Synthesis of NiO/rGO Nanocomposites for Energy Storage Applications through Structural, Optical and Electrochemical Characterization with Machine Learning-based Prediction
Abstract
Nickel oxide/reduced graphene oxide (NiO/rGO) nanocomposites have attracted considerable interest as promising electrode materials of high electrical conductivity, tunable electronic structure, and excellent electrochemical characteristics. In this work, a hydrothermal route was employed to synthesize NiO/rGO nanocomposites containing 90 wt.% NiO and 10 wt.% rGO. An intelligent predictive framework integrating fuzzy logic (FL), machine learning (ML), and experimental characterization was developed to optimize the structural, optical, and electrochemical properties of the synthesized material. A Mamdani fuzzy inference system with three input variables (NiO content, rGO loading, and precursor molarity) and three output variables (crystallite size, optical band gap, and photoluminescence (PL) wavelength) was established to estimate material responses. Four ML algorithms, namely Artificial Neural Network, Random Forest, Gaussian Process Regression (GPR), and a Hybrid model were trained using a dataset of 1000 samples with an 80:20 training–testing split to predict crystallite size, optical band gap, PL wavelength, charge storage capacity, charge transfer resistance, energy density, and power density. X-ray diffraction confirmed the formation of crystalline cubic NiO with an average crystallite size of 16.85 nm. Field emission scanning electron microscopy with energy dispersive X-Ray spectroscopy verified the homogeneous distribution of NiO nanoparticles over wrinkled rGO sheets, while Fourier transform infrared analysis confirmed the interaction between NiO and rGO. Photoluminescence analysis revealed an emission wavelength of 430 nm with a band gap of 3.18 eV. The predicted values obtained from FL and ML showed excellent agreement with the experimental measurements, with the Hybrid and GPR models providing the highest prediction accuracy. The proposed integrated experimental–computational methodology offers a reliable and efficient strategy for designing high-performance NiO/rGO nanocomposites for advanced energy-storage applications.