Leo E. Thompson and Jack L. McElroy
Geosafe Corporation

The Maralinga Site of South Australia is a former nuclear weapons test site used by the British in the 1950's and early 1960's for above ground testing. At Taranaki, Maralinga's most heavily contaminated area, a series of minor trials involving the explosive dispersal of plutonium and uranium resulted in extensive contamination of surface soil and generated substantial amounts of heavily contaminated debris [1]. Primary contaminants include plutonium, uranium, beryllium, lead, and barium. The contaminated soil and debris from the trials were subsequently buried in a series of 21 shallow pits at Taranaki. The amount of plutonium in each of the pits is expected to be on the order of 100-200 GBq. The Commonwealth Government's Department of Primary Industries and Energy is undertaking a program to rehabilitate the most heavily contaminated areas at the site. Geosafe's In Situ Vitrification (ISV) technology was selected for use to treat the 21 pits because it offered advantages of improved occupational, public, and environmental safety together with superior containment of the radioactive materials in the vitreous and crystalline product, which would be substantially more durable compared to alternative stabilization methods. This paper describes the progress of the ISV project.

In Situ Vitrification is a mobile, thermal treatment process that involves the electric melting of contaminated soils, sludges, or other earthen materials and debris in situ for the purposes of permanently destroying, removing, and/or immobilizing hazardous and radioactive contaminants. The batch process involves forming a pool of molten soil at the surface of a treatment zone between an array of four graphite electrodes. Continued application of energy results in the melt growing deeper and wider until the desired volume has been treated. When electrical power is shut off, the molten mass solidifies into a vitreous and crystalline, rock-like monolith with unequaled physical, chemical and weathering properties compared to alternative solidification/stabilization technologies. Individual melts up to 7 m deep, 12 m in diameter and up to 1,000 tonnes can be formed. Off-gases generated by the process are contained under a steel hood covering the treatment area and are drawn to an off-gas treatment system.

Upon cooling a highly durable vitrified product is formed that consists of a mixture of glass and crystalline materials. The resulting product is typically ten times stronger than concrete and is extremely leach resistant. Organic contaminants such as dioxins, pesticides and PCBs are destroyed by the process. Heavy metals and radionuclides are retained in the melt and immobilized in the resulting product. The ISV process has been used commercially to successfully treat all contaminant types (volatile organics, semivolatile organics, heavy metals, and radionuclides) and in all types of soil media (sands, silts, clays and sludges). ISV is also distinguished by its ability to tolerate significant amounts of debris within the treatment zone. Types of debris previously processed by ISV in commercial operations include scrap metal, steel drums, concrete, asphalt, wood, plastic, paper, protective clothing, HEPA filters, automobile tires, and general construction demolition debris. Individual full-scale melts can typically accommodate tens to hundreds of tonnes of debris.

The overall Maralinga Rehabilitation Program involves two parts. The first part, which is complete, primarily involved the collection of contaminated surface soil at three different sites at Maralinga followed by burial in large pits which have been covered with five meters of uncontaminated soil. Over 500,000 cubic meters of contaminated soil have been collected and buried. The second primary part of the Programme involves the application of ISV to the Taranaki Pits.

Geosafe has been under contract for the past several years to the Commonwealth Department of Primary Industries and Energy to conduct a four phase project to use the ISV technology to remediate the Taranaki burial pits.

Phase 1, completed in 1994, involved an initial site assessment, site characterization activities, and treatability tests on soils from the Maralinga site.

Phase 2, completed in 1996, primarily involved a series of on-site treatability tests including two multi-ton intermediate-scale demonstrations involving uranium and plutonium-contaminated blast debris [2].

Phase 3 involves the design and construction of the full-scale ISV treatment plant. Phase 3 also included commissioning tests of the plant which have just been completed and the mobilization of the plant to the Maralinga site.

Phase 4 involves the actual vitrification operations to treat the 21 burial pits. Approximately 38 melts will be required to treat the 21 pits.

Phases 1 and 2 were conducted by Geosafe Corporation. Phases 3 and 4 are being conducted by Geosafe Australia Pty. Ltd., a Geosafe Corporation subsidiary established to carry out the full-scale operations. Geosafe has one major Australian subcontractor, AMEC Engineering Pty. Ltd., supporting both the Phase 3 and 4 engineering and operations requirements.

