Iron Phosphate Glass-Containing Hanford Waste Simulant [report]

Gary J. Sevigny, Marcia L. Kimura, Christopher M. Fischer, M. J. Schweiger, Carmen P. Rodriguez, Dong-Sang Kim, Brian J. Riley
2012 unpublished
P un I H G M C M C D B J repared for the nder Contract D ron P Hanfo GJ Sevigny ML Kimura CM Fischer MJ Schweig CP Rodrigue D Kim BJ Riley anuary 201 U.S. Departmen DE-AC05-76RL0 Phosp ord W ger ez 12 nt of Energy 01830 phate Waste S Glass Simu E s-Co lant PNNL-2 EMSP-RPT ntain 0670, Rev. T-005, Rev. ing . 1 . 1 Summary Resolution of the nation's high-level tank waste legacy requires the design, construction, and operation of large and technically complex one-of-a-kind processing waste
more » ... sing waste treatment and vitrification facilities. While the ultimate limits for waste loading and melter efficiency have yet to be defined or realized, significant reductions in glass volumes for disposal and mission life may be possible with advancements in melter technologies and/or glass formulations. This test report describes the experimental results from a small-scale test using the research-scale melter (RSM) at Pacific Northwest National Laboratory (PNNL) to demonstrate the viability of iron-phosphate-based glass with a selected waste composition that is high in sulfate (4.37 wt% SO 3 ). The primary objective of the test was to develop data to support a cost-benefit analysis related to the implementation of phosphate-based glasses for Hanford low-activity waste (LAW) and/or other high-level waste streams within the U.S. Department of Energy complex. The testing was performed by PNNL and supported by Idaho The RSM is a small, joule-heated melter capable of processing melter feed on a continuous basis. The melter is equipped with Inconel 693 electrodes, Monofrax K-3 refractory, and an Inconel 690 pour spout. For the experiments described here, an electric kiln surrounded the melter body and minimized heat loss from the melter body during operation. The RSM was equipped with an off-gas treatment system that employed quenching, wet scrubbing, and high-efficiency mist elimination. The glass discharge section was heated to facilitate pouring of the glass. The melter cavity was 15 cm in diameter with a nominal glass depth of 7.6 cm. The melter was operated with a glass target temperature of 1030°C and a plenum temperature between 300 and 600°C. The RSM test was broken into five segments to determine the effects of adding sugar to the feed and subsurface air injection through the molten glass, both of which are included in the current Waste Treatment Plan design. The test segments were: 1. No sugar, no air bubbler 2. No sugar, air bubbler short duration 3. Sugar, no air bubbler 4. Sugar, air bubbler 5. No sugar, air bubbler long duration Overall, the test produced 124 kg of glass. The average glass density was 2.77 g/cm 3 . At the conclusion of the test, the melter and exhaust lines were inspected for particulate deposition and corrosion. The melter electrodes and bubbler tube were removed from the glass in the RSM. The electrodes were examined with an optical microscope and a scanning electron microscope (SEM). The electrodes appeared discolored and without the loss of significant amounts of metal. Before and after measurements indicated a very small reduction in electrode dimension. Differences in the thickness measurements indicated a corrosion rate less than 0.3 mm/year and maximum length differences indicated a maximum corrosion rate of 2.1 mm/yr. iv Samples of glass and off-gas condensate were collected during the test for mass balance calculation and glass durability measurements. The data indicate an average sulfur retention in the glass of 70 wt% which is within the requirements for the low-level waste glass. Retention of Re (Tc surrogate) was 80 wt% based on the feed-to-off-gas values and 30 wt% based on the feed-to-glass values. The unaccounted for Re is most likely attributed to inefficiencies of the off-gas scrubbing equipment and scrub solution. The off-gas condensate data, along with the feed-rate and concentration data, also indicate that air bubbling likely decreased overall retention in the melter, but especially reduced retention of S in the melter. The processing of iron-phosphate-based glass was similar to borosilicate glass, although the melter was operated at a lower average temperature (1030°C) with 4.4 wt% SO 3 compared to typical borosilicate melts at 1150°C and <1.5 wt% SO 3 . The average RSM glass production rate was from 0.31 to 1 kg/h resulting in a melter surface area normalized glass generation rates of 411 to 1330 kg/day·m 2 . The addition of sugar and/or air bubbling increased the processing rate as expected, with the glass production rate more than doubling with the introduction of both sugar and air bubbling. The product glass met the LAW glass product consistency test (PCT) requirement for both quenched and canister centerline cooling (CCC) treated glasses, with lower PCT performance for CCC treated glasses likely caused by crystallization. The CCC treated samples had greater alteration rate variability than the quenched samples, with three of the CCC treated samples failing the vapor-phase hydration test (VHT) requirement. Whereas all quenched glasses showed little sign of alteration. The failure of some CCC samples to meet VHT requirement may be related to the high crystallinity and low reproducibility of crystallization in CCC samples. The corrosion of melter components was acceptable, and losses to the melter exhaust were typical of other waste glasses. v
doi:10.2172/1034577 fatcat:u5lgsv2nwverlfi4opgpqbueme