Experimental demonstration of SnO₂ nanofiber-based memristors and their data-driven modeling for nanoelectronic applications

This paper demonstrates the fabrication, characterization, data-driven modeling, and practical application of a 1D SnO2 nanofiber-based memristor, in which a 1D SnO2 active layer is sandwiched between silver (Ag) and aluminum (Al) electrodes. This device yields a very high ROFF:RON∼104 (ION:IOFF∼105) with an excellent activation slope of 10 mV/dec, low set voltage VSET∼1.14 V, and good repeatability. This paper physically explains the conduction mechanism in the layered SnO2 nanofiber-based memristor. The conductive network comprises nanofibers that play a vital role in the memristive action, as more conductive paths could promote the hopping of electron carriers. Energy band structures experimentally extracted using ultraviolet photoelectron spectroscopy (UPS) strongly support the claims reported in this paper. We have developed an ML-assisted, data-driven model of the fabricated memristor using different popular algorithms such as Polynomial Regression, Support Vector Regression (SVR), k Nearest Neighbors (kNN), and Artificial Neural Network (ANN) to model the data of the fabricated device. We have proposed two types of ANN models (type I and type II) algorithms, illustrated with a detailed flowchart, to model the fabricated memristor. Benchmarking with standard ML techniques shows that the type II ANN algorithm provides the best Mean Absolute Percentage Error (MAPE) of 0.0175 with a 98% R2 score. We have further validated the proposed data-driven model with the characterization results of similar new memristors fabricated using the same fabrication recipe, giving satisfactory predictions. Lastly, we have applied the ANN type II model to design and implement simple AND & OR logic functionalities using the fabricated memristors with expected, near-ideal characteristics.

Перевод пока недоступен