Methane yield optimization using mix response design and bootstrapping: application to solid-state anaerobic co-digestion process of cattle manure and damp grass

Abstract

Optimization of solid-state anaerobic digestion on cattle manure and damp grass were performed simultaneously and combined to a bootstrapping tool to significantly decrease the number of experiments for a methane yield optimization. 15 batch reactors at lab scale were launched two times respecting a mix surface response methodology. A numerical method called Bootstrapping was used to verify results significance. Results have shown a significant influence with a p-value between 10−8 and 10−11: the optimal parameter values depend on the substrate composition with a maximal p-value of 5.60.10−2. The methane yield reached 156.19 NL · kgVS−1 for a mixture of cattle manure and damp grass, and 142.92 NL · kgVS−1 for cattle manure only. The bootstrapping were validated with a standard error lower than 3% in comparison with ANOVA method, which confirms that the mix surface response methodology combined to bootstrapping is an innovative and efficient way to optimize solid-state anaerobic digestion process.