Electrostatic Ion Trapping in Organic Electrochemical Transistors for Record Neuromorphic Memory Performance

Area of Science:

• Materials Science
• Organic Electronics
• Neuromorphic Computing

Background:

• Organic electrochemical transistors (OECTs) show promise for memory devices but are limited by poor ion retention.
• Previous methods for ion trapping in OECTs resulted in limited memory windows (< 5.3 V).
• Achieving stable ion retention is crucial for enhancing OECT performance in memory and neuromorphic applications.

Purpose of the Study:

• To design a molecular strategy for enhanced ion retention and hysteresis control in OECTs.
• To develop a dual-functional zwitterionic crosslinker for creating electrostatic ion-trapping channels.
• To investigate the impact of this design on OECT performance for neuromorphic computing.

Main Methods:

• Synthesis and incorporation of a dual-functional zwitterionic crosslinker into polymer semiconductors.
• Characterization of OECTs using techniques like GIWAXS (Grazing Incidence Wide-Angle X-ray Scattering) under electrical biasing.
• Evaluation of device performance, including memory window, hysteresis strength, on/off ratio, and conductance retention.
• Testing of neuromorphic functionalities, including synaptic metrics and performance on MNIST and biosignal processing.

Main Results:

• The zwitterionic crosslinker created a deep electrostatic trap (2.03 eV barrier) and a repulsive barrier, enhancing ion retention.
• Achieved record hysteresis strength (96.4 V) and memory window (8.65 V) in OECTs, with an on/off ratio of ~10^6.
• Demonstrated excellent device stability with 86.4% conductance retention after 200,000 pulses and successful application in neuromorphic models and biosignal processing.

Conclusions:

• The designed zwitterionic crosslinker provides molecular-level control over hysteresis, significantly improving OECT memory capabilities.
• The Z-FPA scheme offers a polymer-agnostic approach for developing high-fidelity neuromorphic OECTs.
• These findings offer critical insights into material design for advancing next-generation neuromorphic OECTs.