Beyond geometrical screening in predicting two-dimensional materials

Shota Ono · 2026

Read Beyond geometrical screening in predicting two-dimensional materials by Shota Ono (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-10462.txt.

Introduction

Shota Ono surveys how novel two-dimensional materials are predicted, and argues the field must go beyond geometrical screening. A few thousand 2D materials, van der Waals and non-vdW kinds alike, have been predicted as exfoliable or stable. Yet only a few hundred have been made, leaving a wide gap between theory and experiment. This perspective reviews the families involved and their promise. It then focuses on recent progress in predicting which non-vdW 2D materials can actually be synthesized, the step needed to close that gap.

Catalog record

CREATORShota Ono
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25236
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.10462

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Beyond geometrical screening in predicting two-dimensional materials Shota Ono Abstract This perspective overviews the family of two-dimensional (2D) materials, which have attracted significant attention due to their properties and potential applications, and discusses how novel 2D materials including van der Waals (vdW) and non-vdW 2D materials have been predicted so far. A few thousand 2D materials have been predicted to be exfoliable or dynamically/thermodynamically stable, whereas a few hundred 2D materials have been synthesized so far, highlighting a gap between the theoretical prediction and experiments. This perspective introduces the recent developments in predicting the synthesis of non-vdW 2D materials. I Introduction Since the exfoliation of monolayer graphene from graphite, two-dimensional (2D) materials have attracted much attention due to their physical and chemical properties that are markedly different from those of 3D counterparts. Graphene has a linear dispersion at the K point around the Fermi level, providing a platform for exploring the Dirac fermion physics, and its unique properties enable numerous applications 35 . 2D transition metal dichalcogenides MX_{2} , where M is the transition metal element and X is the chalcogen atom, adopt several phases such as 2H, 1T, and 1T ′ , governing the electronic, magnetic, and mechanical properties. Thus, phase engineering by strain and doping as well as the combination of M and X has been extensively studied, leading to next-generation electronics and valleytronics 48 . Beyond graphene, several 2D elemental materials termed Xenes have been explored. Most of them are classified as non-van der Waals (non-vdW) 2D materials because their 3D counterparts have non-layered structure. First-principles calculations have predicted that silicene (Si) and germanene (Ge) have buckled honeycomb structure that is different from the planar geometry of graphene 9 . Following the theoretical prediction, the experimental synthesis of silicene 44 ; 13 and germanene 26 ; 6 ; 12 were reported by several groups. Realization of stanene (Sn) 51 and plumbene (Pb) 47 were also reported. Borophene (B) was cleaved from \beta -rhombohedral boron 11 . The stability of 2D metals has been systematically investigated using first-principles approach 34 ; 37 . More recently, goldene has been synthesized, exhibiting a hexagonal monolayer structure derived from the Au(111) surface 23 . A recent review discusses the details of experimental synthesis routes for Xenes 31 . The physical properties of non-vdW 2D materials have been investigated, such as quantum spin Hall states in group IV 2D materials 1 ; 7 and in-plane transport properties of goldene 49 . The family of 2D materials has rapidly expanded in recent years, driven by the development of computational materials databases constructed from high-throughput first-principles calculations with the help of machine-learning approaches. Several thousand candidate 2D materials have been predicted 25 ; 3 ; 33 ; 17 ; 16 , and several databases have been used to explore their functional properties. However, only a few hundred 2D materials have been experimentally synthesized (see X2DB 2 ), highlighting limitations in current screening strategies. In particular, reliable approaches for predicting non-vdW 2D materials have not been established, as the stabilization mechanisms are fundamentally different from those of vdW 2D materials (see Fig. 1 for the classification of 2D materials). Therefore, it is timely to reexamine the construction principle underlying existing 2D materials databases. In this perspective, we review how 2D materials have been explored and predicted through screening materials space. This will help us to understand which 2D materials are actually synthesized experimentally, which is necessary for any practical applications. Therefore, we focus on structural stabilities in the following sections. Figure 1: Relationship between 3D and 2D materials. Through high-throughput screening of about 50,000 bulk materials, a few thousand potential vdW 2D materials are predicted to be exfoliable from layered materials. On the other hand, non-vdW 2D materials are identified by studying the stringent stability check for their monolayer structures. A few hundred 2D materials are synthesized experimentally. The crystal structure is visualized by using VESTA 32 . II How to predict 2D materials The Inorganic crystal structure database (ICSD), the world’s largest materials database, contains more than 327,833 materials (on October 21, 2025). The Materials Project (MP) contains about 154,879 materials, among which 49,283 materials have an ICSD number (on September 25, 2025) 20 . Several approaches have been proposed to estimate the number of possible 2D materials that can be created from 3D structures. Before considering the approaches used in the prediction of 2D materials, it should be emphasized how the materials prediction has been done in the field of computational materials science. The “stable material” should have (i) non-positive formation energy to avoid the decomposition into different constituents and (ii) positive vibrational frequencies over the entire Brillouin zone to avoid the phase transition into different crystal structures at zero and finite temperatures 30 . In general, the dynamical stability at zero and finite temperatures are investigated through phonon dispersion calculations and molecular dynamics simulations, respectively. To accurately predict the quadratic dispersion of out-of-plane vibrational modes, it is important to consider the invariance and equilibrium conditions for interatomic force constants 27 . If a material satisfies the conditions (i) and (ii), such a material is regarded as stable. However, this does not guarantee the experimental realization of the predicted material: although the two conditions may allow the existence of the local minimum in the potential energy surface (PES) characterized by many degrees of freedom (lattice parameters, atomic positions, and atomic species), their conditions do not provide how to reach the energy minimum. It is desirable to include kinetic effects and growth pathways of the materials synthesis in the computational simulation, which remains an open question in this research field. II.1 Geometrical screening Graphene and MX_{2} are classified into vdW 2D materials, where the interlayer bonds are much weaker than the intralayer bonds in the parent 3D crystal. The vdW 2D materials not only exhibit interesting properties themselves, but also open up the Moiré physics in multilayered heterostructures. Exploring novel vdW 2D materials is crucial to advance the 2D materials physics. We first discuss a top-down approach that is applicable to the prediction of vdW 2D materials. Lebègue et al. have focused on the geometry of the crystal structures contained in 3D materials database 25 . They identified layered structures by investigating whether the packing ratio is small (0.15–0.5) and the interlayer distance along the c axis is sufficiently large ( >2.4 Å). This screening approach was improved by several research groups 3 ; 24 ; 33 ; 17 ; 16 . For example, a vdW gap is identified if the interatomic distance d_{ij} between atoms i and j satisfies the inequality \displaystyle d_{ij}>k(r_{i}+r_{j}), (1) where r_{i} is the atomic radius of atom i and k is a parameter 24 . On the other hand, d_{ij}<k(r_{i}+r_{j}) indicates the presence of covalent bond between atoms i and j . In this way, the chemical bonds are identified for each 3D material. A different inequality such as d_{ij}>r_{i}+r_{j}+\Delta has also been used, where \Delta is a parameter 33 . Once different atoms are bonded in the unit cell, a topology-scaling approach (TSA) developed by Ashton et al. was used to determine the material’s dimensionality 3 . In TSA, one identifies the cluster in the unit cell

Read the complete text