TWMCC Conference, Spring 2007 Modeling and Optimization of CO2 removal in power plants Sepideh Ziaii Fashami Dr. Thomas F.Edgar Dr. Garry T.Rochelle Chemical Engineering Department University Of Texas at Austin ,Austin ,TX 78712 Modeling and Optimization of CO2 removal in power plants Summary The public interest in CO2 removal from flue gases has recently increased, due to more restrictive regulations on emissions of greenhouse gases. The use of a classic absorption/stripping process downstream of power plants is the only option so far. Aqueous MEA is the most common solvent used to absorb CO2. Its problem is that the energy required for solvent regeneration is high. Modeling and Optimization of CO2 removal in power plants Process description The absorption/stripping process consists of an absorber, a stripper, a cross heat exchanger, rich and lean solvent pumps and lean solvent cooler. In this base case, flue gas contacts the aqueous solvent in a countercurrent, packed absorber. The source of flue gas is a coal-fired plant giving 13 mol% CO2.The rich solvent is stripped by steam in a countercurrent reboiled column to produce pure CO2. Heat is recovered from the hot lean solvent by cross heat exchanger with cold rich solvent. The lean solvent is typically cooled before entering the absorber. Modeling and Optimization of CO2 removal in power plants Process flow sheet Modeling and Optimization of CO2 removal in power plants Reactions •Equilibrium reactions HCO3- + H2O ↔ CO3-- + H3O+ MEAH+ + H2O ↔ MEA + H3O+ 2 H2O ↔ H3O+ + OH- •Non-equilibrium reactions + − + ⎯ CO 2 + MEA + H 2O ⎯ ⎯ ⎯ → MEA + HCO 3 K 1, MEA CO2 + MEA + H 2O ⎯⎯,MEA→ MEACOO − + H 3O + K2 ⎯ ⎯ + MEACOO − + H 3O + ⎯⎯−⎯ → CO2 + MEA + H 2O ⎯ K 2 , MEA − MEA + + HCO 3 ⎯ ⎯−⎯ → CO 2 + MEA + H 2O ⎯ K 1 , MEA Modeling and Optimization of CO2 removal in power plants Objective • Obtaining valid kinetic expressions for non-equilibrium reactions of CO2 with (MEA). •Modeling Absorption /stripping process to remove 90% CO2 from a coal-fired flue gas with MEA in Aspen plus software. • Optimizing required energy of CO2 capture plant Model Development Absorber : Radfrac Packed column, rate based calculation, Equilibrium and non-equilibrium reactions Stripper : Rebiled column with one equilibrium stage excluding Reboiler Thermodynamic model : Electrolyte-NRTL Kinetic Model: (Concentration based) 2152 Hikita (1977) Log 10 k 2 , MEA = 10 . 99 − 5771 T Little (1971) ln k1,MEA = 21 − T Modeling and Optimization of CO2 removal in power plants Model Matching The concentration range of Hikita model is 0.015-0.018, much lower than the 5 M used industrially and studied in this work. Therefore, Hikita model is corrected to match kinetic experimental data provided by Aboudheir (2002) and Little model is modified to get consistent bicarbonate composition in the absorber. -7 2.2x10 -7 2.0x10 -7 1.8x10 kPa cm ) 2 0.30 loading . -7 1.6x10 . k ' (mol/s -7 1.4x10 5m MEA 7m MEA 9m MEA 11m MEA g o 0.40 loading 1.2x10 -7 60 C o -7 50 C 1.0x10 o 40 C -8 8.0x10 0.00300 0.00305 0.00310 0.00315 0.00320 - 1 Temp Figure 2.Calculated mass transfer coefficients of MEA solutions at 0.30 and 0.40 loading by Ross(2006) using Aboudheir kinetic data. Result: Activity based kinetic model, calculated and used in this work, 2628 5771 Log 10 k 2 , MEA = 26 . 7 − ln k 1 , MEA = 29 − T T Modeling and Optimization of CO2 removal in power plants Sensitivity analysis Sensitivity analysis is done to minimize the equivalent work of modeled plant. As a result, Absorber with 22.5 M packing height and MEA with 0.4 lean loading is selected as an optimum condition. Equivalent Work : ⎛T + 10 − 40 ⎞ W = 0 .75Q reb ⎜ reb ⎜ T ⎟+W ⎟ +W reb + 10 ⎠ richpump leanpump ⎝ W = Equivalent work Qreb= Reboiler Duty Treb= Reboiler Temperature (◦C) Figure 3. Sensitivity analysis Modeling and Optimization of CO2 removal in power plants Conclusion and future work In this work, lean loading is the only one variable used for optimization. Basically, in addition to lean loading, other parameters such as stripper pressure, packing height and cross heat exchanger temperature approach are effective and can be accounted for optimization. In next work, another process configuration (Double matrix) will be modeled and optimized. We are determined to implement some proper optimization techniques to consider all the variables that are effective in optimization and also implement economic issues to have more complete objective function.
Pages to are hidden for
"Modeling and Optimization of CO2 removal in power plants"Please download to view full document