2019年第66回応用物理学会春季学術講演会

講演情報

一般セッション(口頭講演)

9 応用物性 » 9.3 ナノエレクトロニクス

[12p-W934-1~10] 9.3 ナノエレクトロニクス

2019年3月12日(火) 13:15 〜 16:00 W934 (W934)

田中 啓文(九工大)、内藤 泰久(産総研)

15:45 〜 16:00

[12p-W934-10] Short-term and long-term memory of random aggregation device using Ag-Ag2S nanoparticles

〇(D)Hadiyawarman Hadiyawarman1、Masanori Eguchi2、Hirofumi Tanaka1 (1.Kyushu Inst. of Tech.、2.Kure Nat. Col. of Tech.)

キーワード:Memristive behavior, Ag-Ag2S core-shell nanoparticles, Short-term Memory

[Introduction] Neuromorphic devices are expected to have a high-performance arithmetic circuit with very low power consumption to be applied in many fields, such as brain-like computer. Especially it is important to control the structure of random network for reservoir computing application. In the present study, we demonstrated a cross-talk, learning behavior among the Ag-Ag2S core-shell nanoparticles aggregation and neuromorphic behavior.

[Experimental] The Ag-Ag2S core-shell nanoparticles were synthesized by modified Brust-Schiffrin procedure [1] at room temperature with Ag/S molar ratios of 0.25/1 as follows: A toluene solution of allylmercaptane (0.37 ml, 4.71 mmol) is mixed with a solution of silver nitrate (200 mg, 1.18 mmol). Then, tetraoctylammonium bromide in deionized (DI) water (360 mg, 0.658 mmol) was added to the solution and followed by adding sodium borohydride solution in DI water (260 mg, 6.87 mmol). The mixture was allowed to react for 2 h at room temperature with magnetic stirring. The water phase was then removed from obtained solution and then was centrifuged at 4000 rpm for 20 min to separate the nanoparticles from the liquid phase.

[Results and discussion] The structural properties of Ag-Ag2S nanoparticles were investigated by XRD, XPS and TEM as shown in Figure 1. It clearly showed that the particles have the range 20-100 nm of diameter and consisted of two different crystal structures, namely Ag as the core and Ag2S as the shell. To create aggregation of particles, the highly concentrated nanoparticles in ethanol were drop-casted on to 50 ℃ of 4-electrodes device and characterized the learning properties by repeated sweeping the bias as shown in Figure 2. Furthermore, the memorization process, such a short-term and long-term memorization was studied by applying rectangular wave input. The details will be presented at the conference.