The Maralinga Site of South Australia is a former nuclear weapons test site used by the British in the 1950's and early 1960's for above ground testing. At Taranaki, Maralinga's most heavily contaminated area, a series of minor trials involving the explosive dispersal of plutonium and uranium resulted in extensive contamination of surface soil and generated substantial amounts of heavily contaminated debris [1]. Contaminated soil and debris from the trials were buried in a series of 21 shallow pits at Taranaki. Reinforced concrete caps were subsequently emplaced over the pits during a clean-up operation which was conducted in 1967. The pits range from approximately 3 to 4 m in depth and up to 10 m wide and 80 m in length.

For each of the minor trials, a steel structure, called a feather bed, was erected to hold the test device. The feather bed typically consisted of a heavy steel I-beam framework supporting ten 2-in thick steel plates, each plate 4 x 8-ft in size. Walls of baryte and lead shielding bricks encased in steel were also included in the design. Following each detonation, fragments and other related debris such as instruments and cables were dumped into each pit. Then the heavy steel debris was dismantled with the assistance of a crane and the steel debris and shielding walls were pushed into each pit using a bulldozer. A bitumen emulsion was sprayed over the adjacent surface soil and the mixture eventually bladed into the pit.

It is known that the Plutonium was dispersed as fine oxide dust, as sub-millimeter sized particles, and as surface contamination on larger fragments of debris [1]. The amount of plutonium in each of the pits is expected to be on the order of 100-200 GBq. The amount of Pu used for each of the Vixen B (VB) trials is well documented and typically ranged from a low of 1338 grams in a VB1 trial to a high of 2150 grams for one of the VB2 trials. The amount of Pu in each pit is estimated to be up to 20% of the original inventory based on measurements made at the time of each trial, from aerial surveys which were subsequently conducted to measure the activity in the plumes, and from studies of similar weapons trials conducted in the U.S.

During Phase 2, a series of on-site tests and demonstrations were conducted, including two intermediate-scale demonstration trials involving radioactive materials. The two radioactive trials involved the treatment of scaled pits filled with soil, steel debris and other debris including bitumen-stabilised soil, lead, plastic, electrical cable and barytes bricks. One kilogram of uranium oxide was buried in each pit to serve as a surrogate for plutonium. For each demonstration melt, the uranium oxide was contained in a plastic bag and located in the centre of the pit to serve as a highly localised area of contamination. The second radioactive demonstration included a steel plate, originating from the weapons tests, that was contaminated with approximately one GBq of plutonium oxide (predominantly ²³⁹Pu with about 3% being ²⁴¹Pu).

Following the two demonstration trials, the resulting monoliths were excavated for examination, weighing, and sampling. Figure 1 shows the excavation of the glass monolith. The mass of each monolith was approximately 4,000 kg. Results indicate that all demonstration objectives were met and that the process could tolerate the types and amounts of debris present in the pits. Health physics surveys of the equipment established that the insides of the off-gas containment hood and off-gas piping were free of detectable contamination above background levels (< 0.25 Bq alpha and beta combined per 100 cm²). Consequently, decontamination of the equipment was not required. Based on isokinetic off-gas sampling, it was determined that 99.99997% of the plutonium and 99.9998% of the uranium were retained in the melts [2]. Radiochemistry analyses showed that the radioactive materials were uniformly distributed in the vitreous products (due to the convective mixing currents that exist in ISV melts). The metal phase at the base of each melt resulting from the steel debris was determined to be free of plutonium and uranium based on qualitative analyses [3].

Long-term leach tests of the Pu-containing vitrified product have been underway for approximately two years [3]. The tests have been structured to cover a wide range of conditions and are based on Materials Characterization Center Method 1 (MCC-1) and Product Consistency Tests (PCT).

A series of leach tests were conducted on samples of the Pu-bearing vitrified product. Product Consistency Tests (PCT) were carried out on powdered specimens at 26 deg. C and 90 deg. C in deionised water. A series of PCT tests were also carried out using a pH 10 buffer at 90 deg. C as under these alkaline conditions, Pu releases are expected to be the highest. The PCT tests were carried out for 0-7 and 0-28 days.

Monolith samples were tested using Materials Characterization Center leach tests (modified MCC-1) at 90 deg. C in pH 7 buffer, in the pH 10 buffer, and in de-ionised water and also in di-ionised water at 26 degrees C. MCC tests were carried out for up to six months and other longer term tests are continuing at this time.

In all tests, concentrated nitric acid was used to strip activity from the vessel walls and the solution was diluted and combined with the leachate for analysis to provide a total activity.

The elemental releases in the leachates from the PCT tests indicate that the durability of the ISV product is outstanding. The critical releases are those of Na and Si, as these elements have been found to be the best indicators of glass quality. The data from the tests show that the releases were approximately two orders of magnitude below US-DOE acceptance criteria. Si releases for 0-7 days at 90 deg. C were 0.046 g m-2 and Na releases under those same conditions were less than 0.0087 g m-2.

