A research team led by Prof. Joon Hak Oh at Seoul National University has now developed a strategy to address this fundamental limitation by introducing a reversible redox reaction into the OECT architecture. The researchers incorporated a redox-active azobenzene layer into a Li-ion gel electrolyte transistor, creating an azobenzene-introduced Li-ion gel electrolyte transistor, termed AB-LGET.
F ull Text:
Organic electrochemical transistors gain long-term synaptic memory through a reversible azobenzene redox layer that suppresses spontaneous de-doping.
The rapid expansion of artificial intelligence and data-intensive computing has increased the demand for hardware that can process and store information more efficiently than conventional computing architectures. Neuromorphic electronics, which seek to emulate the parallel information processing and adaptive memory functions of biological neural systems, have emerged as a promising approach toward energy-efficient computing.
Among the various device platforms being explored for neuromorphic computing, organic electrochemical transistors (OECTs) offer several attractive features. Their operation is based on coupled ionic and electronic transport, allowing them to modulate channel conductivity at relatively low voltages while exhibiting behaviors analogous to biological synapses. However, a major challenge remains. Once the electrical stimulus is removed, ions introduced into the organic semiconductor channel can spontaneously diffuse back into the electrolyte, causing rapid de-doping and loss of the stored conductance state. This thermodynamically driven relaxation makes it difficult for conventional OECTs to maintain robust long-term synaptic plasticity.
In the device, application of a negative presynaptic voltage drives TFSI⁻ anions from the Li-ion gel electrolyte into the poly(3-hexylthiophene) (P3HT) semiconductor channel. The resulting electrochemical doping increases the conductivity of P3HT and represents synaptic potentiation. At the same time, Li⁺ ions migrate toward the gate side of the device.
In a conventional Li-ion gel electrolyte transistor, this redistribution of ions creates a charge imbalance within the electrolyte. Once the external voltage is removed, the system tends to recover its electrical neutrality by driving the TFSI⁻ anions back out of the P3HT channel, resulting in spontaneous de-doping and rapid loss of the potentiated state.
The researchers introduced azobenzene to fundamentally change this process. Li⁺ ions arriving at the gate side participate in a reversible lithiation reaction with the azobenzene layer. This Faradaic redox reaction provides localized charge compensation and helps maintain electrical neutrality within the electrolyte. As a result, the thermodynamic driving force for spontaneous TFSI⁻ back-diffusion is reduced, allowing the electrochemically doped state of P3HT to persist for substantially longer periods.
This redox-mediated charge compensation produced pronounced improvements in synaptic behavior. Compared with the conventional LGET, the AB-LGET exhibited enhanced paired-pulse facilitation and substantially improved long-term potentiation. Following repeated stimulation, the enhanced conductance state remained evident for more than 800 s after the electrical stimulus was removed. The device also displayed stable multistate conductance modulation and highly linear potentiation and depression characteristics, which are particularly important for reducing weight-update errors in artificial neural networks.
To investigate the physical origin of the enhanced memory retention, the researchers used grazing-incidence wide-angle X-ray scattering (GIWAXS) to examine structural changes in the P3HT channel before and after electrical stimulation. The analysis indicated that structural signatures associated with molecular doping were more effectively retained within the crystalline regions of P3HT in the AB-LGET. These observations provide structural support for the proposed mechanism in which azobenzene-mediated charge compensation stabilizes the electrochemically doped state of the organic semiconductor.
The team further evaluated whether these improved device characteristics could translate into better neuromorphic computing performance. Experimentally measured potentiation and depression characteristics were incorporated into an artificial neural network simulation for recognition of handwritten digits from the Modified National Institute of Standards and Technology (MNIST) dataset. The AB-LGET-based system achieved a maximum recognition accuracy of 93.02%, demonstrating the potential of the redox-stabilized organic synapse for neuromorphic information processing.
The significance of the approach lies in how it addresses ionic instability. Rather than simply attempting to physically restrict ion transport, the researchers actively regulate charge balance through a reversible interfacial redox reaction. This strategy provides a new route for suppressing spontaneous de-doping while preserving the ion-electron coupling that makes OECTs attractive as artificial synaptic devices.
The study therefore establishes redox-mediated electrical-neutrality control as a promising design principle for organic neuromorphic electronics. By combining stable memory retention with controllable synaptic weight modulation, this approach may contribute to the development of more reliable organic artificial synapses and future neuromorphic computing hardware.
Introduction of the Research Group
This work was carried out in the Organic & Polymer Electronics Lab led by Prof. Joon Hak Oh in the Department of Chemical and Biological Engineering at Seoul National University, Republic of Korea. Prof. Joon Hak Oh served as the corresponding author of this study.
The Organic & Polymer Electronics Lab conducts research on organic and polymer electronic materials and their integration into advanced electronic devices. Building on fundamental studies of molecular and polymeric semiconductors and their structure-property-device relationships, the group has expanded its research beyond conventional organic field-effect transistors (OFETs) toward a broad range of advanced electronic and optoelectronic platforms.
The group actively investigates electronic skin (E-skin), neuromorphic electronics, and chemical, biological, optical, and pressure sensors by integrating functional organic and polymeric materials into advanced device architectures. Its research activities also extend to emerging areas of optoelectronics, including chiral electronics, perovskite electronics, and plasmonics, with a particular focus on developing new functional materials and understanding how their molecular and structural characteristics determine device performance.
In the present study, the group extended its expertise in organic electronics to organic electrochemical transistors and neuromorphic computing. By exploiting the reversible redox chemistry of azobenzene to regulate ion-electron coupled processes and electrical neutrality within the electrolyte, the researchers developed a new artificial-synapse design strategy that suppresses spontaneous de-doping and enhances long-term synaptic plasticity and stable weight modulation in OECT-based neuromorphic devices.
More recently, the Organic & Polymer Electronics Lab has also expanded its research activities toward next-generation batteries and energy-storage materials, building on its expertise in functional materials, interfacial engineering, and electrochemical processes.
DOI Link:
https://doi.org/10.26599/NR.2026.94909008
About Nano Research
Nano Research is a peer-reviewed, open access, international and interdisciplinary research journal, sponsored by Tsinghua University and the Chinese Chemical Society, published by Tsinghua University Press on the platform SciOpen. It publishes original high-quality research and significant review articles on all aspects of nanoscience and nanotechnology, ranging from basic aspects of the science of nanoscale materials to practical applications of such materials. After 18 years of development, it has become one of the most influential academic journals in the nano field. Nano Research has published more than 1,000 papers every year from 2022, with its cumulative count surpassing 8,000 articles. In 2025 InCites Journal Citation Reports, its 2025 IF is 9.4 (8.3, 5 years), and it continues to be the Q1 area among the four subject classifications. Nano Research Award, established by Nano Research together with TUP and Springer Nature in 2013, and Nano Research Young Innovators (NR45) Awards, established by Nano Research in 2018, have become international academic awards with global influence.
Nano Research
Redox-Mediated Charge Compensation Stabilizes Organic Electrochemical Synapses for Neuromorphic Computing
12-Aug-2026