This paper is a contribution to the Physical Review Applied collection titled Physics-Inspired Computing.

One promising approach for combating rapidly escalating computational demands is to use networks of naturally stochastic units called probabilistic bits ($p$-bits). To date, hardware implementations of $p$-bits have predominantly relied on thermally unstable nanomagnets. However, the search continues for alternative material platforms exhibiting easily accessible intrinsic forms of stochasticity. In this work, we demonstrate hitherto unreported $p$-bit functionality in epitaxial thin films of ${\mathrm{La}}_{0.67}{\mathrm{Sr}}_{0.33}{\mathrm{Mn}\mathrm{O}}_3$, a material showing an electrically triggered metal-insulator transition, grown on twin-textured ${\mathrm{La}\mathrm{Al}\mathrm{O}}_3$ substrates. By leveraging the combined effects of phase and structural inhomogeneities, we show two distinct modes of voltage-tunable stochastic operation: clocked binary switching and unclocked multibit switching. Our findings suggest that the differences result from variations in the energy landscape near the phase transition. This tunability of the energy landscape is promising for designing diverse stochastic behavior within the same material, highlighting its potential for applications in true random-number generation for cryptography and probabilistic computing.

Close

PrevNext

View figure in article

• Exchange interaction
• Metal-insulator transition
• Probabilistic computing
• Resistive switching
• Spintronics
• Magnetic thin films
• Manganites
• Strongly correlated systems