LAUSR

With the exploration of ferroelectric materials, researchers have a strong desire to explore the next generation of non-volatile ferroelectric memory with silicon (Si)-based epitaxy, high density storage and algebraic operations. Here, we report a silicon-based epitaxial memristor in vertically aligned nanostructures BaTiO3 -CeO2 based on La0.67 Sr0.33 MnO3 (LSMO)/ SrTiO3 (STO)/Si substrate. The ferroelectric polarization reversal is optimized through continued exploring growth temperature, and the epitaxial structure is obtained, thus it improves the resistance characteristic the multi-value storage function of five states is achieved, and the robust endurance characteristic can reach 109 cycles. In the synapse plasticity process by pulse modulation, and the neural modulation function of spiking-time-dependent plasticity (STDP) and paired-pulse facilitation (PPF) is simulated successfully. More importantly, the algebraic operations of addition, subtraction, multiplication and division are realized by using fast speed pulse of the width ∼ 50 ns. Subsequently, we constructed a Convolutional Neural Network (CNN) for identifying the CIFAR-10 dataset to simulate the performance of our device, the online and offline learning recognition rate reached 90.03% and 92.55%, repectively. This study pave the way for silicon-based epitaxy ferroelectric memristors to realize multi-value storage, algebraic operations and neural computing chips application. This article is protected by copyright. All rights reserved.