Defect engineering in two-dimensional materials for photocatalysis: A mini-review of first-principles design

Abstract

Two-dimensional (2D) materials have emerged as a significant class of materials promising for photocatalysis, and defect engineering offers an effective route for enhancing their photocatalytic performance. In this mini-review, a first-principles design perspective on defect engineering in 2D materials for photocatalysis is provided. Various types of defects in 2D materials, spanning point, line, and planar defects are explored, and their influence on the intrinsic properties and photocatalytic efficacy of these materials is highlighted. Additionally, the use of theoretical descriptors to characterize the stability, electronic, optical, and catalytic properties of 2D defective systems is summarized. Central to the discussion is the understanding of electronic structure, optical properties, and reaction mechanisms to inform the rational design of photocatalysts based on 2D materials for enhanced photocatalytic performance. This mini-review aims to provide insights into the computational design of 2D defect systems tailored for efficient photocatalytic applications.

Graphical abstract

Keywords

photocatalysis / first-principles / defect engineering / descriptors / two-dimensional materials

1 Introduction

Photocatalysis has gained significant attention due to its potential as an alternative energy source and its ability to address the energy crisis [1]. Among various materials considered for photocatalysis, two-dimensional (2D) materials are of great significance due to their excellent properties, including large specific surface area, high electrical conductivity, abundant reactive sites, and suitable electronic structures for photocatalytic reactions such as water splitting and CO₂ reduction [2–8]. The explosive popularity of 2D materials in catalytic applications can be attributed not only to their intrinsic properties but also to the flexibility to tune their properties. Among various tuning methods, defect engineering holds great promise as it can effectively tailor the electronic structures, promote charge separation, and increase the active sites of 2D materials [9–12].

The effective and rational design of 2D defects for photocatalysis necessitates a thorough understanding of the electronic structures, optical properties, and reaction mechanisms of the materials. Besides, to facilitate the exploration of novel photocatalysts, it is essential to establish descriptors that enable the computational screening of materials. First-principles calculations employing the density functional theory (DFT) can be utilized to gain insights into the mechanism and guide the material design and discovery for photocatalysis [13,14]. Furthermore, by using appropriate descriptors, the discovery of 2D defective photocatalysts can be accelerated by screening, especially when combined with machine learning (ML) techniques [15].

In this mini-review, defect engineering for 2D photocatalysis is explored from a first-principles design perspective. 2D defects and their capability to tune the intrinsic properties are briefly introduced, followed by useful descriptors to appropriately describe the stability, electronic, optical, and catalytic properties of 2D defect systems, as well as their limitations. In addition, the application of ML techniques in accelerating photocatalyst design is also explored. While the primary focus is on photocatalytic water splitting, the insights provided here apply to other redox reactions as well. This review aims to offer a comprehensive understanding of the computational design of 2D materials for photocatalysis.

2 Defects in 2D materials

2D materials for photocatalysis can be broadly classified into three categories: 2D metal oxides, including TiO₂, Bi₂WO₆, and perovskite oxides; 2D metal chalcogenides, such as MoS₂, SnS₂, and WSe₂; and metal-free photocatalysts like graphitic carbon nitride (g-C₃N₄) and hexagonal boron nitride (h-BN) [16–19]. The fabrication of 2D materials inevitably introduces imperfections in the structures. Common defects in 2D materials include point defects (e.g., vacancies, dislocations, and dopants), line defects (e.g., grain boundaries, phase boundaries, and lateral heterostructures), and planar defects (e.g., van der Waals heterostructures and Janus structures) (Fig.1). These defects play an important role in modifying the intrinsic properties of 2D materials and ultimately impacting their photocatalytic performance.

2.1 0D defects

2D materials, such as 2D transition metal dichalcogenides (TMDCs), have been extensively investigated for catalyzing water splitting [20,21]. However, theoretical and experimental work have shown that their active sites are located at the edges other than the basal plane [22,23]. The introduction of point vacancies in 2D materials can activate the basal plane and enhance catalytic activity [24–26]. For instance, introducing sulfur vacancies modifies the electronic structure of 2H-MoS₂, creating defective states in the gap near the Fermi level that contribute to the enhanced hydrogen binding at the vacancy sites [27]. In addition to modifying catalytic activity, vacancies in atomically thin metal oxide sheets also regulate the band edge position, bandwidth, and photo-induced carrier migration kinetics to improve the photocatalytic activity [28–30]. Doping is also an effective method to enhance the photocatalytic performance of 2D materials [31,32]. For example, Co doping in In₂S₃ enhances photocatalytic water splitting by increasing the density of states at the conduction band minimum (CBM), which significantly boosts visible-light absorption [33]. While doping introduces foreign atoms into the lattice, creating new energy states within the band structure, vacancies arise from the absence of atoms, typically generating defect states within the band gap. These differences lead to varying modifications in electronic structures, and recent studies suggest that a synergistic combination of vacancies and dopants can further enhance photocatalytic performance [31,34].

