First-principles studies on the electronic and photocatalytic water splitting properties of surface functionalized Y₂C-based MXenes

Abstract

Recently, MXenes, an emerging family of two-dimensional (2D) materials, have attracted increasing interest for photocatalytic water splitting due to their various excellent physical and chemical properties, such as large specific surface area, good hydrophilicity, and remarkable light absorption ability. However, the photocatalysts of MXenes with symmetric structures are limited by rapid recombination of photo-generated carriers and the prerequisite of a large band gap no less than 1.23 eV. Differently, Janus MXenes with different surface functional groups facilitate the separation of photo-generated electrons and holes with the help of the intrinsic electric field. And, at the same time, there is no prerequisite for the band gap of Janus MXene photocatalysts as long as they possess appropriate band edge positions. Here, we explored the structural, electronic and photocatalytic water splitting properties of symmetric Y₂CT₂ and Janus Y₂CTT′ MXenes (T, T′ = H, F, Cl, OH) using the density functional theory (DFT) method. Our calculations show that all the investigated Y₂CT₂ are not suitable photocatalysts for photocatalytic water splitting at all pH values (pH = 0, 7, and 14). In contrast, all the investigated Janus Y₂CTT′ MXenes are good water splitting photocatalysts with high optical absorption coefficients and remarkable solar-to-hydrogen (STH) efficiencies larger than 18% at pH = 14. Moreover, the STH efficiencies are larger than 18% even at all investigated pH values for Y₂CHCl (18.5–22.6%), Y₂ CFCl (∼18.7%), and Y₂ C(OH)Cl (∼19.4%). Based on the first-principles calculations, we here for the first time propose an easy strategy to design Janus MXene photocatalyst candidates with possible high STH efficiency according to the electronic properties of their symmetric counterparts. Our study is helpful for the future design of Janus MXenes and more generally Janus 2D photocatalysts for water splitting with high STH efficiency.

---

1 Introduction

With the rapid development of human society, we are facing increasingly serious energy and environmental problems. Researchers are devoted to finding renewable clean energy to replace traditional fossil fuels. Due to the high energy density and harmless combustion products (H₂O), hydrogen has been considered as the ultimate solution to current energy and environmental crises. However, producing hydrogen is not an easy task, and many researchers have developed many kinds of hydrogen production methods such as methanol thermochemical conversion and electrolysis of water.¹ Among all these hydrogen production methods, photocatalytic water splitting is definitely an ideal way to obtain hydrogen economically, environmentally friendly and safely. In general, a good photocatalyst for water splitting should meet the following three conditions. (1) Photo-generated electron and hole carriers should be spatially separated efficiently. (2) It should have appropriate band alignments with the redox potentials of H⁺/H₂ and H₂O/O₂, namely, the potential of its conduction band minimum (CBM) should be higher (more negative) than the hydrogen reduction potential (0 V vs. standard hydrogen electrode (SHE)), while the potential of its valence band maximum (VBM) should be lower (more positive) than the water oxide potential (1.23 V vs. the SHE). This means that the band gap should be larger than 1.23 eV for structurally symmetric photocatalysts. (3) It should have good sunlight absorption characteristics to fully utilize the solar energy. Since the discovery of TiO₂ photocatalysts, many other semiconductors have been studied for photocatalytic water splitting.^2–4 But these photocatalysts have disadvantages of inefficient carrier separation and low sunlight absorption.⁵ Therefore, searching efficient photocatalysts for photocatalytic water splitting has long been a vital task in producing hydrogen.

In 2004, graphene was successfully prepared in experiments, which opened a new era of two-dimensional (2D) materials.⁶ Since then, plenty of 2D materials have been newly discovered and prepared;^7–12 among which 2D transition metal carbides, nitrides and carbonitrides (MXene) have attracted widespread attention because of their large specific surface area, variety of compositions and structures and other remarkable properties.^13,14 MXenes can be formulated as M_n+1X_nT_x, in which M denotes transition metals (e.g. Sc, Ti, V, Cr, Mn, Y, Zr, Nb, Mo, Hf, Ta, W, or their combination), X denotes C or/and N, and T denotes surface functional groups (such as –H, –F, –O, –OH, and –Cl).^15,16 So far, hundreds of MXenes have been experimentally synthesized or theoretically predicted with diverse applications, such as energy storage devices,^17,18 gas sensors,¹⁹ supercapacitors,²⁰ electromagnetic interference shielding^21,22 and photocatalysts.^23–26

In 2019, the stacked sheets of Y₂CF₂ were successfully synthesized by Druffel et al.²⁷ In the next year, Maeda et al. succeeded in reducing the protons into hydrogen using the stacked sheets of Y₂CF₂.²⁸ According to the two studies mentioned above, although Y₂CF₂ has a band gap of about 1.90 eV and its CBM potential is more negative than the hydrogen reduction potential, the VBM potential of Y₂CF₂ is more negative than the water oxidation potential, which forbids the process of water oxidation (H₂O to O₂). This means the electrons excited to conduction bands have sufficient energy to reduce protons H⁺ to H₂ but the energy of holes that remained in valence bands is not low enough for them to accept electrons from H₂O to produce O₂. Therefore, other reductants must be added to enable the redox reaction process to proceed, which will bring unnecessary cost and other products besides H₂ and O₂. On the other hand, photo-excited carriers in symmetric Y₂CF₂ can easily recombine and the relatively large band gap also limits the utilization of sunlight, resulting in low photocatalytic efficiency of water splitting.

