Advancing Empirical Models for the Hydrothermal Liquefaction of Microalgae

Abstract

Hydrothermal liquefaction (HTL) is a sustainable-energy technology used to convert microalgae into an energy-dense biofuel precursor known as biocrude oil. This dissertation illuminates myriad effects of HTL process inputs on product distribution and composition that provide the foundation for advancing new quantitative models of HTL. Employing fast-heating reactors with measured temperature profiles for the HTL of Nannochloropsis oculata enabled kinetic modeling over significantly shorter timescales (10 s – 10 min) than previously established. This improved model demonstrated that the kinetics of biocrude and aqueous co-product (ACP) formation at 300 °C occur on the timescale of just seconds, significantly shorter than previously thought. Regression models of biocrude properties as functions of feedstock characteristics enabled quantification of species identity effects for the first time, which ranged from 11 to 40 % of those of biochemical composition.

The aforementioned insights, along with identified gaps in the literature, informed a more comprehensive and rigorous design of experiments probing the effects of temperature, reaction time, slurry concentration, biochemical composition, and species identity on HTL product yields and elemental compositions. All examined factors affected the yield and makeup of the biocrude, aqueous, solid, and gas products, especially temperature and biochemical composition. The data suggested that increased slurry concentration promotes Maillard reactions between amino acids and saccharides that result in increased biocrude yield, C content, and N content and the inhibition of aqueous ammonium recovery (a nutrient for recycling). Fast HTL (300 °C, 3.2 min) of high-lipid, low-concentration slurries provided recoveries of high-value saturated, monounsaturated, and polyunsaturated fatty acids in the biocrude of up to 89, 80, and 65 wt%, respectively. The same slurries reacted at 200 °C for 31.6 min maximized ACP recyclability while limiting N and S recovery in the biocrude to less than 5 and 8 %, respectively.

These empirical results enabled the development of a novel gravimetric, elemental, and multiphase kinetic model for microalgal HTL. This model leverages known classes of reactions that occur during HTL, such as hydrolysis, Maillard reactions, and deamination, to construct a reaction network with 16 unique pathways. These pathways established a system of coupled rate equations governing the temporal evolution of total, carbon, and nitrogen yields for 22 unique lumped-product fractions. The model captures many empirical trends over a broad range of reaction conditions and feedstock biochemical profiles. In particular, slurry concentration and Maillard reaction effects are quantified for the first time. Agreement between calculated and observed quantities was particularly high for the biocrude and ammonia fractions, substantiating the utility of the model for optimizing important HTL process metrics that will ultimately enhance overall process sustainability and energy return on investment.