2.2 1D defects

In addition to point defects, line defects also have a significant impact on activating the basal plane of 2D materials [11,35–37]. Similar to point defects, the presence of boundaries in 2D materials influences the electronic properties, thereby modifying the chemical properties [11]. More importantly, introducing boundaries is an effective approach for creating separate domains for different functions on the surface, while still maintaining structural integrity. One good example is multiphasic 2D MoS₂ nanosheets for photocatalytic HER [38]. In this study, light absorption and charge generation were achieved by the 2H phase, efficient charge separation occurred at the 1T'/2H interface, and HER was favored in the 1T' phase. The well-balanced compositions of different domains on MoS₂ sheets result in an optimal photocatalyst that exhibits excellent optical, electronic, and chemical properties.

2.3 2D defects

Another advantage of 2D materials is their ability to form van der Waals heterostructures [39,40]. 2D materials serve as building blocks for heterostructures, and when held together by van der Waals forces, they can create materials with desirable physical and chemical properties for photocatalysis. Such face-to-face construction allows an intimate interface to promote the transfer and separation of photogenerated charge carriers, thus improving photocatalytic efficiency [41–43]. Moreover, these multi-stacked structures can include layers with various functions, such as semiconductor or cocatalyst layers. The reactivity of cocatalyst layers can be further enhanced by introducing point or line defects in their basal planes [44].

Among various defects in 2D materials, point defects, such as vacancies, are most extensively studied due to their controllability and versatility [45,46]. However, the impact of any given defect on photocatalytic efficiency is highly material-specific and depends on several factors, including defect concentration, interactions with the electronic structure of the material, and synergistic effects among multiple defects. Therefore, a comprehensive understanding of these factors is essential for optimizing photocatalytic performance in 2D materials via defect engineering.

3 Descriptors for defect engineering in 2D materials for photocatalytic applications

To computationally design materials with enhanced photocatalytic performance through defect engineering, it is essential to identify key material properties for accurate and efficient analysis. This is particularly important for high-throughput computational screening in the design of 2D photocatalysts. To utilize 2D defects for photocatalytic applications, certain conditions must be met, and suitable descriptors are needed to characterize these properties (Fig.2).

3.1 Structural properties

Stability is the first essential requirement for designing materials for any application. One direct and effective descriptor for stability is the formation energy. The formation energy of a 2D defect comprises two significant components: the formation energy of the monolayer from the bulk materials, and the formation energy of the defect on the 2D materials. The calculation of the formation energy of a specific 2D defect using DFT is well-established and straightforward [49–51]. Similar to defects in bulk materials, 2D defects are typically modeled in a supercell and repeated periodically throughout space. However, for charged defects, this supercell approach may introduce errors due to electrostatic interactions between the defect and its periodic images as well as the background charge (Fig.3(a)). As a result, careful model construction and energy correction are required to obtain accurate formation energies of 2D charged defects [51,52]. Additionally, while each defect can be accurately addressed and calculated individually, the computational exploration of a broad range of potential 2D host materials and defect configurations is still challenging. This difficulty arises from the extensive chemical space, the high computational cost due to the large supercell required for defect calculations, and the lack of defect generation methods and data sets. In recent years, considerable effort has been devoted to evaluating formation energies for 2D materials through high-throughput screening [53–57], and several databases for 2D defects have been established [58,59]. In addition to thermodynamic stability, it is important to consider the corrosion of 2D materials in aqueous solutions for photocatalytic applications. The corrosion of 2D materials is usually examined using Pourbaix diagrams (Fig.3(b)) [60], and the interaction between the surface and solution can be accounted for using implicit or explicit solvation models [61,62]. Additionally, other key descriptors, such as phonon dispersion for evaluating dynamic stability and bond length deviations for assessing structural changes, can offer further insights into the structural properties of materials [63, 64].

3.2 Electronic properties

The potential of 2D defect systems as photocatalysts relies heavily on their electronic structures, specifically the band gaps and band edge positions [13]. For efficient photocatalysis in 2D systems, the band gap must be greater than 1.23 eV, which is the reaction energy for water splitting [65]. Furthermore, the energy of the CBM should be higher than the reduction potential of H⁺/H₂ (−4.44 eV at pH = 0), and the energy of the valence band maximum (VBM) should be lower than the oxidation potential of O₂/H₂O (−5.67 eV at pH = 0) [20]. It is important to note that these potentials shift by approximately 0.059 eV per pH unit as pH increases, which can either enhance or hinder photocatalytic efficiency depending on the alignment of the CBM and VBM with the redox potentials. A wide range of 2D materials exhibit suitable band gaps and band edge positions for water splitting and CO₂ reduction (Fig.3(c)). These properties can be calculated using DFT. However, it is known that there is a trade-off between efficiency and accuracy when it comes to the exchange-correlation functionals used. Standard functionals such as local density approximation (LDA) and Perdew-Burke-Ernzerhof (PBE) can calculate electronic properties with low computational cost, but they tend to significantly underestimate band gaps [66,67]. On the contrary, traditional hybrid functionals such as PBE0 and HSE06 can provide more accurate results [68] but are computationally more expensive, especially for models that involve defects, which typically contain at least dozens of atoms. Recently developed functionals such as SCAN meta-GGA have the potential to improve the accuracy of band gap calculations while maintaining affordable efficiency [69]. In addition to these energy-based considerations, other properties such as charge separation and charge carrier mobility are also important electronic factors for photocatalytic systems, and several descriptors have been developed to evaluate these properties [70].