In general, assuming that both light absorbance and quantum efficiency can reach 100% and considering that the appropriate band gap is about 2.0 eV for conventional water splitting photocatalysts, then the theoretical solar-to-hydrogen (STH) efficiency (a parameter often used to evaluate the conversion rate from solar energy to hydrogen) for conventional water splitting photocatalysts is at most ca. 18%.²⁹ In fact, the experimentally reported STH efficiencies by far rarely exceed a criterion of 10% for practical applications.^30–34 Therefore, exploring photocatalysts for photocatalytic water splitting with high STH efficiency is of great importance. For this aim, Yang et al. suggested asymmetric 2D materials as a new class of photocatalysts for photocatalytic water splitting as long as the potential of the VBM contributed from one surface is lower than the oxidation potential of H₂O to O₂ and the potential of the CBM contributed from the other surface is higher than the reduction potential of H⁺ to H₂.³⁵ This proposed mechanism of photocatalytic water splitting is very similar to that in Z-scheme heterostructures which has been verified experimentally.^36–38 They have revealed that the intrinsic electric field (IEF) in structurally asymmetric 2D materials facilitates the carrier separation; meanwhile, it is no longer a prerequisite for water-splitting photocatalysts to have a band gap larger than 1.23 eV due to different band edge positions for two surfaces. Recently, lots of 2D Janus materials have been predicted to possess high photocatalytic efficiency for water splitting.^29,39–43 For example, Fu et al. theoretically investigated the characteristics of 2D M₂X₃ (M = Al, Ga, In; X = S, Se, Te) for photocatalytic water splitting.²⁹ All the studied systems are promising photocatalysts for photocatalytic water splitting and moreover, the calculated STH efficiencies of Al₂Te₃, Ga₂Se₃, Ga₂Te₃ (with 3% strain), In₂Se₃, and In₂Te₃ are lager than 18%. Similarly, Janus MXenes with different functional groups on two surfaces are also expected to be good candidates for water-splitting photocatalysts.^44,45 For example, Zhang et al. theoretically reported that Sc₂C(OH)Cl, Sc₂CFCl, and Sc₂CHCl are promising photocatalysts for photocatalytic water splitting.⁴⁴

Though Janus MXenes have the advantages of efficient separation of light-generated electrons and hole carriers and no prerequisite for the band gaps to be good photocatalysts for photocatalytic water splitting, they should still have appropriate band alignments with the redox potentials of water and good sunlight absorption characteristics. Therefore, it is still not a trivial task to search such Janus MXenes satisfying the above two prerequisites. In order to search more Janus MXenes as excellent water splitting photocatalysts with high STH and more importantly to explore the underlying design strategy for Janus MXene photocatalysts for photocatalytic water splitting, here we studied the structural, electronic and photocatalytic properties of Y₂C-based MXenes Y₂CT₂ and Y₂CTT′ (T, T′ = H, F, Cl, OH) by using the density functional theory (DFT) method. Our first-principles calculations reveal that though all the investigated Y₂CT₂ are not suitable photocatalysts for photocatalytic water splitting at all investigated pH values (pH = 0, 7, and 14), all the Janus Y₂CTT′ MXenes are good water splitting photocatalysts at pH = 14 with STH efficiencies of larger than 18%. Furthermore, the STH efficiencies of Y₂CHCl, Y₂CFCl, and Y₂C(OH)Cl are larger than 18% at all pH values. Finally, we propose an easy strategy to design Janus MXene photocatalysts with possible high STH efficiency by simply inspecting the band edge positions of their symmetric counterparts.

2 Computational details

All the calculations in this work were based on density functional theory (DFT) implemented in the Vienna ab initio simulation package (VASP).^46,47 The projector-augmented wave (PAW) pseudopotentials were employed to describe the interactions between valence electrons and cores.⁴⁸ The valence electron density was expanded in the plane wave basis set with an energy cutoff of 500 eV. As sketched by the red dashed lines in Fig. 1, the primitive cell was used to perform the calculations. The periodic boundary condition was used within the plane of 2D MXenes. For the direction perpendicular to the 2D MXene plane, a lattice parameter of 30 Å was used (i.e. a vacuum of about 23–25 Å). And, at the same time, a dipole correction was applied during the calculations for all Janus MXenes. For the structural relaxation, the generalized gradient approximation (GGA) of Perdew–Burke–Ernzerhof (PBE) was chosen as the exchange–correlation functional,⁴⁹ and a criterion of 0.01 eV Å⁻¹ for the residual force on each atom was set. On the other hand, the Heyd–Scuseria–Ernzerhof (HSE06) hybrid functional was used to calculate the electronic and optical properties. The k-point meshes of 11 × 11 × 1 and 15 × 15 × 1 generated using the Monkhorst–Pack scheme were used for structural optimizations and electronic self-consistent calculations, respectively. For the dynamic stability calculations, 3 × 3 × 1 supercells were constructed based on the optimized primitive cells and the force constants were calculated by density functional perturbation theory (DFPT) using the PHONOPY code,^50,51 for which a k-point sampling of 5 × 5 × 1 was used instead.

Fig. 1 Top and side views of structural models for (a) Y₂CT₂ MXenes and (b) Y₂CTT′ Janus MXenes. The primitive cells are indicated by the red dashed lines.

