COST-EFFECTIVE METHOD FOR PRODUCING SELF SUPPORTED PALLADIUM ALLOY MEMBRANES FOR USE IN EFFICIENT PRODUCTION OF COAL DERIVED HYDROGEN

Abstract

Over the last quarter, we developed procedures for producing free-standing, defect free films using rigid silicon and glass substrates over areas up to 12 square inches. Since formation of contiguous Pd-Cu films in the 2-3 {micro}m-thick range is ultimately governed by the size of the particle contamination on the supporting substrate surface, we have adopted techniques utilized by the semiconductor industry to reduce and eventually eliminate particle contamination. We have found these techniques to be much more effective on rigid substrates and have made a down select decision on removal methods (a key milestone) based on these results and the performance of membranes fabricated by this technique. The path to fabricating even larger membranes is straightforward and will be demonstrated in the coming months. Hydrogen permeation tests were also conducted this quarter on as-deposited, Pd-Cu membranes, between 6-14 {micro}m-thick. In the case of a 6 {micro}m-thick film, the pure hydrogen flux at 20 psig and {approx}260 C was 36 cm{sup 3}(STP)/cm{sup 2} min. This flux corresponds to a pure hydrogen permeability of 7.4 {center_dot} 10{sup -5} cm{sup 3} cm cm{sup -2} s{sup -1} cm Hg{sup -1/2} at 250 C. This value is within 20% of the pure hydrogen permeability at 250 C reported in the McKinley patent. In the case of a 14 {micro}m-thick membrane tested at 350 C, the pure hydrogen flux, measured before initiating a pinhole-size leak, was 2.1 {center_dot} 10{sup -5} cm{sup 3}(STP) {center_dot} cm/cm{sup 2} {center_dot} s {center_dot} cm Hg{sup 0.5}. This value is considerably lower than the expected permeability of Pd{sub 60}Cu{sub 40} materials at 400 C. To date, essentially all of the sputtered deposited Pd-Cu thin film membranes have had palladium compositions that were as much as 3% greater than the ideal 60 weight percent composition (this is a direct consequence of sputtering from a 60/40, Pd/Cu alloy target). As the concentration of Pd is increased beyond the optimum 60% value, a less desirable two-phase structure forms at the higher temperatures (in this case, above 260-280 C). As we continue development of procedures for producing thinner Pd-Cu films next quarter, we will also be optimizing alloy composition and corresponding hydrogen permeation flux as well.