Stacking-Mediated Type-I/Type-II Transition in Two-Dimensional MoTe₂/PtS₂ Heterostructure: A First-Principles Simulation

Abstract

Recently, a two-dimensional (2D) heterostructure has been widely investigated as a photocatalyst to decompose water using the extraordinary type-II band structure. In this work, the MoTe₂/PtS₂ van der Waals heterostructure (vdWH) is constructed with different stacking structures. Based on density functional calculations, the stacking-dependent electronic characteristic is explored, so that the MoTe₂/PtS₂ vdWH possesses type-I and type-II band structures for the light-emitting device and photocatalyst, respectively, with decent stacking configurations. The band alignment of the MoTe₂/PtS₂ vdWH is also addressed to obtain suitable band edge positions for water-splitting at pH 0. Furthermore, the potential drop is investigated, resulting from charge transfer between the MoTe₂ and PtS₂, which is another critical promotion to prevent the recombination of the photogenerated charges. Additionally, the MoTe₂/PtS₂ vdWH also demonstrates a novel and excellent optical absorption capacity in the visible wavelength range. Our work suggests a theoretical guide to designing and tuning the 2D heterostructure using photocatalytic and photovoltaic devices.

1. Introduction

Hydrogen (H₂) is regarded as an important energy source to alleviate environmental pollution and energy shortage, because the combustion productions are almost water. Comparing the bulk photocatalysts, its use of two-dimensional (2D) materials as a photocatalyst to decompose the water is advantageous characteristic [1,2,3]. Since the graphene was proposed to have novel performances [4], tremendous efforts were made to develop other 2D materials [5,6,7,8]. 2D semiconductors show a broad specific surface area [9], suggesting extraordinary electronic [10,11], thermal [12,13] and optical [14,15] performances. Especially, the large surface area can provide more active sites in the photocatalytic process, which also contributes to the photoexcited carrier motion [16]. Furthermore, some 2D materials have been proved to be potential photocatalysts [17,18,19,20].

At the same time, when the single semiconductor is used as a photocatalyst to decompose water, the photogenerated charges move to the surface of the 2D materials. Even if the larger specific surface area of the 2D materials provides a shorter path, the rapid recombination between the photogenerated electrons and holes also hinders photocatalytic efficiency. Thus, the 2D heterostructure constructed by two different semiconductors is adopted to separate the photogenerated electrons and holes using an extraordinary type-II band structure. More importantly, some strategies are further conducted to modulate the novel properties of the heterostructure, such as external electric field [21], external strain [22] and imperfection [23]. When the 2D heterostructure is constructed, there are possible stacking configurations. These stacking structures possess a similar binding energy (E_b) but different properties. For example, the charge transfer of the MoS₂/WS₂ heterostructure is strongly dependent on the interlayer stacking configurations using optical, two-color, ultrafast pump−probe spectroscopy [24]. The valence band-splitting is remarkable. It is induced in a multilayer heterostructure based on transition-metal dichalcogenide (TMD) using stacking engineering in spintronics [25]. Additionally, the different stacking styles of the WSe₂/WS₂ heterostructure can be prepared by vapor growth, which also affects the optical properties [26]. These results suggest that the stacking configurations of the heterostructure have a promising impact on the electronic and optical performances of the 2D heterostructure when used as a photocatalyst.

Recently, a novel transition metal dichalcogenides (TMDs) material, 2D MoTe₂, was prepared with a chemical vapor deposition synthesis method using promising nanoelectronics [27]. MoTe₂ also possesses excellent electronic [28], carrier transport [29] and thermoelectric [30] properties. Another TMDs, 2D PtS₂ was also investigated as a heterostructure, such as PtS₂/InSe [31], HfS₂/PtS₂ [32] and PtS₂/arsenene [33], which are potential photocatalysts for water-splitting. The synthesized MoTe₂ and PtS₂ provide the possibility of preparing the MoTe₂/PtS₂ heterostructure using potential photocatalytic, photovoltaic, and optical devices. The stacking tuning of the electronic performances of the 2D materials is also a popular method. Therefore, in this report, we aim to study these novel TMDs materials, using first-principles simulations, to explore the response of the structural, electronic and optical performances to the stacking configuration of the MoTe₂/PtS₂ heterostructure as a latent further nano-device, which could provide theoretical guidance for the design of the 2D heterostructure.

