Maximizing biohydrogen production from watermelon peels using Clostridium butyricum NE133: a statistical optimization approach with Plackett–Burman and Box–Behnken designs

Abstract Background Biohydrogen production from agricultural waste is a promising strategy to address climate change and energy challenges. This study aimed to optimize the process parameters for biohydrogen production from watermelon peels (WMP) by Clostridium butyricum NE133 using statistical optimization techniques. Initial screening of eight significant variables influencing hydrogen production including, initial pH, incubation temperature, WMP concentration, inoculum volume, yeast extract, tryptone, sodium acetate, and ammonium acetate concentration was conducted by a Plackett–Burman (PB) design. Results The results showed that four variables including, initial pH (P < 0.001), WMP concentration (P < 0.001), sodium acetate (P = 0.023), and ammonium acetate (P = 0.048) had statistically significant effects on hydrogen production. The model curvature (P = 0.040) indicated that it was significant. Box–Behnken (BB) design under response surface methodology (RSM) was employed to optimize the four selected variables to maximize hydrogen production. The optimal conditions for maximizing hydrogen production from WMP by C. butyricum were: initial pH of 8.98, WMP concentration of 44.75%, sodium acetate 4.49 gL−1, and ammonium acetate 1.15 gL−1 at with predicted Hmax of 4703.23 mLL−1. The determination coefficient R2 of the model was 0.9902 with the lack of fit F-value was 1.86. Conclusions The confirmation experiment revealed only a 0.59% difference between the predicted and experimental hydrogen production, indicating that the optimum conditions were actual with the least error. Improvement of about 103.25% in hydrogen production from WMP by C. butyricum NE133 was achieved after the optimization process. Graphical Abstract