(626f) CO2 Selective Membranes for Carbon Capture in Water-Gas-Shift Reactions

Introduction During the last century, many new technologies have been developed and the industrialized world has grown fast. This has had and still has an enormous impact on the environment, especially through the greenhouse gas emissions associated with energy production in accordance with the Kyoto Protocol [1]. This global warming is mostly due to greenhouse effect through gases like CO₂. Apart from energy saving and durable sources, a way to reduce these emissions is to capture and store CO₂. Post-combustion capture technologies such as MEA scrubbing [2] and using polymeric membranes [2] already exist. Pre-combustion capture would have a lower capture penalty on the efficiency and takes place at higher temperatures where polymers do no longer function as a membrane. When starting out with a coal gasifier, a water gas shift membrane reactor can be used either with hydrogen selective membranes or CO₂ selective membranes. CO₂ selective membranes enable keeping hydrogen pressurized at a high temperature and is, therefore, an attractive alternative for noble metal alloy based hydrogen separating membranes [3,4,5,6]. Our research target within the Global Climate and Energy Project (GCEP [7]) is, therefore, to develop CO₂ separating membranes. In the Sorption Enhanced Reforming Project (SERP) [8,9,10], as well as in other projects [11,12,13,14], hydrotalcite materials are investigated as CO₂ sorbents. These materials seem to be good candidates for adsorbing CO₂ although the desorption process is still rather slow. The presumption is that these materials might be applicable either as a dense or a porous membrane. Therefore, we decided to deeply study these materials and to look at all properties required for a viable membrane process. Hydrotalcites are clay-like materials consisting of brucite-type layers separated by layers containing water and anions. The hydrotalcite, with the general formula M²⁺_xM³⁺_1-x(OH)₂(CO₃)_(1-x)/2.xH₂O (with M²⁺ being Mg²⁺ and M³⁺ being Al³⁺ in this case), has a crystal structure in which, via its interlayer containing CO₃^2- anions and water, is presumed capable of having CO₂ molecules enter the structure on one side and get out of it on the other side; thus acting as a highly selective dense membrane. Porous membranes are presumed to transport CO₂ through the inter grain pores. Neither possibility has been proven yet. It has been claimed in the literature that hydrotalcites could be synthesized with any Mg/Al ratio [15,16,17,18,19] and that it decomposes with increasing temperature, but so far no decomposition pathway has been described. We present our results here concerning hydrotalcite characterization and decomposition. We also propose a way to produce hydrotalcite membranes for pre-combustion CO₂ capture. Materials and techniques Commercial materials from Sasol and materials prepared in-house following the urea hydrothermal hydrolysis [20] with different Mg/Al ratios have been compared using X-Ray diffraction (XRD) and Scanning Electron Microscopy (SEM) with Emission Dispersive X-ray (EDX). Neutron Diffraction (ND) has been used to evaluate the exact Mg/Al ratio in hydrotalcite. The decomposition pathway has been established using ThermoGravimetric Analysis (TGA) attached to a Mass Spectrometer (MS), in-situ XRD and in-situ Diffuse Reflectance Infrared Fourier Transformed (DRIFT) spectroscopy. Experimental Characterization A comparison between the commercial materials and the in-house produced ones has been performed with XRD. From this, it can be deduced that the commercial materials are poorly crystalline compared to the in-house produced ones. Furthermore, the closer the Mg/Al ratio is to 2, the more crystalline the materials are. Since the number of electrons of Mg²⁺ and Al³⁺ are equal, XRD can not be used to find the ratio between these ions, however, with ND one can. Three samples of different Mg/Al ratios have then been analyzed at the GEM beam line at ISIS. The pattern could be refined using the GSAS program suite, in which CO₃^2- and H₂O are treated as rigid bodies. The Mg/Al ratio could be determined to be about 1.8. Deviations from this value yielded either Al rich (Boehmite) or Mg rich (hydromagnesite) impurity phases. Thermal stability The water-gas-shift reaction takes place at about 400°C. TGA-MS has been performed on the commercial material MG61. Four decomposition steps could be seen. The first step at ≈100°C represents the loss of adsorbed water, the second at ≈240°C represents the interlayer water disappearing, the third at ≈310°C represents the hydroxides decomposing into