Lecture_23_2023.pdf

Carbon Capture during Power Generation Fuel N2 CO2 Separation Power Generation Post Combustion CO2 Air H2O Air/O2/Steam Air Pre-combustion Fuel Gasification Reformer + CO2 Separation Power Generation N2 H2O CO2 H2O Fuel Power Generation CO2 H2O Air O2 Separation N2 Oxy-fuel Co...

Carbon Capture during Power Generation Fuel N2 CO2 Separation Power Generation Post Combustion CO2 Air H2O Air/O2/Steam Air Pre-combustion Fuel Gasification Reformer + CO2 Separation Power Generation N2 H2O CO2 H2O Fuel Power Generation CO2 H2O Air O2 Separation N2 Oxy-fuel Costs of Carbon Capture (Power Generation) The primary costs in implementing CO2 capture technology are • Capital cost (total design, purchase and installation costs) • Product cost (e.g. electric power) • Cost of CO2 avoided (difference in product cost per kWh-1 with and without CO2 capture as a fraction of the difference in total mass of CO2 emissions per kWh-1 with and without capture) Cost of CO avoided (€/tonne) 2 = (€ / kWh ) capture − (€ / kWh )reference ( tonne CO2 / kWh )reference − ( tonne CO2 / kWh )capture • Cost of CO2 captured (difference in product cost per kWh-1 with and without CO2 capture as a fraction of the total mass of CO2 captured per kWh-1) Costs of Carbon Capture • Capture is the most expensive step (2/3 of the added cost of CCS) • Costs must be expressed in CO2 avoided and not CO2 captured. • Cost must be assessed based on added cost of electricity (CoE) produced (criterion for selecting technology and plant construction) PC plant with capture 2096 -1 Capital cost ($ kW ) without capture 1286 with capture 73 Product cost ($ MWh-1) without capture 46 Cost CO2 captured $/tCO2) 29 Cost Cost CO2 avoided $/tCO2) Change plant efficiency Ε 41 0.76 Plant NGCC IGCC Hydrogen 998 1825 568 1326 54 62 9.1 37 47 7.8 44 20 12 53 0.85 25 0.83 15 0.92 Summary of some modern power plant cost performance indicators (Metz et al. (2005)) Summary of Case Studies for Carbon Capture (SEI (2005)) See also: https://bit.ly/3N0pI4H [accessed 14/10/22] CO2 : Post-combustion Capture Carbon Capture during Power Generation Fuel N2 CO2 Separation Power Generation Post Combustion CO2 Air H2O Air/O2/Steam Air Pre-combustion Fuel Gasification Reformer + CO2 Separation Power Generation N2 H2O CO2 H2O Fuel Power Generation CO2 H2O Air O2 Separation N2 Oxy-fuel Column internals (packing) (2) Gas out Liquid in Raschig Rings Pall Rings Packing Gas in (1) Liquid out Berl Saddles Adsorbent Liquid Liquid cA , xA z + z pA , yA c Ai p Ai Fluid Gas LA G B z Gas Interface (i) G, L = local gas and liquid total molar Species ‘A’; conc. in liquid ‘c’; partial pressure, flowrates per unit cross-section of the tower ‘p.’ ‘i’ = interface. (mol. s-1. m-2 ); x, y = local mole fractions in the liquid and gas phases respectively; d (Gy A ) d ( Lx A ) = = − N Aa dz dz NA = local molar flux of A across the gasliquid interface (mol.s-1.m-2); (Gy A ) − (Gy A )1 = ( Lx A ) − ( Lx A )1 ( ( d (Gy A ) − N Aa = = − k ac y − y G G A Ai dz d ( Lx A ) or − N A a = = − k ac x − x L L Ai A dz ) ) a = Interfacial area per unit tower volume (m2.m-3); kG, kL = Mass transfer coefficients (m.s-1) cG, cL = total concentration in gas or liquid, respectively. Using the equilibrium relationship 1 p *A HA c *A = then Henry’s Law (I) N A = KG ( p A − p*A ) = KG H A (c*A − c A ) Overall mass Transfer coefficient 1 RT H A = + KG kG kL If the concentration of A at all points in the column in both the liquid and gas phases is dilute then all of the parameters are constant. Using (Gy A ) − (Gy A )1 = ( Lx A ) − ( Lx A )1 equilibrium line gives, under dilute conditions controls slope (cA − cA,1 ) = G cT ( p A − p A,1 ) L p Eq. (I) cA cA,1 1 operating line (II) 2 Eq. (II) pA,1 pA Height of packing required (h) cA cA,1* 1 Eq. (I) Eq. (II) 2 G h= pK G a pA2  p A1 c A2 dp A L dc A = p*A − p A cT H A K G a cA1 c A − c*A pA,1 pA The slope of the dashed blue line gives the maximum allowed value for G/L i.e. p (c − cA,2 ) G   =  L Max cT ( pA,1 − pA,2 ) * A,1 Note that if cA,1 = cA,1* (at which point we also have pA,1 = pA,1*) then the column height would be infinite. This condition defines the maximum ratio for the flows G/L (or the minimum L/G) from Eq (II) at a given p and cT. Solving the integral for the height of the absorber yields   G  h=    GH AcT  pK G a      Lp     