Hydrogen Evolution Reaction of Electrodeposited Ni-W Films in Acidic Medium and Performance Optimization Using Machine Learning

Ni−W alloy films were electrodeposited from a gluconate aqueous bath at pH=5.0, at varying current densities and temperatures. While there is little to no difference in composition, i. e., all films possess ~12 at.% W, their activity at hydrogen evolution reaction (HER) in acidic medium is greatly influenced by differences in surface morphology. The kinetics of HER in 0.5 M H₂SO₄ indicates that the best performing film was obtained at a current density of −4.8 mA/cm² and 50 °C. The Tafel slopes (b) and the overpotentials at a geometric current density of −10 mA/cm² (η₁₀) obtained for 200 cycles of linear sweep voltammetry (LSV) from a set of films deposited using different parameters were fed into a machine learning algorithm to predict optimum deposition conditions to minimize b, η₁₀, and the degradation of samples over time. The optimum deposition conditions predicted by the machine learning model led to the electrodeposition of Ni−W films with superior performance, exhibiting b of 33–45 mV/dec and an η₁₀ of 0.09–0.10 V after 200 LSVs.