Colloidal Interaction of Cellulose with Solid-acid Catalyst and Its Implications for Hydrolysis

Zhang, Ziyang

Etd

Colloidal Interaction of Cellulose with Solid-acid Catalyst and Its Implications for Hydrolysis

公开

Lignocellulose based biomass has become one of the most promising renewable energy sources for potentially replacing fossil fuel-based energy source such as coal and natural gas. Solid-acid catalysts, e.g. carbon, polystyrene or metal oxides are proposed to decompose cellulose into reduced sugares due to their advantages such as ease of separation, recyclability, reusability and cost-effectiveness. However, current design principals for solid-acid mostly focuses on molecular binding of functionalized solid with cellulose. In fact, both cellulose and solid acid catalysts are colloidal particles, meaning interfacial colloidal interaction between cellulose and solid acid catalysts are important rather than molecular interpretation. Unfortunately, interfacial surface interaction at colloidal level remains largely unknown. The Derjaguin and Landau, Verwey and Overbeek model (D.L.V.O.) is widely used for analyzing colloid surface interactions and designing a colloid system either promoting colloid stability or instability. To apply DLVO analysis to cellulose-solid acid catalysts interaction, van der Waals attractions have to be analyzed beforehand. Surface energy has been used to indirectly determine the Hamaker constant, a constant that dictates strength of van der Waals attraction. However, a common approach that uses pure liquids contact angles has problems associated with surface compatibility, toxicity and evaporation. Here, we have developed a new method which uses contact angles of water organic solvent (dimethyl sulfoxide, formamide and ethylene glycol) to precisely estimate polymer surface energies components using theories including Owens-Wendt and van Oss. New method has precisely and accurately measured surface energies for polymer surfaces such as PDMS, PMMA and PVC. The success of applying mixtures method on polymer surfaces has motivated us to extend it to biomass particles such as cellulose and chitin. Analysis shows that mixture approach can result in similar surface energy components to values reported in literatures. Using determined surface energies values, we calculate the Hamaker constant and apply DLVO theory to rationally design solid-acid catalysts with enhanced capacity for adsorbing cellulose. Results show that cellulose-solid acid particles interaction is favored for the cases where catalysts have a large Hamaker constant (or high surface energy) and low surface charges. A series of recommendations are made for enhancing cellulose-solid acid catalysts adsorption, including using weak acid e.g. pKa > 2-3), high temperature (T > 150oC), bifunctional catalysts with roughly equal amount of acid/base site density and high shear rate (e.g. > 10 s-1). To overcome electrostatic repulsion and increase coagulation, we introduced a cationic polyelectrolyte, poly-diallyldimethylammonium chloride to weaken solid acid surface negative charge (e.g. Amberlyst-15, ZSM-22). Our results show that addition of polyDDA increases product (glucose) yield by roughly 10%. We then investigate interfacial surface interaction between cellulose with liquid of reaction medium, to enhance surface reactivity. Cellulose is typically mechanically pretreated to remove crystallites prior to acid hydrolysis. However, certain liquids, such as water, can trigger cellulose recrystallization and reduce reactivity. Hydrophobic effects at the water-cellulose interface has motivated us to engineer and control interfacial interactions by adding inorganic/organic salts. Our studies show that salts contribute differently to glucose yield: guanidinium chloride increases glucose yield by around 16% compared to reaction without salts added, while ammonium chloride, lithium chloride increase glucose yield slightly by less than 5%. Further cellulose structure characterization analysis such as X-Ray diffraction and Raman reveals that guanidinium chloride may be able to suppress crystal growth and reduces cellulose crystal size, increasing accessibility while other salts such as lithium chloride and ammonium chloride have little effect on cellulose crystal size growth. Further, molecular dynamic (MD) simulation analysis shows that guanidinium chloride and HCl suppresses cellulose crystal growth the most while ammonium chloride has helped the growth of cellulose crystal the most. MD simulations have also shown that guanidinium chloride can influence glycosidic bonds angle distribution, revealing its interferes with cellulose chain structures.

Creator