3.3 Optical properties

The efficiency of light harvesting is one of the most important factors that determines the overall photocatalytic performance. Materials used for photocatalysis should be able to absorb a substantial fraction of the visible light spectrum, as it accounts for over 40% of solar energy [60]. The evaluation of light absorption typically requires the determination of optical band gaps. However, to accurately compute the optical band gaps, hybrid functionals or GW approximations are necessary. Furthermore, excitonic effects are generally more pronounced in 2D systems due to quantum confinement, and thus it is important to include the excitonic effects (Fig.3(d)) [71–74]. One key metric for evaluating the excitonic effects is the exciton binding energy (EBE) [75]. A lower exciton binding energy promises easier generation of free electrons and holes, which is advantageous for photocatalytic processes. The accurate calculations of EBE have been achieved by the Bethe-Salpeter Equation (BSE) formalism for both bulk and 2D materials [76,77]. However, the BSE formalism demands significant computational resources. Given the inevitably increased size of the 2D defect system, the challenge in obtaining its optical properties lies in selecting accurate methods that are also computationally affordable.

3.4 Catalytic properties

In terms of the catalytic activity of materials, several descriptors have been established in the past few decades, including adsorption energies, work function, coordination numbers, and band centers [78]. Detailed discussions of these descriptors have been covered in several review papers [79–81]. Among these descriptors, the band center is particularly important for understanding and predicting the catalytic activity of materials [24,82,83]. The key properties of a catalyst are fundamentally determined by its electronic structure, as described by the band center. Typically, a band center closer to the Fermi level results in stronger adsorbate-surface interactions, leading to enhanced adsorption. This model provides valuable insights into how electronic properties dictate catalytic performance and is widely used in catalyst design. However, due to the complexity of the defect configurations, descriptors that were successfully applied to simple models may no longer be suitable for 2D defect structures. One example is the failure of band center models to accurately describe the hydrogen adsorption at the defect sites of 2D TMDCs, highlighting the need for new descriptors for adsorption (Fig.3(e)) [11,24,84]. This failure can be attributed to the different contributions of atomic orbitals at the defect site to the adsorbed hydrogen atom, and the variability in the hydrogen adsorption geometry. In fact, this could be beneficial for catalyst design, as the failure of simple descriptors sometimes also indicates the breaking of ‘scaling relations’ for reactions, which limits the enhancement of catalytic activity [85]. However, the failure of descriptors is also a challenge for rational design and screening of the 2D defect systems. It is necessary to understand the catalytic reactions on 2D defects and develop new descriptors for them.

4 Machine learning (ML) techniques for accelerating 2D material design for photocatalysis

While descriptors offer an effective approach for the high-throughput design of 2D materials for photocatalysis, the substantial computational costs of first-principles methods remain challenging. ML presents a promising solution to these issues. For example, ML techniques have been employed to enhance prediction accuracy for electronic and optical properties of 2D materials, bypassing the need for time-consuming functionals [87,88]. Recent studies show that ML-assisted models, whether using PBE band gaps as features or not, can predict band gaps of 2D materials with accuracy comparable to the GW method, but at significantly lower computational cost [89–91]. Additionally, ML models can serve as efficient pre-screening tools, enabling rapid prediction of material properties, such as vacancy formation energies, without exhaustively sampling the entire configurational and compositional space of defect systems [92,93]. Furthermore, machine-learned interatomic potentials, including descriptor-based and moment tensor potentials, have been developed to predict energies and forces in atomic structures with high precision across various material systems [94–96]. These potentials substantially reduce computational costs while maintaining accuracy, facilitating the exploration of large defect systems [97–99].

By leveraging these advanced ML techniques, researchers can efficiently screen and design 2D materials with desired photocatalytic properties. However, challenges in applying ML to 2D materials with defects include the need for high-quality, diverse data sets for training robust models, and the development of interpretable models that offer insights into underlying physical mechanisms. Several efforts have been devoted to addressing these challenges [58,100].

5 Discussion and future prospects

Defect engineering offers an effective method for tuning the electronic, optical, and chemical properties of 2D materials for photocatalytic applications. In this mini-review, a brief introduction to various defects in 2D materials and their impact on photocatalytic properties was provided. Common descriptors used for the computational design of photocatalysts were discussed, focusing on their stability, electronic, optical, and catalytic properties, as well as their limitations. Additionally, ML techniques for accelerating catalyst design were explored.