The elemental releases from the MCC tests generally decrease with time. After six months, the leach rate for all elements was less than 0.1 g m-2 day-1 and in most cases approached 0.01 g m-2 day-1. As expected, the releases at 26 deg. C are approximately an order of magnitude less than the releases at 90 deg. C.

Leached and unleached specimens were examined using scanning electron microscopy (SEM) and Energy Dispersive X-ray Spectrometry (EDS). In all experiments, after leaching, it was determined that precipitates formed in situ on the wollastonite crystalline phases. In the pH 7 buffered system, it appeared that the precipitate was a mixture of several or more amorphous, phosphate bearing phases. In the pH 10 buffer and in de-ionised water, calcite precipitated on the wollastonite. The maximum depth of attack was only about 10 microns and this attack occurred on the wollastonite crystalline phases. Degradation of the glassy phase, where the plutonium was located, was relatively slight compared with the wollastonite phase. No Pu was detected in any of the leached phases or precipitates.

Phase 3, to be completed in early 1998, involves the design and construction of a full-scale ISV treatment plant. Phase 3 also includes establishing necessary infrastructure at Taranaki such as a 4.8 MW diesel powered electrical power generating plant.

Construction of the ISV treatment plant has been completed. Commissioning tests of the plant were successfully conducted in December of 1997 to verify that the plant satisfied design requirements. The acceptance testing involved an actual melt test of the entire system including the diesel power plant. A maximum power input of 4.2 MW was achieved during the test, which is the highest power input ever achieved for Geosafe's GeoMelt technologies. Figure 2 shows the ISV treatment plant being placed in operation during the commissioning tests.

The ISV system has been designed to meet the site specific technical and regulatory needs of the project. Because the site is located in the Great Victoria Desert, summer temperatures can often exceed 50 deg. C. In addition, because the surrounding vegetation was removed during the soil removal project, significant concerns exist regarding dust. Consequently, the ISV plant has been designed to accommodate the extremely hot dusty environment.

Because the only regulatory off-gas emission of concern is plutonium, and because any plutonium emissions are associated with particulates, a dry off-gas filtration system is being employed to treat the off-gas. The system consists of a novel single stage, high capacity, self-cleaning filter system that can tolerate off-gas temperatures in excess of 500 deg. C. The off-gas treatment system includes a series of continuous emission monitors including a system to monitor alpha emissions.

Because the off-gas generation rate will vary depending on the waste configuration found in each pit, a variable speed fan has been incorporated into the design of the off-gas treatment system. The variable speed drive fan is designed to normally operate in the range of 200 to 300 standard cubic meters per minute. The control system can automatically increase the flow rate up to 600 standard cubic meters per minute, which is approximately 12 times the typical flow rate used by ISV systems employed in the US.

The power generation plant for soil melting consists of three 1.6 MW Caterpillar diesel powered generators. The three 2000 HP generators are configured in parallel and are controlled by synchronizing and load sharing equipment to optimize their operation. Since there is no safety consequence related to a loss of any of these generators that supply power for soil melting, there are no back-up generators.

Power from the diesel generators for soil melting is converted from three phase-AC input to a four-phase AC output by a 5 MW scott-tee connected electrical transformer. The transformer is divided between two enclosures mounted on two separate trailers. Saturable reactors are used to dampen electrical transients that can be created from electrical short circuits caused by pooled metal in the melt. Although substantially larger and heavier, the saturable reactors are more compatible with the diesel generators compared to electronic silicon controlled rectifiers.

Power for all auxiliary equipment including process control, off-gas treatment, lighting and HVAC are provided by a 480 kW diesel powered generator. An uninteruptable power supply (UPS) provides back-up emergency power for critical systems in case the auxiliary power generator fails. The UPS provides power to critical systems to allow operating staff time to either fix the problem or to bring the plant to a secure stand-by condition. In addition, a 150 kW generator is available to provide back-up power in case of an extended outage.

Two off-gas containment hoods are being employed for the project. Two hoods are used to increase the efficiency of the operation. While one hood is being used to support a melt operation, the second hood is prepared for the next melt. The hoods are of a modular design and can be quickly disassembled and reassembled. The hoods are moved on-site in pieces, which eliminates the need for a large 100+ tonne capacity crane. A hydraulic electrode feed system which provides greater lifting and clamping strength and is well suited to the hot dusty environment is being employed for the project.