oxides, and the fourth at ≈450°C represents the last hydroxides and carbonates decomposing into oxides. To characterize the structural evolution upon decomposition, in-situ XRD has been performed on the same material (fig.1). Fig.1: in-situ XRD pattern of MG61 as a function of temperature Figure 1 shows that the three first steps found with the TGA are confirmed by the XRD. Between 200°C and 250°C, there is a shift towards smaller d-spacing, indicative of the collapse of the interlayer at ≈240°C. Between 300°C and 350°C, the crystallinity of the material is strongly decreasing, while the hydroxides are decomposing into oxides. More detailed information on the process has been obtained using DRIFT. This technique permits infrared spectroscopy under different atmospheres and temperature. Combining XRD, TGA-MS and DRIFT results yields the following decomposition pathway: Mg₄Al₂(CO₃)(OH)₁₂.xH₂O.yH₂O → Mg₄Al₂(CO₃)(OH)₁₂.xH₂O+yH₂O (150°C)(1) Mg₄Al₂(CO₃)(OH)₁₂.xH₂O → Mg₄Al₂(CO₃)(OH)₁₂+xH₂O (230°C) (2) Mg₄Al₂(CO₃)(OH)₁₂ → MgCO₃.3Mg(OH)₂.Al₂O₃+3H₂O (320°C) (3) MgCO₃.3Mg(OH)₂.Al₂O₃ → 3MgO.MgAl₂O₄+CO₂+3H₂O (440°C) (4) Membranes The results of the previous paragraph clearly show that the dense membrane option is not feasible since the interlayer does not exist anymore at the operational temperature. A porous membrane is still an option. In this case, very small particles have to be produced that can be deposited on a support as thin layers. Using a conventional co-precipitation synthesis route made it possible to obtain nano-particles of 15nm diameter. Conclusions and perspectives Using XRD, TGA-MS, ND, DRIFT and SEM-EDX resulted in the formulation of the decomposition pathway as well as the Mg/Al ratio in hydrotalcites. It has been shown that hydrotalcite materials cannot be used as dense membranes. However, the option to make hydrotalcite based-porous membranes still seems viable. It has been shown that the small nano-particles needed to make such membranes can be produced. The next step is to disperse them to form a colloidal suspension and to coat them on a support. Another solution would be to use a sol-gel method or to make these particles grow in situ. Acknowledgements The authors thank the GCEP program (http://gcep.stanford.edu/) for financial support, NWO for access to the ISIS facilities and ISIS staff for measurements of the samples, and also Vera Smit-Groen, Andre van Zomeren and Sacha Valster for their help with XRD, TGA/MS and DRIFT experiments. References 1. Convention-cadre des Nations Unies sur les changements climatiques (1992) 2. B. METZ, O. DAVIDSON, H. C. CONINCK DE, M. LOOS, and L. A. MEYER, IPCC special report on carbon dioxide capture and storage (2005) 3. Y. H. MA, in Proceedings of the 9th International Conference on Inorganic Membranes,Lillehammer, June 2006, p. 28. 4. Y. H. MA, I. P. MARDILOVICH, and E. E. ENGWALL, Annals of the New York Academy of Sciences 984 (2003) p. 346. 5. H. KLETTE, T. A. PETERS, A. MEJDELL, and R. BREDESEN, in Proceedings of the 8th International Conference on Greenhouse Gas Control Technologies, 2006 6. http://gcep.stanford.edu/ 7. M. V. CIOCCO, B. D. MORREALE, O. IYOHA, R. M. ENICK, R. P. KILLMEYER, and B. H. HOWARD, in Proceedings of the International Technical Conference on Coal Utilization & Fuel Systems 30th, 2005, p. 303. 8. J. HUFTON, S. MAYORGA, T. GAFFNEY, S. NATARAJ, M. RAO, and S.SIRCAR, Sorption Enhanced Reaction Process (SERP) for the Production of Hydrogen (2005) 9. J. HUFTON, W. WALDRON, S. WEIGEL, M. RAO, S. NATARAJ, and S. SIRCAR, Sorption Enhanced Reaction Process (SERP) for the production of hydrogen PA 18195 (2000) 10. Y. DING and E. ALPAY, Chemical Engineering Science 55 (2000) p. 3929. 11. Y. DING and E. ALPAY, Trans IChemE Part B 79 (2001) p. 45. 12. Z. YONG, V. MATA, and A. E. RODRIGUES, Ind. Eng. Chem. Res. 40 (2001) p. 204. 13. Z. YONG and A. E. RODRIGUES, Energy Conversion and Management 43 (2002) p. 1865. 14. J.-I. YUNG and J.-N. KIM, Korean Journal of Chemical Engineering 23 (2005) p. 77. 15. M. TSUJI, G. MAO, T. YOSHIDA, and Y. TAMAURA, Material Research Society 8 (1993) p. 1137. 16. K. TAKEHIRA, T. KAWABATA, T. SHISHIDO, K. MURAKAMI, T. OHI, D. SHORO, M. HONDA, and K. TAKAKI, Journal of Catalysis 231 (2005) p. 92. 17. T. YAMAMOTO, T. KODAMA, N. HASEGAWA, M. TSUJI, and Y. TAMAURA, Energy Conversion and Management 36 (1995) p. 637. 18. R. MARCHIDAN, D. CRISAN, E. IVANA, M. GABROVSKA, R. EDREVA-KARDJIEVA, and A. ANDREEV, Bulgarian Chemical Communications 30 (1998) p. 217. 19. E. KANEZAKI, Mat. Res. Bull. 33 (1998) p. 773. 20. M. M. RAO, B. R. REDDY, M. JAYALAKSHMI, V. S. JAYA, and B. SRIDHAR, Materials Research Bulletin 40 (2005) p. 347.