While this mini-review primarily focuses on photocatalytic water splitting, the modification methods and descriptors discussed are fundamental to photocatalyst design and are broadly applicable to various redox reactions. Defect engineering has received extensive attention and has been widely explored in various photocatalytic fields, including CO₂ reduction, nitrogen fixation, and organic synthesis [101–104]. A notable example is the use of defect-rich ultrathin ZnAl-layered double hydroxide nanosheets as photocatalysts for CO₂ reduction to CO [105]. The introduction of oxygen vacancies generates coordinatively unsaturated Zn⁺ centers in the nanosheets, leading to high photocatalytic activity. For these photocatalytic reactions, key descriptors such as band gaps and band centers have been investigated to better understand and predict the photocatalytic performance of 2D materials [6,106].

It is worth noting that the limitations of first-principles methods in modeling defect properties can be mitigated through multiscale modeling [107,108]. Molecular dynamics (MD) simulations enable the exploration of larger systems and longer timescales than DFT, offering insights into nanoscale phenomena like molecule adsorption and diffusion [109,110]. Additionally, kinetic Monte Carlo (kMC) simulations allow for simulations of system dynamics over extended timescales that are challenging for traditional MD methods, effectively modeling material properties, including defect diffusion, charge transport, and reaction kinetics [111–113].

In prospect, there are still several challenges that need to be addressed in this field. First, it is crucial to bridge the gap between theory and experiments. While computational methods are mature and widely used to predict the atomic structures and properties of materials, identifying true atomic structures under realistic working conditions remains a significant challenge [106]. Secondly, although water splitting represents a fundamental and relatively simple reaction, the complexity of most reactions is far beyond the scope of correlations and descriptors, which hinders the feasibility of large-scale computational screening [114]. Considering defects in the atomic models adds complexity to the computations, necessitating cautious use and development of new descriptors. A promising avenue for future research involves leveraging ML techniques to predict properties and design descriptors for defects in 2D materials, which could provide promising approaches to accelerate the discovery of photocatalytic materials. Nonetheless, challenges such as the lack of defect databases and the limited interpretability of ML models remain significant obstacles that need to be addressed. Overall, gaining a deeper understanding of the electronic structures of defects in 2D materials and how they impact the mechanisms of photocatalytic reactions is crucial. Only then can it be possible to identify the origins of these phenomena and guide the future computational design of 2D defective systems for photocatalysis.

References

[1] Moniz S J A, Shevlin S A, Martin D J. . Visible-light driven heterojunction photocatalysts for water splitting—A critical review. Energy & Environmental Science, 2015, 8(3): 731–759

[2] Morozov S V, Novoselov K S, Katsnelson M I. . Giant intrinsic carrier mobilities in graphene and its bilayer. Physical Review Letters, 2008, 100(1): 016602

[3] Peigney A, Laurent C, Flahaut E. . Specific surface area of carbon nanotubes and bundles of carbon nanotubes. Carbon, 2001, 39(4): 507–514

[4] Lee C, Wei X, Kysar J W. . Measurement of the elastic properties and intrinsic strength of monolayer graphene. Science, 2008, 321(5887): 385–388

[5] Novoselov K S, Geim A K, Morozov S V. . Electric field effect in atomically thin carbon films. Science, 2004, 306(5696): 666–669

[6] Peng B, Ang P K, Loh K P. Two-dimensional dichalcogenides for light-harvesting applications. Nano Today, 2015, 10(2): 128–137

[7] Mortazavi B, Fan Z, Pereira L F C. . Amorphized graphene: A stiff material with low thermal conductivity. Carbon, 2016, 103: 318–326

[8] Chen Y, Meng F, Bie X. . Atomistic and continuum modeling of 3D graphene honeycombs under uniaxial in-plane compression. Computational Materials Science, 2021, 197: 110646

[9] Di J, Hao G, Jiang W. . Defect chemistry in 2D atomic layers for energy photocatalysis. Accounts of Materials Research, 2023, 4(11): 910–924

[10] Ou P, Zhou X, Meng F. . Single molybdenum center supported on N-doped black phosphorus as an efficient electrocatalyst for nitrogen fixation. Nanoscale, 2019, 11(28): 13600–13611

[11] Chen Y, Ou P, Bie X. . Basal plane activation in monolayer MoTe₂ for the hydrogen evolution reaction via phase boundaries. Journal of Materials Chemistry. A, Materials for Energy and Sustainability, 2020, 8(37): 19522–19532

[12] Zhao Y, Chen Y, Ou P. . Basal plane activation via grain boundaries in monolayer MoS₂ for carbon dioxide reduction. ACS Catalysis, 2023, 13(19): 12941–12951

[13] Li Y, Li Y L, Sa B. . Review of two-dimensional materials for photocatalytic water splitting from a theoretical perspective. Catalysis Science & Technology, 2017, 7(3): 545–559

[14] Samanta B, Morales-García Á, Illas F. . Challenges of modeling nanostructured materials for photocatalytic water splitting. Chemical Society Reviews, 2022, 51(9): 3794–3818

[15] Ge L, Ke Y, Li X. Machine learning integrated photocatalysis: Progress and challenges. Chemical Communications, 2023, 59(39): 5795–5806

