Optimization of hypochlorous acid generation by HCl electrolysis through response surface methodology and artificial neural networks

Abstract

Hypochlorous acid (HOCl), also known as electrolyzed water (EW), is a sustainable and environment-friendly disinfectant, which can comply with food safety regulations. In this study, HOCl was produced through HCl electrolysis using an acrylic electrolyzing reactor without a diaphragm between iridium oxide electrodes coated with titanium. The Box–Behnken design (BBD) was adopted to study the effects of feed water flow rate (FR), pH, temperature, electric voltage, acid concentration (AC), and acid dosing pump stroke (APS) on HOCl production measured as available chlorine concentration (ACC). Then, optimum parameters were explored using response surface methodology (RSM). The proposed quadratic models defined the relationship between HOCl production and the process variables. The models were statistically significant as shown by analysis of variance. The optimum operating parameters for HOCl production were 6.5 L/min, pH 7.5 at 20 °C, 3.8 volts, 6% AC, and APS at 3%. The FR, AC, and APS were significant parameters, and no substantial effect was observed for pH and temperature of feed water or the variation in current–voltage. High FR caused a conspicuous drop, and the high voltage showed a slight increase in ACC. However, an increase in AC and APS significantly enhanced the ACC, although the pH dropped below 5.0–6.5, indicating the presence of other chlorine-based compounds apart from HOCl. RSM model validation through experimental results indicated that the proposed model could effectively provide a relationship between ACC and process variables. Additionally, an artificial neural network model was tested to predict the ACC that showed good fitness to the experimental results.