capture Peter Brder, Hallvard F. Svendsen Norwegian University of - PowerPoint PPT Presentation

Post Combustion Capture Conference 2011, Abu Dhabi 17-19 May 2011 Solvent comparison for postcombustion CO 2 capture Peter Brúder, Hallvard F. Svendsen Norwegian University of Science and Technology

Basic characterization of the amine – CO 2 systems  Density (loaded / unloaded)  Viscosity (loaded / unloaded)  Amine volatility  Physical solubility of CO 2 in the water-amine solution  Reaction rate amine – CO 2  Maximum rich and minimum lean loading (cyclic capacity)  Stripping specific heat consumption  Partial CO 2 pressure over the loaded solution  Environmental impact of the amine  Thermal and oxidative degradation of the amine

Screening equipment

Screening procedure - The absorption temp. is 40 0 C, the stripping occurs at 80 0 C - The system is flushed with N 2 before the measurements - The calibration is done for 0 % and 10 % CO 2 - The outlet gas from the sample is analysed by IR detector - After the end of absorption and desorption the density of the samples were measured, and samples were taken for titration of CO 2

Tested chemicals Structure Comment Solvent CAS p K a 2-Amino-2- (hydroxymethyl)- 77-86-1 8.1 1,3-propanediol (TRIS) DMMEA 108-01-0 9.2 AMP 124-68-5 9.7 DEEA 100-37-8 9.8 Piperazine 110-85-0 9.8 Ethanolamine (MEA) 141-43-5 9.5 3-amino-1- methylaminopropan 6291-84-5 10.5 e (MAPA)

Tested concentrations (A:B) A B Piperazine MEA MAPA 5 5 TRIS 3:1 3:1 3:1 DMMEA 3:1, 5:2 5:2 3:2, 5:2, 5:1 AMP 3:1 3:1 DEEA 3:1 5:2 5:2, 4:4

Results

Absorption rates 3M AMP + 1M PZ 08.08 2008 0,6 3 M AMP + 1 M MAPA 06.10 2008 3 M DMMEA + 1 M PZ 20.08.2008 0,5 3 M DMMEA + 2 M MAPA 09.10 2008 5 M DMMEA + 2 M PZ 21.08 2008 0,4 5 M DMMEA + 2 M MEA 3.09 2008 Absorbtion rate [Nl/min] 5 M DMMEA + 2 M MAPA 4.09 2008 0,3 5 M DMMEA + 1 M MAPA 23.03 2010 3 M DEEA + 1 M PZ 15.09 2008 5 M DEEA + 2 M MEA 01.10 2008 0,2 3 M TRIZMA + 1 M PZ 17.09 2008 3 M TRIZMA + 1 M MEA 22.09 2008 0,1 3 M TRIZMA + 1 M MAPA 23.09 2008 30 % MEA 02.10 2008 0 5 M MAPA 13.10 2008 0 1 2 3 4 5 load [mol CO2/kg solution]

Lean loading - to final 1 kPa CO 2 , 80 0 C (titration) Concentration [mol/l] System Lean loading [mol CO 2 /kg sol] 5 MAPA 1.08 3 : 2 DMMEA : MAPA 1.68 5 : 2 DMMEA : PZ 2.41 5 : 2 DMMEA : MAPA 1.93 3 : 1 AMP : PZ - 3 : 1 AMP : MAPA 1.65 3 : 1 DEEA : PZ - 5 MEA 0.95 5 : 2 DMMEA : MEA 1.79 3 : 1 DMMEA : PZ 1.82 3 : 1 TRIS : MAPA 0.79 3 : 1 TRIS : PZ 1.06 5 : 2 DEEA : MEA - 3 : 1 TRIS : MEA 0.77

Cyclic loading calculation - The cyclic loading from screening values are always bigger than the values from the titration (average ~ 11 %) - Experimentally for MEA ~ 90 % of the equilibrium loading can be reached during absorption, corresponding an absorber bottom mass transfer rate of about 0.2 Nl/min - Two values for cyclic loading were calculated, one based on the end point of the absorption curves, and the other based on an end pointdetermined by a minimum mass transfer rate of 0.2 Nl/min

System comparison total amine mol/l Absorbed CO2 titration mol/kg sol Cyclic capacity total mol/kg sol Cyclic capacity 80 C (Loading 0,2 Nl/min - Lean loading) mol/kg sol 8 7 6 5 4 3 2 1 0 5MAPA 3DMMEA/2MAPA 5DMMEA/2MAPA 3AMP/1MAPA 5MEA 5DMMEA/2MEA 3TRIS/1MAPA 5DEEA/2MEA 3TRIS/1MEA 5DMMEA/2PZ 3AMP/1PZ 3DEEA/1PZ 3DMMEA/1PZ 3TRIS/1PZ

Absorbent evaluation - MAPA has the lowest cyclic capacity between the tested promotors compared to the total loading - 5 DMMEA : 2 MEA has approximately the same cyclic and total capacity as 3 DMMEA : 1 PZ, but with a much slower kinetics - AMP with PZ has much better kinetics than MAPA - Tris and DMMEA perform worst with MEA from the three promotors - Despite of the lower amine concentration, the 3 DMMEA : 2 MAPA has higher kinetics and total absorption in the final stage than the 5 DMMEA : 2 MAPA/PZ systems, probably due to the lower viscosity, however the cyclic concentration is lower

Absorption rates for phase change systems 0,6 0,5 5 M DEEA + 2 M MAPA 16.09 2008 30 % MEA 02.10 2008 0,4 Absorbtion rate [Nl/min] 4 M DEEA + 4 M MAPA 08.10 2008 5 M MAPA 13.10 2008 0,3 0,2 0,1 0 0 1 2 3 4 5 6 load [mol CO2/kg solution]

DEEA:MAPA 5:2 Upper phase Lower phase Unloaded solution Loaded solution with separated phases

DEEA:MAPA 5:2 The prepared unloaded solution in %: DEEA : MAPA : H2O Mol %: 33 : 13 : 54 Mass %: 64 : 20 : 16 Vol %*: 65 : 20 : 15 ~ 60 vol % ~ 0,91 kg/l ~ 0,25 mol CO 2 /kg ~ 8,1 mol NH 2 /kg ~ 40 vol % ~ ??? kg/l ~ 4,8 mol CO 2 /kg ~ 9,9 mol NH 2 /kg * Density for the calculation has been taken from chemical description

DEEA:MAPA 5:2 At 40 0 C - The 2 phases (layers) are separated immediately after the end of the experiment, but to reach better separation, needs 1 – 2 days at 20 0 C - Seems that two phases are formed for DEEA : MEA too, but after mixing only 1 phase remains, and this is stable - The system 4:4 was also tested, but the lower phase is very viscous and in other cases the viscosity doesn’t allow formation of 2 layers

Conclusion • Mixed solvents seems to have better performance than the single ones • The kinetics of the reaction can be effected significantly with the increasing viscosity during the CO 2 absorption • These two results imply significantly reduced stripping steam consumption, sensible heat loss, and around half of the solvent circulation rate in a plant operation compared to MEA. • The liquid phase separated systems seems to have a better kinetics, probably due to the increased contact area

Acknowledgements The research leading to these results has received funding from the European Community’s Seventh Framework Program FP7/2007-2011 under grant agreement nº 213569 (CESAR).

Thank you !