3 Results and discussion

3.1 Geometric structures

According to previous studies,^52–55 the surface functional groups prefer to locate at the hollow sites of the Y₂C MXene. There are two kinds of hollow sites in the Y₂C MXene. One is above the hollow of three neighboring C atoms of the middle C atomic layer, aligned with a Y atom of either Y atomic layers (denoted as hollow site A). The other is above the hollow of three neighboring Y atoms of either Y atomic layers, aligned with a C atom of the middle C atomic layer (denoted as hollow site B). When the two surfaces of the Y₂ C MXene are functionalized with the same functional group (i.e., Y₂CT₂ MXenes), as seen from Fig. 1(a), there are three possible structure models for this case. For Model-1, the functional groups at each surface locate at the A hollow sites. For Model-2, the functional groups locate at the B hollow sites. For Model-3, the functional groups locate at the A hollow sites on one surface but at the B hollow sites on the other surface. Obviously, Model-1 and Model-2 are structurally symmetric while Model-3 is structurally asymmetric. Therefore, when it turns to the cases of Janus functionalized MXenes, two different structural models will derive from the above asymmetric Model-3 by exchanging the functional groups on the two surfaces, which are respectively denoted as Model-3 and Model-4 as shown in Fig. 1(b).

To figure out which structural model is energy preferable, we calculated the total energies of considered structural models for each Y₂CT₂/Y₂CTT′ MXene, which are listed in Table 1. It can be clearly seen from Table 1 that for both Y₂CT₂ and Y₂CTT′ MXenes, Model-1 is the most stable structure with the lowest energy, which is in good agreement with previously reported Sc₂C and Ti₂C MXenes.^44,56 To further confirm the dynamic stability of Model-1, phonon dispersion of all investigated MXenes with the Model-1 structure has been inspected by using the density functional perturbation theory (DFPT) implemented in the PHONOPY code. As seen from Fig. 2 for Y₂CH₂ and Y₂CFCl (refer to Fig. S1 of the ESI† for the other 8 systems), there are only few imaginary frequencies near the Γ point, which are of acoustic nature^8,57 and expected to be eliminated by using a larger supercell in the phonon dispersion analysis (but accompanying a dramatic increase in computation).⁵⁸ At this point, it can be concluded that all the investigated systems with the Model-1 structure are also dynamically stable. Therefore, we used the optimized Model-1 structure for the subsequent analyses. The optimized lattice parameters for all investigated systems are given in Table 2. Besides, we also optimized the structure of stacked sheets of Y₂CF₂, which was experimentally synthesized recently.^27,28 The calculated lattice constant of stacked sheets of Y₂CF₂ is 3.63 Å, which agrees well with an experimental value of 3.66 Å.^27,28 This suggests that the DFT-GGA method is good enough for geometry optimization. From Table 2, we can see that the lattice constant of the Y₂CF₂ monolayer is 3.57 Å, which is very close to the value of stacked sheets. Also, it can be seen that the lattice constant is closely related to the surface functional groups. More specifically, as shown in Table 2, MXenes with the –Cl functional group have relative larger lattice constants. In contrast, the lattice constants for those without the –Cl functional group are relatively smaller and they share a common value of around 3.58 ± 0.1 Å. This can be attributed to the fact that the size of –H, –F and –(OH) groups are much smaller than that of the –Cl group.

Table 1 Total energies (in eV) for Y₂CT₂ and Y₂CTT′ MXenes with different structural models. The energy for the Model-1 structure is chosen to be zero for each system

Systems	Model-1	Model-2	Model-3	Model-4
Y2CH2	0	0.627	0.294	—
Y2CF2	0	1.045	0.507	—
Y2CCl2	0	0.627	0.308	—
Y2C(OH)2	0	0.607	0.305	—
Y2CHF	0	0.894	0.495	0.309
Y2CHCl	0	0.651	0.256	0.366
Y2C(OH)H	0	0.637	0.294	0.322
Y2CFCl	0	0.844	0.551	0.261
Y2C(OH)F	0	0.838	0.301	0.559
Y2C(OH)Cl	0	0.623	0.335	0.267

---

Fig. 2 Phonon dispersion spectra of (a) Y₂CH₂ and (b) Y₂CFCl.

Table 2 The lattice constant a (Å), band gap E_gap (eV), band gap type, work function Φ (eV), work function difference ΔΦ (eV), and intensity of intrinsic electric field E_IEF (V m⁻¹) of investigated MXenes

Systems	a (Å)	E gap (eV)	Direct/indirect gap	Φ (eV)		ΔΦ (eV)	E IEF (V Å^−1)
Y2CH2	3.59	1.663	I	4.478		0	0
Y2CF2	3.57	1.952	I	4.615		0	0
Y2CCl2	3.71	1.630	I	6.223		0	0
Y2C(OH)2	3.58	0.684	D	1.427		0	0
Y2CHF	3.58	1.793	I	4.780(H)	4.710(F)	0.070	0.014
Y2CHCl	3.65	1.578	I	4.636(H)	6.186(Cl)	1.550	0.281
Y2C(OH)H	3.58	1.098	I	1.809(OH)	4.729(H)	2.920	0.481
Y2CFCl	3.64	1.747	I	4.452(F)	6.292(Cl)	1.840	0.327
Y2C(OH)F	3.58	1.127	D	2.122(OH)	4.402(F)	2.270	0.367
Y2C(OH)Cl	3.64	0.896	D	1.689(OH)	6.099(Cl)	4.410	0.654

