Economic evaluation of CO2 capture from flue gas by membrane permeation.

Igor Wiesberg(1), José L. de Medeiros(1), Ofélia de Queiroz Fernandes Araújo(1), Rita Maria de Brito Alves(2), Paulo Coutinho(3)

(1) Department of Chemical Engineering, Federal University of Rio de Janeiro, RJ, Brazil (2)University of São Paulo, Polytechnic School â?? Department of Chemical Engineering, São Paulo, Brazil (3)BRASKEM S.A., São Paulo Brazil

ofélia@eq.ufrj.br, rmbalves@usp.br
Carbon dioxide is recognized as an important greenhouse gas (GHG). Its concentration in atmosphere has increased since industrial revolution, mainly due energy generation from fossil fuels. Thereby, environment issues has driven a global research for emissions reduction technologies to stabilize the CO2 concentration in the atmosphere. One important way to approach this technological challenge is the Carbon Capture, Utilization, and Storage (CCUS) which is a group of technologies to capture CO2 from industrial sources, e.g., flue gas, and then utilizing physically (e.g., EOR) or chemically (e.g., as feedstock to the chemical industry), or safely storing in geological reservoirs. The capture is the most expensive step of the CCUS supply chain, holding an important share in the costs of carbon sequestration. In a thermal power plant, it can be separated in a post-combustion phase, preferred due to fewer changes in the process, or in a pre-combustion process, where the fuel is converted into syngas and then CO2 is separated from H2, which, upon combustion, yields water. The most common post- combustion capture process is the absorption with amines, like the monoethanolamine (MEA), the only commercially proven post-combustion technology. Nevertheless, its pitfalls resides in the high costs with the reboiler for solvent recovery. Alternatively, membranes permselective to CO2 appear in this scenario as a potential technology to reduce capture costs, since it does not require energy for phase change. Usually, the flue gas feed is compressed to a high pressure prior to the separation in the membranes to assure significant transmembrane difference of partial pressures, which acts as the separation driving force, so that the permeate stream can be at atmospheric pressure. Clearly, such technology replaces thermal energy by mechanical energy and process configurations should explore alternatives to reduce compression efforts. A recent configuration utilizes permeation to vacuum with the feed at low pressure, potentially reducing compression costs at the expense of increasing membrane area.
This work analyses the economic feasibility of carbon dioxide capture with membranes, compared to capture by chemical absorption with MEA as reference, from flue gases. The main objective of this study is to identify membrane conditions, as selectivity and separation factor, which would result in the same costs (CAPEX plus OPEX) of chemical capture with MEA. In addition, it also determines the best of two membrane configurations: (i) CASE 1 - feed at high pressure and permeate at atmospheric pressure; (ii) CASE 2 - feed at low pressure and permeate at sub-atmospheric pressure. Process simulations were performed using a user developed module in ASPEN HYSYS for membranes, with a flue gas stream of 498330 kg/h containing 12.4% of CO2 (molar) and a defined target for output streams, of 1.4bar in rich CO2
stream and 0.03% of CO2 in the retentate. Membrane area is defined in order to achieve these performance specifications. Simulation results supported equipment sizing for subsequent economic analyses.
Based on a review of the state of the art, silica/silica-zirconia/alpha-alumina was chosen as the reference material for the membrane, since it has the highest permeance and CO2/N2 selectivity, of 9.0*10-7 mol m-2 s-1 Pa-1 and 25, respectively. With the membrane material defined, CAPEX and OPEX were calculated for CASES 1 and 2. Additionally, a sensitivity analysis was performed to verify which configuration was more sensitive to membrane permeance to CO2. CASE 2 exhibited best performance and was selected as the membrane process configuration in subsequent analyses, to avoid combinatorial explosion. CASE 2 resulted in
35$/t of CO2 captured against 45 $/t of CO2 in CASE 1. Moreover, CASE 2 was also the more sensitive to membrane permeance changes. Membrane costs accounted for more than 20% of CAPEX and 17% of OPEX in CASE 2, while for CASE 1 membrane module represented 2% of CAPEX and 2% of OPEX. Therefore, CASE 2 was taken as the membrane process configuration for further analyses.
For the selected membrane configuration, a cost (CAPEX and OPEX) sensitivity analysis was carried out where membrane selectivity was varied for three costs of membrane material: 50,
100 and 150 $/m². The baseline cost was set as the capture cost by chemical absorption with MEA, so that the minimum membrane selectivity was identified as the selectivity at which membrane process cost equals chemical absorption cost. Last, the membrane was changed to a lower performance material, and the analysis with 100$/m² was conducted for comparison purpose.
Itâ??s noteworthy that the membrane material was selected as material with best reported performance in the literature while the MEA capture was not optimized to decrease its heat load in the regenerator, taken as 4.2 GJ/t CO2, which accounted for more than 50% of the OPEX, resulting in a cost of 36$/t CO2.
The minimum separation factor required for membrane prices of 50, 100 and 150 $/m² were 7,
10 and 13, respectively, for silica/silica-zirconia/alpha-alumina as membrane material (MATERIAL 1). For a second membrane material (MATERIAL 2), taken as 100$/m², the calculated separation factor was about 40. MATERIAL 1 results in lower operations costs than MEA for membrane prices inferior to 100$/m². It should be noticed that other membrane materials could also have lower costs, however, the separation factor should be high enough or else larger membrane area would be required, eliminating the advantage of the lower unitary cost. Furthermore, membranes with such performances are still hard to be obtained in industrial scale, and hence classical methods for CO2 capture are still the best economic option.