Techno-economic comparison of combined cycle gas turbines with advanced membrane configuration and MEA solvent at part load conditions

Abstract

This work compares the part load techno-economic performance of CO2 capture from a CCGT using a membrane configuration with selective CO2 recycle and using MEA solvent, under the assumption of flexible power plant dispatch. This is the first time that the techno-economic performance of CO2 capture technologies is compared assuming a flexible dispatch profile, and the assessment was done using a comprehensive, new, part load assessment approach. Analysing the part load performance of CCS technologies is relevant because of significant changes in our power systems, dramatically reducing the utilisation of thermal power plants. The technical performance of the configurations with and without CCS was simulated at steady state, at operating points between maximum continuous rating (100% gas turbine loading), and minimum stable load (35% gas turbine loading). The performance at these operating points was then aggregated into weighted averages to produce single performance indicators (specific CO2 intensity, specific primary energy per tonne of CO2 avoided (SPECCA), and levelised cost of electricity (LCOE)) over the dispatch profile of the power plant. The technical performance of the MEA configuration was favourable over the membrane configuration over the whole CCGT loading range. The MEA SPECCA increased from 3.02 GJ/t CO2 at 100% GT loading, to 3.65 GJ/t CO2 at 35% GT loading; the membrane SPECCA increased from 3.35 to 4.20 GJ/t CO2. The higher SPECCA of the membrane configuration is caused by the reduced gas turbine efficiency, due to the selective recycling of CO2 to the GT. When equal GT efficiency was assumed for combustion with normal air and with CO2 enriched air, the membranes' technical performance was comparable with that of MEA. The capital costs of the CCGT with membrane configuration were 35% higher than the CCGT with MEA configuration. That, and the 6-year replacement frequency of the membranes led the membrane LCOE to be 10 €/MWh higher than the MEA LCOE, when calculated with the part load approach. The membrane LCOE was 8 €/MWh higher when full load was assumed. The new part load approach proved instrumental in highlighting performance (differences) at flexible dispatch conditions, and aggregating those into easy to understand performance indicators.