Presentation Information

[8p-N304-11]Enhanced Reservoir Computing Performance with Staggered Polarization Switching in MFIFM HZO Capacitor

〇(D)DONGHEE LEE1, Kasidit Toprasertpong1, Ryosho Nakane1, Mitsuru Takenaka1, Shinichi Takagi1 (1.Univ. Tokyo)

Keywords:

Reservoir computing,Ferroelectric HZO,MFIFM capacitor

This study demonstrates an MFIFM HZO capacitor for reservoir computing based on staggered polarization switching. Compared with a single-layer MFM HZO capacitor, the asymmetric MFIFM stack with two HZO layers separated by an ultrathin Al2O3 interlayer exhibits multi-peak switching current characteristics. These diverse transient responses improve the memory and nonlinear components required for reservoir computing, resulting in enhanced short-term memory and parity-check performance.