Exploring pretreatment with ionic liquids for enzymatic cellulose hydrolysis

The efficient hydrolysis of cellulose, which is one major component of plant biomass, is crucial for the economic production of 2nd and 3rd generation biofuels. Enzymatic cellulose hydrolysis allows for the selective conversion of cellulose to glucose at moderate reaction conditions. The efficiency of enzymatic cellulose hydrolysis is primarily limited by the accessibility of the cellulose polymer to cellulases due to the highly organized and recalcitrant cellulose structure. Therefore, cellulose is pretreated to increase the availability of cellulose for cellulases. For cellulose pretreatment ionic liquids, as new generation of solvents, are attractive because some of them can dissolve cellulose and thereby open the highly organized cellulose structure. Dissolved cellulose can then be precipitated by addition of water, leading to an amorphous regenerated cellulose structure with large pores. Therefore, regenerated cellulose can be hydrolyzed much more efficiently with strongly increased reaction rates. The enzymatic heterogeneous hydrolysis of undissolved, regenerated cellulose with a commercial cellulase preparation was characterized and the effect of residual ionic liquid from the pretreatment was analyzed. The ionic liquid strongly reduced the cellulase activity but the enzyme stability was maintained. Therefore, regenerated cellulose was hydrolyzed much faster, also in the presence of residual ionic liquid. However, with increasing amounts of ionic liquid, the yields were reduced. To improve the hydrolysis of regenerated cellulose a semi-empiric mathematical model was developed. This was successfully used to rationally optimize the cellulase mixture and increase the yield by 10% points. An even further improved process was achieved by combining a chemical hydrolysis step with the enzymatic hydrolysis to achieve nearly quantitative conversion in only 5 hours. In addition to the heterogeneous hydrolysis of undissolved cellulose, the homogeneous hydrolysis of dissolved cellulose in ionic liquid was evaluated. Since common cellulases are not active at such high ionic liquid concentrations, two new analytical techniques based on sugar formation and viscosity decrease were developed to quantify cellulase activity in this ionic liquid reaction system. Using a new cellulase from the extremophile archaeon Sulfolobus solfataricus, it was possible to measure cellulase activity in nearly pure ionic liquid for the first time. Therefore, it was demonstrated that the homogeneous enzymatic hydrolysis of cellulose dissolved in ionic liquid is generally feasible. For a holistic understanding of enzymatic cellulose hydrolysis, mechanistic modeling is a powerful tool. A mechanistic population balance model for homogeneous cellulose hydrolysis was developed that describes changes in chain length distributions. The parameters of a chemical hydrolysis model were estimated to verify the applicability of population balance modeling to describe cellulose hydrolysis. Afterwards, models for the enzymatic cellulose hydrolysis were developed and parameter variations revealed influencing factors of homogeneous enzymatic cellulose hydrolysis. For a further in-depth analysis of cellulose hydrolysis and to provide the required experimental data for future population balance modeling, a new method to analyze cellulose chain length distributions was developed. This method simplifies sample preparation and thereby, allows to measure cellulose chain length distributions much faster. This method was applied to measure and compare chain length distributions of different cellulose substrates, in original form and regenerated from ionic liquid, during the enzymatic hydrolysis. It was found that the cellulase hydrolysis reaction was significantly influenced by the accessibility of the different substrates. Based on the work presented here, fundamental new aspects of ionic liquid assisted enzymatic cellulose hydrolysis were identified. Future work based on these results can ultimately lead to a fully mechanistic model of cellulose hydrolysis that allows for the optimization of enzymatic cellulose hydrolysis, irrespective of biomass origin or pretreatment method. Furthermore, it will be possible to identify targets for the optimization of cellulases and pretreatment method. Therefore, this will be a central element for the economic conversion of cellulose in future biorefineries.