2. Computing Method

Using the Vienna ab initio simulation package, the first-principles simulations were employed using the density functional theory [34,35]. The core electrons were considered using the projector augmented wave potentials (PAW) [36], which were explored using the Perdew–Burke–Ernzerhof (PBE) functional together with the generalized gradient approximation (GGA) [37]. In all calculations, the Grimme was conducted using a DFT-D3 method to describe the weak dispersion forces [38]. The Heyd–Scuseria–Ernzerhof (HSE06) hybrid functional was utilized [39]. Energy cut-off was set at 550 eV; Monkhorst–Pack k-point grids were15 × 15 × 1. The vacuum space was selected to be 25 Å to hold back the forces between the nearby slabs. The convergence force in the simulations was 0.01 eV·Å⁻¹; at the same time, the energy converged to 0.01 meV. The binding energy (E_b) of the MoTe₂/PtS₂ heterostructure was decided by:

E_b = E_MoTe2/PtS2 − E_MoTe2 − E_PtS2,

(1)

where the E_MoTe2/PtS2, E_MoTe2 and E_PtS2 are the total energy of the MoTe₂/PtS₂ heterostructure, pristine MoTe₂ and PtS₂ monolayers, respectively. The charge density difference (Δρ) was calculated by:

Δρ = ρ_MoTe2/PtS2 − ρ_MoTe2 − ρ_PtS2,

(2)

where the ρ_MoTe2/PtS2, ρ_MoTe2 and ρ_PtS2 are total charge density of the MoTe₂/PtS₂ heterostructure, pristine MoTe₂ and PtS₂ monolayers, respectively. The optical absorption property of the monolayered was obtained as follows:

$$\alpha (\omega )=\frac{\sqrt{2}\omega }{c}{\left\{{\left[{\epsilon }_1^2(\omega )+{\epsilon }_2^2(\omega )\right]}^{1/2}-{\epsilon }_1(\omega )\right\}}^{1/2}$$

(3)

where the angular frequency, absorption coefficient and speed of light are used by ω, α and c, respectively. ${\epsilon }_1(\omega )$ and ${\epsilon }_2(\omega )$ were used as real and imaginary elements in the dielectric constant.

3. Results and Discussion

Before constructing the MoTe₂/PtS₂ heterostructure, the structure of MoTe₂ and PtS₂ monolayers were optimized by the 3.529 Å and 3.564 Å, respectively. The obtained band energy of the MoTe₂ and PtS₂ monolayers are exhibited in Figure S1 in the Supporting Information, with bandgaps of 1.22 eV and 2.60 eV, respectively. Then, the MoTe₂/PtS₂ heterostructure was formed by six different representative structures, shown in Figure S2, named from P1 to P6. For example, the P1 configuration was formed by putting the Mo atoms and the Te atoms on top of the Pt atoms and the upper S atoms. P2 was built so that the Mo atoms are on top of the Pt atoms, while the Te atoms were located on top of the lower S atoms. The following investigations are all based on these six configurations.

The binding energy of these MoTe₂/PtS₂ heterostructures is summarized in

Table 1. The obtained binding energy (E_b, eV), bond length (B, Å), the interface across the interface (H, Å), the bandgap (E_g, eV), the charge transfer (Δρ, electron) and potential drop (ΔV, eV) in the optimized MoTe₂/PtS₂ heterostructure constructed with different stacking styles.

		Eb	BMo–Te	LPt–S	H	Eg	Δρ	ΔV
P1	Type-II	−17.08	2.73	2.40	3.79	1.27	0.047	4.416
P2	Type-II	−26.39	2.73	2.39	2.85	1.20	0.051	4.578
P3	Type-II	−17.40	2.73	2.40	3.77	0.95	0.017	4.417
P4	Type-I	−25.98	2.73	2.40	3.10	1.35	0.033	4.584
P5	Type-I	−26.39	2.73	2.40	3.06	1.31	0.035	4.558
P6	Type-II	−28.10	2.73	2.39	2.87	1.44	0.047	4.672

, which suggests the van der Waals interactions in these interfaces [40]. The bond lengths of Mo–Te and Pt–S are almost small changes compared with those in pristine MoTe₂ (2.74 Å) and PtS₂ (2.40 Å) monolayers, which further demonstrates the MoTe₂/PtS₂ vdW heterostructures (vdWH). In addition, the interface across the interface of these MoTe₂/PtS₂ vdWH is also calculated by Table 1.

The projected band energy of the MoTe₂/PtS₂ vdWH with different stacking configurations is calculated in Figure 1. One can see that P1-, P2-, P3- and P6-MoTe₂/PtS₂ vdWHs possess a type-II band structure, with the conduction-band minimum (CBM) resulting from PtS₂ and the valence-band maximum (VBM) from MoTe₂. While the P4- and P5-MoTe₂/PtS₂ vdWHs have a type-I band alignment by the CBM and VBM located at the MoTe₂ monolayer. Furthermore, the bandgap obtained with these MoTe₂/PtS₂ vdWHs is explained in Table 1. P3-MoTe₂/PtS₂ vdWH has a narrow bandgap of about 0.95 eV. It is worth noting that the direct bandgaps in the MoTe₂/PtS₂ vdWH from the MoTe₂ are 1.33 eV, 1.30 eV, 1.34 eV, 1.34 eV, 1.31 eV and 1.35 eV for from P1 to P6, respectively.

