Polymer-based probabilistic bits for thermodynamic computing

Abstract

Probabilistic bits (p-bits) are stochastic hardware elements whose output probability can be tuned by an input bias, offering a route to energy-efficient architectures that exploit, rather than suppress, fluctuations. Here we report p-bit generation in an organic memristive device, establishing polymers as the first class of soft-matter systems to realize probabilistic hardware. The active element is a dithieno[3,2-b:2',3'-d]pyrrole (DTP)-backbone polymer with pendant triphenylamine (TPA) groups, whose stochastic resistance fluctuations are converted into binary outputs by a simple voltage-divider/comparator circuit. The resulting probability distributions follow logistic transfer functions, characteristic of stochastic binary neurons. Separately, ensembles of pulsed IV measurements were analyzed to construct binned current distributions, from which the discrete Shannon entropy was calculated. Peaks in this entropy coincide with bias conditions that maximize variability in the memristor voltage drop, directly linking device-level stochasticity to intrinsic material properties. Dielectric analysis shows that pendant TPA units provide dynamically active relaxation modes, while energy-resolved electrochemical impedance spectroscopy and density functional theory calculations indicate that the frontier orbitals of DTP, TPA and ITO align within the transport gap to produce a bifurcated percolation network. The correspondence between microscopic relaxation dynamics, electronic energetics and macroscopic probabilistic response highlights how organic semiconductors can serve as chemically tunable entropy sources, opening a polymer-based pathway toward thermodynamic computing.