Harnessing Stochasticity in 2D Materials—A First-Principles Study of Phonon-Assisted Defect Charge Dynamics in WS₂

Stochasticity plays an important role in cryptography, stochastic and neural-like computing, facilitating low-cost, error-tolerant solutions to complex problems such as integer factorization, invertible logic and optimization. Specifically, we investigate phonon-assisted defect charge dynamics in 2D materials, e.g. a sulfur vacancy in WS₂ monolayer (V_S-WS₂), using first principles calculations [1]. We find that a true random bit generation scheme based on V_S-WS₂ optimally balances bitrate, energy efficiency, and tunability of bit probability. The entropy source based on V_S-WS₂ enables unbiased random bit generation at bitrates of 30 Mbps, which can be further optimized up to terabits per second by minimizing the energy barrier (E_b) for defect-to-band electron transition through 3% tensile strain of the host-material. Further advantages of V_S-WS₂ include a large current contrast between two states, significant tunability of bit probability, and reduced energy consumption. Furthermore, we explore the conditions needed to achieve a low E_b, unveiling new possibilities in 2D material-based entropy sources. The essential insights into phonon-assisted defect charge dynamics combines with practical designs for high-performance, energy-efficient stochastic devices.

Supported by the Office of Naval Research (N00014-22-1-2753)

[1] Chenmu Zhang, Evgeni S. Penev, Boris I. Yakobson. Exploring the limits of random bits generation in two-dimensional semiconductors from first principles. submitted (2024).