Energy Procedia. Viability of mixed conducting membranes for oxygen production and oxyfuel processes in power production

Transcription

1 Available online at Energy Procedia 00 1 (2009) (2008) Energy Procedia GHGT-9 Viability of mixed conducting membranes for oxygen production and oxyfuel processes in power production M. den Exter a J.F. Vente a D. Jansen a W.G. Haije a, * a Energy research centre of the Netherlands, PO Box 1, 1755 ZG Petten, The Netherlands Elsevier use only: Received date here; revised date here; accepted date here Abstract This paper is devoted to the state of the art of the most important aspects of high temperature ceramic air separation membranes for oxygen production and oxidation processes. Materials choice, membrane and module design, seal development and fabrication issues will be discussed Elsevier Ltd. Open access under CC BY-NC-ND license. Keywords: MIEC; membrane geometry;module design;stability;processing 1. Introduction It has been generally accepted by now that CO 2 emissions are the origin of the ever increasing global temperature rise [1]. One of the means to reduce these emissions is capturing this greenhouse gas at point sources like power plants, as long as the transition to a durable energy supply is not completed and fossil fuels still have to be used to bridge this gap. Among the three options for carbon capture i.e. oxy-fuel combustion, post-combustion capture and pre-combustion capture, production of oxygen via high temperature mixed conducting membranes is energetically more favorable than the conventional cryogenic process. Application of mixed ionic electronic conducting (MIEC) membranes in power production so as to burn fuel in pure oxygen is ideally the easiest way of separating CO 2 out of the flue gasses, the other component being just water. Clearly both processes pose different demands as to reduction stability of the membrane material. Suitable materials can be found in the family of cubic perovskite oxides ABO 3 and oxides of the structural archetype of the K 2 NiF 4 type compounds (Figure 1). Selection criteria are high flux, high mechanical, chemical and thermal stability, ease of fabrication, and cost. Furthermore membrane geometry, module design, high temperature high oxygen pressure resistant sealing and ceramic processing are important issues for a viable technology. These issues will be discussed further in this paper. * Corresponding author. Tel.: ; fax: address: Haije@ecn.nl doi: /j.egypro

2 456 M. den Exter et al. / Energy Procedia 1 (2009) Author name / Energy Procedia 00 (2008) ABO 3-δ A 2 BO 4+ δ Figure 1 Left: the perovskites structure with oxide vacancies indicated, and right: the A2BO4+δ structure with surplus oxygen regions indicated 2. Materials and materials processing Suitable compositions for application in the two mentioned areas of application are Sr 0.97 Ce 0.03 Fe 0.8 Co 0.2 O 3-δ (SCFC) and La 2 Ni 0.9 Co 0.1 O 4+δ for materials under reducing conditions and SrCo 0.8 Fe 0.2 O 3-δ (SCF) and Ba 0.5 Sr 0.5 Co 0.8 Fe 0.2 O 3-δ (BSCF) for materials under non reducing conditions [2,3,4]. SCF and LSCF have been shown to exceed the official flux target of 10 ml/cm 2 min [5]. This is ascribed to the high Co content. Under reducing conditions, however, in high Co content materials the Co-oxide part will eventually be reduced to metallic Co, therefore a high Fe content is used instead at the cost of a lower oxygen flux. J (ml/cm 2 min) SCFC (PD) vs SOFC o.g. SCFC (PD) vs He SCFC (D) vs He LNC (D) vs He T (K) Figure 2 Left: permeate side of a BSCF membrane after three weeks of testing (see text) and right: oxygen fluxes for SCFC and LNC (see text) The left part of Figure 2 shows the effect on the permeate side after 3 weeks permeating oxygen through a BSCF membrane where a He sweep was used to provide the driving force. Already in this mild case structural and compositional stability appear to be an issue although the flux did not show a decline yet. Near the helium inlet the partial pressure of oxygen is very low and close to the outlet the oxygen partial has been built up and no sign of deterioration can be found. The right part shows flux results for 16 cm 2 flat membranes made of materials to be used under more reducing conditions: the lowest curve is for LNC with a helium sweep using a dense 200 micron thick membrane, the next one shows the same for SCFC, the third one shows a measurement on a 10 micron thick dense layer on a 200 micron thick porous support and in the fourth one instead of a He sweep a gas mixture representative of an SOFC off-gas, containing H 2, CO, CO 2, and H 2 O, has been used. At this stage clearly SCFC is preferred: the flux is much higher than for LNC and for a real life application one does not have to work with the highly carcinogenic Ni compound. The Co rich materials have been shown to exhibit substantial creep at operation temperatures [6], which hampers, among other things, sealing (see below).

