🔬 mesoscale physics

Electric field controlled spin transport in a topological insulator interfaced with a ferroelectric antiferromagnet

This study demonstrates electric-field-controlled spin-charge conversion in a Bi $_2$ Te $_3$ /BiFeO $_3$ heterostructure, revealing that topological surface-state-dominated spin transport remains robust above a 10 nm thickness but vanishes at 5 nm due to hybridization-induced trivial insulating phases.

Original authors: Yogesh Kumar, Pushpendra Gupta, Xinyan Li, Richa Mudgal, Ashish Omar, Ryan Chen, Mito Funatsu, Maya Ramesh, Nicholas Reiterer, Yuanqi Lyu, Yiping Zeng, Darrell G. Schlom, Alessandra Lanzara, Robert J.

Published 2026-07-16

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Yogesh Kumar, Pushpendra Gupta, Xinyan Li, Richa Mudgal, Ashish Omar, Ryan Chen, Mito Funatsu, Maya Ramesh, Nicholas Reiterer, Yuanqi Lyu, Yiping Zeng, Darrell G. Schlom, Alessandra Lanzara, Robert J. Birgeneau, James G. Analytis, Ramamoorthy Ramesh, Sajid Husain

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 a world where computers don't just process information with electricity, but with the "spin" of tiny particles called electrons. Think of electron spin not as a physical spinning top, but as a tiny, invisible compass needle attached to every electron. In the next generation of electronics, known as spintronics, scientists want to use these compass needles to carry data. It's like switching from sending messages by shouting (electricity) to sending them by waving colored flags (spin). The problem is, spinning these flags usually requires strong magnets or heavy metals, which are bulky and eat up a lot of power.

To solve this, scientists are looking at special materials called Topological Insulators. You can think of these as "magic highways" for electrons. Inside the material, electrons are stuck and can't move, but on the very surface, they zip around freely. The coolest part? On this surface, the direction an electron moves is locked to the direction its compass needle points. If it moves forward, its needle points left; if it moves backward, it points right. This "spin-momentum locking" makes them incredibly efficient at turning spin into electricity and vice versa. However, most experiments so far have needed big, external magnets to control this traffic, which isn't practical for the tiny chips inside your phone. The big question is: Can we control this magical traffic using just an electric field, like flipping a switch, without needing a giant magnet nearby?

This paper takes a bold step toward answering that question by building a new kind of "spin highway." The researchers created a sandwich-like structure using two special materials. The bottom layer is a magnetic material called BiFeO3, which is unique because it's an insulator (it doesn't let electricity pass through) but still has magnetic properties that can be flipped around using an electric field. It's like a silent, invisible switch that can change the direction of the magnetic compass needles without any wires touching it. On top of this, they placed the "magic highway" material, Bi2Te3, a topological insulator.

The team built tiny devices to test how well electrons could travel across this interface. They injected a current into one end and measured the voltage at the other, looking for signs that the spin was being converted into an electrical signal. The results were exciting: they found that they could indeed control the spin traffic using an electric field. First, they used an electric pulse to flip the internal magnetic state of the bottom layer. Then, they turned the electric field off and measured the signal while the bottom layer stayed in its new "remanent" state. When they had flipped the switch earlier, the voltage signal on top flipped its direction too. It was as if they had a remote control for the electron compasses that worked even after the remote was put away.

However, the thickness of the top "highway" mattered a lot. The researchers tested layers ranging from 70 nanometers down to just 5 nanometers thick. They found that as long as the layer was thicker than about 10 nanometers, the spin traffic flowed smoothly and the signal was strong. But once they tried to make the layer thinner than 10 nanometers, the signal started to fade, and at 5 nanometers, it vanished completely. The paper suggests this happens because when the layer gets too thin, the top and bottom surfaces of the material start to "talk" to each other and mix, which ruins the special magic that makes the surface so efficient.

The study also ruled out some other possibilities. They checked to make sure the signal wasn't just caused by heat or by electricity leaking through the bottom layer, and they confirmed it wasn't. The signal was purely due to the spin properties of the top material interacting with the magnetic switch below. Furthermore, they showed that this new setup works just as well as, or even better than, the heavy metals currently used in labs, but without needing any external magnets.

In short, this paper demonstrates a working prototype for an electrically controlled spin device. It proves that you can use a magnetic insulator to flip the spin of electrons on a topological insulator using only an electric field to set the state, and then measure the effect without any field applied. While the effect disappears if the material gets too thin, the findings suggest a clear path toward building faster, smaller, and much more energy-efficient electronic devices that don't rely on bulky magnets. It's a significant step toward turning the "magic highways" of quantum physics into real-world technology.

