PNNL

Abstract

Privatized services are being procured to vitrify low-activity tank wastes for eventual disposalin a shallow subsurface facility at the Hanford Site. Over 500,000 metric tons of low-activitywaste glass will be generated, which is among the largest volumes of waste within the U.S.Department of Energy (DOE) complex and is one of the largest inventories of long-livedradionuclides planned for disposal in a low-level waste facility. Before immobilized waste canbe disposed, DOE must approve a"performance assessment," which is a document that describesthe impacts of the disposal facility on public health and environmental resources. Because therelease rate of radionuclides from the glass waste form is a key factor determining these impacts,a sound scientific basis for determining their long-term release rates must be developed if thisdisposal action is to be accepted by regulatory agencies, stakeholders, and the public. In part, thescientific basis is determined from a sound testing strategy.The foundation of the proposed testing strategy is a well accepted mechanistic model that isbeing used to calculate the glass corrosion behavior over the geologic time scales required forperformance assessment. This model requires that six parameters be determined, and the testingprogram is defined by an appropriate set of laboratory experiments to determine theseparameters, and is combined with a set of field experiments to validate the model as a whole.Three general classes of laboratory tests are proposed in this strategy: 1) characterization, 2)accelerated, and 3) service condition. Characterization tests isolate and provide specificinformation about processes or parameters in theoretical models. Accelerated tests investigatecorrosion behavior that will be important over the regulated service life of a disposal systemwithin a laboratory time frame of a few years or less. Service condition tests verify that thetechniques used in accelerated tests do not change the alteration mechanisms. The recommendedcharacterization tests are single-pass flow-through tests using a batch reactor design, Acceleratedand service conditions tests include product consistency and pressurized unsaturated flow (PUF)tests. Nonradioactive glasses will be used for the majority of the laboratory testing (-80%), withthe remainder performed with glasses containing a selected set of key radionuclides.Additionally, a series of PUF experiments with a natural analog of basaltic glass is recommendedto confirm that the alteration products observed under accelerated conditions in the PUF tests aresimilar to those found associated with the natural analog. This will provide additionalconfidence in using the PUF test results to infer long-term corrosion behavior.Field tests are proposed as a unique way to validate the glass corrosion and contaminanttransport models being used in the performance assessment. To better control the test conditions,the field tests are to be performed in lysimeters (corrugated steel containers buried flush with theground surface). Lysimeters provide a way to combine a glass, Hanford soil, and perhaps otherengineered materials in a well-controlled test, but on a scale that is not practicable in thelaboratory. The recommended field tests include some experiments where a steady flow rate ofwater is artificially applied. These tests use a glass designed to have a high corrosion rate so thatit is easier to monitor contaminant release and transport. Existing lysimeters at the Hanford Sitecan be used for these experiments or new lysimeters that have been equipped with the latest inmonitoring equipment and located near the proposed disposal site.