---

3.2 Electronic properties

It has been known that the electronic density of states (DOS) and band gaps of MXenes can be modified by the surface functional groups so that their electronic properties can be tuned in a large range from dielectrics to semiconductors, semimetals, and metals.¹⁴ Here, we examined the electronic properties of Y₂CT₂ and Y₂CTT′ MXenes. From Table 2, one can see that all the investigated Y₂C-based MXenes are semiconductors. It should be mentioned that for stacked sheets of Y₂CF₂ our calculations from the HSE06 functional give a more reliable band gap (1.987 eV) than the pure GGA functional (1.138 eV) in comparison to an experimental observed value of 1.90 eV.²⁸ This is why we used the HSE06 functional to calculate the electronic and optical properties of Y₂C-based MXenes in our study. The band gap of Y₂C-based MXenes ranges from 0.684 to 1.952 eV, which closely depends on the surface functional groups. Roughly speaking, the –(OH) functional group tends to lower the band gap while the –F functional group is in favor of increasing the band gap. By this, Y₂CF₂ has the biggest band gap of 1.952 eV, while Y₂ C(OH)₂ has the smallest one of 0.684 eV. This can also be confirmed by the band structures shown in Fig. 3 (also refer to Fig. S2 of the ESI†) as well as the density of states (DOS) in Fig. S3 of the ESI.† The conduction bands are mainly donated by 4d orbitals of Y atoms while the valence bands are dominated by Y 4d and C 2p orbitals for the investigated Y₂ C-based MXenes. For MXenes with the –(OH) functional group, the CBM contributed from 2p orbitals of the O atom in the –(OH) group at the Γ point can be seen, which significantly reduces the band gaps and hence Y₂C(OH)₂, Y₂C(OH)Cl, and Y₂C(OH)F are changed from indirect band semiconductors to direct ones.

Fig. 3 Band structures of (a) Y₂CF₂, (b) Y₂C(OH)₂, (c) Y₂CHF, and (d) Y₂ C(OH)Cl. The dashed lines indicate the Fermi level E_F, which is put at the middle of the band gap and is set to zero.

As one of the basic physical characteristics, the work function Φ also plays an important role in the electronic properties, which is closely related to the surface functional groups for MXenes.⁵⁶ Therefore, we calculated Φ for each investigated MXene, which is defined as the energy difference between E_F and vacuum level (electrostatic potential far away from the surface). As seen in Table 2, the calculated Φ values for Y₂CH₂, Y₂CF₂, Y₂CCl₂, and Y₂C(OH)₂ are 4.478 eV, 4.615 eV, 6.223 eV, and 1.427 eV, respectively. It can be seen that surfaces with the –Cl functional group have the highest Φ value. In contrast, surfaces with the –(OH) functional group have the lowest Φ value. The calculated very low work function of Y₂C(OH)₂ agrees with previous theoretical and experimental reports for other OH-terminated MXenes.^56,59,60 This is because the change in work functions of functionalized MXenes has a positive linear correlation with the change in surface dipole moments, while the intrinsic dipole moment of the polar –(OH) group makes a large and negative contribution to the total surface dipole moments.⁵⁶ When it turns to Y₂CTT′ Janus MXenes, each individual surface has its own Φ. As shown in Fig. 4 (also see Table 2), there is a work function difference ΔΦ for each Y₂CTT′ Janus MXene, which in turn contributes to an IEF perpendicular to the plane of the Janus MXenes. The intensity of IEF (E_IEF) relates to ΔΦ by the following equation:³⁵

Fig. 4 The calculated electrostatic potential energy differences of Janus MXenes: (a) Y₂CHF, (b) Y₂CHCl, (c) Y₂C(OH)H, (d) Y₂CFCl, (e) Y₂C(OH)F, and (f) Y₂(OH)Cl. The dotted and dashed lines represent the vacuum level of each sides and the Fermi level which is set to zero, respectively.

Therefore, it is expected that Janus MXenes with the –(OH) group on one side and the –Cl group on the other side (i.e., Y₂C(OH)Cl) will have the largest ΔΦ and hence the strongest E_IEF, which can be easily confirmed from Table 2 and Fig. 4. The IEF in Janus MXenes facilitates the spatial separation of photo-excited electrons and holes onto different functional groups of two surfaces, which can be verified in Fig. S4 in the ESI.† The spatial separation of photo-excited electrons and holes is in favor of the photocatalytic water splitting on the surfaces.

3.3 Photocatalytic properties

For photocatalytic water splitting, the key processes are reduction of H⁺ to H₂ and oxidation of H₂O to O₂. For these two processes to proceed, it needs electrons with high energy and holes with low energy. Because electrons with high energy show strong reducibility and can easily reduce H⁺ to H₂, while holes with low energy have strong oxidability and can readily oxidize H₂ O to O₂. The reduction potential of H⁺ to H₂ is 0 V vs. the SHE and oxidation potential of H₂ O to O₂ is 1.23 V vs. SHE, which align with the band energy values of −4.44 and −5.67 eV vs. absolute vacuum scale (AVS), respectively. Therefore, a photocatalyst candidate for water splitting should simultaneously have a CBM of higher than −4.44 eV and a VBM of lower than −5.67 eV. Here, the absolute band edge positions vs. AVS were calculated to study the redox ability of Y₂CT₂ and Janus Y₂CTT′ MXenes.

