The recovery and recycling of a deep eutectic solvent after biomass fractionation with cellulosic membranes

Abstract

Lignocellulosic biomass is a renewable alternative to fossil resources, providing cellulose, hemicellulose, and lignin as a source for fuels, chemicals, and materials. However, its complex interlinked structure complicates efficient valorization. Deep eutectic solvents (DESs) have emerged as promising green fractionation agents due to their tunability, low toxicity, and ability to solubilize lignin. Among them, choline chloride–lactic acid (CC-LA) DES has shown high selectivity to lignin, though its high acidity and viscosity are challenging. While DES purification and recovery strategies – such as antisolvent precipitation – have been studied, efficient and scalable methods remain underexplored. Additionally, the lignin precipitated from spent DES by antisolvent addition may exhibit a wide molecular weight distribution depending on process conditions. This heterogeneity can complicate lignin valorization and utilization, particularly where consistent molecular features are required. Therefore, optimizing both DES composition and recovery methods is key to maximizing the value and applicability of lignin.

To expand the understanding of strategies for the purification and recycling of spent DESs, this study focused on the use of cellulosic membranes for the purification of CCLA DES. Cellulose membranes were selected due to their expected chemical resistance to the acidic CC-LA DES system. One of the anticipated benefits of membrane-based treatment is splitting spent DES into concentrate and purified fractions. If lignin is to be recovered from the concentrate fraction, that requires less antisolvent compared with the direct addition of antisolvent to spent DES.

The filtration of the viscous CC–LA DES through the RC70PP membrane was possible when assisted by ethanol dilution (60% spent DES in ethanol) and mild heating (45 °C). Stability tests revealed that the RC70PP membrane, along with other tested cellulosebased membranes (Ultracel 5 kDa and a laboratory-cast cotton membrane), maintained their filtration performance – assessed via pure water flux and model compound retention – for up to four weeks. However, changes in the cellulose functional groups were observed after just two weeks of DES exposure, suggesting the limited long-term suitability of cellulose membranes for industrial DES purification. Membrane-based DES recycling without antisolvent addition led to a gradual decline in delignification performance with each reuse cycle. In contrast, the traditional antisolvent-based method provided better DES purification and higher delignification efficiency, though it required significantly more solvent. The membrane-based process reduced antisolvent consumption by a factor of 1.9 compared with the antisolvent approach. High-purity lignin (>87%) was produced with both the antisolvent-based and membrane-based methods.

While cellulosic membranes could be used for DES purification at laboratory scale, the further development of chemically robust membrane materials and their combinations with addition of antisolvents is needed. Nevertheless, this study contributes valuable insights toward designing greener, more resource-efficient separation processes for future lignocellulosic biorefineries.