3 M. den Exter et al. / Energy Procedia 1 (2009) Author name / Energy Procedia 00 (2008) cm 5 cm Figure 3 Left: supported membrane made using a slightly wrong temperature profile and right: using the appropriate one. Furthermore quite a lot of effort is needed for the development of recipes for support and dense layer fabrication as well as the sintering procedures. Binders and pore formers have different combustion temperatures and enthalpies. Careful design of the heating program is decisive for success or failure (see Figure 3). 3. Membrane geometry, module design and sealing The design of an adequate air separation unit highly depends on the type of membrane configuration chosen. Several conceptual options are available: hollow fibers, multi-channel monoliths, single tube, and tube-and-plate configurations. Each membrane configuration has its advantages and disadvantages when taking reachable surface area, sealing technology and possible use of a sweep gas into account. Hollow fibers provide the highest membrane surface area but are relatively fragile and difficult to seal. Sealing of multichannel monoliths, especially when small channels are used for maximizing the amount of surface area, is also difficult. Single tube configurations, on the other hand, are less difficult to seal while up-scaling principles to one meter length has already been proven for other tubular membrane systems. An extensive study of the pros and cons of the aforementioned membrane geometries has been performed, taking into account both gas velocities and the space needed for seals and flanges, and using vessels that are commonly used in industry, clearly showed that the single tube design outperforms the other options [7]. A schematic is shown in Figure 4 (left, middle). The membrane length is about 2.5 m and the module length and diameter are about 4.5 m and 2.0 m respectively. About 40 of these modules are needed for an oxygen production of 40,000 Nm 3 /hr, equivalent to the size of a conventional large cryogenic oxygen plant. On the right of Figure 4, the principle of compression sealing of membranes with a tubular configuration is depicted. The membrane tube is mounted in a metal end-cap and the space in between end-cap and membrane filled by the sealing material. Compressive forces are obtained upon increasing the temperature and the magnitude of these forces can be tuned by proper selection of the metal type and sealing material, based on their Thermal expansion coefficients (TEC). This tuneable principle is beneficial since the perovskite membranes will likely have moderate strength compared to steel. Other sealing material options can be found in ceramic glues with tuneable TEC, melts of the membrane materials or systems with a combination of binder with membrane material. Last two options have the advantage of a proper TEC but are less preferred from an economical point of view since manifolding will be more complicated.

4 458 M. den Exter et al. / Energy Procedia 1 (2009) Author name / Energy Procedia 00 (2008) Figure 4 Left and middle: module design for oxygen generation with tubular MIEC membranes: double shell: inner shell hot, outer cool (pressure vessel), insulation in between, membrane area 150m2. Right: Schematic of the patented ECN compression seal. 4. Concluding remarks From the discussion in this paper, the conclusion can be drawn that the actual implementation of ITM technology cannot be expected in the short term. Although quite some hurdles have been taken, a number of technology challenges remain that still need to be met. Sealing technology, chemical and mechanical stability of the different compositions envisioned, are still not meeting the target. Reduction of high temperature creep is prerequisite for application. However, the small amount of public information from two large leading consortia in the US, led by Air Products and Praxair, hampers a thorough and reliable assessment of the current state-of-the art ITM technology. Our vision on the viability high-temperature ITM technology can thus be unnecessarily pessimistic. Acknowledgment This research is part of the CAPTECH programme. CAPTECH is supported financially by the Dutch Ministry of Economic Affairs under the EOS programme. More information can be found on References 1. IPCC, 2007: Summary for policymakers. In: Metz, B, Davidson OR, Bosch PR, Dave R, Meyer LA, editors. Climate change 2007: mitigation. Contribution of Working Group III to the Fourth Assessment, Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York: Cambridge University Press; J.F. Vente, W.G. Haije, Z.S. Rak. Performance of functional perovskite membranes for oxygen production. J Membr Sci S. McIntosh, J.F. Vente, W,G. Haije, D.H.A. Blank, H.J.M. Bouwmeester. Phase stability and oxygennonstoichiometry of SrCo 0.8 Fe 0.2 O 3-δ measured by in-situ neutron diffraction. Solid State Ionics

5 M. den Exter et al. / Energy Procedia 1 (2009) Author name / Energy Procedia 00 (2008) J.M. Paulsen. Thermodynamics, Oxygen stoichiometric effects and transport properties of ceramic materials in the system Sr-Ce-M-O (M = Co, Fe), Ph.D. Thesis 1998, Technical University of Dresden. 5. R. Bredesen, K. Jordal, O. Bolland. High-temperature membranes in power generation with CO 2 capture. Chem Eng Process J.F. Vente, S. McIntosh, W.G. Haije, H.J.M. Bouwmeester,Properties and performance of Ba x Sr 1-x Co 0.8 Fe 0.2 O 3-δ materialsfor oxygen transport membranes, J. Solid State Electrochemistry, J.F. Vente, W.G. Haije, R. IJpelaan, F.T. Rusting. On the full-scale module design of an air separation unit using mixed ionic electronic conducting membranes. J Membr Sci