3.3.1 Photocatalytic properties of symmetric Y₂CT₂ MXenes. We first discuss the photocatalytic properties of symmetric Y₂CT₂ MXenes (i.e., Y₂CH₂, Y₂CF₂, Y₂CCl₂, and Y₂C(OH)₂). As seen in Table 2, their band gaps are 1.663, 1.952, 1.630, and 0.684 eV, respectively. Obviously, as a structurally symmetric system, Y₂C(OH)₂ has a band gap of smaller than 1.23 eV, which indicates that it is not suitable for photocatalytic water splitting. In fact, as shown in Fig. 5, at pH = 0, though the CBM potential of Y₂C(OH)₂ is more negative than the hydrogen reduction potential, its VBM potential is also more negative than the water oxidation potential, forbidding the process of H₂O to O₂. It can be easily seen from Fig. 5 that the cases of Y₂CH₂ and Y₂CF₂ are similar to that of Y₂C(OH)₂ even though both of them have large enough band gaps. That is, at pH = 0, both Y₂CH₂ and Y₂CF₂ can only reduce H⁺ to H₂ but cannot oxidize H₂O to O₂ as well. The theoretically predicted performance of Y₂CF₂ agrees with the experimental observation for stacked sheets of Y₂CF₂.²⁸ In contrast, Y₂CCl₂ can only oxidize H₂O to O₂ but cannot reduce H⁺ to H₂ at pH = 0. It should be mentioned that our calculated VBM and CBM for Y₂CF₂ are very close to those observed in the experiment for stacked sheets of Y₂CF₂ (theoretically −1.65 and 0.3 V while experimentally −1.5 and 0.4 V for potentials of the CBM and VBM vs. SHE, respectively), suggesting that band energies calculated from the HSE06 functional are reliable.

Fig. 5 The calculated band edge positions of (a) Y₂CH₂, (b) Y₂CF₂, (c) Y₂CCl₂, (d) Y₂C(OH)₂, (e) Y₂CHF, (f) Y₂CHCl, (g) Y₂C(OH)H, (h) Y₂CFCl, (i) Y₂C(OH)F, and (j) Y₂C(OH)Cl vs. AVS (left) and SHE (right). The red, purple, and blue dash dot (dash) lines represent the reduction (oxidation) potential of water at pH = 0, 7, and 14, respectively.

We further checked the photocatalytic properties for other pH values (here taking pH = 7 and pH = 14 as examples). As is known, the redox potentials (E_P) of water change with the pH value, which follows the Nernst equation.⁶¹

E_P(pH) = E_{P(pH = 0)} − 0.0591 × pH (2)

As shown in Fig. 5, though pH changes the redox potentials (E_P) of water, the relative relationships of their alignments with the VBMs and CBMs of Y₂CT₂ MXenes are not fundamentally altered. Therefore, the four structurally symmetric Y₂CT₂ MXenes still cannot act as photocatalysts for water splitting at pH = 7 and 14.

Though the structurally symmetric Y₂CT₂ MXenes are not suitable photocatalysts for water splitting at all pH values, for Y₂CH₂ (noting that the VBM of Y₂CH₂ is merely 0.02 eV lower than the water oxidation potential at pH = 14, which is believed not large enough to efficiently drive the water oxidation), Y₂CF₂, and Y₂C(OH)₂, their electrons in conduction bands have relatively high energies, showing very large over-potentials, especially Y₂C(OH)₂. This will make the process of hydrogen evolution much faster. On the other hand, Y₂CCl₂ has a very positive valence band potential, which makes the process of oxygen evolution much faster instead. Therefore, with the addition of suitable reactants, Y₂CT₂ MXenes can be utilized to individually produce hydrogen or oxygen.

3.3.2 Photocatalytic properties of Janus Y₂CTT′ MXenes. The above discussion shows that for a structural symmetric system a band gap of larger than 1.23 eV does not guarantee a photocatalyst for water splitting. Instead, it simultaneously requires the potential of the VBM to be lower than the oxidation potential of H₂O to O₂ and the potential of the CBM to be higher than the reduction potential of H⁺ to H₂. Even so, for symmetric systems, the conduction and valence bands are distributed symmetrically on two sides, which indicates that the electrons and holes can easily recombine at the surfaces. This also severely limits the applications of symmetric systems in photocatalytic water splitting. Compared with symmetric Y₂CT₂, Janus Y₂CTT′ can act as photocatalysts for water splitting as long as the potential of the VBM contributed from one surface is lower than the oxidation potential of H₂ O to O₂ and the potential of the CBM contributed from the other surface is higher than the reduction potential of H⁺ to H₂. On the other hand, the states of the CBM have a significant distribution on one surface, while the states of VBM have a significant distribution on the other surface (refer to Fig. S4 in the ESI†). At the same time, the direction of IEF is always from the surface dominated by the CBM to the other surface dominated by the VBM, which facilitates the photo-excited electrons and holes staying in the higher-energy conduction band states and lower-energy valence band states, respectively.

