New reactor system for supercritical water oxidation and its application on phenol destruction

Abstract

A new reactor system for supercritical water oxidation which can treat reaction variables independently was developed. With this system, phenol oxidation experiments were carried out at temperatures 380–440°C and pressures 190–270 atm. Reaction time was varied from 12 to 120 s, and corresponding conversion was 11–99%. The initial phenol concentration was below 8.8 mM based on reaction volume. The initial oxygen concentration ranged from 100 to 1750% of the stoichiometrically required amount for complete oxidation of phenol. According to the results of kinetic experiments, oxidation rate was dependent on temperature and concentration of water, oxygen and phenol. However, the oxidation rate was not dependent on the pressure itself. As the ratio of oxygen concentration to phenol concentration increases, the reaction rate increases asymptotically. In the case where oxygen concentration was greater than 300%, the oxygen content did not affect the reaction rate any more. Water has a significant effect on the reaction rate, and oxidation rate increased as water concentration increased. Various reaction mechanisms in which water can affect the reaction rate were considered, and this consideration suggested that the water participates in the reaction as a reactant to generate active radicals. The global reaction order was 1.38 ± 0.24 for water and 1 for phenol. The effect of oxygen on reaction rate was expressed as a Langmuir-type equation with the ratio of oxygen to phenol concentration as a variable, whose constant was 2.89 ± 1.43. The corresponding activation energy was 23.8 ± 2.2 kcal/mol, and this was larger than that of a gas-phase reaction, but smaller than that of a liquid-phase reaction. A previously suggested reaction model and equation were used to predict phenol conversion. The multi-step reaction model suggested by Tufano cannot predict the conversion well, but it seems to explain the effects of each parameter qualitatively. The equation proposed by Gopalan and Savage predicts conversion fairly well when the concentration of oxygen is adjusted to 300% of the stoichiometrically required amount.