Mild steel is widely used due to its low cost, ease of fabrication, and versatility, but its susceptibility to surface degradation requires protective coatings to enhance performance and service life. Electroless nickel–phosphorus (ENi–P) coating is preferred as it provides uniform, adherent, and wear-resistant deposits on complex geometries without external electrical current. This study presents a two-phase optimization of an electroless Ni–P bath for mild steel, progressing from single-factor analysis to multivariate modelling. In Phase I, a One-Factor-at-a-Time (OFAT) approach was used to study nickel chloride, sodium hypophosphite, and trisodium citrate individually, each showing a bell-shaped response with an intermediate optimum. The highest coating mass (0.34 g), along with 97.31% Ni, 2.69% P, and 507 HV hardness, was obtained at the central condition (A = 30 g/L, B = 40 g/L, C = 25 g/L). Quadratic fits showed excellent agreement with experimental data, and Solver-based refinement further identified nearby optimal conditions for each parameter. Microstructural analysis confirmed that balanced bath compositions produced compact, defect-free coatings with higher hardness, while deviations led to porosity and reduced performance.
In Phase II, a Box–Behnken Design (RSM) was developed within the Phase I concentration limits to build a second-order predictive model. Response surface analysis confirmed that the optimum lies near the center of the design space and revealed a critical interaction effect: simultaneous increase of sodium hypophosphite and trisodium citrate beyond their optimal levels leads to a sharp reduction in coating mass, reaching a minimum of 0.23 g, which could not be captured through single-factor analysis. The desirability-based optimization identified the optimum at A = 30.0 g/L, B = 38.8 g/L, and C = 26.9 g/L with a predicted coating mass of 0.3240 g (desirability = 0.854). This closely matched the OFAT and Solver results, and the consistent convergence across all methods confirms the robustness of the optimization and highlights.

Mild steel is widely used due to its low cost, ease of fabrication, and versatility, but its susceptibility to surface degradation requires protective coatings to enhance performance and service life. Electroless nickel–phosphorus (ENi–P) coating is preferred as it provides uniform, adherent, and wear-resistant deposits on complex geometries without external electrical current. This study presents a two-phase optimization of an electroless Ni–P bath for mild steel, progressing from single-factor analysis to multivariate modelling. In Phase I, a One-Factor-at-a-Time (OFAT) approach was used to study nickel chloride, sodium hypophosphite, and trisodium citrate individually, each showing a bell-shaped response with an intermediate optimum. The highest coating mass (0.34 g), along with 97.31% Ni, 2.69% P, and 507 HV hardness, was obtained at the central condition (A = 30 g/L, B = 40 g/L, C = 25 g/L). Quadratic fits showed excellent agreement with experimental data, and Solver-based refinement further identified nearby optimal conditions for each parameter. Microstructural analysis confirmed that balanced bath compositions produced compact, defect-free coatings with higher hardness, while deviations led to porosity and reduced performance.
In Phase II, a Box–Behnken Design (RSM) was developed within the Phase I concentration limits to build a second-order predictive model. Response surface analysis confirmed that the optimum lies near the center of the design space and revealed a critical interaction effect: simultaneous increase of sodium hypophosphite and trisodium citrate beyond their optimal levels leads to a sharp reduction in coating mass, reaching a minimum of 0.23 g, which could not be captured through single-factor analysis. The desirability-based optimization identified the optimum at A = 30.0 g/L, B = 38.8 g/L, and C = 26.9 g/L with a predicted coating mass of 0.3240 g (desirability = 0.854). This closely matched the OFAT and Solver results, and the consistent convergence across all methods confirms the robustness of the optimization and highlights.