Industrial & Engineering Chemistry Research, Vol.47, No.4, 1310-1317, 2008
Dynamic measurement of carbon dioxide volumetric mass transfer coefficient in a well-mixed reactor using a pH probe: Analysis of the salt and supersaturation effects
This paper presents a new method for the evaluation of volumetric mass transfer coefficients (k(L)a) using a pH probe response recorded during the absorption of carbon dioxide in a well-mixed reactor. For this method of evaluation, it is not necessary to know the reaction equilibrium constant, the experiment start time, or the initial and final steady-state pH probe readings. The experimental procedure is based on Hill's recent publication (Ind. Eng. Chem. Res. 2006, 45, 5796). Hill reported a decrease in k(L)a following the addition of salt during carbon dioxide absorption in a well-mixed reactor and explained it as the result of ionic charge effects reducing the ability of carbon dioxide molecules to diffuse away from the surface. Generally, oxygen absorption from air is preferred because the solubility of oxygen is low and, thus, so will be the gas-phase depletion. In such experiments, the mass transfer coefficients are less affected by gas-phase residence time distribution and by bubble-size distribution: The solubility of carbon dioxide is 26 times higher than that of oxygen, which can lead to significant gas concentration changes. Thus, the k(L)a values, measured using absorption of diluted carbon dioxide, are more likely to be distorted by driving force errors caused by the use of an inappropriate, gas-mixing model. Using a stirred cell, the interfacial area of which is known, the mass transfer coefficients of oxygen and carbon dioxide in water and in a salt solution were compared. The mass transfer coefficients obtained for oxygen agreed with, or were slightly superior to, the coefficients obtained for carbon dioxide, which corresponds with the lower molecular diffusivity Of CO2. The mass transfer coefficients of carbon dioxide obtained in salt solution were not significantly lower than those obtained in pure water, which is in strong agreement with the literature but contradicts the results reported by Hill. It is shown that Hill's findings may be the result of his use of both an inaccurate gas-mixing model (no depletion of gas) and an imprecise reaction term in his equations.