Abstract
Abstract Poly(benzodifurandione) (PBFDO) is a promising n-type mixed conductor for organic electrochemical transistors (OECTs), but its high intrinsic conductivity results in excessive operating currents and energy consumption for neuromorphic computing. Here, we combine molecular-weight engineering and ionic doping to overcome this limitation. Benzofuranone end-capping produces a reduced chain length polymer (PBFDO-BF) with substantially lower intrinsic conductivity, while LiTFSI doping enhances ion-mediated conductance modulation and synaptic functionality. PBFDO-BF + LiTFSI provides enhanced OECT modulation and spike-dependent plasticity while reducing operating currents by approximately 1 order of magnitude compared with pristine PBFDO. In the Modified National Institute of Standards and Technology (MNIST)-based convolutional neural network simulations, the device achieves 97.8% training and 98.6% inference accuracy, with the lowest cumulative energy consumption to reach ≈90% accuracy. These results establish molecular-weight control combined with ionic doping as an effective strategy for developing energy-efficient PBFDO-based neuromorphic OECTs without compromising stability or solution processability.