Measurement and Modeling of Phase Equilibria in Systems Containing Water, Xylose, Furfural, and Acetic Acid

Abstract

Wood hydrolysates obtained in biotechnological processes are typically aqueous solutions that contain, among others, sugars, acetic acid, and furfural. Only little is known on the influence of the sugars on the phase equilibria in those mixtures. Therefore, liquid–liquid equilibria (LLE), solid–liquid equilibria (SLE), and solid–liquid–liquid equilibria (SLLE) in the system (water (W) + xylose (X) + furfural (F)) were studied in the present work at 298.15 and 333.15 K. Additionally, the LLE in the system (W + X + F + acetic acid (AA)) was studied at 298.15 K. The results show that, up to the solubility limit of xylose, adding xylose to mixtures of (W + F) hardly influences the width of the miscibility gap, and that there is practically no xylose in the furfural-rich phase. However, the miscibility gap in the ternary system (W + F + AA) is slightly widened by the addition of xylose. The experimental data on the phase equilibria from the present work were described using the nonrandom two-liquid model. The model describes the experimental data well.

Phase equilibria modeling of biorefinery-related systems: a systematic review

“…This small difference can be attributed to the fact that being a 3-parameter model, the NRTL model adapted itself better to the experimental data. Galeotti et al (2019) also pointed out this importance of adjusting data, showing how NRTL model was sufficient for some regions, but as soon as there was a shortage of data due to the imposition of some difficulty, such as increased temperatures, concentrated systems or the extrapolation for three or more components, it started to depart from the experience. Zarei et al (2019), in turn, showed that Flory-Huggins method was worse than the two GE models mentioned above for ionic liquids and sugars, with an average RSMD of 3.13 against 1.64 for NRTL and 1.79 to UNIQUAC.…”