Characterization and performance prediction of jet pulse electrodeposited Ni-SiC nanocomposites by means of artificial neural networks

Abstract

In this study, Ni-SiC nanocomposites were successfully deposited on Q325 steel substrates using jet pulse electrodeposition. Morphologies, microstructures, microhardness values and corrosion properties of Ni-SiC nanocomposites were examined by scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), Rockwell hardness testing, and electrochemical apparatus. Microhardness and corrosion properties of Ni-SiC nanocomposites were then predicted by BP artificial neural network and compared to experimental values. Results demonstrated that as-prepared Ni-SiC nanocomposites at pulse current density of 4 A/dm2, SiC particle concentration of 5 g/l and jet rate of 5.5 m/s exhibited maximum microhardness reaching up to ~ 884.2 HV. By contrast, Ni-SiC nanocomposite obtained at current density of 4 A/dm2, SiC particle concentration of 5 g/l and jet rate of 5.5 m/s showed smaller racemule-like surface morphology with smooth, fine, and uniform microstructures. Average grain sizes of Ni grains and SiC nanoparticles were estimated to 53.4 nm and 28.7 nm, respectively. Concentrations of Ni and Si in Ni–SiC nanocomposite fabricated at current density of 3 A/dm2, SiC particle concentration of 3 g/l and jet rate of 4 m/s were recorded as 71.4 at% and 11.7 at%, respectively. Corrosion current density of Ni–SiC nanocomposite deposited at current density of 4 A/dm2, SiC particle concentration of 5 g/l and jet rate of 5.5 m/s revealed minimum corrosion current density of 5.1 × 10−5 A/cm2 and maximum impedance value, demonstrating the optimal anticorrosion ability. Maximum MEs of microhardness and corrosion mass loss of Ni–SiC nanocomposite predicted by proposed BP model were estimated to 3.1% and 3.4%, respectively. These findings suggested that BP model could effectively predict microhardness and corrosion mass loss of Ni–SiC nanocomposites.