in the carrier mobility of two-dimensional
[29] Ma X, Wu X, Wang H, Wang Y. A janus MoSSe monolayer: a potential wide solar- materials: a theoretical perspective. ACS Omega 2020;5:14203–11.
spectrum water-splitting photocatalyst with a low carrier recombination rate. [50] Gao Y, Jiang X, Qiu Z, Zhao J. Photoexcitation induced magnetic phase transition
J Mater Chem A 2018;6:2295–301. and spin dynamics in antiferromagnetic MnPS3 monolayer. npj Comput Mater
[30] Kumar R, Das D, Singh AK. C2N/WS2 van der waals type-II heterostructure as a 2023;9:107.
promising water splitting photocatalyst. J Catal 2018;359:143–50. [51] Chu W, Zheng Q, Prezhdo OV, Zhao J, Saidi WA. Low-frequency lattice phonons in
[31] Wang J, Guan Z, Huang J, Li Q, Yang J. Enhanced photocatalytic mechanism for halide perovskites explain high defect tolerance toward electron-hole
the hybrid g-C3N4/MoS2 nanocomposite. J Mater Chem A 2014;2:7960–6. recombination. Sci Adv 2020;6:eaaw7453.
[32] Hu L, Yi W, Rao T, Tang J, Hu C, Yin H, et al. Two-dimensional type-II g-C3N4/ [52] Zhang L, Chu W, Zheng Q, Benderskii AV, Prezhdo OV, Zhao J. Suppression of
SiP–GaS heterojunctions as water splitting photocatalysts: first-principles electron-hole recombination by intrinsic defects in 2D monoelemental material.
predictions. Phys Chem Chem Phys 2020;22:15649–57. J Phys Chem Lett 2019;10:6151–8.
[33] Yan XJ, Li WY, Nan GT, Zou X, Liu LL, Wang A, et al. Comparative study of janus [53] Long R, Fang W, Akimov AV. Nonradiative electron-hole recombination rate is
B2XY (X, Y = S, Se, Te) and F-BNBN-H monolayers for water splitting: revealing the greatly reduced by defects in monolayer black phosphorus: ab initio time domain
positive and negative roles of the intrinsic dipole. Phys Chem Chem Phys 2022;24: study. J Phys Chem Lett 2016;7:653–9.
20980–7. [54] Long R, Fang W, Prezhdo OV. Moderate humidity delays electron-hole
[34] Nan G, Zhang W, Yan X, Qin X, Wu S, Tang R, et al. ZnGeSe2 monolayer: water- recombination inhybrid organic-inorganic perovskites: time-domain ab initio
splitting photocatalyst with ultrahigh solar conversion efficiency. Phys Chem Chem simulations rationalize experiments. J Phys Chem Lett 2016;7:3215–22.
Phys 2023;25:24594–602. [55] Wang S, Luo Q, Fang WH, Long R. Interfacial engineering determines band
[35] Deng X, Zhang W, Zhou X, Wang Z, Tang J, Hu L, et al. InSe based Janus VI-III-IV-V alignment and steers charge separation and recombination at an inorganic
monolayers as water-splitting photocatalysts: role of vacuum level difference. Int J perovskite quantum Dot/WS2 junction: a time domain ab initio study. J Phys Chem
Hydrogen Energy 2021;46:35271–9. Lett 2019;10:1234–41.
[36] Huang X, Shu X, Li J, Cui Z, Cao S, Chen W, et al. DFT study on type-II [56] Li L, Long R, Bertolini T, Prezhdo OV. Sulfur adatom and vacancy accelerate charge
photocatalyst for overall water splitting: g-GaN/C2N van der waals heterostructure. recombination in MoS2 but by different mechanisms: time-domain ab initio
Int J Hydrogen Energy 2023;48:12364–73. analysis. Nano Lett 2017;17:7962–7.
[37] Zhou Y, Zhou L, He J, Frauenheim T. Inartificial two-dimensional Ge4Se9 janus [57] Jin H, Wang T, Gong ZR, Long C, Dai Y. Prediction of an extremely long exciton
structures with appropriate direct band gaps and intrinsic polarization boosted lifetime in a Janus-MoSTe monolayer. Nanoscale 2018;10:19310–5.
charge separation for photocatalytic water splitting. J Phys Chem Lett 2020;11: [58] Dong H, Zhao J, Yang H, Zheng Y. Tunable long-lived exciton lifetime in single-
3095–102. layer two dimensional LiAlTe2. Phys Rev Mater 2022;6:104001.
[38] Toroker MC, Kanan DK, Alidoust N, Isseroff LY, Liao P, Carter EA. First principles [59] Zhou X, Wang P, Li M, Wu M, Jin B, Luo J, et al. Synergistic effect of phosphorus
scheme to evaluate band edge positions in potential transition metal oxide doping and MoS2 co-catalysts on g-C3N4 photocatalysts for enhanced solar water
photocatalysts and photoelectrodes. Phys Chem Chem Phys 2011;13:16644–54. splitting. J Mater Sci Technol 2023;158:171–9.
[39] Hu L, Wei D. Janus group-III chalcogenide monolayers and derivative type-II [60] Zhang W, Chai C, Fan Q, Yang Y, Sun M, Palummo M, et al. Two-dimensional
heterojunctions as water-splitting photocatalysts with strong visible-light borocarbonitrides for photocatalysis and photovoltaics. J Mater Chem C 2023;11:
absorbance. J Phys Chem C 2018;122:27795–802. 3875–84.

255