PNNL

Abstract

This document presents results of experiments conducted to measure release fractions during certain tank retrieval processes. The tests were performed in a ¼ - scale model of a waste storage tank. The retrieval processes simulated were: 1. Discharging liquid or slurry from the mouth of a vertically oriented two-in. Schedule 40 pipe. The discharging material was in free-fall from the mouth of the pipe near the top of the tank into a liquid or slurry pool at the bottom of the tank. 2. The jet from a 9/16-in.-diameter nozzle transferring liquid or slurry waste from one side of the tank to the other. The discharging liquid was aimed at the opposite side of the tank from the nozzle and either impacted the tank wall or fell into a liquid or slurry pool in the bottom of the tank. 3. A high pressure fan jet of liquid striking a steel plate or simulated waste from a stand-off distance of a few inches. For each process, a water-soluble fluorescent dye was added to the liquid fraction as a tracer. Kaolin clay was used to represent the solids. The tank was covered and there was no forced ventilation in the tank during the tests. Six air samples were collected during each test. The air samples were collected at fixed positions in the tank. The air sample filters were dried and weighed to determine the solids collection. The fluorescent dye was then leached from each filter and quantified with a fluorometer to determine the collection of liquid. Samples of the slurry and liquid simulants were also collected to determine the quantities of simulant used in each test. To calculate the release fraction, the quantity collected on each air sample was adjusted for the fraction of the tank volume sampled and divided by the quantity of material exposed in the simulation. The method was not as sensitive for the solids content as it was for the liquid content, but in those instances where a solids release fraction was determined, it was in relatively good agreement with that of the liquid phase. Release fractions are commonly used to make conservative estimates of emissions from processes. Usually, rather gross assumptions are made in such estimates, such as the total failure of abatement equipment and the use of maximum inventory values. Consequently, it is common practice to report bounding release fraction values with single digit accuracy. The release fractions for the top of the unventilated tank ranged from 9 x 10-7 to 8 x 10-5 depending on the process simulated.