Enzymatic hydrolysis is limited by biomass–water interactions at high-solids: improved performance through substrate modifications

BACKGROUND: To improve process economics for production of fuels and chemicals from lignocellulosic biomass, high solids concentrations are applied in enzymatic hydrolysis, to increase product concentration and reduce energy input. However, increasing solids concentrations decrease cellulose conversion yields, the so called ‘high-solids effect.’ Previous work suggests that product inhibition and mixing contribute, but an understanding of how biomass properties influence the high-solids effect, is lacking. RESULTS: Cellulose hydrolysis yields with an industrial cellulase (Ctec2) were measured on pretreated wheat straw and spruce from 5 to 30% dry matter (DM), and compared to yields of an older industrial cellulase mixture (Celluclast 1.5L/Novozym188). For Ctec2, yield was independent of DM below 15–18% DM, while yields decreased with increasing DM above this range, but at different rates for each biomass. For Celluclast 1.5L/Novozym188, yields decreased already from the lowest DM, suggesting that the high-solids effect was more a function of product inhibition, while the yields of the newer Ctec2 mixture were driven more by biomass–water interactions. LF-NMR relaxometry showed that the onset of the high-solids effect for Ctec2 corresponded to the disappearance of free water from the system, and a decrease in water self-diffusion rates. While the spruce had higher yields at low-solids, the wheat straw had higher yields at high-solids conditions, exhibiting that relative yields at low and high-solids are not related. Higher yields corresponded to increased water constraint by the biomass at high-solids conditions. Modifications to the pretreated wheat straw resulted in improved yields, and changes to the inflection point and intensity of the high-solids effect, showing that this effect can be reduced. CONCLUSIONS: The high-solids effect is both enzyme and substrate dependent, and can be reduced by modifying the pretreated biomass, suggesting that pretreatment processes can be designed to achieve similar effects. Yields at low and high-solids concentrations do not correlate for a given biomass, and thus industrial evaluation of biomass recalcitrance should be carried out at high-solids conditions. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s13068-018-1339-x) contains supplementary material, which is available to authorized users.