Rate-based modeling of CO₂ absorption in amine solutions, evaluation of mass transfer, kinetics and scale-up

Abstract

Post-combustion CO2 capture (PCC) with chemical solvents is a near-term option for commercial deployment of CO2 capture from power plants. However, the technology is energy intensive and costly, and there is no experience with the deployment of CO2 capture at large power plant scale. Therefore, proper knowledge of CO2 capture mechanisms is required to solve the technological challenges. Also, due to lack of experience with large-scale CO2 capture, the reasonable way for evaluation of the process is simulation and precise modeling. In the design of an absorber, the largest capital unit, important parameters such as mass and heat transfer, wetting and active area, kinetics for the solvent chosen and effects of operational conditions need to be properly understood before moving from small to large-scale. Due to lack of proper understanding of the complex transfer phenomena occurring in the absorber, modeling the absorber is based on empirical and semi-empirical correlations with parameters obtained from fitting to experimental data. Therefore, using these correlations in largescale column applications requires good confidence in the parameters which affect the performance of the system.

In this thesis, several correlations for prediction of the hydrodynamics and mass transfer through structured packed columns were compared and the studies showed that there is a large uncertainty associated with applying these correlations to large-scale packed columns. None of the proposed correlations in the literature are verified for large-scale CO2 capture processes. The uncertainty can be explained by three main factors: (i) the models which the correlations are based on may contain many assumptions that are not generally valid; (ii) the data basis for fitting parameters and for validation of the models may not be extensive enough, or of sufficient quality, and (iii) the complexity of the models may need a numerical approach for solving equations where accuracy may be an issue.

A series of rate-based model simulations (RateSep models) were carried out in Aspen plus in order to perform a sensitivity analysis and compare the effect of different design correlations (e.g. mass transfer and interfacial area), physiochemical properties and kinetic models on the performance of an absorber in a large-scale CO2 capture process with a chemical solvent. The RateSep models were validated using pilot data, the important design parameters such as pressure drop, temperature profiles and CO2 removal were compared to the pilot data. The results showed that the accuracy of the Aspen RateSep model depends heavily on the reaction kinetic model, selected mass transfer and interfacial area correlations, and more confidence is required for applying the available kinetics and mass transfer models in rigorous rate based modeling of industrial absorber units. Model validation using pilot plant or commercial data for the absorption system chosen is highly recommended.

Multi-scale simulations of the absorber were also investigated using Aspen RateSep in order to perform cost and sensitivity analysis for CO2 capture process from a coal-fired and a gas-fired power plant both with a net power output of 400MWe . Possible redesign of the absorber was studied and the results show that large electrical energy savings in the feed gas blower can be obtained when a design is chosen giving reduced absorber pressure drop. This can be achieved if the column cross sectional area is increased. This is offset by extra capital cost by a need for slightly more volume of packing. The operational design flexibility in a CO2 capture plant for a gas-fired power plant is higher than for a coal-fired power plant.