Completed Research
Techno-Economic Analysis of the Carbon Dioxide Capture Process in Pre-Combustion Applications
Husain E. Ashkanani, PhD, 2021
(Thesis: University of Pittsburgh ETD)
Aspen Plus v.8.8 was used to perform techno-economic analysis (TEA) of the CO2 capture process in pre-combustion applications. The capital expenditure (CAPEX), operating expenditure (OPEX), and levelized cost of CO2 captured (LCOC) were calculated to assess the feasibility of the process. 35 physical solvents (11 ionic liquids, 5 hydrocarbons, 7 oxygenated-hydrocarbons, 2 nitrogenized-hydrocarbons, 5 cyclic-hydrocarbons, 3 polymers, and 2 subcooled) were used for CO2 capture from actual shifted fuel gas at 54.1 bar in 543 MW power plant. The CO2 capture process included a countercurrent packed-bed absorber containing random or structured packing, followed by three flash drums for solvent regeneration, and multi-stage compressors for CO2 sequestration. The key process constraints were 90% CO2 capture from the shifted gas and less than 0.5 mol% fuel gas and maximum 600 ppm water in the CO2 sequestration stream. The PC-SAFT Equation-of-State was used to model the solvent density and VLE of the gas-liquid systems used, while other physico-chemical properties were acquired from experimental data, literature, and Aspen Plus database. Aspen Plus calculated results indicated that (1) structured packing, Mellapak 250Y, of large specific surface area improved gas-liquid mass transfer, which lowered LCOC; (2) operating at low temperatures increased CO2 solubility and decreased solvent loss, which lowered LCOC; (3) volatile solvents exhibited significant solvent loss; (4) ionic liquids with negligible vapor pressure have high viscosity and would be suitable for warm/hot temperature CO2 capture; (5) the CAPEX and OPEX decreased with decreasing plant power capacity, however, LCOC increased due to the small mass of CO2 captured; and (6) among the 35 solvents used, diethyl sebacate provided the lowest LCOC at $7.14 per ton CO2 captured. An artificial neural network (ANN) was developed using the Aspen Plus calculated CAPEX, OPEX, and LCOC of the CO2 capture process. 320 randomly selected cases were used for training and 481 cases were used for testing the ANN. The input to the ANN included plant power capacity, operating temperature, solvent properties, and packing specific surface area. The ANN was able to predict the calculated CAPEX, OPEX, and LCOC with high coefficient of determination (R2) of 0.9961, 0.9994, 0.9995, respectively.