Figure 1. (a) Cellulose retention, (b) xylan removal, and (c) delignification of WT and TG sorghum stover before and after binary & ternary DES pretreatments with different compositions. Pretreatment conditions: 140 ℃, 3 h.

The primary goal of biomass pre-processing, including pretreatment and fractionation, is reducing and removing the recalcitrance factor from biomass. Hemicellulose and lignin physically hinder the enzyme accessibility to cellulose and decrease the enzyme efficiency by productive and non-productive bindings. Therefore, the removals of hemicellulose and lignin are important parameters to measure processing effectiveness. The numerical data for the cellulose retention, xylan removal, and delignification (Figure 1 ) are tabulated inTable S1 in the Supplementary Materials. Cellulose retention was over 90% after DES pretreatments with both WT and TG sorghum stover (Figure 1a ). There was no significant difference in cellulose retention between WT and TG as well as among three DES pretreatments. AsFigure 1b shows, xylan removals from WT and TG sorghum stover showed similar trends. Binary DES (ChCl-DHBA) and ChCl-DHBA-water (ChCl-DHBA-W) had approximately 20% higher than the ternary DESs with EG (ChCl-DHBA-EG). Also, binary DES and ChCl-DHBA-W resulted in similar xylan removal with WT sorghum stover (75% and 71%, respectively) as well as with TG biomass (77% and 78%, respectively). However, ChCl-DHBA-EG was not as efficient as the former two DESs in terms of xylan removal (54-56%). Li et al. reported that the content of EG in a ternary DES showed contradictory effects on xylan removal.³⁶ EG content in ChCl-oxalic acid-EG DES showed positive effects on xylan removal when the molar ratio between EG and ChCl was below 3:1, while xylan removal was reduced with increasing EG content when the molar ratio of EG and ChCl was above 3:1.

The third constituent in the ChCl-DHBA DES influenced more significantly the delignification of biomass. The additional solvents as the third constituent enhanced the delignification. Binary ChCl-DHBA DES showed 22% of delignification with WT biomass while resulting in only 9% with TG biomass. Water-incorporated ternary DES (ChCl-DHBA-W) increased delignification of WT biomass to 28%, and the one with TG biomass was improved to 34%. In the previous study, ethanol-incorporated ChCl-polyol DES systems showed higher solubility of aromatics compared to the binary DES.³⁷ Since ethylene glycol leads to higher lignin solubility than water, higher delignification with ChCl-DHBA-EG was expected. Interestingly, delignification yields of WT and TG sorghum stover by ethylene glycol-incorporated ternary DES (ChCl-DHBA-EG) were 38% and 34%, respectively, which were higher than binary DES. However, compared to the delignification with ChCl-DHBA-W, the yield of ChCl-DHBA-EG was not notably improved, especially with TG biomass. It is possible that other factors like the viscosity of the DES could be more significant than lignin solubility in this ternary DES system; therefore, the enhancement of pretreatment effects could be explained by the improved mass and heat transfer during the pretreatment process. Nonetheless, the third constituent clearly altered the fractionation performance of the DES, as evidenced by the changed delignification and xylan removal with the tested three DESs.