ln  p A,1 − p A,2   GH AcT  − 1 + 1     *      p A,1 − p A,1   Lp     − 1           c A,1 − c A,2 G   = ln  *   c A,1 − c A,1   GH AcT      pK G a   − 1       Lp     1 −   Lp   GH AcT     + 1   Typically, the operational liquid flow is determined by specifying the flow ratio to be a multiple (e.g. 1.5 times) of the minimum L/G flow ratio. The inlet conditions are also known and the required outlet concentration in the gas is usually specified. The remaining unknown outlet concentration, cA,1, is then calculated from the overall material balance. Chemical absorption systems are currently the preferred options, particularly at low CO2 concentrations, and have been in use in selected industrial applications (but not widely in power plants) since the 1980s. The reaction of primary interest here is CO2 + 2HOC2H4NH2 → HOC2H4NHCOO- + HOC2H4NH3+ i.e. A + bB → products This represents the overall reaction (see Kim et al (2009)) as shown on the next slide. The first three of these reactions combine to give the overall result above. (1) CO2 + 2H2O → HCO3- + H3O+ (K1 < 10-7) (2) HOC2H4NH2 + HCO3- → HOC2H4NHCOO- + H2O (K2 > 100) (3) HOC2H4NH2 + H3O+ → HOC2H4NH3+ + H2O (4) HCO3- + H2O → CO32- + H3O+ (5) 2H2O → H3O+ + OH- (K3 > 1010) (K4 < 10-11) (K5 < 10-15) The rate of reaction is largely determined by a combination of reactions (1) and (2) CO2 + HOC2H4NH2 + H2O → HOC2H4NHCOO- + H3O+ This is known to be second order leading to the overall rate described by rA = kcAcB (for example k(315K) = 5.183x107 m3/kmol.hr) Because of the very high rate constant in this absorption process we can generally employ a design based on fast reaction within a thin layer close to the liquid surface. Liquid cB Adsorbent Liquid do pA , yA c Ai cA = 0 Fluid Gas Gas z + z LA p Ai G B z Compared to the case of simple absorption, here we consider that within the bulk liquid phase for each mole of A absorbed (‘increase’ in A) reaction occurs to remove b moles of B (i.e a decrease in B). The local overall balance is therefore modified to read d (Gy A ) 1 d ( LxB ) = − N Aa = − dz b dz If we assume that the MEA concentration remains relatively high (pseudo-first order kinetics) then it may be shown (see Levenspiel (1980), Freguia and Rochelle (2003)) that NA = 0 1 RT + k AG HA DAL kcB ( p A − H Ac A ) Note that here we consider cB to be effectively constant i.e. an average between points 1 and 2. The local balance may then be integrated to give  p A,1 G h= ln  pK eff a  p A, 2   where 1 = RT +  K eff k AG  HA DALkcB A range of other cases are discussed by Levenspiel (1980). The above conditions however correspond closely to conditions for MEA/CO2 absorption in many situations. One such analysis is reported by Froment and Bischoff (1990). Column operating conditions (Froment and Bischoff): MEA composition in feed liquid = 4.4 mol% (2.22 kmol/m3) T = 315K, P = 14.3 bar, pA,1 = 1.954 bar, pA,2 = 0.000715 bar G = 535.5 kmol/hr.m2, L = 4976 kmol/hr.m2; Ac = 0.866 m2. Parameter values: DAL = 0.711 x 10-5 m2/hr; k = 5.183 x 107 m3/(kmol.hr) cB = Average of top and bottom = 1.3275 kmol/m3 HA = 48.6 m3 bar/kmol; RT/kG = 1.0325 m2hr bar/kmol a = 90.74 m2/m3 These data provide h = 10.55 m The result from an exact analysis (variable flows, variable concentration of B, mathematically rigorous analysis etc.) gives h = 10.95 m. The above separation but for physical absorption only: T = 315K, p = 14.3bar, pA,1 = 1.954 bar, pA,2 = 0.000715 bar; G = 535.5 kmol/hr.m2; Ac = 0.866 m2; cA,2 = 0, cT = 55 kmol/m3; HA = 48.6 m3 bar/kmol; RT/kG = 1.0325 m2hr bar/kmol; kLa ~ 40 hr-1 (C. Judson King (1980); a = 90.74 m2/m3 Then cA,1* = 1.954/48.6 = 0.0402 and the maximum G/L = 0.00535. Take the operating G/L = max/1.5 = 0.00357 (with GAc = 463.7 kmol/hr this implies L = 129,900 kmol/hr i.e. an enormous flow of water for absorption and stripping requiring a large tower diameter) Then (GHAcT/Lp) = 0.6673 and cA,1 = GcT/Lp(pA,1 – pA,2) (Eq.