Cold War Legacy: Low-Level Nuclear Weapons Waste

SwRI engineers develop sensor technology to facilitate waste characterization and long-term storage

by Glenn M. Light, Ph.D.      image of PDF button


For almost 20 years, Institute Scientist Dr. Glenn Light has developed sensors, systems, and techniques for the nondestructive evaluation of materials and structures for the nuclear power industry. He is shown with an SwRI-fabricated decontamination module, which will be attached to low-level nuclear waste storage tanks to clean the LDUA as it is removed after tank inspections.


In the years following the second World War, the U.S. produced large amounts of new and upgraded nuclear weapons at its Department of Energy laboratories across the country. To fuel these weapons, it was necessary to manufacture as much weapons-grade nuclear material as possible, in as safe a manner as possible. Measures were taken to ensure that operating procedures were safe for facility workers and area inhabitants, but the emphasis was on production, with less consideration given to potential environmental impact. Waste generated during the fabrication process was stored in large underground tanks at the processing plants. These wastes consisted primarily of chemical solutions and precipitates containing various radionuclides (radioactive elements and compounds), as well as contaminated solvents and process residuals such as organic and inorganic solvents, metals, and some fissionable material.


The design drawing above is of a robotic end-effector, a specialized tool that is attached to the end of a light duty utility arm (LDUA) designed for low-level nuclear waste storage tank inspection. End-effectors can be used to inspect tank walls and to visualize, grasp, and analyze materials within storage tanks.

The underground storage tanks range in size from 10 to 75 feet in diameter and from 30 to 35 feet high, and their volume capacity ranges from approximately 50,000 gallons to more than one million gallons. Most of the tanks are single shell, carbon steel structures, with concrete surrounding all the surfaces, and some are double shell, with an annulus between the two shells.

Before storage, much of the waste contained acids and other corrosive materials. To minimize corrosion problems, the wastes were chemically treated to be more basic (pH of approximately 11). However, it is not known what effects the treated waste may have had on the carbon steel in the tanks over the last five decades. The material in the tanks is now in the form of sludge, hard salt cake, and some liquid.

The end of the cold war brought a diminished need for a large nuclear weapon production capability as well as increased concerns about the safe storage of nuclear waste. A major concern has been the possibility that the waste might be released into the environment and contaminate the water, soil, and air around a storage facility. To guard against such an occurrence, stored waste must be characterized, retrieved from storage, treated, separated into low- and high-level waste streams, and finally put into a disposal form that effectively encapsulates the waste and isolates it from the environment for a long period of time.

Southwest Research Institute is assisting the U.S. Department of Energy, through a multiphase, multiyear contract initiated in 1994 with the Westinghouse Hanford Company, to ensure that hazardous waste storage facilities are adequate and that technology to characterize waste is available. While the high-level waste will remain in storage until a permanent disposal facility, such as the one proposed for Yucca Mountain, Nevada, is placed into service, the Westinghouse Hanford site in Washington will serve as a central repository for vitrified low-level nuclear weapons waste.

The Institute has teamed with Los Alamos Technical Associates, British Nuclear Fuels, Ltd., and TRW to provide a full-service, cost-effective solution for the Department of Energy. SwRI has developed nondestructive evaluation (NDE) sensors and sensor system technology to help ensure the safe and adequate handling of the waste. In addition, SwRI engineers are designing and fabricating hardware for a robotic, light duty utility arm (LDUA) that can lift up to 75 pounds and to which specialized tools, or end-effectors, will be attached that can visualize, grasp, and analyze core samples of materials inside the tanks.

Southwest Research Institute sensor system technology will be employed in a number of ways to detect and monitor storage tank corrosion damage. To nondestructively and remotely detect corrosion effects, Institute engineers are developing mechanically manipulated sensors to examine the outer walls of double shell tanks in the annulus area. For single shell tanks, NDE sensors attached to robotic end-effectors will assess interior tank wall conditions. A wide range of sensors is also needed to verify that equipment inserted into the waste is operating properly and not causing any hazardous effects, such as damage to tank walls.

