Modification of the molten carbonate in dual-phase membranes for improved carbon dioxide separation

Abstract

Solid metal and metal oxide supports for high-temperature dual-phase molten salt membranes often contain high-cost metals such as samarium, cobalt or silver. These metals are used due to their oxide and/or electron-conductive properties, which assist in carbon dioxide transport by enhancing the rate of carbonate ion formation at the feed-side surface. A cheaper alternative approach would be to use a low-cost, abundant high-temperature stable ceramic support, such as alumina, and improve membrane performance by using small quantities of metals to modify or add functionality to the membrane. First, CO2 release from the permeate-side surface was found to be slower than the uptake of CO2 on the feed-side surface. Therefore, two different doping methods were explored to improve the rate of CO2 release from the permeate side. However, due to difficulties with solubility and precipitation, both approaches proved unsuccessful. To navigate around these issues, an effective dopant would need to still provide functionality after precipitation, one such material is silver, as it maintains its electron conductivity after precipitation from molten carbonate. Therefore, by using silver as the dopant, precipitation was used advantageously to coat the walls of the support to produce self-forming pathways that contributes to the support’s functionality. The molten carbonate was used as a carrier to redistribute silver that was doped into the membrane using multiple controlled techniques. Due to the concentration gradient across the dual-phase membrane, the silver was redistributed to form electron-conducting pathways that span the cross-section of the membrane. This process was performed in-situ during permeation and transformed a low-flux alumina supported membrane (Al2O3-MC) into a high-flux, low-silver content alumina/silver supported membrane (Ag/Al2O3-MC). The self-forming Ag/Al2O3-MC membrane presented in this thesis provided a CO2 flux of 1.25 mL min-1 cm-2 at 650°C. This flux is higher than any oxide or electron-conducting supported membrane at the same temperature while using a CO2 and O2 feed or CO2 alone. Furthermore, when normalising the flux to account for gas feed partial pressure, membrane thickness and amount of silver used (mol m-1 s-1 Pa-1 mol Ag-1), the Ag/Al2O3-MC membrane had a permeability per mol of silver one order of magnitude higher than silver-supported membranes in the literature. The membrane also demonstrated stability at temperatures exceeding 650°C, where silver-supported membranes have been known to suffer as a result of carbonate retention issues. Computed tomography investigations indicated that could be due to silver not coating the entire pore wall. The Ag/Al2O3-MC membrane reached a permeance and permeability in the range reported as required for economically competitive carbon capture.