A strategy to predict the current conduction mechanisms into Al/PVP:Gr-BaTiO3/p-Si Schottky structure using Artificial Neural Network

Abstract

In this work, Artificial Neural Network (ANN) algorithm is used to predict the current conduction mechanisms into the metal-semiconductor (MS) and metal-nanocomposite-semiconductor (MPS) structures along with their primary electronic parameters, such as the leak current (I0), potential barrier height (ΦB0), ideality factor (n), series/shunt resistance (Rs/Rsh), rectifying ratio (RR), and interface states density (Nss) by analyzing the I–V characteristics. The polyvinylpyrrolidone (PVP), barium titanate (BaTiO3) and graphene (Gr) nanoparticles are mixed together to create the interfacial nanocomposite layer. Training data for ANN algorithm is gathered using the thermionic emission hypothesis. In order to study the efficacy of the ANN model, the predictive power of the ANN technique for predicting the current conduction mechanisms and electronic properties of SDs has been assessed by comparing the predicted and experimental results. The ANN predictions of the current conduction mechanisms at the forward/reverse bias and the fundamental electronic specifications of the MS and MPS structures are a high level of agreement with the experimental results. Furthermore, the results show that the RR and Rsh rise whereas the n, Rs, and Nss for MS structure decrease when the PVP:Gr-BaTiO3 nanocomposite interlayer is employed.