Journal of Semiconductors, Volume. 45, Issue 9, 092402(2024)
InGaZnO-based photoelectric synaptic devices for neuromorphic computing
Fig. 1. (Color online) (a) Schematic illustration of the device structure and fabrication process flow. (b) Schematic diagram of the synaptic information transmission in neurons. (c) The X-ray photoelectron spectroscopy (XPS).
Fig. 2. (Color online) (a) Current−voltage (I−V) characteristics under negative voltage scanning from 0 to −2 V. (b) I−V characteristics under positive voltage scanning from 0 to +2 V. (c) LTP and LTD characteristics, elicited by a series of positive and negative voltage pulses, respectively. (d) LTP response induced by a sequence of light spikes.
Fig. 3. (Color online) (a) Schottky barrier at the Ni−IGZO junction under different voltage biases applied to the nickel electrode. (b) Model illustrating the switching mechanism of the device.
Fig. 4. (Color online) (a) EPSC response to light pulses of various durations: 0.1, 0.3, 0.5, 1, 2, 3 s. (b) PPF response to two consecutive light pulses with a 1-s interval. (c) Transition from STP to LTP as the number of pulses increases from 3 to 30 within 60 s. (d) Simulation of learning-forgetting-relearning mechanism in human brain using light pulses.
Fig. 5. (Color online) (a) Schematic diagram of a three-layer ANN for handwritten digit recognition. (b) Progression of recognition accuracy in correlation with the number of training epochs. (c) Average confusion matrix under 10th, 100th, and 1000th training epoch.
Get Citation
Copy Citation Text
Jieru Song, Jialin Meng, Tianyu Wang, Changjin Wan, Hao Zhu, Qingqing Sun, David Wei Zhang, Lin Chen. InGaZnO-based photoelectric synaptic devices for neuromorphic computing[J]. Journal of Semiconductors, 2024, 45(9): 092402
Category: Articles
Received: Apr. 26, 2024
Accepted: --
Published Online: Oct. 11, 2024
The Author Email: Jialin Meng (JLMeng), Lin Chen (LChen)