Improved hydrolysis yields and silica recovery by design of experiments applied to acid-alkali pretreatment in rice husks.

[Display omitted] • 2 sequential approaches of experimental design were used to evaluate 5 variables. • Optimized conditions have a sugar release 8 times higher than rice husks in natura. • Silica of 97 % purity can be isolated from the alkaline liquor at high yields. • Lignin and hemicellulose were removed from rice husks in the acid pretreatment step. • Surprisingly, the alkali step removed mainly silica (not lignin) from the substrate. In this work, a two-step pretreatment using acid and alkali was optimized for rice rusks (Oryza sativa) using a 25−1 fractional factorial design (FFD), followed by a central composite design (CCD) to further optimization of enzymatic saccharification. The effect of five variables was simultaneously evaluated: H 2 SO 4 concentration (from 0–5.4 % w/w); NaOH concentration (0–6 % w/w); temperature (85–125 °C); time (20–100 min) and solid to liquid ratio (S/L = 5–12.5 % w/w). The best pretreatment conditions were: 1.8 % w/w of H 2 SO 4 in the first step and 6 % w/w of NaOH at 85 °C for 100 min at a S/L = 12.5 % (w/w) in the alkaline step, which resulted in 58.7 mg of glucose/g substrate, an 8-fold increase compared to the sample in natura (7 mg/g). In rice husks, in contrast to the results commonly found in literature, NaOH extracts mainly silica instead of lignin, while H 2 SO 4 has an important role in lignin removal. High purity silica (97 %) was isolated at high yields (70 %) from the alkaline liquor by a simple and scalable process, which could contribute to making ethanol production from this biomass economically viable. [ABSTRACT FROM AUTHOR]