Solvent liquefaction (SL) is a promising technology for converting biomass to renewable fuels and chemicals. However, many technical barriers currently prevent the technology from commercial advancement. This research focused on the technical challenges that were confronted during the development of a continuous SL process development unit (PDU).

A 1 kg hr-1 pilot plant was designed to evaluate the technical feasibility of a continuous hydrocarbon-based SL process. The process was demonstrated to convert Loblolly pine to liquid products at a yield of 51.2 wt%. The liquid products were low in oxygen content (23.2 wt%) and moisture (13.4 wt%). In addition to validating product yields and quality, several unit operations were also evaluated. Online solids filtration continuously separated over 99% of the solid residue from the product stream. Acetone, injected into the process to aid in solids filtration, was continuously recovered with only 3 wt% loss to the process. Bio-oil fractionation via a distillation column was also demonstrated. A medium oil cut, suitable for use as a recycle solvent, was recovered using this system. This cut accounted for approximately 93 wt% of the initial solvent.

The effect of moisture on SL of biomass in a hydrocarbon solvent (tetralin) was evaluated to help determine the extent to which the feedstock should be dried for the SL PDU. These experiments were conducted in a quasi-batch reactor with independent pressure control and external vapor recovery. It was found that increasing the feedstock moisture from 1 to 50 wt% resulted in a reduction in liquid yield of 25, 21, and 35 wt%, for pine, cellulose, and lignin, respectively. Analysis of the solid residue, which increased proportionally to the decrease in liquid yield, indicated acid-catalyzed polymerization of liquid products was the likely mechanism for solids-formation. The measured reduction of monomeric content in the liquid products supported this hypothesis. This behavior was attributed to the ionic dissociation of water at the reaction conditions, which resulted in increased acidic behavior. Although water was less than 20 wt% of the solvent loading in these experiments, it strongly influenced SL of biomass.

A response surface methodology statistical model was constructed to evaluate the influence of process conditions on the SL of lignin in a phenolic solvent. The goal of this work was to develop a process capable of producing high liquid yields from lignin with a high selectivity for phenolic monomers suitable for use as a recycle solvent. The effects of hydrogen donor solvent blend ratio, reaction temperature, solids loading, and residence time were studied. Hydrogen donor solvent and reaction temperature were found to be the most significant factors. Maximum liquid yield (58.6 wt%) was achieved at temperatures as low as 260 à °C and hydrogen donor solvent blend ratios of less than 30 wt%. As much as 40 wt% of the liquid products were volatile below 340 à °C, which indicated a significant production of distillable phenolic monomers suitable for use as a recycle solvent. Thus, the use of a phenolic solvent at moderate reaction temperatures was demonstrated to be suitable for the design basis of a continuous SL process.

The thermal stability of fractionated fast pyrolysis bio-oil was evaluated to determine the effect of rapid heating on bio-oil quality. These goal of this work was to develop a method with which to simulate rapid heating similar to the heating required for conventional refining processes, and to investigate the factors that most impacted thermal stability. Bio-oil fractions were produced from fast pyrolysis of Loblolly pine at 500 à °C. Each fraction was heated to 100, 200, and 300 à °C in under 120s. Bio-oil acidity, as measured by total acid number (TAN), was found to be the most significant indicator of thermal instability. Samples with higher TAN values exhibited increased tendency to undergo polymerization reactions upon rapid heating. These findings were later extended to determine the thermal stability of bio-oil produced from the SL PDU. The goal of this was to evaluate the impact of thermal fractionation on bio-oil quality.

In order to design and build the SL PDU, alternative methods were developed to determine selected thermophysical properties of bio-oil. The goal of this work was to utilize readily available laboratory equipment, instead of costly ASTM test methods and specialty apparatus. Specific heat, thermal conductivity, viscosity, surface tension, and enthalpy of vaporization were measured for six unique stage fractions and a mixture of these fractions from the fast pyrolysis of Loblolly pine. The results and methods obtained from this work were then used as the design basis for the SL PDU.