Abstract

Although heavy metal materials, due to their large spin-orbit interaction, would be expected to act as an efficient converter of spin current to charge current, Bismuth does not obey this rule. Heterostructures of yttrium iron garnet (YIG)/Bi are investigated by means of the spin pumping technique and the results show that Bi exhibits negligible spin-charge conversion. In addition to not converting spin to charge, Bi exhibits an exotic spin relaxation mechanism, which is driven by its large diamagnetic response. By considering a nonlinear stochastic model of diffusing spin particles, the results show that the Bi layer acts as a binary classifier device that categorizes the pumped spins into relaxed spins and those that flow upwards and eventually reach the top layer of Pt. This shows that the investigation reported here opens different perspectives involving fundamental aspects in the fields of machine learning and condensed matter physics.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call