Spin transport in InSb nanowires offers a way to control electron spin using local magnetic contacts rather than relying solely on a magnetic field applied across an entire device. In experiments using iron contacts, researchers observed spin-dependent transport in quasi-one-dimensional InSb channels and found that electrostatic gating and source-drain bias could tune the measured magnetic signals. The results establish a platform for studying spin-orbit physics in nanowires, while pointing to a possible route toward quantum devices in which spin degeneracy is lifted locally.
Why integrate ferromagnets with InSb nanowires?
InSb nanowires combine strong spin-orbit interaction with properties useful for quantum transport. In Majorana-device concepts, an external magnetic field is used to lift spin degeneracy and create helical states, where an electron’s spin is linked to its direction of motion. When superconductivity is induced in that regime, the system may support Majorana zero modes. But a global field can also affect superconductivity and impose constraints on device geometry and future scaling.
Ferromagnetic contacts offer another way to interact with electron spins. They can inject spin-polarized carriers into a semiconductor and provide a means of detecting spin-dependent transport. If magnetic interactions can be localized near the nanowire, they may help researchers investigate device architectures that do not depend on applying a global magnetic field. The reported experiments examine this possibility; they do not demonstrate Majorana zero modes or a working zero-field Majorana device.
Ballistic transport provides a foundation for spin measurements
The devices used two iron contacts separated by an InSb nanowire, with contact spacings ranging from 200 nanometers to 1.2 micrometers. Measurements across six devices showed transport features associated with a clean, confined one-dimensional channel. In shorter devices, conductance plateaus appeared as the back-gate voltage changed. These plateaus, approximately quantized in multiples of e2/h, are signatures of transport through a small number of one-dimensional modes.
At lower source-drain bias, conductance oscillations formed diamond-like patterns as both bias and gate voltage varied. The pattern is consistent with Fabry-Pérot interference: electron waves partially reflect at the contact interfaces, travel through the channel, and interfere. Gate voltage changes the electron wave vector and can shift the device between constructive and destructive interference. This phase-coherent, ballistic transport is important because spin signals can be affected by how electrons propagate and reflect within the channel.
Gate and bias tune the spin-valve response
When the magnetization directions of the two iron contacts switch, the measured conductance changes hysteretically. The researchers interpreted this spin-valve-like response as evidence that the contacts inject, transport, and detect spin-polarized current through the nanowire. The size of the signal depended on both gate voltage and source-drain bias. In particular, the relative magnetoconductance was generally larger at lower bias and lower transmission, and it tended to decrease as gate voltage increased the conductance.
One possible explanation is an energy barrier at the ferromagnet–nanowire interface. Such a barrier can influence how effectively spin polarization transfers from a metal contact into a semiconductor. Adjusting gate voltage or bias changes the channel and interface transmission, and therefore can change the observed spin signal. Gate voltage may also affect the strength of Rashba spin-orbit interaction and spin relaxation.
Conductance resonances associated with Fabry-Pérot interference also tracked changes in magnetoconductance. This relationship suggests that quantum interference and the relative coupling between the two contacts can influence the magnitude—and sometimes the sign—of the spin-valve response. In most measurements, the conductance was higher for antiparallel contact magnetizations than for parallel magnetizations, an inverse response compared with the common spin-valve pattern. The sign could change with a small gate adjustment, underscoring the role of transmission and interference.
Distinguishing spin transport from other magnetic effects
Hysteresis alone does not establish spin injection. Switching ferromagnetic contacts can also affect conductance through stray magnetic fields or the magneto-Coulomb effect, in which magnetic changes in a contact indirectly shift the channel’s electrostatic potential. The study compared the gate dependence of the signal with the expected behavior of these alternatives. The inferred stray-field contribution varied strongly with gate voltage, unlike a field change tied to contact magnetization switching. Estimates of the magneto-Coulomb contribution were also much smaller than the measured magnetoconductance response.
Contact-spacing measurements provided another check. Across devices with spacings from 200 to 1,000 nanometers, the reciprocal magnetoconductance followed the approximately linear length dependence expected for the study’s ballistic spin-transport model. From that fit, the researchers estimated a spin-relaxation length of 1.2 to 1.5 micrometers, assuming interface transmission between 0.8 and 1. The longest device, at 1.2 micrometers, departed from the trend; the authors noted that device-specific effects or a breakdown of the ballistic picture could contribute.
A distinct spin-filtering regime
In the device with the shortest contact spacing, 200 nanometers, tuning the gate voltage into a conductance dip produced a different hysteresis pattern. Rather than simply distinguishing parallel from antiparallel contact states, the response favored one magnetization direction over the other. This is consistent with a spin-filtering regime, in which transmission depends on the spin state of the carriers.
The filtering was asymmetric with respect to bias polarity and was most apparent in the few-subband regime. Strong spin-orbit coupling and interference at the interfaces may contribute, but the precise mechanism was not resolved. The result should therefore be viewed as an observed transport effect with an open explanation, rather than as a fully characterized spin-filter design.
Implications for quantum-device engineering
For engineers and researchers developing semiconductor quantum devices, the work highlights the importance of treating the ferromagnet–nanowire interface, gate control, and channel length as connected design variables. Interface barriers can affect spin injection, while phase-coherent reflections can modify both conductance and magnetoresistance. Device-to-device variations also matter: the measured hysteresis shapes differed, likely reflecting mesoscopic effects.
Integrating ferromagnets with InSb nanowires creates a platform for exploring local spin control and spin-orbit-based functions such as filters or modulators. It may also inform approaches to Majorana-device architectures that seek to reduce reliance on a global magnetic field. Achieving those outcomes will require further work, including improved interfaces and investigation of how magnetic and superconducting properties interact.
In practical terms, the key finding is that spin-dependent transport in InSb nanowires can be electrically tuned and remains measurable in the ballistic regime. The spin-filtering observation is promising but not yet fully explained. Together, the results offer useful design guidance for future experiments and a foundation for evaluating locally controlled spin in nanoscale quantum devices.