A TWO-PARAMETER MODEL FOR THE DETERMINATION OF THE PARTITION COEFFICIENT FOR TOTAL REDUCE COMPOUNDS IN A CULTURE MEDIUM

A two-parameter model was applied to describe the activity coefficient for a culture medium, and, compared with Henry’s law for water, Henry's law fitted to experimental data, the extended UNIQUAC model, and experimental data obtained from an experimental setup system, consisting of a liquid culture media Thiobacillus (ATCC 290) with hydrogen sulfide (H2S), dimethyl sulfide (DMS), me-thyl mercaptan (MM), and dimethyl disulfide (DMDS), separately. The R² and R²_adj coefficients obtained from the analysis show that Henry's Law for water gives lower or null determinations than Henry’s Law adjusted. Adjusted Henry’s Law shows lower determination co-efficients than the two-parameter model (with differ-ences in R²_adj between 0.69% and 3.64%), except for DMS (R²_adj = 95.88% for adjusted Henry’s Law and R²_adj = 94.00% for two-parameter model). The ex-tended UNIQUAC model presents lower determina-tions than the other models for all compounds except H2S. On the other hand, this species has the worst comparative fits in all the models reviewed, even when Henry's law was adjusted, which may be asso-ciated with its effective solubility in the complex cul-ture medium where the study was carried out. The ANOVA test for model discrimination shows that, for H₂S, MM, and DMDS, the two-parameter model significantly represents better the liquid-vapor distribution for such components, precisely when the gas concentration was upper 300 [ppm]. In complex culture media very different from pure water, adjusted models to describe the liquid-vapor equilibrium are necessary due to the high deviations compared to pure water. In the case of the culture me-dium, using Henry's Law is not sufficient to describe the equilibrium for H₂S, MM, and DMDS.