Abstract

Valorizing biomass by-products and reducing their accumulation are critical current issues. Nevertheless, the large volume of such materials requires continuous and enhanced strategies to mitigate potential environmental problems. The valorization of biomass through the recovery of valuable compounds must be sustainable and optimized, emphasizing the selection of optimal and greener solvents for this purpose. In this context, to reduce experimental efforts, material resources, and energy expenses, predictive in silico tools enable efficient preliminary screening, highlighting the tools based on Hansen Solubility Parameters (HSP) and COnductor-like Screening MOdel for Real Solvents (COSMO-RS). Considering their widespread application, this review critically analyzes both computational methods, highlighting their main concepts, applications, and limitations. Finally, we propose a simplified guide from solvent selection to the purified (or not) fraction. © 2025 Elsevier B.V.