One prevailing issue for assessing the performance of hydrothermal liquefaction is understanding the role of the extraction solvent used for product separation. This study evaluated the extraction agent's impact on the hydrothermal liquefaction products and energy efficiency. Three representative solvents (acetone, dichloromethane, and toluene) were chosen with three representative high-carbohydrate, protein, and ash content feedstocks (Chlorella sp., Nannochloropsis sp., and Enteromorpha pr., respectively). Extraction of the oil using dichloromethane led to the highest biocrude oil yield (dry biomass) for Chlorella sp. (48.8%), toluene for Nannochloropsis sp. (23.3%), and acetone for Enteromorpha pr. (9.8%). The solvent selection led to a maximum variation of 20.4% for all oil yields. Dichloromethane produced high energy recovery values (maximum: 67.1%) and low energy consumption ratios (minimum: 0.06) regardless of the feedstock chemical composition. Dichloromethane also led to consistently high net energy values and high fossil energy ratios amongst all feedstocks. We speculate that the solvent polarity, chemical structure, hydrogen bonding, and dipole-dipole interactions influenced output parameters by the selective isolation and extraction of the chemical compounds in the biocrude oil. This study suggested that the extraction solvent selection should be carefully considered and normalized for the reporting of hydrothermal liquefaction yields and energy efficiency values. (C) 2018 Elsevier Ltd. All rights reserved.