[16] Luo B, Liu G, Wang L. Recent advances in 2D materials for photocatalysis. Nanoscale, 2016, 8(13): 6904–6920

[17] Hisatomi T, Kubota J, Domen K. Recent advances in semiconductors for photocatalytic and photoelectrochemical water splitting. Chemical Society Reviews, 2014, 43(22): 7520–7535

[18] Yu J, Xu C Y, Ma F X. . Monodisperse SnS₂ nanosheets for high-performance photocatalytic hydrogen generation. ACS Applied Materials & Interfaces, 2014, 6(24): 22370–22377

[19] Dong X, Cheng F. Recent development in exfoliated two-dimensional g-C₃N₄ nanosheets for photocatalytic applications. Journal of Materials Chemistry. A, Materials for Energy and Sustainability, 2015, 3(47): 23642–23652

[20] Zhang X, Zhang Z, Wu D. . Computational screening of 2D materials and rational design of heterojunctions for water splitting photocatalysts. Small Methods, 2018, 2(5): 1700359

[21] Faraji M, Yousefi M, Yousefzadeh S. . Two-dimensional materials in semiconductor photoelectrocatalytic systems for water splitting. Energy & Environmental Science, 2019, 12(1): 59–95

[22] Hinnemann B, Moses P G, Bonde J. . Biomimetic hydrogen evolution: MoS₂ nanoparticles as catalyst for hydrogen evolution. Journal of the American Chemical Society, 2005, 127(15): 5308–5309

[23] Jaramillo TF, Jørgensen KP, Bonde J. . Identification of active edge sites for electrochemical H₂ evolution from MoS₂ nanocatalysts. Science, 2007, 317(5834): 100–102

[24] Ouyang Y, Ling C, Chen Q. . Activating inert basal planes of MoS₂ for hydrogen evolution reaction through the formation of different intrinsic defects. Chemistry of Materials, 2016, 28(12): 4390–4396

[25] Li H, Tsai C, Koh A L. . Corrigendum: Activating and optimizing MoS₂ basal planes for hydrogen evolution through the formation of strained sulphur vacancies. Nature Materials, 2016, 15(3): 364

[26] Li L, Qin Z, Ries L. . Role of sulfur vacancies and undercoordinated Mo regions in MoS₂ nanosheets toward the evolution of hydrogen. ACS Nano, 2019, 13(6): 6824–6834

[27] Li H, Tsai C, Koh A L. . Activating and optimizing MoS₂ basal planes for hydrogen evolution through the formation of strained sulphur vacancies. Nature Materials, 2016, 15(1): 48–53

[28] Guan M, Xiao C, Zhang J. . Vacancy associates promoting solar-driven photocatalytic activity of ultrathin bismuth oxychloride nanosheets. Journal of the American Chemical Society, 2013, 135(28): 10411–10417

[29] Bi W, Ye C, Xiao C. . Spatial location engineering of oxygen vacancies for optimized photocatalytic H₂ evolution activity. Small, 2014, 10(14): 2820–2825

[30] Liu Y, Xiao C, Li Z. . Vacancy engineering for tuning electron and phonon structures of two-dimensional materials. Advanced Energy Materials, 2016, 6(23): 1600436

[31] Dou Y, He C T, Zhang L. . Approaching the activity limit of CoSe₂ for oxygen evolution via Fe doping and Co vacancy. Nature Communications, 2020, 11(1): 1664

[32] Feng C, Wu Z P, Huang K W. . Surface modification of 2D photocatalysts for solar energy conversion. Advanced Materials, 2022, 34(23): 2200180

[33] Lei F, Zhang L, Sun Y. . Atomic-layer-confined doping for atomic-level insights into visible-light water splitting. Angewandte Chemie International Edition, 2015, 54(32): 9266–9270

[34] Chen K, Zhou S, Jiang T. . Synergy of Fe dopants and oxygen vacancies confined in atomically-thin cobaltous oxide sheets for high-efficiency CO₂ photoreduction. Journal of Materials Chemistry. A, Materials for Energy and Sustainability, 2021, 9(39): 22353–22363

[35] Lee Y, Ling N, Kim D. . Heterophase boundary for active hydrogen evolution in MoTe₂. Advanced Functional Materials, 2022, 32(10): 2105675

[36] Zhao N, Wang L, Zhang Z. . Activating the MoS₂ basal planes for electrocatalytic hydrogen evolution by 2H/1T′ structural interfaces. ACS Applied Materials & Interfaces, 2019, 11(45): 42014–42020

[37] Zhu J, Wang Z C, Dai H. . Boundary activated hydrogen evolution reaction on monolayer MoS₂. Nature Communications, 2019, 10(1): 1348

[38] Peng R, Liang L, Hood Z D. . In-plane heterojunctions enable multiphasic two-dimensional (2D) MoS₂ nanosheets as efficient photocatalysts for hydrogen evolution from water reduction. ACS Catalysis, 2016, 6(10): 6723–6729

[39] Duong D L, Yun S J, Lee Y H. van der Waals layered materials: Opportunities and challenges. ACS Nano, 2017, 11(12): 11803–11830