A Janus MXene can be simply considered as a combination of its two symmetric counterparts. And if the potential of the VBM for one counterpart is lower than the oxidation potential of H₂O to O₂ and the potential of the CBM for the other counterpart is higher than the reduction potential of H⁺ to H₂, then this Janus MXene is more likely to have appropriate alignments of its band edge positions with the redox potentials of water. According to this simple strategy for finding photocatalyst candidates, we predicted that Y₂CHCl, Y₂CFCl, and Y₂C(OH)Cl are promising photocatalyst candidates for water splitting at pH = 0, which can be easily verified in Fig. 5. Even though the band gap of Y₂C(OH)Cl is only 0.896 eV, the redox over-potentials are still large demonstrating a fast rate of the photocatalytic water splitting reaction.

Since Y₂CH₂, Y₂CF₂, and Y₂C(OH)₂ belong to the same kind symmetric MXene with only the potential of the CBM higher than the reduction potential of H⁺ to H₂ at pH = 0, the hybridizations of them do not provide effective photocatalyst candidates for water splitting. However, according to the Nernst equation (eqn (2)), the increase of the pH value can move the water redox potentials towards a negative direction, which makes their hybrid Janus MXenes possible photocatalyst candidates of water splitting. For example, at pH = 7, except for Y₂CHCl, Y₂CFCl, and Y₂C(OH)Cl, two additional Janus MXenes Y₂CHF and Y₂C(OH)H can act as photocatalysts for water splitting. When the pH value increases to 14, all the investigated Janus MXenes become photocatalysts for water splitting.

To explore the energy conversion, we calculated the upper limits of STH efficiency for all investigated Janus MXenes at pH = 0, 7 and 14, which are listed in Table 3 and Tables S1, S2 of the ESI,† respectively (refer to the calculation method of STH efficiency in the ESI†). We can see from Table 3 that the corrected STH efficiencies at pH = 0 for Y₂CHCl, Y₂CFCl, and Y₂C(OH)Cl are 22.6%, 18.7%, and 19.4%, respectively, all of which are lager than a theoretical limit of about 18% for structurally symmetric photocatalysts. It is worth mentioning that the corrected STH efficiencies of Y₂CHCl, Y₂CFCl, and Y₂C(OH)Cl remain nearly unchanged and are always larger than 18% when the pH value is changed from 0 to 14 due to their large over-potentials for hydrogen and oxygen evolution reactions, illustrating their stable photocatalytic abilities. The corrected STH efficiencies of Y₂CHF and Y₂C(OH)H at pH = 7 are only 10.1% and 12.2% (refer to Table S1 of the ESI†), respectively, due to their small over-potentials of oxygen evolution. However, when the pH value increases to 14, the corrected STH efficiencies for all the investigated Janus MXenes are larger than 18% (refer to Table S2 of the ESI†).

Table 3 The calculated over-potential at pH = 0 for the hydrogen evolution reaction χ(H₂) and the oxygen evolution reaction χ(O₂) in eV, the energy conversion efficiency of light absorption η_abs, carrier utilization η_cu, STH η_STH and corrected STH η′_STH in %

Systems	χ(H2)	χ(O2)	η abs	η cu	η STH	η′STH
Y2CHF	1.02	−0.39	—	—	—	—
Y2CHCl	0.83	1.06	58.6	53.9	31.6	22.6
Y2C(OH)H	3.13	−0.34	—	—	—	—
Y2CFCl	1.15	1.2	48.7	53.2	25.9	18.7
Y2C(OH)F	2.88	−0.71	—	—	—	—
Y2C(OH)Cl	3.17	0.91	88.4	71.7	63.4	19.4

---

As stated before, a good photocatalyst should also have a good sunlight absorption characteristic so that the solar energy can be utilized at most. To further check the ability of solar energy utilization, the optical absorption coefficients α(ω) of all investigated Janus Y₂CTT′ MXenes are calculated and shown in Fig. 6 according to the following equation.⁶²

where ε₁ and ε₂ represent the real and imaginary parts of the dielectric function, respectively. It can be easily seen from Fig. 6 that all the investigated Janus Y₂CTT′ MXenes have high absorption coefficients at the wavelengths of high solar irradiance according to the Air Mass 1.5 Global (AM1.5G) standard solar spectrum.⁶³

Fig. 6 The calculated absorption rates of Janus Y₂CTT′, and the gray shadow background is the solar spectral irradiance as Air Mass 1.5G.

4 Conclusions

In summary, based on the first-principles calculations, the structural, electronic and photocatalytic properties of surface functionalized Y₂CT₂ and Janus Y₂CTT′ MXenes (T, T′ = H, F, Cl, OH) have been systematically studied. The lattice constant, band gap, work-function, and photocatalytic properties of Y₂C-based MXenes are closely related to the surface functional groups. The –Cl group increases the lattice constant of Y₂C-based MXenes in comparison with the –H, –F or –(OH) group. For photocatalytic water splitting, all the four structurally symmetric Y₂CT₂ MXenes are not suitable photocatalysts for water splitting at all pH values (pH = 0, 7, and 14). In contrast, all the investigated Janus Y₂CTT′ MXenes can act as good photocatalysts for water splitting since they have high absorption coefficients and meanwhile have significant STH efficiencies of larger than 18% at pH = 14. Furthermore, Y₂CHCl, Y₂CFCl, and Y₂CFCl are predicted to be good photocatalysts for water splitting with STH efficiencies larger than 18% at all investigated pH values. Finally, for the first time, we propose an easy strategy to design Janus MXene candidates for photocatalysts with possible high STH efficiency by preliminarily inspecting the band structures and work functions of their symmetric counterparts. This work is helpful for the future design of Janus MXenes and more generally Janus 2D photocatalysts for photocatalytic water splitting with high STH efficiency.