To safely retrieve waste from storage prior to its placement in the repository, certain physical properties must be determined. For example, does the waste form strata in the tank, are there organics or explosive substances in the waste, is the waste pumpable, which radionuclides are present, and what is the moisture content? Since the waste is radioactive, all characterization must be done remotely. Two characterization approaches are being considered. The first is to physically retrieve a sample from the waste tank and send it to a laboratory for analysis. The second approach is to use in situ techniques to characterize the waste. Two types of devices exist for this approach - penetrometers and mini-lab robotic end-effectors. Both devices allow physical parameters to be measured by instrumentation inserted into the tank, without having to send samples to remote laboratories for analysis.

The Institute will soon begin work for Westinghouse Hanford on the development of end-effector components to aid placement and use of the LDUA. Other end-effectors will also be developed to collect samples in the mini-lab, to detect moisture using nuclear magnetic resonance or eddy current techniques, and to determine the presence of radionuclides with nuclear particle detection equipment. Institute engineers will fabricate prototypes and production units to implement these sensor technologies.


Low-level nuclear waste storage tanks range in size from 10 to 75 feet in diameter and from 30 to 35 feet tall, and their volume capacity ranges from approximately 50,000 to more than one million gallons. Surrounded by concrete, the tanks are either single- or double-shell carbon steel. This cutaway shows the LDUA in place for tank inspection and other robotic operations.


If a remote sampling approach is used, sensor technology can help locate the surface of the waste inside the tank and can verify that a sample has been effectively retrieved. The Institute has participated in one development in this area that employs ultrasonics, weight, and neutron thermalization to verify that a sample has been collected. In a second project, SwRI engineers are evaluating an existing rotary-mode waste sampling system and recommending ways to improve its reliability.

Once the waste has been characterized, it must be retrieved from storage and transferred to a holding tank. The retrieval of sludge and some salt cake can be accomplished with a process called sluicing, in which a high-pressure water jet loosens the waste and transform it into a slurry that can be pumped. The sluicing process can erode storage tank walls, so NDE tools such as ground penetrating radar, electrical resistivity tomography, neutron probes, and radio-imaging will be deployed in boreholes to detect leaks underneath tanks. Institute engineers helped develop a procedure for tank operators to follow when detecting leaks.

The stored nuclear weapons waste is believed to be a mixture of a relatively small amount of highly radioactive material within a large amount of sludge, salt cake, and fluids, including lubricants, that were not originally radioactive. To minimize disposal costs, it is important to separate this waste into high- and low-level waste streams. The purpose of tank waste treatment is to accomplish this separation, but sensors are needed to monitor separating systems. Once the waste has been separated, the low-level waste can be vitrified and placed in the Hanford repository. The goal of vitrification is to encapsulate the low-level waste in a form that will prevent any hazardous components from escaping into the water and air for up to 10,000 years in accordance with U.S. Environmental Protection Agency Title 40 of the Code of Federal Regulations, Chapter 1, Part 191 (7/1/92 edition). To achieve this goal, the vitrification process must be monitored to ensure a quality product, and the vitrified waste must be inspected to verify that it is defect-free. Defects could lead to cracking, which provides more surface area on which water can react to leach radionuclides out of the glass. In addition, NDE technologies will be needed to inspect the vault in which the waste will be stored.

One of the most challenging requirements for future sensor technology is that it be able to monitor the condition of the glass and vault to verify their integrity for a period of 100 years. This means that sensors, power supplies such as batteries, and data acquisition systems must be operable for the same amount of time. To meet this requirement, scientists and engineers must develop equipment that requires minimal maintenance and that can be remotely repaired or replaced. The Institute's experience in materials, sensors, and integrated systems can help address these challenges.

Published in the Spring 1996 issue of Technology Today®, published by Southwest Research Institute. For more information, contact Joe Fohn.

Spring 1996 Technology Today SwRI Publications SwRI Home