[17p-S4_202-1~12] 9.5 New functional materials and new phenomena

2:15 PM - 2:30 PM JST(5:15 AM - 5:30 AM UTC)

[17p-S4_202-2] In-materio Physical Reservoir Computing Using a Photoresponsive Single-Walled Carbon Nanotube/Porphyrin Derivative Network

〇Koki Nietani¹, Deep Banerjee¹, Hiroyuki Furuta^2,3, Yuki Usami^1,4, Hirofumi Tanaka^1,4 (1.LSSE, Kyushu Inst. Tech., 2.Grad. Sch. of Eng., Kyushu Univ., 3.College of Sci., SHU, 4.Neumorph Center, Kyushu Inst. Tech)

Keywords:

Reservoir Computing,Edge AI

We investigate the effect of optical irradiation on in-materio physical reservoir computing using photoresponsive single-walled carbon nanotube (SWNT)/porphyrin derivative composite networks. Under 457-nm laser illumination, the electrical characteristics and waveform generation performance were evaluated. The SWNT/NCP composite, which exhibits optical absorption at this wavelength, showed enhanced hysteresis in the I–V characteristics and improved accuracy in waveform generation tasks. In contrast, no significant changes were observed for the SWNT/TTP composite, which does not absorb at 457 nm. These results indicate that optical absorption modulates the electronic states and charge transport dynamics in the SWNT/NCP network, leading to enhanced nonlinearity and improved computational performance.