🔬 mesoscale physics

Current-induced spin-orbit torque on the surface of a transition metal dichalcogenide connected to a two-dimensional ferromagnet CrI $_3$ : Effects of twisting and gating

This study employs the steady-state Boltzmann equation to demonstrate that current-induced spin-orbit torque in TMDC/CrI $_3$ bilayers can be significantly enhanced, reversed, and tuned by up to an order of magnitude through the strategic manipulation of doping type, twist angle, and transverse gate electric fields.

Original authors: Leyla Majidi, Azadeh Faridi, Reza Asgari

Published 2026-08-04

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Leyla Majidi, Azadeh Faridi, Reza Asgari

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the world of tiny computers as a bustling city where information travels like cars on a highway. For decades, these cars have been made of electricity—streams of electrons zipping around. But electricity is messy; it generates heat and wastes energy, like a car engine that overheats. Scientists are now trying to build a smarter city using "spin" instead of just charge. Think of spin not as a car, but as a tiny, spinning top that each electron carries. If you can control the direction these tops spin, you can store and process information much faster and with less heat. This is the exciting world of spintronics.

To make these spinning tops do what we want, we need to give them a nudge. Usually, this nudge comes from a magnetic field, but that's like trying to push a whole parade by shouting at it. A better way is to use a "spin-orbit torque." Imagine a crowded dance floor where the floor itself is slippery and twisted. If you slide across it, the friction doesn't just slow you down; it makes you spin. In the microscopic world, when electrons slide across certain special materials, the material's internal structure forces them to spin in a specific direction. This creates a "torque" (a twisting force) that can flip magnetic switches, turning data on and off. The big question for scientists is: how can we make this spinning force stronger, more controllable, and easier to tune?

This paper dives into a specific, high-tech playground to answer that question. The researchers are looking at a sandwich made of two ultra-thin, two-dimensional materials: a transition metal dichalcogenide (TMDC), which acts like the slippery dance floor, and a magnetic layer called Chromium Iodide (CrI3), which acts like the target we want to spin. They are specifically testing two types of dance floors: WSe2 and MoSe2. The team uses computer simulations (like a super-advanced video game physics engine) to see what happens when they run an electric current through this sandwich. They aren't just watching; they are playing with the controls, twisting the layers relative to each other like a Rubik's cube and adjusting a "gate" voltage (like turning a faucet) to see how these changes affect the spinning force.

Here is what they found in their simulations. First, the type of material matters a lot. When they used MoSe2, the "damping-like" torque (a specific kind of spin push that helps flip the magnet) became incredibly strong—up to 1,000 times stronger than in the WSe2 version. In fact, in the MoSe2 sandwich, this damping force became almost as strong as the other main force, the "field-like" torque. This is a big deal because usually, one force dominates and the other is weak. They also discovered that the direction of the spin push depends heavily on whether the material is "n-doped" (filled with extra electrons) or "p-doped" (missing electrons), acting like a switch that changes the sign of the force.

The most playful part of their discovery involves twisting. Imagine holding two sheets of paper with a pattern on them and rotating one relative to the other. The researchers found that changing the "twist angle" between the layers is a powerful tool. Depending on the material and the chemical settings, twisting the layers could flip the direction of the spin torque entirely. For instance, in a MoSe2 sandwich, twisting it could reverse the sign of the force. Furthermore, they found that applying a transverse electric field (a gate voltage) could tune the strength of this torque by nearly ten times and even cause it to flip signs at a specific twist angle of 10.16 degrees.

In short, the paper suggests that by carefully choosing the material (MoSe2 over WSe2), adjusting the chemical doping, and precisely twisting the layers, we can create a highly tunable system for generating spin-orbit torque. While these results come from simulations and not yet from physical experiments, they offer a promising roadmap for designing future spintronic devices where the magnetic switches can be controlled with high precision using simple twists and electrical gates.

Technical Summary: Current-Induced Spin-Orbit Torque in Twisted TMDC/CrI3 Bilayers

Problem Statement and Motivation
The manipulation of magnetization via spin-orbit torque (SOT) is a central objective in spintronic memory and logic devices. While conventional heavy metal/ferromagnet (HM/FM) and topological insulator/ferromagnet (TI/FM) bilayers have demonstrated significant SOT efficiencies, transition metal dichalcogenides (TMDCs) offer a promising alternative due to their strong intrinsic spin-orbit coupling (SOC) and diverse symmetry features. Recent studies have highlighted the tunability of proximity exchange in TMDC/ferromagnet (FM) heterostructures, specifically between TMDCs (such as MoSe2 or WSe2) and the 2D ferromagnet chromium iodide (CrI3), via gating and twisting. However, a comprehensive microscopic theory describing current-induced SOT in these twisted TMDC/CrI3 bilayers beyond the linear response regime, particularly regarding the interplay of doping, twist angle, and gate voltage, remains an open area of investigation.

