Bending Resistant Multibit Memristor for Flexible Precision Inference Engine Application

Abstract

This work reports 2-bits/cell hafnium oxide-based stacked resistive random access memory devices fabricated on flexible polyimide substrates for neuromorphic applications considering the high thermal budget. The ratio of low-resistance state current ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${I}_{ \mathrm{\scriptscriptstyle ON}}$ </tex-math></inline-formula> ) to high-resistance state current ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${I}_{ \mathrm{\scriptscriptstyle OFF}}$ </tex-math></inline-formula> ) or <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${I}_{ \mathrm{\scriptscriptstyle ON}}/{I}_{ \mathrm{\scriptscriptstyle OFF}}$ </tex-math></inline-formula> for the fabricated devices was above <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.4\times10$ </tex-math></inline-formula> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> with a low device-to-device variation at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$100 \boldsymbol {\mu }\text{A}$ </tex-math></inline-formula> current compliance. The mechanical stability over 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sup> bending cycles at a 5 mm bending radius and endurance over 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> WRITE cycles makes these devices suitable for online neural network training. The data retention capability over 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sup> s at 125°C also infuses these devices’ long-term inference capability. Furthermore, the performance of the devices has been verified for neuromorphic applications by system-level simulations with experimentally calibrated data. The system-level simulation reveals only a 2% loss in inference accuracy over ten years from the baseline.