Hybrid Electrolyte‐Doped OFETs Enabling Synaptic Plasticity and Physical Reservoir Computing

ABSTRACT

Emerging applications such as brain‐machine interfaces, bio‐compatible prosthetics, and adaptive soft robotics rely on artificial neuromorphic devices that can interface directly with biological systems. To address this need, we demonstrate organic field‐effect transistors (OFETs) doped with a hybrid electrolyte, 1,4‐di‐ tert ‐butylbenzene‐2,5‐bis(1‐propoxy‐3‐sulfonate) lithium salt ( BBOPSO ₃ Li ), as novel artificial synaptic devices. Incorporation of BBOPSO ₃ Li into poly(3‐hexylthiophene) (P3HT) channels enables precise modulation of shallow trap states through the close energetic alignment of BBOPSO ₃ Li with the P3HT HOMO (∼0.1 eV offset), resulting in tunable threshold voltage, controllable hysteresis, and enhanced carrier mobility. Doped devices exhibit outstanding synaptic functions, including excitatory postsynaptic current (EPSC), paired‐pulse facilitation (PPF), and long‐term potentiation–depression (PD) with analog weight updates. An optimal doping level of 0.070 mol% provides the highest excitability, balanced temporal memory characteristics, and the widest conductance window. System‐level validation demonstrated improved classification accuracy (up to 89.8% on MNIST‐like datasets) and robust time‐series prediction in a physical reservoir computing framework, achieving a normalized mean square error of −32 dB. These findings demonstrate that hybrid electrolytes capable of self‐doping can be used to introduce well‐defined, reversible trap states, allowing control of synaptic plasticity and temporal dynamics in OFETs, advancing their potential as efficient neuromorphic computing platforms.