🔬 mesoscale physics

Large bias-tunable magnetoresistance from spin-dependent interlayer hybridization in van der Waals antiferromagnet CrSBr-based heterostructures

This study reveals that the large, bias-tunable magnetoresistance in CrSBr-based van der Waals heterostructures arises from spin-dependent interlayer hybridization that linearly modulates the barrier band-edge offset with magnetization angle, rather than from electrode spin polarization as described by the Jullière model.

Original authors: Sadeed Hameed (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany), Aditya Kumar (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany), Chengjie Yu (School

Published 2026-08-13

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Sadeed Hameed (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany), Aditya Kumar (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany), Chengjie Yu (School of Physics and Electronic Engineering, Jiangsu University, Zhenjiang, China), Aravind P. Balan (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany, Department of Materials Science and NanoEngineering, Rice University, Houston, Texas, USA), Xinran Wang (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany), Lichuan Zhang (School of Physics and Electronic Engineering, Jiangsu University, Zhenjiang, China), Yuriy Mokrousov (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany, Peter Grünberg Institut and Institute for Advanced Simulation, Forschungszentrum Jülich and JARA, Jülich, Germany), Mathias Kläui (Institute of Physics, Johannes Gutenberg University Mainz, Mainz, Germany, Centre for Quantum Spintronics, Department of Physics, Norwegian University of Science and Technology, Trondheim, Norway)

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 electronics as a bustling city where tiny packets of information, called electrons, zip through wires to power our phones and computers. For decades, this city has relied on a clever trick called "spintronics." Instead of just using the electron's charge (like a battery), scientists realized they could also use its "spin"—a tiny, invisible magnetic property that makes the electron act like a miniature compass needle pointing either up or down. By sorting these needles, we can store data more efficiently and build faster, smarter devices.

However, there's a catch. Most of these magnetic tricks rely on materials that are sensitive to air or require complex setups. Recently, scientists discovered a new family of materials called "van der Waals" crystals. Think of these like stacks of sticky notes; the layers stick together gently, allowing us to peel them apart and stack them in new, custom ways without breaking them. Among these, a material called CrSBr (Chromium Sulfide Bromide) has become a star. It's an "antiferromagnet," which is a fancy way of saying its internal magnetic needles point in opposite directions, canceling each other out, but they can be coaxed into aligning with a magnetic field. The big question scientists have been asking is: How exactly does this material control the flow of electricity when we twist its magnetic needles? Is it acting like a simple filter, or is something more complex happening inside?

The Mystery of the M-Shaped Curve

In this study, a team of researchers built a tiny sandwich-like device to solve this puzzle. They took a flake of CrSBr and placed it between two layers of graphene (a super-thin form of graphite) and a protective layer of hexagonal boron nitride (hBN). This setup acts like a gatekeeper for electrons. When they ran electricity through this gate, they noticed something strange and exciting: the resistance (how hard it is for electricity to flow) didn't just go up or down smoothly as they changed the voltage. Instead, it formed a giant "M" shape on their graphs. At specific voltages, around ±0.5 V, the resistance spiked dramatically, reaching a massive 350% change at cold temperatures (20 K).

This "M" shape was the clue. The researchers wanted to know why the electricity behaved this way. Was it because the material was acting like a simple filter, only letting "up-spin" electrons through while blocking "down-spin" ones? Or was the material's internal energy structure changing in a more sophisticated way?

The Detective Work: Twisting the Magnets

To find the answer, the scientists used a magnetic field as a dial. They could twist the magnetic needles inside the CrSBr layer from pointing in opposite directions (the natural, "antiferromagnetic" state) to pointing in the same direction (the "ferromagnetic" state). They didn't just snap them from one state to the other; they slowly rotated them through every angle in between, like turning a dimmer switch on a light.

As they twisted the magnets, they watched the "M" shape shift. They discovered that the energy barrier the electrons had to jump over changed in a very specific, predictable way. The height of this barrier didn't follow a complicated curve; it changed in a straight line as they rotated the magnets. This linear relationship was the smoking gun.