The type-II band alignment in P1-, P2-, P3- and P6-MoTe₂/PtS₂ vdWHs provides the ability to separate the photoexcited charges, as shown in Figure 2a. With the assistance from the conduction-band offset (CBO), the photoexcited electrons at the conduction-band (CB) of the MoTe₂ continue to flow to the CB of the PtS₂, while the photoexcited holes at the valence-band (VB) of the PtS₂ migrate to the VB of the MoTe₂ under the valence-band offset (VBO). Thus, the photoexcited electrons and the holes induce the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) at PtS₂ and MoTe₂, respectively, which can be used as a potential photocatalyst. In addition, the intrinsic type-I band structure of P4- and P5-MoTe₂/PtS₂ vdWHs results in a unidirectional flow mode for excited charges, which is a promising light-emitting device.

Furthermore, the band edge positions of these MoTe₂/PtS₂ vdWHs are obtained by HSE06 calculations in Figure 3, and compared with the HER and OER potential energy for the water-splitting at pH 0. One can see that the P-4 and P-6 MoTe₂/PtS₂ vdWHs possess a decent band alignment, which can promote the HER and OER for water-splitting at pH 0. Additionally, together with a type-II band structure, the P-6 MoTe₂/PtS₂ vdWH can be considered as an advantageous photocatalyst to decompose water. Although the P-4 MoTe₂/PtS₂ vdWH has a type-II band alignment to separate the photogenerated electrons and holes, the suitable band edge positions still have the ability to induce redox for water-splitting. Furthermore, the type-I heterostructure is also reported to show a novel photocatalytic performance [41,42].

The interfacial characteristics of the MoTe₂/PtS₂ vdWHs are investigated using the charge density difference and potential drop across the interface. Using Equation (2), the charge density difference in these MoTe₂/PtS₂ vdWHs is calculated, showing that the PtS₂ layer obtained the electrons from the MoTe₂ layer (Figure 4). Additionally, the charge transfer between the MoTe₂ and PtS₂ was investigated by Bader charge analysis [43], as shown in Table 1, which suggests a maximum charge transfer in P6-MoTe₂/PtS₂ vdWH of about 0.047 electrons. Furthermore, the potential drop (shown by Figure 5) in the interface for MoTe₂/PtS₂ vdWHs is addressed in Table 1. One can see that MoTe₂/PtS₂ vdWHs possesses a pronounced potential drop across the interface, ranging from 4.41 eV to 4.67 eV, which is larger than that in the WSSe/BSe vdW heterostructure [44]. Importantly, this potential drop can act as a significant motivating force to separate the photogenerated electrons and holes in type-I and type-II heterostructures [45,46].

The visible-light absorption capacity of these MoTe₂/PtS₂ vdWHs is further investigated, as in Figure 6, which is also a crucial performance using a photocatalyst. The calculated visible-light absorption ability demonstrates that the maximum absorption peaks of the P1-, P2-, P3- and P6-MoTe₂/PtS₂ vdWHs are obtained by 6.40 × 10⁵ cm⁻¹, 7.10× 10⁵ cm⁻¹, 6.26 × 10⁵ cm⁻¹, 6.28 × 10⁵ cm⁻¹, 6.57 × 10⁵ cm⁻¹ and 6.85 × 10⁵ cm⁻¹, which are located at 369 nm, 379 nm, 369 nm, 378 nm, 380 nm and 376 nm, respectively. Importantly, some absorption peaks also exist, in the range 500−600 nm, for these MoTe₂/PtS₂ vdWHs. The calculated light absorption peak in these MoTe₂/PtS₂ vdWH is also larger than that of other heterostructures that are used as photocatalysts, such as CdO/Arsenene (8.47 × 10⁴ cm⁻¹) [47], AlN/Zr₂CO₂ (3.97× 10⁵ cm⁻¹) [48] and Hf₂CO₂/AlN (3.63× 10⁵ cm⁻¹) [49]. It is worth noting that the strongest absorption peak in these MoTe₂/PtS₂ vdWHs is 8.12 × 10⁵ cm⁻¹ at 333 nm, for P5-MoTe₂/PtS₂ vdWH. The results show that the MoTe₂/PtS₂ vdWHs possesses a fantastic and tunable optical performance using the stacking configuration.

4. Conclusions

Using first-principles simulations, the structural electronic natures of the 2D MoTe₂/PtS₂ vdWHs are addressed. These are formed by different stacking configurations. P-4 and P-5 MoTe₂/PtS₂ vdWHs possess a type-I band structure for a light-emitting device, while others have type-II band alignment to separate the photogenerated electrons and holes. Furthermore, the band edge positions of these MoTe₂/PtS₂ vdWHs are investigated, and the P-6 MoTe₂/PtS₂ vdWHs have suitable potential to induce the redox reactions for water-splitting at pH 0 when used as a promising photocatalyst. The MoTe₂/PtS₂ vdWHs also show stacking-dependent interfacial and excellent optical properties. All these results suggest a theoretical method that could be used to design and tune the performances of a 2D heterostructure.