[40] Novoselov K S, Mishchenko A, Carvalho A. . 2D materials and van der Waals heterostructures. Science, 2016, 353(6298): aac9439

[41] Luo Y, Wang S, Ren K. . Transition-metal dichalcogenides/Mg(OH)₂ van der Waals heterostructures as promising water-splitting photocatalysts: A first-principles study. Physical Chemistry Chemical Physics, 2019, 21(4): 1791–1796

[42] Ren K, Yu J, Tang W. A two-dimensional vertical van der Waals heterostructure based on g-GaN and Mg (OH)₂ used as a promising photocatalyst for water splitting: A first-principles calculation. Journal of Applied Physics, 2019, 126(6): 065701

[43] Yu H, Dai M, Zhang J. . Interface engineering in 2D/2D heterogeneous photocatalysts. Small, 2023, 19(5): 2205767

[44] Xu J, Gao J, Qi Y. . Anchoring Ni₂P on the UiO-66–NH₂/g-C₃N₄-derived C-doped ZrO₂/g-C₃N₄ heterostructure: Highly efficient photocatalysts for H₂ production from water splitting. ChemCatChem, 2018, 10(15): 3327–3335

[45] Ma Y, Qiu B, Zhang J. . Vacancy engineering of ultrathin 2D materials for photocatalytic CO₂ reduction. ChemNanoMat: Chemistry of Nanomaterials for Energy, Biology and More, 2021, 7(4): 368–379

[46] Wang X, Wu J, Zhang Y. . Vacancy defects in 2D transition metal dichalcogenide electrocatalysts: From aggregated to atomic configuration. Advanced Materials, 2023, 35(50): 2206576

[47] Aras M, Kılıç Ç, Ciraci S. Lateral and vertical heterostructures of transition metal dichalcogenides. Journal of Physical Chemistry C, 2018, 122(3): 1547–1555

[48] Bölle FT, Mikkelsen AEG, Thygesen KS. . Structural and chemical mechanisms governing stability of inorganic Janus nanotubes. npj Computational Materials, 2021, 7(1): 41

[49] Freysoldt C, Grabowski B, Hickel T. . First-principles calculations for point defects in solids. Reviews of Modern Physics, 2014, 86(1): 253–305

[50] Li A, Pan J, Dai X. . Electrical contacts of coplanar 2H/1T′ MoTe₂ monolayer. Journal of Applied Physics, 2019, 125(7): 075104

[51] Komsa H P, Berseneva N, Krasheninnikov A V. . Charged point defects in the flatland: Accurate formation energy calculations in two-dimensional materials. Physical Review X, 2014, 4(3): 031044

[52] Dabo I, Kozinsky B, Singh-Miller N E. . Electrostatics in periodic boundary conditions and real-space corrections. Physical Review B: Condensed Matter and Materials Physics, 2008, 77(11): 115139

[53] Schleder G R, Acosta C M, Fazzio A. Exploring two-dimensional materials thermodynamic stability via machine learning. ACS Applied Materials & Interfaces, 2020, 12(18): 20149–20157

[54] Mounet N, Gibertini M, Schwaller P. . Two-dimensional materials from high-throughput computational exfoliation of experimentally known compounds. Nature Nanotechnology, 2018, 13(3): 246–252

[55] Zhou J, Shen L, Costa M D. . 2DMatPedia, an open computational database of two-dimensional materials from top-down and bottom-up approaches. Scientific Data, 2019, 6(1): 86

[56] Choudhary K, Kalish I, Beams R. . High-throughput identification and characterization of two-dimensional materials using density functional theory. Scientific Reports, 2017, 7(1): 5179

[57] Ashton M, Paul J, Sinnott S B. . Topology-scaling identification of layered solids and stable exfoliated 2D materials. Physical Review Letters, 2017, 118(10): 106101

[58] Huang P, Lukin R, Faleev M. . Unveiling the complex structure-property correlation of defects in 2D materials based on high throughput datasets. npj 2D Materials and Applications, 2023, 7(1): 6

[59] Bertoldo F, Ali S, Manti S, Thygesen K S. Quantum point defects in 2D materials—The QPOD database. npj Computational Materials, 2022, 8(1): 56

[60] Singh A K, Mathew K, Zhuang H L. . Computational screening of 2D materials for photocatalysis. Journal of Physical Chemistry Letters, 2015, 6(6): 1087–1098

[61] Mathew K, Sundararaman R, Letchworth-Weaver K. . Implicit solvation model for density-functional study of nanocrystal surfaces and reaction pathways. Journal of Chemical Physics, 2014, 140(8): 084106

[62] Steinmann S N, Sautet P, Michel C. Solvation free energies for periodic surfaces: Comparison of implicit and explicit solvation models. Physical Chemistry Chemical Physics, 2016, 18(46): 31850–31861

[63] Pan B C, Yang W S, Yang J. Formation energies of topological defects in carbon nanotubes. Physical Review B: Condensed Matter, 2000, 62(19): 12652–12655