(II)) = 0.02682  h = 1022 m A less stringent separation with only 90% recovery of the CO2. T = 315K, p = 14.3 bar, pA,1 = 1.954 bar, pA,2 = 0.1954 bar Then cA,1* = 1.954/48.6 = 0.0402 and in this case maximum G/L = 0.00594 and actual G/L = max/1.5 = 0.00396. Then (GHAcT/Lp) = 0.740 and cA,1 = GcT/Lp(pA,1 – pA,2) = 0.02678  h = 213 m Column diameter: If the estimated liquid rate is too low to ensure wetting of the packing, then the minimum wetting rate as determined by correlations in Morris and Jackson (1953) can be used to specify the liquid flow requirements within the tower. Using the result obtained for L’/G’ then column diameter is determined on the basis of the flooding criterion as depicted in the diagram on the right (Treybal (1980)). This diagram provides an appropriate value for the superficial gas mass velocity G’ within the tower which should be maintained to achieve pressure drop conditions below flooding. Approximate flooding Y Gas pressure drop N/m2 m X L'   G  X = G'   L − G G '2 C f  L0.1   ; Y = G (  L − G )  0.5 See Treybal (Table 6.3) for Cf Solvent Regeneration The absorbed carbon dioxide is removed (at least in part) to regenerate the solvent for recycle to the absorption tower. This regeneration is achieved in the stripping (or desorption) column the design procedure for which is very similar to that for the absorber. The major differences are: 1. Absorption is carried out at low temperatures (typically ambient conditions or marginally above) to enhance the absorption of CO2. 2. To regenerate the solvent, the temperature must be increased (>100oC) in the desorber. As demonstrated overleaf for the equilibrium constant K for CO2 + 2HOC2H4NH2  HOC2H4NHCOO- + HOC2H4NH3+ there is a significant shift in the overall absorption reaction to the left at higher temperatures thus releasing CO2. Equilibria under Reaction Conditions The equilibrium expression for the overall reaction is 1.0E+008 cHOC H NHCOO− cHOC H NH + 2 4 2 4 3 2 cCO 2 cHOC 2 H 4 NH 2 Where f is a factor accounting for non-idealities in the liquid mixture and is 1.0 for simple ideal systems. Equilibrium Constant K K = fcT 1.0E+009 1.0E+007 1.0E+006 1.0E+005 1.0E+004 1.0E+003 1.0E+002 1.0E+001 2 2.4 2.8 3.2 103/T 3.6 4 Energy Requirements The large energy requirement for the regeneration process is one of the major drawbacks in alkanolamine carbon capture processes. While the absorption process involves energy considerations (mainly cooling to maintain the liquid within the tower at or close to ambient conditions), heat/energy integration is a key to proper management of the capture process. A major demand on the energy requirements of the alkanolamine capture process is the desorption energy needed in the stripping tower. At a temperature high enough to achieve adequate stripping (e.g. ~120oC for MEA/CO2), this contribution is q(kW) = (H r / b)(( LxB )2 − ( LxB )1 ) Hr is the heat of reaction per mole of CO2 and is obtained from the equilibrium constant using H r = − R(d ln K / d (1 / T ) ) The industrial example of Froment and Bischoff • At 120oC the heat of reaction is approximately 110kJ/mol; • The liquid solution (containing MEA predominantly in the complex form HOC2H4NHCOO-) is fully regenerated from 0.435 (0.894%) to 2.22 kmol/m3 (4.4%) MEA; • With an average LAc = 1.197 kmol/s, the energy required is 2.31 MW. Conditions for a 1 GW power generation facility • The flue gas molar flowrate and CO2 composition to be treated after removal of SOx and NOx are 41x103 mol/s and 16.5 mol%; • It may be shown that for a fully regenerated MEA solution concentration of xB1 = 0.112 (30 wt%) the MEA solution flowrate required for gas cleaning in this case is ~ 50 kmol/s i.e. the energy needed to desorb the CO2 is ~ 0.15 GW (xB2 = 0.056) → 0.3 GW (xB2 = 0.0) (xB2 is the MEA composition in the liquid stream entering the desorber from the absorption unit). Breakdown in energy consumption for MEA/CO2 capture Breakdown in regeneration energy for MEA A. Kothandaraman et al, Energy Procedia 1 (2009) 1373-1380. Minimum Power Requirements (EU) for Capture (GW) Absolute Minimum Power Requirements for Postcombustion CO2 Capture 100 80 Based on total CO2 emissions of 1,836 kmol/s (EU27 in 2020) with fully reversible separation and compression of the pure CO2 product to 100 bar at 298K CO2 fractional capture f = 1.0 60 f = 0.75 40 Realistic processes would operate at ~30-40% efficiency or less (minimum 3 times more power required) f = 0.5 20 0 0.0001 0.001 0.01 0.1 CO2 Mole Fraction in Gas 1 EU27 power demand (total) = 2320 GW EU27 electric power demand ~ 450 GW Minimum Power = FT G Minimum Power   1 − yCO2 f    p2   yX   RTFCO (GW ) =  f ln   − ln yCO2 + ln 1 − yCO2 f + (1 − f ) ln  2  yCO (1 − f )   yCO2   p1   2    ( )