Development and application of high temperature CO 2 sorbents

Transcription

1 1 Development and application of high temperature CO 2 sorbents B. Arstad, E. Bakken, J. Prostak, Christine Wüst, R. Blom SINTEF Materials & Chemistry P.O. Box 124 Blindern, N-0314 Oslo, Norway bjornar.arstad@sintef.no; richard.blom@sintef.no

2 2 Outline Background and principles Our approach: Continuous lab scale reactor testing Recent results using natural Ca-based sorbents Search for new sorbents Search based on thermodynamic properties Conclusions and further prospects.

3 Background 3 Separation of CO 2 from gas mixtures using solid sorbents at high temperature ( C) represents an alternative to processes at ambient or sub-ambient temperatures. Two important examples: The carbonate looping process (CL) [1] Sorption enhanced -steam methane reforming/-water-gas shift reactions (SE-SMR) [2]. 1) J.C. Abanades, E.J. Anthony, J. Wang, J.E. Oakey, Environ. Sci. Technol., 39, (2005), ) D.P. Harrison, Ind. Eng. Chem. Res., 47, (2008), 6486

4 4 Application principles The sorbents are typically a metal oxide that reacts with CO 2 and form a carbonate. A subsequent regeneration decomposes the carbonate and thus releases CO 2 as a gas. The newly reformed metal oxide is then ready for use again. MO + CO C MCO C

5 Carbonate looping 5 Flue gas CO 2 CO 2 Carbonation MCO 3 MO Regeneration MCO 3 Fuel burning in this reactor provides heat for regeneration of sorbent MO Flue gas C Gas/coal/biomass + O 2 1) J.C. Abanades, E. J. Anthony, J. Wang, J. E. Oakey, Environ. Sci. Technol. 39 (2005) 2861.

6 6 Sorption-enhanced methane steam reforming Hydrogen production from natural gas with simultaneous capture and separation of CO 2. One possible concept: Sorption enhanced methane steam reforming, SE-SMR [1]. SE-SMR has been proposed to be an economical competitive alternative to conventional steam methane reforming with CO 2 capture [2]. SE-SMR unit Natural gas Steam Steam reforming + water-gas shift + CO 2 capture H 2 The hydrogen market Sorbent regeneration CO 2 Sequestration 1) D.P.Harrison, Ind. Eng. Chem. Res. 47 (2008) ) E. Ochoa-Fernandez, G. Haugen, T. Zhao, M. Rønning, I. Aartun, B. Børresen, E. Rytter, M. Rønnekleiv, D. Chen, Green Chem. 9 (2007) 1.

7 Riser for transport Cyclone Lab-scale testing of SE-SMR H 2 Reforming: CH 4 (g) + 2H 2 O(g) + MO(s) Feed gases: CH 4 + steam + N 2 MCO 3 (s) + 4H 2 (g) Near thermoneutral Loop seal 2 Fluidization N 2 Regeneration: MCO 3 (s) N 2 CO 2 MO(s) + CO 2 (g) Endothermic Loop seal 1 N 2 N 2

8 SE-SMR (feed reformer: 50 ml/min CH 4, 200 ml/min H 2 O, 1860 ml/min N 2 diluent) - Effect of catalyst/sorbent ratio: 8 CO2 sep. efficiency % % % % % % 3060 % ml/min ml/min CH4 H2 CH4 CO CO % % 0 0 % ,0 1,0 2,0 Time [min] 3,0 4,0 5,0 T [h] CO CO2 CO2 sep. ml/min efficiency 100 % % 80 % % 60 % % ml/min % 30 % 2050 % 10 % 0 % 0 ml H2 ml CH4 ml CO ml CO2 0,0 2,0 4,0 T [h] 6,0 8,0 Time [h] Catalyst/sorbent= 20/80 Catalyst/sorbent= 50/ ml CH4 ml CO ml CO2 0,0 2,0 4,0 6,0 8,0 Time [h]

9 SE-SMR (feed reformer: 50 ml/min CH 4, 200 ml/min H 2 O, 1860 ml/min N 2 diluent, initial catalyst/sorbent= 20/80, used and regenerated powder mixture) - Effect of catalyst addition during experiment: and Methane total Cs conversion from R1 and [%], R2 exp - exp C 580 C ml H2 ml H2 200 ml CH % ml CH4 Measured ml H2 ml CO CO % Cs from ml R1CO % Cs from R % % % Time 3 Time [h] [h] Time [h] Volumeflow Methane ml / min conversion [ml/min] H2 measured and calculated from conversion of methane Reformer outlet, [ml/min] exp C 9