Technical Summary: Electric Field Controlled Spin Transport in a Topological Insulator Interfaced with a Ferroelectric Antiferromagnet

Problem Statement
While topological insulators (TIs) are promising candidates for efficient spin-to-charge conversion (SCC) due to their spin-momentum-locked Dirac surface states, previous investigations have relied heavily on metallic ferromagnets as spin sources. These metallic interfaces introduce parasitic charge current shunting, interfacial spin-dependent scattering, and screening effects that obscure the intrinsic response of the TI surface states. Furthermore, the reliance on external magnetic fields as a control knob limits the scalability and integration of such devices into miniaturized, energy-efficient spintronic elements. There is a need to probe spin transport in TIs using a clean, electrically insulating magnetic interface that allows for electric field control without external magnetic fields.

Methodology
The authors fabricated high-quality heterostructures consisting of sputter-deposited Bi $_2$ Te $_3$ (a topological insulator) interfaced with epitaxial BiFeO $_3$ (an antiferromagnetic multiferroic insulator) grown on DyScO $_3$ substrates. The Bi $_2$ Te $_3$ films were grown at a low substrate temperature of 150°C, while BiFeO $_3$ was grown via pulsed laser deposition (PLD) or molecular beam epitaxy (MBE).

The study employed a non-local spin transport geometry to isolate the spin-charge conversion signal from bulk charge currents. Key experimental steps included:

Structural Characterization: X-ray diffraction (XRD), reciprocal space mapping (RSM), and high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) were used to confirm the epitaxial growth, crystallinity, and abruptness of the Bi $_2$ Te $_3$ /BiFeO $_3$ interface.
Transport Measurements: Longitudinal resistivity and magnetoconductance were measured across a range of Bi $_2$ Te $_3$ thicknesses (5 nm to 70 nm) to characterize charge transport and quantum interference effects (specifically weak antilocalization).
Non-Local Spin Transport: Devices were patterned into Hall bars where a DC current ( $I_{inj}$ ) was injected to generate a spin current via spin-orbit coupling. The resulting spin accumulation was detected as a voltage ( $V_{SCC}$ ) at a remote detector.
Electric Field Control: The ferroelectric polarization of the underlying BiFeO $_3$ was switched using electric field pulses. The $V_{SCC}$ was measured in the remanent state (zero external field) to correlate spin transport with the antiferromagnetic state of the BiFeO $_3$ .

Key Results

Interface Quality: The Bi $_2$ Te $_3$ films exhibited graphoepitaxial growth with 12-fold rotational symmetry and a chemically abrupt interface with minimal interdiffusion.
Thickness-Dependent Transport: Resistivity measurements showed a crossover from metallic to semiconducting behavior as film thickness decreased. Magnetoconductance analysis using the Hikami–Larkin–Nagaoka (HLN) model revealed a coherency factor ( $\alpha$ ) consistent with weak antilocalization (WAL) in thicker films (70 nm, $\alpha \approx -0.27$ ), indicating dominant 2D surface transport. As thickness decreased to 5 nm, the WAL cusp vanished ( $\alpha \approx 0$ ), suggesting a transition to a trivial insulating phase due to surface-state hybridization and disorder.
Electric Field Control: The spin-charge conversion voltage ( $V_{SCC}$ ) exhibited hysteresis that exactly followed the ferroelectric polarization of the BiFeO $_3$ . This demonstrates that spin transport is controlled by the electric field via the antiferromagnetic order of the BiFeO $_3$ .
Sign Reversal: Reversing the direction of the injected current ( $I_{inj}$ ) reversed the polarity of $V_{SCC}$ . This is attributed to the current-induced shift of the Fermi surface in the helical Dirac states, which flips the induced spin polarization (Edelstein effect).
Thickness Dependence of SCC: The magnitude of $V_{SCC}$ remained robust for Bi $_2$ Te $_3$ thicknesses down to ~20 nm but decreased rapidly below 15 nm and vanished at 5 nm. This trend correlates directly with the coherency factor derived from WAL, confirming that efficient SCC is dominated by topological surface states.
Efficiency: The measured SCC resistance in Bi $_2$ Te $_3$ was orders of magnitude larger than that of conventional heavy metals like Pt and comparable to or exceeding state-of-the-art oxide spin-orbit metals like SrIrO $_3$ .

Significance and Claims
The paper claims to provide the first direct evidence of electric field-controlled spin-charge conversion in a topological insulator mediated by an antiferromagnetic magnon channel, without the need for external magnetic fields or metallic spin sources. By utilizing an entirely insulating magnetic interface (BiFeO $_3$ ), the study eliminates parasitic shunting and screening, allowing for a clearer observation of the intrinsic topological surface state contributions.

The authors assert that their work establishes the Bi $_2$ Te $_3$ /BiFeO $_3$ system as a clean platform for probing interfacial spin transport. The results demonstrate that spin-momentum-locked surface states dominate interfacial spin transport in the decoupled regime, while hybridization in ultrathin films suppresses this effect. Furthermore, the study highlights sputter-deposited Bi $_2$ Te $_3$ as a high-quality, scalable platform for integrating quantum materials into devices. The non-local spin transport approach presented offers a simple, unprecedented method for directly evidencing spin-charge conversion, paving a practical pathway toward designing energy-efficient, electrically tunable spin-based devices.

More like this