Power generation with CO 2 capture: Technology for CO 2 purification

Abstract

The goal of this paper is to find methodologies for removing a selection of impurities (H 2O, O 2, Ar, N 2, SO x and NO x ) from CO 2 present in the flue gas of two oxy-combustion power plants fired with either natural gas (467 MW) or pulverized fuel (596 MW). The resulting purified stream, containing mainly CO 2, is assumed to be stored in an aquifer or utilized for enhanced oil recovery (EOR) purposes. Focus has been given to power cycle efficiency i.e.: work and heat requirements for the purification process, CO 2 purity and recovery factor (kg of CO 2 that is sent to storage per kg of CO 2 in the flue gas). Two different methodologies (here called Case I and Case II) for flue gas purification have been developed, both based on phase separation using simple flash units (Case I) or a distillation column (Case II). In both cases purified flue gas is liquefied and its pressure brought to 110 atm prior to storage. Case I: A simple flue gas separation takes place by means of two flash units integrated in the CO 2 compression process. Heat in the process is removed by evaporating the purified liquid CO 2 streams coming out from both flashes. Case I shows a good performance when dealing with flue gases with low concentration of impurities. CO 2 fraction after purification is over 96% with a CO 2 recovery factor of 96.2% for the NG-fired flue gas and 88.1% for the PF-fired flue gas. Impurities removal together with flue gas compression and liquefaction reduces power plant output of 4.8% for the NG-fired flue gas and 11.6% for the PF-fired flue gas. The total amount of work requirement per kg stored CO 2 is 453 kJ for the NG-fired flue gas and 586 kJ for the PF-fired flue gas. Case II: Impurities are removed from the flue gas in a distillation column. Two refrigeration loops (ethane and propane) have been used in order to partially liquefy the flue gas and for heat removal from a partial condenser. Case II can remove higher amounts of impurities than Case I. CO 2 purity prior to storage is over 99%; CO 2 recovery factor is somewhat lower than in Case I: 95.4% for the NG-fired flue gas and 86.9% for the PF-fired flue gas, reduction in the power plant output is similar to Case I. Due to the lower CO 2 recovery factor the total amount of work per kg stored CO 2 is somewhat higher for Case II: 457 kJ for the NG-fired flue gas and 603 kJ for the PF-fired flue gas.