[64] Malyi O I, Sopiha K V, Persson C. Energy, phonon, and dynamic stability criteria of two-dimensional materials. ACS Applied Materials & Interfaces, 2019, 11(28): 24876–24884

[65] Chen Z, Dinh H N, Miller E. Photoelectrochemical Water Splitting. New York: Springer, 2013

[66] Xiao H, Tahir-Kheli J, Goddard W A III. Accurate band gaps for semiconductors from density functional theory. Journal of Physical Chemistry Letters, 2011, 2(3): 212–217

[67] Crowley J M, Tahir-Kheli J, Goddard W A III. Resolution of the band gap prediction problem for materials design. Journal of Physical Chemistry Letters, 2016, 7(7): 1198–1203

[68] Jain M, Chelikowsky J R, Louie S G. Reliability of hybrid functionals in predicting band gaps. Physical Review Letters, 2011, 107(21): 216806

[69] Saßnick H D, Cocchi C. Electronic structure of cesium-based photocathode materials from density functional theory: Performance of PBE, SCAN, and HSE06 functionals. Electronic Structure., 2021, 3(2): 027001

[70] Mourino B, Jablonka K M, Ortega-Guerrero A. . In search of covalent organic framework photocatalysts: A DFT-based screening approach. Advanced Functional Materials, 2023, 33(32): 2301594

[71] Wang H, Jin S, Zhang X. . Excitonic effects in polymeric photocatalysts. Angewandte Chemie International Edition, 2020, 59(51): 22828–22839

[72] Botti S, Sottile F, Vast N. . Long-range contribution to the exchange-correlation kernel of time-dependent density functional theory. Physical Review B: Condensed Matter and Materials Physics, 2004, 69(15): 155112

[73] Kronik L, Stein T, Refaely-Abramson S. . Excitation gaps of finite-sized systems from optimally tuned range-separated hybrid functionals. Journal of Chemical Theory and Computation, 2012, 8(5): 1515–1531

[74] Shu H, Li Y, Niu X. . Greatly enhanced optical absorption of a defective MoS₂ monolayer through oxygen passivation. ACS Applied Materials & Interfaces, 2016, 8(20): 13150–13156

[75] Biswas T, Singh AK. Excitonic effects in absorption spectra of carbon dioxide reduction photocatalysts. npj Computational Materials, 2021, 7(1): 189

[76] Ugeda M M, Bradley A J, Shi S F. . Giant bandgap renormalization and excitonic effects in a monolayer transition metal dichalcogenide semiconductor. Nature Materials, 2014, 13(12): 1091–1095

[77] Qiu D Y, da Jornada F H, Louie S G. Optical spectrum of MoS₂: Many-body effects and diversity of exciton states. Physical Review Letters, 2013, 111(21): 216805

[78] Abild-Pedersen F, Greeley J, Studt F. . Scaling properties of adsorption energies for hydrogen-containing molecules on transition-metal surfaces. Physical Review Letters, 2007, 99(1): 016105

[79] Wang X, Zhang G, Yang L. . Material descriptors for photocatalyst/catalyst design. Wiley Interdisciplinary Reviews. Computational Molecular Science, 2018, 8(5): e1369

[80] Liu F, Shi C, Guo X. . Rational design of better hydrogen evolution electrocatalysts for water splitting: A review. Advanced Science, 2022, 9(18): 2200307

[81] Østergaard F C, Bagger A, Rossmeisl J. Predicting catalytic activity in hydrogen evolution reaction. Current Opinion in Electrochemistry, 2022, 35: 101037

[82] Hammer B, Nørskov J K. Electronic factors determining the reactivity of metal surfaces. Surface Science, 1995, 343(3): 211–220

[83] Nilsson A, Pettersson L G M, Hammer B. . The electronic structure effect in heterogeneous catalysis. Catalysis Letters, 2005, 100(3): 111–114

[84] Shu H, Zhou D, Li F. . Defect engineering in MoSe₂ for the hydrogen evolution reaction: From point defects to edges. ACS Applied Materials & Interfaces, 2017, 9(49): 42688–42698

[85] Pérez-Ramírez J, López N. Strategies to break linear scaling relationships. Nature Catalysis, 2019, 2(11): 971–976

[86] Huang Y, Nielsen R J, Goddard W A III. . The reaction mechanism with free energy barriers for electrochemical dihydrogen evolution on MoS2. Journal of the American Chemical Society, 2015, 137(20): 6692–6698

[87] Knøsgaard N R, Thygesen K S. Representing individual electronic states for machine learning GW band structures of 2D materials. Nature Communications, 2022, 13(1): 468

[88] Tawfik S A, Isayev O, Stampfl C. . Efficient prediction of structural and electronic properties of hybrid 2D materials using complementary DFT and machine learning approaches. Advanced Theory and Simulations, 2019, 2(1): 1800128

[89] Zhu Z, Dong B, Guo H. . Fundamental band gap and alignment of two-dimensional semiconductors explored by machine learning. Chinese Physics B, 2020, 29(4): 046101