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

This work was supported by the financial support from the National Natural Science Foundation of China (Grant No. 22173052 and 21933002).

References

1. G. Garcia, E. Arriola, W.-H. Chen and M. D. De Luna, Energy, 2021, 217, 119384.
2. T. Simon, N. Bouchonville, M. J. Berr, A. Vaneski, A. Adrovic, D. Volbers, R. Wyrwich, M. Döblinger, A. S. Susha, A. L. Rogach, F. Jäckel, J. K. Stolarczyk and J. Feldmann, Nat. Mater., 2014, 13, 1013–1018.
3. X. Chen, L. Liu, P. Y. Yu and S. S. Mao, Science, 2011, 331, 746–750.
4. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
5. J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo and D. W. Bahnemann, Chem. Rev., 2014, 114, 9919–9986.
6. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666–669.
7. B. Lalmi, H. Oughaddou, H. Enriquez, A. Kara, S. Vizzini, B. Ealet and B. Aufray, Appl. Phys. Lett., 2010, 97, 223109.
8. S. Cahangirov, M. Topsakal, E. Aktürk, H. Sahin and S. Ciraci, Phys. Rev. Lett., 2009, 102, 236804.
9. M. E. Dávila, L. Xian, S. Cahangirov, A. Rubio and G. Le Lay, New J. Phys., 2014, 16, 095002.
10. H. Liu, A. T. Neal, Z. Zhu, Z. Luo, X. Xu, D. Tománek and P. D. Ye, ACS Nano, 2014, 8, 4033–4041.
11. L. Li, Y. Yu, G. J. Ye, Q. Ge, X. Ou, H. Wu, D. Feng, X. H. Chen and Y. Zhang, Nat. Nanotechnol., 2014, 9, 372–377.
12. M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi and M. W. Barsoum, Adv. Mater., 2011, 23, 4248–4253.
13. C. Tan, X. Cao, X.-J. Wu, Q. He, J. Yang, X. Zhang, J. Chen, W. Zhao, S. Han, G.-H. Nam, M. Sindoro and H. Zhang, Chem. Rev., 2017, 117, 6225–6331.
14. A. VahidMohammadi, J. Rosen and Y. Gogotsi, Science, 2021, 372, abf1581.
15. Z. Fu, N. Wang, D. Legut, C. Si, Q. Zhang, S. Du, T. C. Germann, J. S. Francisco and R. Zhang, Chem. Rev., 2019, 119, 11980–12031.
16. C. Zhan, W. Sun, Y. Xie, D.-E. Jiang and P. R. C. Kent, ACS Appl. Mater. Interfaces, 2019, 11, 24885–24905.
17. M. Ghidiu, M. R. Lukatskaya, M.-Q. Zhao, Y. Gogotsi and M. W. Barsoum, Nature, 2014, 516, 78–81.
18. Y.-H. Liu, C.-Y. Wang, S.-L. Yang, F.-F. Cao and H. Ye, J. Energy Chem., 2022, 66, 429–439.
19. X.-f Yu, Y.-c Li, J.-b Cheng, Z.-b Liu, Q.-z Li, W.-z Li, X. Yang and B. Xiao, ACS Appl. Mater. Interfaces, 2015, 7, 13707–13713.
20. Y. Xia, T. S. Mathis, M.-Q. Zhao, B. Anasori, A. Dang, Z. Zhou, H. Cho, Y. Gogotsi and S. Yang, Nature, 2018, 557, 409–412.
21. F. Shahzad, M. Alhabeb, C. B. Hatter, B. Anasori, S. M. Hong, C. M. Koo and Y. Gogotsi, Science, 2016, 353, 1137–1140.
22. A. Iqbal, P. Sambyal and C. M. Koo, Adv. Funct. Mater., 2020, 30, 2000883.
23. Z. Guo, J. Zhou, L. Zhu and Z. Sun, J. Mater. Chem. A, 2016, 4, 11446–11452.
24. H. Zhang, G. Yang, X. Zuo, H. Tang, Q. Yang and G. Li, J. Mater. Chem. A, 2016, 4, 12913–12920.
25. C.-F. Fu, X. Li, Q. Luo and J. Yang, J. Mater. Chem. A, 2017, 5, 24972–24980.
26. I. Ihsanullah, J. Environ. Chem. Eng., 2022, 10, 107381.
27. D. L. Druffel, M. G. Lanetti, J. D. Sundberg, J. T. Pawlik, M. S. Stark, C. L. Donley, L. M. McRae, K. M. Scott and S. C. Warren, Chem. Mater., 2019, 31, 9788–9796.
28. K. Maeda, H. Wakayama, Y. Washio, A. Ishikawa, M. Okazaki, H. Nakata and S. Matsuishi, J. Phys. Chem. C, 2020, 124, 14640–14645.
29. C.-F. Fu, J. Sun, Q. Luo, X. Li, W. Hu and J. Yang, Nano Lett., 2018, 18, 6312–6317.
30. C. R. Cox, J. Z. Lee, D. G. Nocera and T. Buonassisi, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, 14057–14061.
31. L. Liao, Q. Zhang, Z. Su, Z. Zhao, Y. Wang, Y. Li, X. Lu, D. Wei, G. Feng, Q. Yu, X. Cai, J. Zhao, Z. Ren, H. Fang, F. Robles-Hernandez, S. Baldelli and J. Bao, Nat. Nanotechnol., 2014, 9, 69–73.
