Space-Charge Effects in Silicon Reconfigurable Nonlinear-Processing Units

TL;DR

This study uncovers that space charge effects govern the nonlinearity in silicon-based reconfigurable nonlinear-processing units, providing a physical understanding crucial for scalable nonlinear computing hardware.

Contribution

It offers a comprehensive physical framework explaining the origin of nonlinearity in silicon RNPUs through charge transport mechanisms, validated by experiments and simulations.

Findings

Transport is governed by space charge effects.

Nonlinearity arises from competition between injected carriers and dopants.

Device behavior exceeds classical quadratic dependence due to doping control.

Abstract

Reconfigurable nonlinear-processing units (RNPUs) are multi-terminal electronic devices that act as computational primitives, exploiting intrinsic nonlinear charge transport combined with electrostatic tunability. Silicon-based realizations provide a scalable and technologically relevant platform, yet the physical origin of their room-temperature nonlinearity has remained insufficiently understood. Here, we investigate charge transport using temperature- and length-dependent current-voltage measurements on physical devices, complemented by drift-diffusion simulations, and show that transport is governed by space charge. Interface trap states strongly suppress the equilibrium carrier density, while the functional nonlinearity arises from the voltage-dependent competition between injected mobile carriers and fixed ionized background dopants. The resulting non-equilibrium transport exhibits a transition from an Ohmic regime to a strongly nonlinear regime, and ultimately to a velocity-saturation space-charge-limited current regime, as evidenced by the observed voltage and length scaling. We further show that background doping of opposite polarity to the injected carriers controls the onset and strength of the nonlinearity, leading to behavior exceeding the quadratic dependence of the classical Mott-Gurney law. Agreement between experiment and simulation supports that the spatial distribution of injected carriers and fixed charge governs the internal electric-field profile and device response. These results establish a physical framework for silicon-based RNPUs without requiring disorder or hopping transport, and provide design guidelines for reproducible, scalable, and CMOS-compatible implementations of nonlinear computing hardware.