TY - JOUR
T1 - Exploration of Analog Synaptic Plasticity and Convolutional Neural Network Simulation in Bilayer TiOxNy/SnOx Memristor for Neuromorphic Systems
AU - Ismail, Muhammad
AU - Kim, Doohyung
AU - Lim, Eunjin
AU - Rasheed, Maria
AU - Mahata, Chandreswar
AU - Seo, Yeongkyo
AU - Kim, Sungjun
N1 - Publisher Copyright:
© 2024 American Chemical Society
PY - 2024/8/5
Y1 - 2024/8/5
N2 - In this study, a TiN/SnO2/Pt sandwich structure is explored for its dual functionalities in electronic synapses and multistate memory. The SnO2 layer is fabricated via reactive sputtering, leading to the formation of a TiN/TiOxNy/SnOx/Pt memristor. This configuration, confirmed by HRTEM and XPS analyses, exhibits several advantageous features: consistent bipolar nonvolatile switching at low operating voltages, endurance up to 500 cycles, an on/off ratio of ∼102, and robust data retention. Set and reset times are approximately 300 and 400 ns, with energy consumption of 3.24 nJ and 3.26 nJ, respectively. The memristor achieves multilevel resistance states, simulating synaptic behaviors such as LTP/LTD, SADP, PPF, and PPD. Utilizing LTP and LTD data, CNN simulation achieved 91.3% recognition accuracy, surpassing the 70.5% accuracy of ANN simulation. These findings suggest the TiN/TiOxNy/SnOx/Pt memristor’s potential for artificial neural network applications.
AB - In this study, a TiN/SnO2/Pt sandwich structure is explored for its dual functionalities in electronic synapses and multistate memory. The SnO2 layer is fabricated via reactive sputtering, leading to the formation of a TiN/TiOxNy/SnOx/Pt memristor. This configuration, confirmed by HRTEM and XPS analyses, exhibits several advantageous features: consistent bipolar nonvolatile switching at low operating voltages, endurance up to 500 cycles, an on/off ratio of ∼102, and robust data retention. Set and reset times are approximately 300 and 400 ns, with energy consumption of 3.24 nJ and 3.26 nJ, respectively. The memristor achieves multilevel resistance states, simulating synaptic behaviors such as LTP/LTD, SADP, PPF, and PPD. Utilizing LTP and LTD data, CNN simulation achieved 91.3% recognition accuracy, surpassing the 70.5% accuracy of ANN simulation. These findings suggest the TiN/TiOxNy/SnOx/Pt memristor’s potential for artificial neural network applications.
UR - http://www.scopus.com/inward/record.url?scp=85198367178&partnerID=8YFLogxK
U2 - 10.1021/acsmaterialslett.4c00406
DO - 10.1021/acsmaterialslett.4c00406
M3 - Article
AN - SCOPUS:85198367178
SN - 2639-4979
VL - 6
SP - 3514
EP - 3522
JO - ACS Materials Letters
JF - ACS Materials Letters
IS - 8
ER -