🔬 mesoscale physics

Charge to spin conversion in atomically thin bismuth

This study demonstrates efficient charge-to-spin conversion in a hybrid heterostructure of atomically thin bismuth sandwiched between silicon carbide and epitaxial graphene, characterized by strong spin-orbit coupling and a significantly enhanced in-plane spin torque compared to control samples.

Original authors: Wilson J. Yánez-Parreño, Alexander Vera, Sandra Santhosh, Chengye Dong, Jimmy C. Kotsakidis, Yongxi Ou, Saurav Islam, Adam L. Friedman, Maxwell Wetherington, Joshua Robinson, Nitin Samarth

Published 2026-05-14

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Wilson J. Yánez-Parreño, Alexander Vera, Sandra Santhosh, Chengye Dong, Jimmy C. Kotsakidis, Yongxi Ou, Saurav Islam, Adam L. Friedman, Maxwell Wetherington, Joshua Robinson, Nitin Samarth

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 you have a tiny, invisible river of electricity flowing through a wire. Usually, this river just carries charge (electrons) from point A to point B. But in the world of spintronics—a field trying to build faster, more efficient computers—scientists want to do something magical: turn that river of charge into a river of "spin."

Think of "spin" not as a physical spinning top, but as a tiny magnetic compass needle attached to every electron. If you can make all these compass needles point in the same direction, you can store information or move data without using as much energy.

This paper is about a team of scientists who successfully built a "factory" to turn charge into spin using a very special, ultra-thin material: Bismuth.

Here is the story of how they did it, explained simply:

1. The Challenge: The "Oxidation" Problem

Bismuth is a metal that is naturally very good at converting electricity into spin (thanks to a property called "spin-orbit coupling"). However, it's like a piece of fresh fruit: if you leave it out in the air, it rots (oxidizes) almost immediately. This makes it incredibly hard to study or use in real devices because it gets ruined before you can measure it.

2. The Solution: The "Club Sandwich"

To solve this, the researchers built a protective "club sandwich" using a technique called Confinement Heteroepitaxy.

The Bottom Bun: A silicon carbide (SiC) wafer (a hard, ceramic-like base).
The Filling: A layer of Bismuth atoms.
The Top Bun: A layer of graphene (a super-thin sheet of carbon).

They heated the sandwich up, causing the Bismuth to melt and slide between the bottom bun and the top bun, getting trapped safely inside. Because the graphene acts as a lid, the Bismuth never touches the air. It stays fresh, stable, and "atomically thin" (only two atoms thick).

3. Checking the Sandwich

Before testing the electricity, they had to make sure the sandwich was built right. They used several "microscopes" and scanners:

X-ray Photoelectron Spectroscopy (XPS): Like a chemical fingerprint scanner, this confirmed the Bismuth was actually there and in its metallic form, not oxidized.
Electron Microscopy: They took a cross-section of the sandwich and saw a neat, bright line of Bismuth atoms sitting perfectly between the layers.
Raman Spectroscopy: This is like listening to the material "sing." The Bismuth layer sang a specific low-frequency song that proved it was there, covering about 96.5% of the surface.

4. The Magic Trick: Turning Charge into Spin

Once they confirmed the sandwich was good, they tested if it could do the magic trick: turning electricity into spin.

They placed a tiny magnet (Permalloy) on top of the graphene lid. Then, they sent a radio-frequency electrical current through the sandwich.

The Result: The electricity flowing through the Bismuth layer generated a "push" (torque) on the magnet above it.
The Comparison: They compared this to a control sample (just graphene without the Bismuth). The Bismuth sandwich was 3.75 times more effective at turning charge into spin than the plain graphene.

5. What the Spin Looks Like

The scientists also figured out the direction of this spin. Imagine the electricity flowing North. The spin they created was pointing East (perpendicular to the flow). This is exactly what you want for efficient computer memory.

The Catch (The "Imperfect" Sandwich)

The paper is honest about a flaw: the sandwich wasn't perfect everywhere. In some spots, the Bismuth layer was thicker or thinner than in others. This made the results vary from one tiny device to another (some worked great, others were just okay). It's like baking a batch of cookies where some are perfectly chocolatey and others have a few less chips.

