Hierarchical Ionic Programming of Stretchable Organic Electrochemical Transistors for Dual-Mode Sensing and Computing

Organic electrochemical transistors (OECTs) are promising building blocks for flexible in-sensor computing, yet the conflicting ionic dynamic demands of high-fidelity sensing and nonvolatile memory remain challenging to reconcile within a unified hardware platform. Here, we report a hierarchical compositional programming strategy that addresses this trade-off across a standardized stretchable OECT architecture. Regulating 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-oxidized cellulose nanofiber content in poly(3,4-ethylenedioxythiophene) (PEDOT) tunes channel capacitance and mechanical compliance, while tailoring the supramolecular network in a deep eutectic solvent gel electrolyte dictates ion retention kinetics. This strategy establishes distinct operational regimes via tailored formulations: a rapidly reversible, low-hysteresis state for sensing and a persistent ion trapping state for neuromorphic computing. The sensing OECT enables real-time electrocardiogram acquisition with a 35.6 dB signal-to-noise ratio and 98.9% biometric identification accuracy across 15 subjects. The computing state supports temporal Morse code demodulation and Pavlovian associative learning, demonstrating application-adaptive stretchable bioelectronics.