Preparation and Heat-Treatment of DWPF Simulants With and Without Co-Precipitated Noble Metals

Abstract

The Savannah River National Laboratory is in the process of investigating factors suspected of impacting catalytic hydrogen generation in the Chemical Process Cell of the Defense Waste Processing Facility, DWPF. Noble metal catalyzed hydrogen generation in simulation work constrains the allowable acid addition operating window in DWPF. This constraint potentially impacts washing strategies during sludge batch preparation. It can also influence decisions related to the addition of secondary waste streams to a sludge batch. Noble metals have historically been added as trim chemicals to process simulations. The present study investigated the potential conservatism that might be present from adding the catalytic species as trim chemicals to the final sludge simulant versus co-precipitating the noble metals into the insoluble sludge solids matrix. Parallel preparations of two sludge simulants targeting the composition of Sludge Batch 3 were performed in order to evaluate the impact of the form of noble metals. Identical steps were used except that one simulant had dissolved palladium, rhodium, and ruthenium present during the precipitation of the insoluble solids. Noble metals were trimmed into the other stimulant prior to process tests. Portions of both sludge simulants were held at 97 C for about eight hours to qualitatively simulate the effects of long term storage on particle morphology and speciation. The simulants were used as feeds for Sludge Receipt and Adjustment Tank, SRAT, process simulations. The following conclusions were drawn from the simulant preparation work: (1) The first preparation of a waste slurry simulant with co-precipitated noble metals was successful, based on the data obtained. It appears that 99+% of the noble metals were retained in the simulant. (2) Better control of carbonate, hydroxide, and post-wash trim chemical additions is needed before the new method of simulant preparation will be as reproducible as the old method. (3) The two new simulants were visually very viscous compared to the traditional SB3 simulant. (4) Heat-treatment reduced the viscosity of the two new simulants with and without coprecipitated noble metals, though they were still more viscous than the traditional SB3. (5) The approach of using a 97 C heat-treatment step to qualitatively simulate tank farm aging may not be optimal. A significant change in the base equivalent molarities of both simulants was observed during heat-treatment. (6) Heat-treatment appeared to make phosphates insoluble in water. The following recommendations came out of the work: (1) Washed slurry should be checked for TIC and base equivalents before calculating the final trim chemical additions of sodium carbonate and sodium hydroxide. (2) Final insoluble trim chemicals should be added to the slurry in the cross-flow filtration unit mixing tank, since significant slurry is lost in the CUF equipment. Adding the chemicals here would keep them in the correct proportion relative to the precipitated insoluble solids. (3) A composite wash and decant sample should be prepared containing proportionally weighted masses of each aqueous stream removed during preparation of a co-precipitated noble metal simulant. This sample should then be checked for noble metal losses. This would reduce the sample load, while still confirming that there was no significant noble metal loss. (4) A study of the impact of heat-treatment on existing simulants should be undertaken. If there is a shift in base equivalents, then SRNL acid stoichiometries may be biased relative to real waste. The study should be extended to several real wastes as well.