Mechanism for reconfigurable logic in disordered dopant networks

Abstract

We present an atomic-scale mechanism based on variable-range hopping of interacting charges enabling reconfigurable logic and nonlinear classification tasks in dopant network processing units in silicon. Kinetic Monte Carlo simulations of the hopping process show temperature-dependent current-voltage characteristics, artificial evolution of basic Boolean logic gates, and fitness-dependent gate abundances in striking agreement with experiment. The simulations provide unique insights in the local electrostatic potential and current flow in the dopant network, showing subtle changes induced by control voltages that set the conditions for the logic operation. These insights will be crucial in the systematic further development of this burgeoning technology for unconventional computing. The establishment of the principles underlying the logic functionality of these devices encourages the exploration and utilization of the same principles in other materials and device geometries.

Dopant network processing units as tuneable extreme learning machines

“…On the one hand, they have a smaller footprint and are silicon-based, which may facilitate scaling and integration with conventional electronics. On the other hand, DNPUs do not exhibit time dynamics at the timescales of our measurements, which holds if the measurements are performed below 1 MHz (Tertilt et al, 2022). This makes these systems, in their present form, unsuitable to process data directly in the time domain, but more suitable for the ELM approach.…”

“…It consists of a network of donor or acceptor dopant atoms in a semiconductor host material. Transport is dominated by hopping and Coulomb interactions, giving rise to strongly non-linear IV characteristics, including negative differential resistance (NDR) (Tertilt et al, 2022). Charge carriers hop from one dopant atom to another under the influence of input and control voltages applied at surrounding electrodes.…”