Ferroelectric HfZrO2 With Electrode Engineering and Stimulation Schemes as Symmetric Analog Synaptic Weight Element for Deep Neural Network Training
Autor: | Min-Sheng Liao, K.-T. Chen, Shun-Ping Chang, Y.-J. Tseng, C.-S. Chang, Y.-Y. Lin, J.-P. Chiu, C.-C. Ho, C.-Y. Liao, Min-Hung Lee, Tuo-Hung Hou, C.-Y. Chueh, Yao-Joe Yang, Kuo-Yu Hsiang, Chun-Ming Chang, C.-H. Wu |
---|---|
Rok vydání: | 2020 |
Předmět: |
010302 applied physics
Physics Pulse analysis Analytical chemistry chemistry.chemical_element Polarization modulation 01 natural sciences Ferroelectricity Electronic Optical and Magnetic Materials Atomic layer deposition chemistry 0103 physical sciences Electrode Field-effect transistor Electrical and Electronic Engineering Tin |
Zdroj: | IEEE Transactions on Electron Devices. 67:4201-4207 |
ISSN: | 1557-9646 0018-9383 |
Popis: | Atomic layer deposition (ALD)-based TiN electrode on ferroelectric HfZrO2 metal/ferroelectric/metal (MFM) capacitor and ferroelectric field-effect transistor (FeFET) is demonstrated experimentally with weight transfer, that is, $\Delta {P}$ , 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, $\alpha _{p} / \alpha _{d} = -0.63$ / −0.84, asymmetry $\vert \alpha _{p} - \alpha _{d}\vert =0.21$ , and polarization modulation ratio (Pot./Dep.) = 97%/98% are achieved for MFM capacitor, and $\alpha _{p} / \alpha _{d} = -1.32$ / −1.88, asymmetry $\vert \alpha _{p} - \alpha _{d}\vert =0.56$ , and $G_{\text {max}} / G_{\text {min}} > 10\times $ are delivered for FeFET. |
Databáze: | OpenAIRE |
Externí odkaz: |