The Verdict: It's All About the Hybrid

The researchers found that the old idea of a simple "spin filter" didn't fit the data. If it were just a filter, the resistance would change in a different, curved pattern. Instead, their results pointed to a phenomenon called "spin-dependent interlayer hybridization."

Here is a simple way to picture it: Imagine the CrSBr material is made of two floors, and the electrons are trying to jump between them. In the "antiferromagnetic" state (opposite spins), the energy gap between the floors is wide, making the jump hard. In the "ferromagnetic" state (aligned spins), that energy gap shrinks, making the jump easier. But the magic is in the middle: as you twist the magnets, the energy gap doesn't just snap shut; it slides smoothly and linearly toward the lower value. This gap is the "band-edge offset," and it dictates how easily electricity flows.

The study confirms that this shifting energy gap is the reason for the huge resistance changes and the M-shaped curve. When the voltage is just right, the electrons in the "easy" state can jump the gap, while those in the "hard" state are still stuck, creating a massive difference in resistance. As the voltage gets even higher, the "hard" state catches up, and the difference shrinks, completing the "M" shape.

Why It Matters

The team also noticed that this effect disappears as the device gets warmer. Around 90 K, the "M" shape flattens out, and the resistance stops being so sensitive to the voltage. This happens because the heat gives the electrons enough energy to hop over the barrier randomly, ignoring the neat, magnetic control the scientists were trying to use.

By proving that the resistance is controlled by this shifting energy gap rather than a simple filter, the paper gives us a new blueprint for building future electronic devices. It shows that in these van der Waals materials, we can program how electricity flows simply by adjusting the magnetic angle and the voltage. This isn't just a cool physics trick; it's a potential roadmap for creating new types of memory and sensors that are faster, more efficient, and capable of storing more information in a tiny space. The researchers didn't just find a new effect; they figured out the exact rulebook governing how it works, ruling out the simpler explanations and pointing the way toward a deeper understanding of magnetic electronics.

Technical Summary: Large Bias-Tunable Magnetoresistance from Spin-Dependent Interlayer Hybridization in CrSBr Heterostructures

Problem Statement
The origin of large, non-monotonic magnetoresistance (MR) observed in van der Waals (vdW) insulator junctions, particularly those involving A-type antiferromagnetic semiconductors, remains unclear. While significant MR effects have been reported in materials like CrI $_3$ , NiBr $_2$ , and CrSBr, previous interpretations have been qualitative and often contradictory. Proposed mechanisms include exchange-split conduction-band edges entering the bias window, resonance shifts in layer-by-layer spin valves, and magnon-assisted inelastic channels. A critical gap exists in understanding how the magnetic configuration of the barrier dictates the band edge that limits vertical transport. Furthermore, distinguishing between a spin-filter effect (where electrode polarization dominates) and a barrier-dependent band-edge offset requires testing across a continuous range of interlayer magnetization angles, rather than just at collinear antiferromagnetic (AFM) and ferromagnetic (FM) states.

Methodology
The authors investigated a vertical heterostructure consisting of hexagonal boron nitride (hBN), few-layer graphene (FLG), a 12 nm flake of CrSBr, and FLG contacts, fabricated entirely within an inert argon atmosphere to prevent interface degradation. The device was characterized using magnetotransport measurements in a 3D vector cryostat (1.6 K to 160 K, up to 5 T).

The study employed two independent magnetic tuning knobs:

Bias Voltage and Temperature: Full current-voltage ( $I$ – $V$ ) characteristics were recorded between -1 V and 1 V across a temperature range of 20 K to 160 K.
Magnetic Field Orientation: Fields were applied along two orthogonal axes:
- The easy ( $b$ ) axis, which induces a sharp layer-by-layer spin-flip transition at $\sim$ 0.3 T, switching the barrier between the zero-field AFM ground state and the field-polarized FM state.
- The hard ( $c$ ) axis, which supports a continuous second-order canting transition, allowing for the continuous variation of the relative interlayer magnetization angle $\theta$ from AFM to FM alignment.

