Temperature-dependent plasticity in organic synaptic transistors for adaptive learning and data encryption
Neuromorphic computing hardware offers a promising alternative to traditional von Neumann architecture. However, the influences of environmental factors, such as temperature, on synaptic behaviors in neuromorphic devices have been underexplored. This work presents an organic synaptic transistor that...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | Materials & Design |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0264127525005726 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Neuromorphic computing hardware offers a promising alternative to traditional von Neumann architecture. However, the influences of environmental factors, such as temperature, on synaptic behaviors in neuromorphic devices have been underexplored. This work presents an organic synaptic transistor that exhibits temperature-dependent plasticity based on the ion migration mechanism, enabling both neuromorphic learning and secure data transmission. The device features a bottom-gate top-contact configuration with a hybrid PMMA/TaOx dielectric and a pentacene semiconducting channel. The statistical analyses reveal that the channel current increases by over 175 % and the memory window widens by a factor of ∼ 2.5 as ambient temperature rises from 25 °C to 65 °C. The temperature-enhanced synaptic behaviors are demonstrated, including excitatory postsynaptic current (EPSC), post-tetanic potentiation (PTP), and paired-pulse facilitation (PPF) with its time constant increasing from 68 ms to 245 ms, indicating improved memory retention at elevated temperatures. By leveraging these thermal responses, we establish a temperature-dependent physical unclonable function (PUF) capable of multi-level logic encoding and self-annihilating data storage. The temperature-sensitive organic neuromorphic devices are endowed with adaptive learning and environment-dependent data encryption, paving the potential way for intelligent systems operable under diverse environmental conditions. |
|---|---|
| ISSN: | 0264-1275 |