Assessment of mass transfer correlations in rate-based modeling of a large-scale CO sub(2) capture with MEA

A Rate-based simulation model was utilized for assessing the effect of correlations for mass transfer coefficients (k sub(g), k sub(1)) and effective interfacial area (a sub(e)) on the performance of absorber for large-scale CO sub(2) capture from a gas-fired power plant with a chemical solvent. The mass transfer and interfacial area correlations available in the literature are applied in the RateSep model to demonstrate the uncertainty associated with using these correlations in a large-scale CO sub(2) capture plant. Selecting the best correlations for such a CO sub(2) capture process is outside the scope of this work due to a lack of large-scale data for model validation. The absorption performance is expressed in terms of pressure drop per height of packing, gas and liquid phase temperature and concentration profiles, enhancement factor and necessary packing height. Variations in individual mass transfer coefficients, effective interfacial area, and overall mass transfer coefficient are also shown. The study showed that there is substantial uncertainty associated with applying the proposed correlations for determining the required absorber height. The variation in results, given as packing height needed for 90% CO sub(2) removal, was very high, ranging from 6 to 24 m. These are clearly too large discrepancies to be acceptable. The reason can, at least partly, be explained by the relationship between transport parameters, kinetics, mass transfer coefficients and effective interfacial area. Most packing mass transfer correlations available in the literature were developed and tested for specific packing types, specific chemical systems and mainly for distillation where the resistance lies mostly in the vapor phase. The best way to judge the correctness of these correlations for application in a large-scale CO sub(2) capture processes with chemical solvents is to see how well the correlations predict reality, and validating the correlation's performance with pilot, or full scale data from CO sub(2) capture with the same chemical solvent.