Resolving hidden interface-trap capture times through stochastic current transients in multitrap nanoscale transistors
Year: 2026
Type: article
Abstract: Charge trapping at semiconductor–dielectric interfaces governs threshold-voltage instability, low-frequency noise, and stochastic variability in nanoscale transistors. In the intermediate multitrap regime, the drain-current transient is neither a defect-resolved signal from a single trap nor a fully averaged ensemble response, making the underlying trap kinetics difficult to infer directly. Here, a framework combining pulsed current–voltage measurement and self-consistent stochastic modeling is introduced for extracting statistical capture-time descriptors of interface-trap ensembles from sensing-phase drain-current transients. The approach analyzes repeated subthreshold transients using both the mean current and its cycle-to-cycle fluctuation, while explicitly incorporating trap-to-trap dispersion, sensing-bias-dependent capture kinetics, and electrostatic feedback induced by progressive trap filling. Simultaneous fitting across multiple sensing biases identifies a common model-constrained parameter pair within an assumed log-normal parameterization of the capture-time-constant distribution. Within the assumed model, this analysis distinguishes intrinsic trap-to-trap kinetic disorder from measurement-induced electrostatic renormalization, revealing that the measured transient is not a direct image of the intrinsic time-constant spectrum but a dynamically reshaped response governed by finite-ensemble stochasticity and channel electrostatics. The framework provides a physics-based route for extracting hidden multitrap kinetic descriptors from stochastic electronic transients in nanoscale devices.
Source: Scientific Reports
Institutions: Ewha Womans University, Seoul National University
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
Awards: G1818599991, G3897326803, G809632853
OpenAlex ID: W7225383725