For the vitrification operations, it is expected that approximately 7 million liters of diesel fuel will be required. Diesel fuel storage is provided by a 220,000 liter tank farm. The fuel consumption rate is expected to vary between 700 to 1100 liters per hour depending on the size of the pit being treated. Fuel will be trucked over 800 km to the site. The last 200 km will be over rough dirt roads. Specially built steel tankers are being used to deliver fuel to the site since standard aluminum tankers can't tolerate the stress from the rough dirt roads. Fuel shipments will be required at a frequency of one shipment every two to three days during melting operations.

The Phase 4 project involves site preparation activities followed by the ISV treatment of the Taranaki pits. Approximately 38 melts will be required to treat the 21 pits over a period of about two years. Vitrification operations are scheduled to commence in early 1998.

As part of the site preparation activities, which are currently on-going, most of the 300 mm thick reinforced concrete caps covering each of the pits are being removed. Although a few concrete caps will be left in place and treated along with the burial pits, most of the caps will be removed so that pit boundaries can be verified. When originally emplaced by the British, the caps were not always centered over each of the pits and, in some cases, the caps were made too small, leaving large areas of the pits uncovered. The cap removal operation is significant because it involves substantial concrete cutting in a radiologically contaminated environment, removal of 15-20 tonne sized cap pieces by crane, and requires the handling of highly contaminated soil under the caps in order to determine the pit boundaries. Figure 3 shows the cap removal operations in progress.

The cap removal activities also include pit preparation activities involving the dynamic disruption technique. This technique involves the use of a 15-cm diameter heavy steel probe, about 4 m in length that is attached to a hydraulic hammer attachment mounted to a tracked excavator. Initially, a 300 to 600 mm thick layer of clean soil is emplaced over the pit. The probed is then driven downward through the cover soil into the pit in a grid pattern. The insertion and hammering process is designed to collapse large voids and rupture any sealed drums that are present in the pit. When a large void is encountered, clean fill from above sloughs down into the pit to fill the void. As this occurs, additional cover soil is added. Variations of this dynamic disruption technique have been successfully employed on other remediation projects by Geosafe Corporation. To date the cap removal operations and dynamic disruption activities have been conducted without the spread of any contamination. Figure 4 shows dynamic compaction of a typical pit.

For the vitrification operations, it was initially projected that only 26 melts would be required to treat the 21 pits. However, during the soil removal operations wherein contaminated soil around the pits was collected and buried, it was determined that almost all of the pits were much larger than indicated in the historical records. In most cases, the volume of the pits are several times larger than expected. In a few cases, the pits are 10 to 25 times larger than expected

Cap removal operations are progressing at the site and will continue for much of 1998. To date, three caps have been removed. At this time, equipment and supplies are being prepared for mobilization to the site. Vitrification operations are expected to commence later this Spring with an expected duration of approximately two years.

To provide a means to determine how much plutonium is contained within selected pits, samples of the vitrified product will be analyzed to determine the plutonium content. Prior studies have shown that the melts are well mixed due to convection currents that exist in the melts. Consequently, it is expected that the resulting plutonium concentration will be relatively uniform throughout each melt. Cerium oxide and lanthanum oxides tracers will be incorporated into selected melts by adding the tracer to the uncontaminated soil overburden above each pit. Cerium and lanthanum are good surrogates for plutonium because they are stable as oxides in the melt environment and are incorporated into the vitreous product as opposed to evolving to the off-gas treatment system or partitioning to other phases, such as the reduced metal phase at the base of the melt. The total mass of tracer used in each melt and the concentration of the tracer in the resulting vitrified product sample can be used to precisely determine the mass of the melt. Then, the mass of the melt and the plutonium concentration in vitrified product samples can be used to determine the amount of plutonium that was initially in each pit.

Results to date show that ISV will satisfactorily treat the contaminated soil and debris in the pits. The project is presently on schedule and budget. Phase 3 will be nearly completed and final preparations will be underway for Phase 4 at the time of the conference. At the conference, details will be provided concerning the preparations leading up to the initial Phase 4 treatment operations .

1. Department of Primary Industries and Energy (1990) 'Rehabilitation of Former Nuclear Test Sites in Australia, Report by the Technical Assessment Group' Commonwealth Government Publishing Service, Canberra, ACT
2. L.E. Thompson, Dr. J.M. Costello (1996) 'Vitrification of TRU Contaminated Buried Waste: Results from Radioactive Demonstrations at Taranaki' Waste Management '96, Tucson Az, USA
3. P.J. McGlinn, K.P. Hart, R.A. Day, J.R. Harries, J.A. Weir, & L.E. Thompson (1997) 'Scientific Studies on the Immobilisation of Pu by ISV in Field Trials at Maralinga, Australia' Materials Research Society '97. Pittsburgh PA, USA