Extension of a multi-criterion performance indicator model for post combustion Co2 capture using amine solvents.

Masters Degree, University of KwaZulu-Natal, Durban.Energy generation by carbonaceous fuel combustion has been identified as one of the predominant sources of CO2 emissions. Many scientists and researchers believe that rising CO2 levels have an adverse effect on the environment, therefore research on the capture and storage of CO2 is ongoing. Post-combustion capture with amine-scrubbing has been identified as a practical short-term solution to the problem. The alkanolamine, monoethanolamine (MEA) is the current solvent of choice for this application. However, due to disadvantages connected to its use, there is a need to identify alternative superior solvents or solvent blends. A quick and inexpensive method to identify alternative solvents is via process simulation and modelling. These tools enable the assessment of solvent viability on a large scale and the elimination of unsuitable candidates without the expense of extensive laboratory testing. The main units considered in a post-combustion CO2 capture simulation are the absorber, where the amine solvent is used to remove CO2 from a flue gas stream, and the stripper, which enables the separation of the CO2 from the solvent to facilitate recycle of the solvent for re-use in the absorber. User inputs into these simulations include the flow rate and composition of the flue gas to be treated, the solvent composition, and the desired CO2 capture rate. A multi-criterion performance model for the evaluation of solvents used for CO2 capture from a coal-fired power plant, was previously developed by Daya (2017) within the Thermodynamics Research Unit at the University of KwaZulu-Natal. The inputs to this performance indicator model are primarily solvent flow rates and equipment heat duties, which were obtained from ASPEN Plus® simulations. Among the other inputs required are price data for the various factors considered in the model, which include energy requirements, make-up flows and carbon taxes. The solvents investigated to test the performance model’s viability consisted of primary, secondary, tertiary and sterically-hindered alkanolamine solvents and their blends. MEA was used as the basis of comparison. In this study, the performance indicator model is used to evaluate the performance of the previously studied amines, n-methyldiethanolamine (MDEA) and 2-amino-2-methyl-1-propanol (AMP), in different blends as along with an additional component, piperazine (PZ). Different concentrations of the binary blends MDEA+PZ and AMP+PZ as well as the ternary blend, MDEA+AMP+PZ, were investigated. The solvent selected as the basis for the ratings was also changed from 30 wt.% MEA to 30 wt.% AMP, as AMP was previously proven to outperform MEA. The rating for the benchmark case calculated by the performance model formulae, is one. When the same calculations are applied to the other amine blends investigated, ratings below one show a performance inferior ABSTRACT v than the benchmark, whilst a rating above one show better performance compared to the benchmark. Of the blends studied, the solvent with composition 25 wt.% AMP + 5 wt.% PZ + 70 wt.% H2O was the best performing with an overall performance increase of approximately 35% (which corresponds to a rating of 1.359). This solvent was further studied using alternative process configurations: the absorber intercooling (ICA) and rich solvent splitting (RSS) configurations. These configurations have been reported to noticeably reduce the energy requirements for solvent regeneration, with minimum additional equipment. A rating of 1.483 was obtained for the ICA configuration, which is a 9% improvement on the rating of the conventional configuration with the same solvent. The results for the RSS configuration, however, shows no improvement on the performance of the conventional configuration.Only available in English