MELTER OFF-GAS FLAMMABILITY ASSESSMENT FOR DWPF ALTERNATE REDUCTANT FLOWSHEET OPTIONS

Abstract

Glycolic acid and sugar are being considered as potential candidates to substitute for much of the formic acid currently being added to the Defense Waste Processing Facility (DWPF) melter feed as a reductant. A series of small-scale melter tests were conducted at the Vitreous State Laboratory (VSL) in January 2011 to collect necessary data for the assessment of the impact of these alternate reductants on the melter off-gas flammability. The DM10 melter with a 0.021 m{sup 2} melt surface area was run with three different feeds which were prepared at SRNL based on; (1) the baseline formic/nitric acid flowsheet, (2) glycolic/formic/nitric acid flowsheet, and (3) sugar/formic/nitric acid flowsheet - these feeds will be called the baseline, glycolic, and sugar flowsheet feeds, respectively, hereafter. The actual addition of sugar to the sugar flowsheet feed was made at VSL before it was fed to the melter. For each feed, the DM10 was run under both bubbled (with argon) and non-bubbled conditions at varying melter vapor space temperatures. The goal was to lower its vapor space temperature from nominal 500 C to less than 300 C at 50 C increments and maintain steady state at each temperature at least for one hour, preferentiallymore » for two hours, while collecting off-gas data including CO, CO{sub 2}, and H{sub 2} concentrations. Just a few hours into the first test with the baseline feed, it was discovered that the DM10 vapor space temperature would not readily fall below 350 C simply by ramping up the feed rate as the test plan called for. To overcome this, ambient air was introduced directly into the vapor space through a dilution air damper in addition to the natural air inleakage occurring at the operating melter pressure of -1 inch H{sub 2}O. A detailed description of the DM10 run along with all the data taken is given in the report issued by VSL. The SRNL personnel have analyzed the DM10 data and identified 25 steady state periods lasting from 32 to 92 minutes for all six melter runs (bubbled and non-bubbled runs for each of the three feeds). The steady state selection was made by limiting the standard deviation of the average vapor space temperature readings from two bare thermocouples (TT-03 and TT-05) to less than 5 C in most cases at a constant feed rate. The steady state data thus selected were mass and heat balanced and the off-gas data were re-baselined to assess the flammability potential of each feed under the DWPF melter operating conditions. Efforts were made to extract as much information out of the data as possible necessary to extend the applicability of the existing baseline cold cap and off-gas combustion models to the glycolic and sugar flowsheet feeds. This report details the outcome of these activities.« less