Methodology
The authors develop a microscopic theory to investigate current-induced spin polarization and the resulting SOT in bilayers composed of a TMDC (WSe2 or MoSe2) and monolayer CrI3. The study operates beyond the linear response regime using the steady-state Boltzmann equation within the relaxation time approximation.

The theoretical framework employs a low-energy effective Hamiltonian that incorporates:

Intrinsic valley-dependent SOC ( $\lambda_c, \lambda_v$ ).
Proximity-induced exchange coupling ( $B_c, B_v$ ) arising from the CrI3 layer.
Rashba SOC ( $\lambda_R$ ) induced by inversion symmetry breaking in the heterostructure.
Orbital gap ( $\Delta$ ) and Fermi velocity ( $v_F$ ).

The spin polarization ( $S$ ) is decomposed into two distinct contributions:

Intraband transitions ( $S_{oc}$ ): Arising from changes in electron occupation within the same band due to electric field acceleration. This is calculated up to the second order in the electric field using recursive relations derived from the Boltzmann equation.
Interband transitions ( $S_{in}$ ): Arising from the modification of quasiparticle wave functions (intrinsic contribution), analogous to the intrinsic anomalous Hall effect.

The resulting SOT ( $\tau$ ) acting on the ferromagnetic layer is calculated as $\tau = \frac{2J}{\hbar} \mathbf{m} \times \mathbf{S}$ , where $J$ is the exchange energy and $\mathbf{m}$ is the magnetization vector. The study numerically evaluates these quantities for various chemical potentials ( $\mu$ ), twist angles ( $\theta_{twist}$ ), and Rashba coupling strengths ( $\lambda_R$ ).

Key Contributions and Results

Dominance of Intraband vs. Interband Mechanisms: The study finds that intraband transitions yield a strong field-like torque ( $\tau_{FL} \propto \mathbf{m} \times \mathbf{S}$ ), while interband transitions produce a comparatively weaker damping-like torque ( $\tau_{DL} \propto \mathbf{m} \times (\mathbf{m} \times \mathbf{S})$ ) in the WSe2/CrI3 system.
Material Dependence (WSe2 vs. MoSe2): A significant finding is the dramatic enhancement of the damping-like torque in n-doped MoSe2. In this system, the damping-like component is enhanced by up to three orders of magnitude compared to WSe2, reaching a strength comparable to the field-like torque. Furthermore, the field-like torque in MoSe2 is an order of magnitude larger than in the WSe2-based bilayer.
Asymmetry and Sign Reversal: Both torque components exhibit strong asymmetry between n-type and p-type doping. Notably, in n-doped MoSe2, the damping-like torque undergoes a sign reversal at specific chemical potentials, a behavior not observed in the WSe2 system. This is attributed to the negative intrinsic SOC parameter ( $\lambda_c$ ) in MoSe2, which creates a sharp peak in spin polarization and distinct competition between subbands.
Role of Twisting: The twist angle between the TMDC and CrI3 layers plays a crucial role. Twisting induces a sign reversal in the proximity exchange field within the valence band (occurring near 8° for WSe2 and 16° for MoSe2). Depending on the TMDC material and chemical potential, twisting can reverse the sign of the SOT and significantly modulate its magnitude. Notably, an anomalous behavior and sign reversal in the SOT are observed at a specific twist angle of 10.16°.
Gate Voltage Tunability: The application of a transverse gate electric field enables substantial tunability of the SOT strength (by nearly one order of magnitude). In the presence of twisting, the gate field can also induce sign reversals, such as at a twist angle of 10.16°.
Rashba Coupling Effects: Increasing the Rashba SOC parameter ( $\lambda_R$ ) generally enhances both field-like and damping-like torques in WSe2 and p-doped MoSe2. However, in n-doped MoSe2, increasing $\lambda_R$ reduces the damping-like torque strength and eliminates the sign reversal of the field-like torque, effectively broadening the spin polarization peaks and suppressing the competition between subbands.

Significance and Claims
The paper claims to provide a microscopic theory for current-induced SOT in TMDC/CrI3 bilayers that extends beyond linear response theory. The primary significance lies in demonstrating that the size and orientation of the SOT can be effectively controlled externally through three specific parameters: gate voltage, twist angle, and chemical potential (doping).

The authors assert that replacing WSe2 with MoSe2 leads to a significant amplification of the damping-like term, making it comparable to the field-like term, which is a desirable feature for efficient magnetization switching. Furthermore, the demonstration that twisting can reverse the sign of the SOT and that gate fields can modulate the torque magnitude by an order of magnitude suggests a viable pathway for designing externally controllable spintronic devices based on these van der Waals heterostructures. The study emphasizes that the interplay between intrinsic SOC, proximity exchange, and Rashba coupling dictates the geometric and topological features of the wave functions, which in turn determine the magnitude and sign of the resulting torques.

More like this