Automated Sampling and Sample Pneumatic Transport of High Level Tank Wastes at the Hanford Waste Treatment Plant

This paper describes the development work, and design and engineering tasks performed, to provide a fully automated sampling system for the Waste Treatment Plant (WTP) project at the Hanford Site in southeastern Washington State, USA. WTP is being built to enable the emptying and immobilization of highly active waste resulting from processing of irradiated nuclear fuel since the 1940s. The Hanford Tank Wastes are separated into Highly Level Waste (HLW), and Low Active Waste (LAW) fractions, which are separately immobilized by vitrification into borosilicate glass. Liquid samples must be taken of the waste and Glass Forming Chemicals (GFCs) before vitrification, and analyzed to insure the glass products will comply with specifications established in the WTP contract. This paper describes the non-radioactive testing of the sampling of the HLW and LAW melter feed simulants that was performed ahead of final equipment design. These trials were essential to demonstrate the effectiveness and repeatability of the integrated sampling system to collect representative samples, free of cross-contamination. Based on existing tried and proven equipment, the system design is tailored to meet the WTP project’s specific needs. The design provides sampling capabilities from 47 separate sampling points and includes a pneumatic transport system to move the samples from the 3 separate facilities to the centralized analytical laboratory. The physical and rheological compositions of the waste simulants provided additional challenges in terms of the sample delivery, homogenization, and sample capture equipment design requirements. The activity levels of the actual waste forms, specified as 486E9 Bq/liter (Cs-137), 1.92E9 Bq/liter (Co-60), and 9.67E9 Bq/liter (Eu-154), influenced the degree of automation provided, and justified the minimization of manual intervention needed to obtain and deliver samples from the process facilities to the analytical laboratories. Maintaining high integrity primary and secondary confinement, including during the cross-site transportation of the samples, is a key requirement that is achieved and assured at all times. INTRODUCTION The U.S. Department of Energy (DOE) has responsibility for managing the safe storage, treatment, and disposal of high-level radioactive waste in underground storage tanks at the Hanford Site in southeastern Washington State, USA. There are 177 underground storage tanks containing over 204 million liters (54 million US gallons) of highly radioactive waste and 7.2E9 GBq (195 million WM’06 Conference, February 26–March 2, 2005, Tucson, AZ curies) of radioactivity.[1] This waste was generated primarily as a result of irradiated nuclear fuel processing for defense production activities from 1943 through 1989. Most of the 177 tanks are well beyond their design life and the removal and immobilization of the waste is a Hanford Site priority. In 1996, DOE initiated a project to treat and immobilize the waste and, subsequently, a team of contractors led by Bechtel National Inc.(BNI) was selected to design, build, and commission the Waste Treatment Plant (WTP) for this purpose.[2] The WTP will separate the Hanford radioactive tank waste into High Level Waste (HLW) and Low Activity Waste (LAW) fractions and will separately immobilize these wastes by vitrification into borosilicate glasses.[3] The Immobilized High Level Waste (IHLW) and Immobilized Low Activity Waste (ILAW) glass products must comply with specifications established in the contract governing the vitrification work. IHLW must also comply with specifications in the Waste Acceptance Product Specifications (WAPS) developed by the DOE.[4] Process control aspects of this compliance strategy are focused on measuring and controlling the feeds of waste and Glass Forming Chemicals (GFC’s) for each process batch so that compliant IHLW or ILAW will be consistently produced. It is desired to avoid both the need for routine compliance sampling of the glass product and the possibility of the need to rework non-compliant glass. Therefore various samples, chemical analyses of samples, and measurements will be required to monitor and control the IHLW and ILAW vitrification processes, and it will be necessary to implement and demonstrate a range of qualification and production strategies for complying with the specifications. As part of this strategy, an automated sampling system is needed so that the radioactive samples can be taken and transported to the analytical laboratory safely and under full process control. In June 2004, BNI awarded BNG America, British Nuclear Group’s US subsidiary, the contract to design and supply such a system. The system design is based on that used successfully in several nuclear plants at the Sellafield nuclear processing site in northwest England. It comprises automated sample collection units (“Autosamplers”) together with a pneumatic transport system to transfer the samples to the analytical laboratory. The Autosamplers are shielded, and located adjacent to the processing cells. Liquids are pumped by fluidic devices or mechanical pumps from the process vessels to the Autosamplers, where small sample volumes are transferred to sample bottles which in turn are placed in sample carriers and pneumatically transferred to the analytical laboratories. No manual intervention is necessary to deliver the process liquids to the autosamplers, or to obtain and transfer the samples through the pneumatic transport system. The requirement was for a system to provide capabilities to obtain samples from each of the 3 separate processing facilities within the WTP; the Pretreatment Facility (PTF), the High Level Waste Vitrification Facility (HLW), and the Low Active Waste Vitrification Facility (LAW), and to transfer the samples pneumatically to the centralized Analytical Laboratory where the equipment delivers the samples to the stations to prepare for analysis. There are 47 separate sampling points in the process, and the 3 facilities and the Analytical Laboratories are separated by approximately 366 meters (1200 ft). The scope of the work included the design and manufacture of 10 Autosamplers, 3 Laboratory Receipt stations, ancillary equipment to support the pneumatic transfer of samples, and the design and specification of the complete pneumatic transport system. Tasks also included the planning of operator training, technical support during installation, testing at the site, and start-up support. WM’06 Conference, February 26–March 2, 2005, Tucson, AZ As a part of the design & equipment selection and validation process, is was necessary to demonstrate the effectiveness and repeatability of the integrated sampling system, by demonstration of the collection of representative samples, using waste simulants as defined by the client. The physical and rheological compositions of the waste simulants provided an additional challenge in terms of the sample delivery, homogenization, and sample capture equipment design, mainly because of the non-newtonian nature of some of the sludge slurries to be sampled. In addition, equipment selection, handling, and shielding requirements were influenced by the activity of the actual waste forms, specified as 486E9 Bq/liter (Cs137), 1.92E9 Bq/liter (Co-60), and 9.67E Bq/liter (Eu-154). This paper describes how the system is designed to provide an overall availability greater than 99.34% as mandated, by demonstrating availability through statistical analysis with substantiated assumptions. It further describes the non-radioactive testing performed on the LAW and HLW melter feed simulants, both with and without GFCs (Glass Forming Chemicals), and describes how the statistical difference between the mean values of the percent solids, measured from the vessel of origin, were compared to the sample obtained. The tests demonstrated excellent repeatability, in terms of volumetric accuracy and statistical difference. This provided a high level of confidence that the equipment incorporated into the design will ultimately provide the necessary degree of sampling accuracy.