The Bottom Line

The researchers successfully created a stable, air-proof layer of ultra-thin Bismuth. They proved that even in this tiny, two-dimensional form, Bismuth is a powerhouse for converting electricity into magnetic spin. This is a "proof of concept"—a demonstration that it is possible to measure and use these effects in atomically thin materials, opening the door for future studies on how to make better, energy-efficient electronics.

Technical Summary: Charge to Spin Conversion in Atomically Thin Bismuth

Problem Statement
Bismuth (Bi) is a material of significant interest in spintronics due to its strong spin-orbit coupling (SOC), which theoretically enables efficient charge-to-spin conversion for spin-orbit torque (SOT) devices. However, experimental reports on the spin Hall angle in bulk Bi show large discrepancies, potentially due to the ordinary Nernst effect and the material's highly anisotropic Fermi surface. Furthermore, while 2D-Bi is predicted to host topological phenomena like the quantum spin Hall insulator, synthesizing air-stable, ultra-thin Bi films on insulating substrates without oxidation remains a major challenge, hindering the exploration of spin transport in the true 2D limit.

Methodology
The authors synthesized wafer-scale, air-stable, crystalline 2D-Bi films confined between a silicon carbide (SiC) substrate and epitaxial graphene (EG) using confinement heteroepitaxy (CHet). In this process, Bi atoms sublime from a precursor and diffuse through a defective EG interface to form a layer sandwiched between EG and SiC.

The structural and chemical properties were characterized using:

X-ray Photoelectron Spectroscopy (XPS): To analyze core-level shifts (C 1s, Bi 4f, Si 2p) and confirm intercalation.
Low Energy Electron Diffraction (LEED): To assess surface reconstruction and the elimination of the graphene buffer layer.
Scanning Transmission Electron Microscopy (STEM) and EDX: To visualize the cross-sectional nanostructure and confirm the presence of a bilayer Bi structure.
Raman Spectroscopy: To map surface coverage using ultra-low frequency (ULF) peaks specific to intercalated 2D-Bi.

Electrical transport properties were measured via Hall bar devices to determine carrier density, mobility, and magnetoresistance. Finally, charge-to-spin conversion efficiency was probed using Spin Torque Ferromagnetic Resonance (ST-FMR) on Permalloy (Py)/EG/2D-Bi heterostructures at room temperature.

Key Results

Structural Characterization: XPS and LEED data confirmed the elimination of the graphene buffer layer and the formation of a metallic 2D-Bi layer. STEM imaging revealed a locally ordered bilayer of Bi atoms. Raman mapping indicated a high intercalation coverage of 96.5% across a 10 × 10 µm area. The samples remained stable under ambient conditions.
Electrical Transport: The heterostructure exhibited insulating behavior with resistance increasing as temperature decreased. At 2 K, the system displayed positive magnetoresistance characteristic of weak antilocalization (WAL), indicating strong SOC. This contrasts with CHet-grown 2D-Pb/EG samples (which show weak localization) and standard EG/SiC templates (which show weak localization), suggesting Bi significantly contributes to the electronic transport.
Charge-to-Spin Conversion: ST-FMR measurements on Py/EG/2D-Bi devices revealed a spin torque ratio (the ratio of in-plane damping-like torque to out-of-plane field-like torque) that was 3.75 times higher than in hydrogenated graphene control samples.
- The average torque ratio for six devices was $2.01 \pm 0.66$ .
- Two outlier devices showed a ratio of $3.64 \pm 0.39$ .
- The control hydrogenated graphene samples had a ratio of $0.97 \pm 0.19$ .
Spin Polarization: Angle-dependent ST-FMR measurements confirmed that the spin current polarization is in-plane and perpendicular to the charge current, consistent with the expected symmetry for heavy metals exhibiting the spin Hall or Rashba-Edelstein effects.

Significance and Claims
The paper claims to demonstrate the synthesis of air-stable, wafer-scale 2D-Bi films and provides the first evidence of charge-to-spin conversion in this 2D limit. The authors state that the observed enhancement in the spin torque ratio compared to control samples is a clear indication that the 2D-Bi layer contributes to the conversion process.

The work serves as a proof-of-principle that spin transport can be measured in atomically thin Bi. The authors modestly note that significant device-to-device variation exists, likely due to macroscopic inhomogeneity in Bi intercalation. They conclude that this demonstration motivates further studies to explore the roles of SOC, dimensionality, and topology in 2D-Bi systems, rather than claiming immediate application in commercial memory devices.