Transport spectroscopy was performed using two complementary analyses of the $I$ – $V$ data:

Normalized Differential Conductance: $(dI/dV)/(I/V)$ to identify peak positions ( $V_{peak}$ ) associated with the band-edge offset.
Fowler–Nordheim (FN) Representation: $\ln(|I|/V^2)$ vs. $1/V$ to identify transition voltages ( $V_T$ ) marking the onset of field emission.

Key Results

M-Shaped Magnetoresistance: At 20 K, the MR ratio exhibits a distinct non-monotonic, M-shaped dependence on bias voltage, with symmetric maxima reaching $\sim$ 350% at $\pm$ 0.5 V. This feature persists for both easy and hard axis field sweeps. As temperature increases, the peaks shift to lower bias, merge near 90 K, and the MR magnitude drops below 100% near 100 K, vanishing at the Néel temperature ( $T_N \approx$ 132 K).
Band-Edge Offset Determination: Transport spectroscopy reveals that the bias at which the conductance peaks ( $V_{peak}$ ) and the FN transition voltage ( $V_T$ ) shift systematically between the AFM and FM states. At 20 K, $V_{peak}$ shifts from $\sim$ 0.80 V (AFM) to $\sim$ 0.62 V (FM), corresponding to a band-edge offset difference ( $\Phi_{AFM} - \Phi_{FM}$ ) of approximately 0.11 eV.
Linear Dependence on Interlayer Angle: By utilizing the hard-axis canting transition, the authors mapped the band-edge offset $\Phi$ against the relative interlayer angle $\theta$ . The data follows a linear relationship with $\cos(\theta/2)$ , where $\Phi(\theta) = \Phi_{AFM} - (\Phi_{AFM} - \Phi_{FM})\cos(\theta/2)$ . This linear dependence is a first-order signature of spin-dependent interlayer hybridization.
Temperature-Dependent Transport Regimes: The study identifies a crossover near 90 K ( $\sim$ $\sim$ 0.7 $T_N$ $T_{N}$ ):
- Below 90 K: Transport is dominated by variable-range hopping (VRH) through localized band-tail states (Efros–Shklovskii regime), exhibiting a small, field-independent activation energy ( $\sim$ 6.5 meV).
- Above 90 K: Transport becomes thermally activated over the conduction-band edge. The activation energy becomes field-dependent, showing an AFM-to-FM splitting of $\sim$ 20 meV, consistent with exchange splitting at the band edge.

Key Contributions and Mechanism
The paper establishes that the large, bias-tunable magnetoresistance in CrSBr junctions is dictated by a magnetic-configuration-dependent band edge rather than electrode spin polarization. The authors demonstrate that:

The linear dependence of the band-edge offset on $\cos(\theta/2)$ rules out the Jullière model (which predicts a quadratic $1-h^2$ dependence) and simple spin-filter projections.
The mechanism is identified as spin-dependent interlayer hybridization. In the AFM state, adjacent layer states are degenerate; as the angle $\theta$ changes, the spin-dependent interlayer coupling $t_\perp \propto \cos(\theta/2)$ splits these states, shifting the band edge and the effective barrier height.
The observed energy scale of the offset difference ( $\sim$ 0.11 eV) aligns with first-principles calculations of spin-dependent interlayer hybridization energies in CrSBr.
The disappearance of the M-shaped MR at higher temperatures is explained by the transition from field-emission-dominated transport (where the barrier height difference is critical) to thermally activated transport (where the barrier difference is less pronounced relative to thermal energy).

Significance
The authors conclude that this work resolves the origin of the large, non-monotonic MR in vdW antiferromagnetic junctions. By separating the band-edge offset picture from competing models (layer-by-layer spin valves, magnon-assisted tunneling, and Jullière TMR), the study provides a quantitative framework for understanding transport in these systems. The findings suggest that the bias voltage acts as an energy-resolved probe of the barrier's electronic structure, allowing the resistance to be programmed by the position of the bias relative to the magnetic-configuration-dependent band edges. This offers a template for reading out magnetic order in other semiconducting vdW antiferromagnets, emphasizing that the magnetic state of the barrier itself, rather than the electrodes, sets the transport characteristics.

More like this