Optimisation of Two-step Steam Pretreatment of Softwood for Bioethanol Production

Fuel ethanol can be produced from lignocellulosics by an enzymatic process consisting of steam pretreatment, enzymatic hydrolysis, fermentation and ethanol refining. This thesis deals with the steam pretreatment step in the enzymatic process used for ethanol production from softwood, which is an abundant lignocellulosic raw material in the Northern hemisphere.Steam pretreatment is an effective method of hydrolysing hemicellulosic sugars, as well as enhancing the accessibility of enzymes to the cellulose. Too mild conditions will result in a material with low enzymatic digestibility. On the other hand, too harsh pretreatment conditions will degrade some sugars to by-products causing a decrease in the yield, as well as inhibition in the following process steps.In this thesis a two-step steam pretreatment process is described where hemicellulosic sugars are recovered after the first pretreatment step, performed under mild conditions, and the enzymatic digestibility of cellulose is increased in the second pretreatment step, performed under harsher conditions. Prior to steam pretreatment either SO2 or H2SO4 was added to catalyse the hydrolysis. Impregnation with SO2 in both pretreatment steps resulted in the highest overall ethanol yield, regardless of whether simultaneous saccharification and fermentation or separate enzymatic hydrolysis and fermentation was employed. Experimental data obtained when steam pretreatment and the enzymatic steps were performed in a process development unit resulted in valuable information on overall sugar and ethanol yields. Two-step steam pretreatment results in higher overall ethanol and sugar yields than one-step steam pretreatment. An economic evaluation shows that although the ethanol yield is higher, this advantage is somewhat offset by the higher capital and energy costs. It is possible to perform the second step without a pressure reduction and re-impregnation of SO2, without a reduction in ethanol yield. The production cost for such a process is lower than for a one-step steam pretreatment process. However, the process concept has yet to be proved on pilot scale