High-Throughput Discovery of Two-Dimensional Materials Exhibiting Strong Rashba-Edelstein effect

The Rashba-Edelstein effect (REE), which generates spin accumulation under an applied electric current, quantifies charge-to-spin conversion (CSC) efficiency in non-centrosymmetric systems. However, systematic investigations of REE in two-dimensional (2D) materials remain scarce. To address this gap, we perform a comprehensive symmetry analysis based on the 80 crystallographic layer groups, elucidating the relationship between materials’ symmetries and the geometric characteristics of the REE response tensor. Our analysis identifies 13 distinct symmetry classes for the tensor and reveals all potential material candidates. Considering the requirement of strong spin-orbit coupling for a large REE response, we screen the C2DB database and identify 54 promising 2D materials. First-principles calculations demonstrate that the largest REE response coefficients in these materials exceed those reported for other 2D systems by an order of magnitude, indicating exceptionally high CSC efficiency. Focusing on three representative materials, including HgI2, AgTlP2Se6 and BrGaTe, we show that their large response coefficients can be well explained by effective kp models and the characteristic spin textures around high-symmetry points in momentum space. This work provides a systematic framework and identifies high-performance candidates, paving the way for future exploration of REE-driven CSC in 2D materials.

💡 Research Summary

The Rashba‑Edelstein effect (REE) converts an applied electric current into a non‑equilibrium spin accumulation, providing a direct measure of charge‑to‑spin conversion (CSC) efficiency in systems lacking inversion symmetry. While REE has been extensively studied in model Hamiltonians and heterostructures, a systematic exploration of intrinsic two‑dimensional (2D) materials has been missing. In this work, the authors first perform a comprehensive group‑theoretical analysis of the REE response tensor χᵢⱼ for all 80 crystallographic layer groups (LGs). By classifying the symmetry operations into seven distinct families, they derive the allowed tensor forms and identify 13 unique symmetry classes that dictate which tensor components may be non‑zero. Importantly, because a 2D slab is surrounded by vacuum, an electric field perpendicular to the plane cannot generate spin‑polarized currents; consequently, the third column of χ is identically zero for all 2D systems.

Armed with this symmetry framework, the authors embark on a high‑throughput computational search using the C2DB database, which contains 1 690 052 candidate 2D compounds. They filter out magnetic and dynamically unstable entries, leaving 3 488 non‑magnetic, phonon‑stable structures. Recognizing that strong spin‑orbit coupling (SOC) is essential for a large REE, they further restrict the pool to materials containing elements from periods 4–6 and lacking inversion symmetry, arriving at 1 089 candidates. These are grouped by LG, and representative compounds are selected from each of the 13 tensor‑symmetry categories, with a bias toward structures that appear frequently within a category. After eliminating overly complex unit cells (>20 atoms), the authors perform density‑functional theory (DFT) calculations of the REE tensor for 54 representative materials.

The calculated χ values reveal that many of the screened compounds exhibit REE coefficients an order of magnitude larger than those reported for prototypical 2D systems such as In₂Se₃ (≈1.3 × 10¹⁰ e·cm·V⁻¹) and WTe₂ (≈1.4 × 10¹⁰ e·cm·V⁻¹). The three most striking examples—HgI₂ (LG P4̅m2, point group D₂d), AgTlP₂Se₆ (LG P3m1) and BrGaTe (LG C2)—show χ components reaching 4.5 × 10¹⁰ to 3.7 × 10¹¹ e·cm·V⁻¹ over a broad range of chemical potentials. Detailed analysis of these materials uncovers the microscopic origins of their giant REE. For HgI₂, the D₂d symmetry yields a Dresselhaus‑type spin texture around the Γ point, described by an effective k·p Hamiltonian H = k_yσ_x + k_xσ_y, leading to tangential‑radial spin winding that maximizes the transverse spin response. In AgTlP₂Se₆, the P3m1 symmetry enforces χ_xx = χ_yy, and the spin‑orbit splitting at the K/K′ valleys produces a chiral spin texture that aligns with the direction of the shifted Fermi surface, suppressing cancellation of spin contributions and enhancing χ. BrGaTe, despite being non‑polar, possesses a C₂ axis that permits only off‑diagonal tensor elements (χ_xy = χ_yx), resulting in a highly anisotropic spin accumulation that can be tuned by the direction of the applied current.

The authors also map the dependence of χ on the chemical potential, demonstrating that the REE can be optimized by modest electron or hole doping, which shifts the Fermi level into regions of strong spin‑splitting. This tunability is crucial for device engineering, as it suggests that gate‑controlled CSC can be achieved without altering the material’s crystal structure.

In summary, the paper delivers a two‑pronged contribution: (1) a rigorous symmetry classification of the REE tensor for all 2D layer groups, providing a universal lookup for which tensor components are symmetry‑allowed; (2) a high‑throughput identification of 54 promising 2D materials, with three highlighted compounds exhibiting REE coefficients up to ten times larger than previously known systems. The combination of symmetry analysis, DFT screening, and effective k·p modeling establishes clear design principles for maximizing charge‑to‑spin conversion in 2D crystals. These findings open a pathway toward low‑power, ultrafast spintronic devices such as spin‑orbit torque switches, spin‑charge interconversion layers, and non‑reciprocal transport platforms, where the intrinsic REE can be harnessed directly.