32. J. Liu, Y. Liu, N. Liu, Y. Han, X. Zhang, H. Huang, Y. Lifshitz, S.-T. Lee, J. Zhong and Z. Kang, Science, 2015, 347, 970–974.
33. Q. Wang, T. Hisatomi, Q. Jia, H. Tokudome, M. Zhong, C. Wang, Z. Pan, T. Takata, M. Nakabayashi, N. Shibata, Y. Li, I. D. Sharp, A. Kudo, T. Yamada and K. Domen, Nat. Mater., 2016, 15, 611–615.
34. Y. Goto, T. Hisatomi, Q. Wang, T. Higashi, K. Ishikiriyama, T. Maeda, Y. Sakata, S. Okunaka, H. Tokudome, M. Katayama, S. Akiyama, H. Nishiyama, Y. Inoue, T. Takewaki, T. Setoyama, T. Minegishi, T. Takata, T. Yamada and K. Domen, Joule, 2018, 2, 509–520.
35. X. Li, Z. Li and J. Yang, Phys. Rev. Lett., 2014, 112, 018301.
36. X. Wu, Y. Hu, Y. Wang, Y. Zhou, Z. Han, X. Jin and G. Chen, Appl. Surf. Sci., 2019, 464, 108–114.
37. B.-J. Ng, L. K. Putri, X. Y. Kong, Y. W. Teh, P. Pasbakhsh and S.-P. Chai, Adv. Sci., 2020, 7, 1903171.
38. L. Guo, H. Huang, L. Mei, M. Li and Y. Zhang, Mater. Chem. Front., 2021, 5, 2484–2505.
39. Y. Liu, Y. Wan, B. Li, C. Yang, X. Lv and Y. Shi, J. Mater. Chem. A, 2023, 11, 21713–21720.
40. Z. Gao, X. He, W. Li, Y. He and K. Xiong, Appl. Surf. Sci., 2023, 639, 158146.
41. N. Ghobadi, A. Rezavand, S. Soleimani-Amiri and S. Gholami Rudi, Appl. Surf. Sci., 2023, 639, 158278.
42. G. Nan, W. Zhang, X. Yan, X. Qin, S. Wu, R. Tang, M.-X. Tang, L. Hu, L. Liu, S. Wang, Y. Feng and W. Yi, Phys. Chem. Chem. Phys., 2023, 25, 24594–24602.
43. Y. Zhao, B. Zhang and J. Lin, Phys. Chem. Chem. Phys., 2023, 25, 26666–26678.
44. Y. Zhang, B. Sa, N. Miao, J. Zhou and Z. Sun, J. Mater. Chem. A, 2021, 9, 10882–10892.
45. K. Xiong, Z. Cheng, J. Liu, P.-F. Liu and Z. Zi, RSC Adv., 2023, 13, 7972–7979.
46. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
47. G. Kresse and D. Joubert, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758–1775.
48. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
49. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
50. A. Togo, F. Oba and I. Tanaka, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 78, 134106.
51. X. Gonze and C. Lee, Phys. Rev. B: Condens. Matter Mater. Phys., 1997, 55, 10355–10368.
52. M. Khazaei, M. Arai, T. Sasaki, C.-Y. Chung, N. S. Venkataramanan, M. Estili, Y. Sakka and Y. Kawazoe, Adv. Funct. Mater., 2013, 23, 2185–2192.
53. Y. Lee, S. B. Cho and Y.-C. Chung, ACS Appl. Mater. Interfaces, 2014, 6, 14724–14728.
54. Y. Lee, Y. Hwang, S. B. Cho and Y.-C. Chung, Phys. Chem. Chem. Phys., 2014, 16, 26273–26278.
55. J. Yang, X. Luo, S. Zhang and L. Chen, Phys. Chem. Chem. Phys., 2016, 18, 12914–12919.
56. M. Khazaei, M. Arai, T. Sasaki, A. Ranjbar, Y. Liang and S. Yunoki, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 92, 075411.
57. H. Sahin, S. Cahangirov, M. Topsakal, E. Bekaroglu, E. Akturk, R. T. Senger and S. Ciraci, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 80, 155453.
58. J. He, X. Li, P. Lyu and P. Nachtigall, Nanoscale, 2017, 9, 2246–2252.
59. X. Bai, X.-H. Zha, Y. Qiao, N. Qiu, Y. Zhang, K. Luo, J. He, Q. Li, Q. Huang, J. S. Francisco, C.-T. Lin and S. Du, Nanoscale, 2020, 12, 3795–3802.
60. C. Peng, T. Zhou, P. Wei, H. Ai, B. Zhou, H. Pan, W. Xu, J. Jia, K. Zhang, H. Wang and H. Yu, Chem. Eng. J., 2022, 439, 135685.
61. V. Chakrapani, J. C. Angus, A. B. Anderson, S. D. Wolter, B. R. Stoner and G. U. Sumanasekera, Science, 2007, 318, 1424–1430.
62. J. Zhang, X. Tang, M. Chen, D. Ma and L. Ju, Inorg. Chem., 2022, 61, 17353–17361.
63. C. A. Gueymard, D. Myers and K. Emery, Sol. Energy, 2002, 73, 443–467.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3cp04191c

---