Ferroelectric HfZrO₂ With Electrode Engineering and Stimulation Schemes as Symmetric Analog Synaptic Weight Element for Deep Neural Network Training

Abstract

Atomic layer deposition (ALD)-based TiN electrode on ferroelectric HfZrO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> metal/ferroelectric/metal (MFM) capacitor and ferroelectric field-effect transistor (FeFET) is demonstrated experimentally with weight transfer, that is, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\Delta {P}$ </tex-math></inline-formula> , per pulse analysis through consecutive alternating potentiation/depression (Pot./Dep.) training pulses. The weight training pulse schemes are studied to have symmetric and linear synapse weight transfer to increase the accuracy and accelerate the deep neural network (DNN) training. With ALD TiN inserted, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\alpha _{p} / \alpha _{d} = -0.63$ </tex-math></inline-formula> / −0.84, asymmetry <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\vert \alpha _{p} - \alpha _{d}\vert =0.21$ </tex-math></inline-formula> , and polarization modulation ratio (Pot./Dep.) = 97%/98% are achieved for MFM capacitor, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\alpha _{p} / \alpha _{d} = -1.32$ </tex-math></inline-formula> / −1.88, asymmetry <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\vert \alpha _{p} - \alpha _{d}\vert =0.56$ </tex-math></inline-formula> , and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$G_{\text {max}} / G_{\text {min}} &gt; 10\times $ </tex-math></inline-formula> are delivered for FeFET.