Buried Contouring PTCDI-C13 Layer for Interface Engineering in Dual-Function Optical Synaptic and Memory Transistors
- Authors
- Kim, Yeo Eun; Kang, Seungme; Kim, Hyeonjung; Kim, Young-Joon; Oh, Seyong; Yoo, Hocheon
- Issue Date
- Sep-2025
- Publisher
- AMER CHEMICAL SOC
- Keywords
- optical synaptic; memory transistor; PTCDI-C-13; neuromorphic computing; dual-function; surface roughness; contour layer
- Citation
- ACS APPLIED MATERIALS & INTERFACES
- Indexed
- SCIE
SCOPUS
- Journal Title
- ACS APPLIED MATERIALS & INTERFACES
- URI
- https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/126691
- DOI
- 10.1021/acsami.5c14502
- ISSN
- 1944-8244
1944-8252
- Abstract
- We present a heterojunction based on the n-type organic semiconductor N,N '-ditridecyl-3,4,9,10-perylenetetracarboxylic diimide (PTCDI-C-13) with a PTCDI-C-13/parylene/PTCDI-C-13-layered structure, enabling dual functionality as both an optical synaptic and a memory transistor. The device exploits a buried contouring PTCDI-C-13 layer, where the lower PTCDI-C-13 and intervening parylene layers serve distinct functions in charge trapping and modulation. In memory mode, the buried PTCDI-C-13 serves as a floating gate, while the parylene layer acts as a tunneling barrier, facilitating charge storage and controlled electron tunneling under combined optical and electrical stimulations. In synaptic mode, the thickness of the buried PTCDI-C-13 dictates the surface roughness, which is transferred to the parylene layer, forming a textured interface with abundant charge trap sites that modulate the synaptic behavior. By tuning the PTCDI-C-13 thickness, we controlled the interface roughness and trap density (n t), achieving optimal performance at 82 nm. The device successfully emulated synaptic plasticity and demonstrated transitions to long-term memory. To further verify its neuromorphic capabilities, our device achieved a recognition accuracy of 91.7% in a Modified National Institute of Standards and Technology-based classification simulation, successfully replicating biological synaptic behavior. Additionally, an electrocardiogram-based simulation demonstrated high classification accuracy while effectively processing dynamic, time-dependent signals. By reliably performing both static image recognition and dynamic biosignal processing, our device showcases its potential for real-time biomedical diagnostics, adaptive AI, and bioinspired computing applications.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - COLLEGE OF ENGINEERING SCIENCES > SCHOOL OF ELECTRICAL ENGINEERING > 1. Journal Articles

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.