Volatile organic solvents derived from fossil resources are typically used in extraction processes, but this usually involves high energy consumption for solvent recovery and negative environmental impacts due to solvents’ hazardous, volatile, and flammable nature. This study presents a systematic approach to solvent screening, using molecular and process simulation techniques, data analysis and classification methods applying techno-economic, environmental and safety criteria. This methodology is demonstrated for lipid extraction from wet algae biomass in biofuel production. First, relevant thermodynamic equilibrium data are predicted with the COSMO-RS (Conductor-like Screening Model for Real Solvents) method. The resulting solvents are clustered according to their partition coefficient and selectivity towards the target solute, and screened further, considering their physicochemical properties and health, safety and environmental (HSE) performance. Finally, the lipid extraction process is simulated in Aspen Plus using all screened solvents to obtain technical, economic, and environmental performance data. Out of 88 initial candidates, cyclohexane, limonene, and ethyl tert-butyl ether are identified as potential alternatives to the benchmark solvent, hexane. Whilst these solvents tend to be more expensive and their recovery more energy intensive (higher boiling points) compared to hexane, they have higher selectivity towards lipids thus reducing the solvent intensity of the process, and are less volatile and non-hazardous according to the HSE classification. This methodology can be applied to other extraction process applications or implemented at early stages in process design to evaluate techno-economic, environmental and safety trade-offs when considering and selecting more sustainable alternatives to fossil-derived solvents.
Keywords: COSMO-RS, process simulation, green solvents, biodiesel, sustainability indicators.