[90] Rajan A C, Mishra A, Satsangi S. . Machine-learning-assisted accurate band gap predictions of functionalized MXene. Chemistry of Materials, 2018, 30(12): 4031–4038

[91] Dau M T, Al Khalfioui M, Michon A. . Descriptor engineering in machine learning regression of electronic structure properties for 2D materials. Scientific Reports, 2023, 13(1): 5426

[92] Manzoor A, Arora G, Jerome B. . Machine learning based methodology to predict point defect energies in multi-principal element alloys. Frontiers in Materials, 2021, 8: 673574

[93] Choudhary K, Sumpter B G. Can a deep-learning model make fast predictions of vacancy formation in diverse materials. AIP Advances, 2023, 13(9): 095109

[94] Mueller T, Hernandez A, Wang C. Machine learning for interatomic potential models. Journal of Chemical Physics, 2020, 152(5): 050902

[95] Musa E, Doherty F, Goldsmith B R. Accelerating the structure search of catalysts with machine learning. Current Opinion in Chemical Engineering, 2022, 35: 100771

[96] Mortazavi B, Silani M, Podryabinkin E V. . First-principles multiscale modeling of mechanical properties in graphene/borophene heterostructures empowered by machine-learning interatomic potentials. Advanced Materials, 2021, 33(35): 2102807

[97] Sun S, Singh A, Li Y, et al. Machine learning accelerated atomistic simulations for 2D materials with defects. In: ASME International Mechanical Engineering Congress and Exposition, 2023: American Society of Mechanical Engineers

[98] Goryaeva A M, Dérès J, Lapointe C. . Efficient and transferable machine learning potentials for the simulation of crystal defects in bcc Fe and W. Physical Review Materials, 2021, 5(10): 103803

[99] Dragoni D, Daff T D, Csányi G. . Achieving DFT accuracy with a machine-learning interatomic potential: Thermomechanics and defects in bcc ferromagnetic iron. Physical Review Materials, 2018, 2(1): 013808

[100] Kumar R, Singh AK. Chemical hardness-driven interpretable machine learning approach for rapid search of photocatalysts. npj Computational Materials, 2021, 7(1): 197

[101] Pan R, Liu J, Zhang J. Defect engineering in 2D photocatalytic materials for CO₂ reduction. ChemNanoMat: Chemistry of Nanomaterials for Energy, Biology and More, 2021, 7(7): 737–747

[102] Xiong J, Di J, Xia J. . Surface defect engineering in 2D nanomaterials for photocatalysis. Advanced Functional Materials, 2018, 28(39): 1801983

[103] Shi R, Zhao Y, Waterhouse G I. . Defect engineering in photocatalytic nitrogen fixation. ACS Catalysis, 2019, 9(11): 9739–9750

[104] Sun X, Zhang X, Xie Y. Surface defects in two-dimensional photocatalysts for efficient organic synthesis. Matter, 2020, 2(4): 842–861

[105] Zhao Y, Chen G, Bian T. . Defect-rich ultrathin ZnAl-layered double hydroxide nanosheets for efficient photoreduction of CO₂ to CO with water. Advanced Materials, 2015, 27(47): 7824–7831

[106] Zhang X, Chen A, Chen L. . 2D materials bridging experiments and computations for electro/photocatalysis. Advanced Energy Materials, 2022, 12(4): 2003841

[107] Momeni K, Ji Y, Wang Y. . Multiscale computational understanding and growth of 2D materials: A review. npj Computational Materials, 2020, 6(1): 22

[108] Kovačič Ž, Likozar B, Huš M. Photocatalytic CO₂ reduction: A review of ab initio mechanism, kinetics, and multiscale modeling simulations. ACS Catalysis, 2020, 10(24): 14984–15007

[109] Koparde V N, Cummings P T. Molecular dynamics study of water adsorption on TiO₂ nanoparticles. Journal of Physical Chemistry C, 2007, 111(19): 6920–6926

[110] Raymand D, van Duin A C T, Spångberg D. . Water adsorption on stepped ZnO surfaces from MD simulation. Surface Science, 2010, 604(9–10): 741–752

[111] Liu D J, Garcia A, Wang J. . Kinetic Monte Carlo simulation of statistical mechanical models and coarse-grained mesoscale descriptions of catalytic reaction–diffusion processes: 1D nanoporous and 2D surface systems. Chemical Reviews, 2015, 115(12): 5979–6050

[112] Deskins N A, Rao P M, Dupuis M. Charge carrier management in semiconductors: Modeling charge transport and recombination. In: Bahnemann D, Patrocinio A O T, eds. Springer Handbook of Inorganic Photochemistry. Cham: Springer International Publishing, 2022

[113] Lin F, Huang J, Hin C. Electron transport from quantum kinetic Monte Carlo simulations. Journal of Physical Chemistry C, 2018, 122(35): 20550–20554

[114] Santos E. In search of lost descriptors: Correlations and their risks. Current Opinion in Electrochemistry, 2022, 37: 101194

RIGHTS & PERMISSIONS

Higher Education Press 2024