Effects of biomass composition on hydrogen yield and char formation during supercritical water gasification: A thermodynamic analysis

Biomass is composed of cellulose, hemicellulose and lignin, whose relative proportions influence H 2 production and char formation during supercritical water gasification (SCWG). However, the combined effects of biomass composition and operating conditions have not been systematically quantified over the complete lignocellulosic composition space. In this study, a two-phase thermodynamic model based on Gibbs free energy minimization was developed to evaluate SCWG over the complete lignocellulosic composition space. Temperature, pressure, and biomass concentration were varied from 500 to 900 ℃, 25 to 50 MPa, and 1 to 50 wt%, respectively. To define the boundary for significant char formation, a critical biomass concentration ( w bio,crit ) was introduced. The results showed that w bio,crit ranged from 33 to 49 wt%, and increased markedly with temperature but decreased slightly with pressure. Lignin-rich compositions showed the strongest thermodynamic tendency toward char formation. In contrast, the condition-averaged H 2 yield varied only weakly over the ternary space, exhibiting a highly condition-dependent response. Sobol-ANOVA showed that operating conditions, biomass composition, and their interaction accounted for 34.68 %, 6.10 %, and 59.22 % of the variance in char formation, respectively. For H 2 yield, the corresponding contributions were 97.61 %, 0.40 %, and 1.99 %. Among the operating variables, biomass concentration had the largest total effect on both char formation and H 2 yield, with total-effect indices of 0.89 and 0.70, respectively, followed by temperature at 0.58 and 0.37. Pressure showed only minor effects, with indices of 0.05 and 0.02. Application to corn straw, pine sawdust, and walnut shell further showed that increasing lignin content narrowed the char-free operating window. These results quantify the distinct roles of biomass composition and operating conditions and provide a thermodynamic basis for feedstock screening and operating-window identification in biomass SCWG.