Quantification of feedstock characteristics before and during hydrothermal liquefaction under subcritical conditions of water

Abstract

Hydrothermal liquefaction (HTL) is a thermal process that converts organics in biomass and waste into renewable crude-like oil. Sewage sludge is a typical waste material that can produce crude-like oil via HTL under subcritical conditions due to its organic-rich content. The composition and properties of sewage sludge significantly influence the feedstock's processability, the conversion of organics, and the crude-like oil yields. HTL feedstock properties, specifically viscosity and density, are important flow properties that affect HTL product formation at different reaction conditions. Knowledge of the feedstock viscosity helps estimate slurry transportation through pipelines, pumping power and heat transfer requirements for design purposes. To determine the flow properties before HTL for pipeline specification, the settling characteristics of biosolid slurry was determined using batch settling experiments. Results of the settling test were used to assess the stability of the slurry during transportation. The effect of solid concentration and particle size of biosolids on slurry stability and pumpability was evaluated. Stability of biosolid slurries improved with an increase in solid concentration. The rheological properties of sewage sludge obtained from different parts of the wastewater treatment plant were also estimated. Generally, sludge slurries were determined to be non-Newtonian fluids. Rheological parameters of sludge feedstock, including yield stress, flow behaviour and consistency indices, were obtained from rheological models. A comparative study was made on the rheology and pumping power required for different sludge types to determine the transportability of these slurries based on a plant capacity of 1000 tonnes/year. To determine the flow properties of sewage sludge during HTL process, the real-time viscosity of sewage sludge slurries was quantified using a modified batch reactor. The torque-rotational speed data of the impeller was converted to shear stress-shear rate data to estimate viscosity. The Couette and Metzner-Otto methods were shown to be valid for real-time viscosity measurement under subcritical conditions. Apparent viscosity changes of lipid, proteins and carbohydrate model compounds at different reaction conditions were estimated. Model compounds exhibited unique viscosity profiles based on the mass yields and chemical speciation of HTL products. Apparent viscosity changes in reacting sludge slurries were determined at variable solid concentration, temperature and pressure. Significant differences were observed between the apparent viscosity profiles of sludge slurries and the apparent viscosity profiles of mixtures of model compounds with similar organic compositions. The effects of lignin, inorganics and the dominance of specific macromolecules on the apparent viscosity of sludge were analysed through a comparative study with the apparent viscosity of microalgae and determined to have significant effects on the apparent viscosity profile of sludge slurries. The major contributions from this PhD investigation can be applied to determine the real-time viscosity of fluids and slurries under subcritical conditions. The study on the rheology of sludge is vital for pipeline specification and reactor design purposes. The changes in apparent biomass viscosity can be predicted for design purposes and process monitoring during reactor operations in an HTL plant.