10 SE-SMR (feed reformer: 50 ml/min CH 4, 200 ml/min H 2 O, 1860 ml/min N 2 diluent, initial catalyst/sorbent= 20/80, used and regenerated powder mixture) - Effect of reformer temperature: 10 Volumeflow Methane conversion [ml/min] Methane Reformer conversion outlet, [ml/min] [%], 530 C 100 % % ml H % ml CH % ml CO % ml CO % % % % % 00 0 % Time [h] [h] [h] Volumeflow Methane conversion [ml/min] Reformer Methane conversion outlet, [ml/min] [%], 630 C 100 % % % % % % % % % % % Time Time [h] [h] ml H2 ml CH4 ml CO ml CO2

11 11 Sorbent development The objectives of this WP is to develop solid sorbents for CO 2 capture processes at elevated temperatures ( C) and pressures (20-30 bar) suitable for industrial use. Our long term goal is to identify a production method for new sorbent material, for a large demonstration unit, based on cheap raw materials. MO + CO C MCO C State of the art (2009) high temperature CO 2 sorbents are usually based on Ca-oxides or Li/Na-silicates/zirconates. Li-based: Li 2 ZrO 3 Li 4 SiO 4 Ca-based: CaO Mg 0.5 Ca 0.5 O CaO/support (lime) (from dolomite) ( synthetic )

12 Thermodynamic search for potential novel solid sorbents: 12 using FactSage 6.1 Use thermodynamic calculations to evaluate various oxides, both simple and mixed oxides for carbonate formation at realistic process conditions (SE-SMR: 44% H 2 O, 35% H 2, 11% CH 4, 7% CO 2, 2% CO, 5 bar tot). Sulphur tolerance has been studied: Do sulphide species (as H 2 S) react with the sorbent and if, does it form sulphides (S 2- ) or sulphates (SO 2-4 )? How do such species influence CO 2 capacities and how must these sorbents be regenerated?

13 The CaO example: 13

14 14 Thermodynamic evaluation, example of data: Carbonate formation at 873 K, hydroxides are liquids DrG (kj/mol CO2) LiOH NaOH KOH RbOH CsOH MgO CaO Sr(OH)2 Ba(OH) Gibbs free energy of carbonate formation at 873 K for selected hydroxides and oxides

15 15 Sulphide formation at 873 K, hydroxides are liquids DrG / (kj*mol H2S) LiOH NaOH KOH RbOH CsOH MgO CaO Sr(OH)2 Ba(OH)2-40 Gibbs free energy of sulphide formation at 873 K for selected hydroxides and oxides

16 Summary thermodynamics 16 For single component sorbents based on alkali and alkaline earth oxides/hydroxides the main trends are: Most of the selected oxides/hydroxides have so high CO 2 affinity that thermal regeneration is challenging, and the incorporation of elements with less CO 2 affinity in the sorbent material may be necessary to facilitate a higher CO 2 equilibrium pressure. The alkaline earth metals have a stronger tendency to form both carbonates and sulphides than the alkali metals. The alkali metals are more selective towards carbonate formation. The regeneration should be performed in low O 2 concentration an at a rather high temperature to avoid sulphate formation Rb, seem to have interesting properties with high CO 2 affinity and very low tendency for sulphide and sulphate formation. The modified alkaline earth metals are better suited as CO 2 sorbents in a sweet (H 2 S free) feed gas. Bulk CaO have good thermodynamic properties for cyclic temperature swing sorption-desorption without modification/incorporation of new elements in the sorbent material.

17 17 CO 2 adsorption capacities 650 C, pco 2 =0.15 bar, ph 2 O=0.29 bar, 1 gram, 30 ml/min total flow CO 2 adsorption capacity [mmol/g] Sintef ACAT2811 full capacity Sintef ACAT2405 full capacity REF ECN CaO/Ca-Aluminate reactivated ECN REF CaO/Ca-Aluminate fresh Cycle number

18 18 Summary Finally, a lab scale CFB reactor system is working well for SE-SMR using calcined dolomite as sorbent and NiO/NiAl 2 O 4 as catalyst which can be used to evaluate novel sorbents (and catalysts) at a relatively small scale (approx 200 ml powder) Identification of novel sorbent compositions by the use of thermodynamic calculations using relevant process conditions are in progress. CaO has good thermodynamic properties (no surprise!) Some new leads will be tested.

19 Acknowledgement This publication has been produced within the H2_SER project (NRC grant ) and the BIGCCS Centre, performed under the Norwegian research program Centres for Environment-friendly Energy Research (FME). The authors acknowledge the following partners for their contributions: Aker Solutions, ConocoPhilips, Det Norske Veritas, Gassco, Hydro, Shell, Statkraft, Statoil, TOTAL, GDF SUEZ and the Research Council of Norway (NRC grant /S60). 19 Thank you for your kind attention