Superconducting circuits carry current without electrical resistance. However, their usefulness also depends on controlling the quantum phase, which describes the collective rhythm of paired electrons. Magnetism can alter this phase, offering a potential solution for building superconducting memory and quantum circuits with fewer external magnetic controls. However, combining magnetism with superconductivity is challenging, because the magnetic alignment of electron spins can disrupt the Cooper pairs that carry a supercurrent. Semiconductors could help researchers study and control this interaction because their electrical properties can be adjusted with a gate voltage. Building such devices, however, requires exceptionally clean interfaces between materials with very different properties.
A research team from the University of Tokyo and New York University has developed a hybrid semiconductor device that brings these capabilities together. Led by Hirotaka Hara and Lukas Baker, with supervision by Masaaki Tanaka, Javad Shabani and Le Duc Anh, the team grew a layer of superconducting aluminum above a semiconductor (indium arsenide) channel layer and a magnetic semiconductor (iron-doped gallium antimonide) layer, below it. The layers were grown in an ultrahigh vacuum without exposing to air, to preserve clean interfaces. Electron microscopy confirmed an atomically sharp and clean interface between the aluminum and the semiconductor.
The researchers created a device structure consisting of two aluminum electrodes separated by a 100-nanometers-wide gap, leaving the semiconductor as a bridge between them. This structure, called a Josephson junction, allows superconducting correlations from the electrodes to extend into the semiconductor, supporting a resistance-free current. At 30 millikelvin, only 0.03 degrees above absolute zero, an ungated device carried approximately three microamperes without resistance. Applying a negative gate voltage decreased the number of electrons in the channel, thereby reducing the supercurrent and demonstrating electrical control of superconducting transport.
The devices also exhibited an unusual response to an applied magnetic field. The maximum resistance-free current increased and decreased in an interference pattern depending on the direction in which the magnetic field was swept. This magnetic-history dependence is consistent with a magnetic contribution to the supercurrent, possibly associated with magnetism induced in the indium arsenide channel by the neighboring magnetic semiconductor. In addition, the current limit differed between opposite current directions, producing a superconducting diode effect with an efficiency up to 20% under an applied magnetic field. At a suitable magnitude of current, the junction can therefore remain superconducting in one direction while becoming resistive in the other.
These results provide an electrically tunable platform for studying how superconductivity and magnetism interact in semiconductors. Together, the clean interfaces, gate-controlled supercurrent and signatures of magnetic influence offer a path toward semiconductor-based superconducting devices for quantum electronics.
Figure 1: Combining superconductivity and magnetism in a gate-controlled semiconductor junction. Aluminum (Al) electrodes induce superconducting correlations in an indium arsenide (InAs) channel, while the underlying iron-doped gallium antimonide, (Ga,Fe)Sb, provides a neighboring magnetic layer. The gate voltage adjusts the channel properties. The schematic illustrates the proposed magnetic proximity effect (MPE), through which the magnetic layer influences electrons in the channel. The device exhibits a gate-tunable supercurrent, magnetic-history-dependent interference patterns, and a superconducting diode effect under an applied magnetic field.
Papers
Journal: Communications Materials
Title: Interplay of superconductivity and ferromagnetism in ferromagnetic-semiconductor-based Josephson junctions
Authors: Hirotaka Hara, Lukas Baker, Axel Leblanc, Shingen Miura, Keita Ishihara, Melissa Mikalsen, Patrick J. Strohbeen, Jacob Issokson, Masaaki Tanaka, Javad Shabani and Le Duc Anh
Faster Quantum Measurement Algorithm Enhanced by Symmetry: Toward the practical realization of fault-tolerant quantum simulation
Learning the Nonlinear Dynamics of Soft Actuators with AI: Hysteresis-based Reservoir Computing Paving the Way for Robotic Grippers Capable of Delicate Movements
