Guy G. Loomis
Idaho National Engineering and Environmental Laboratory
Lockheed Martin Idaho Technologies Company
P.O. Box 1625
Idaho Falls, Idaho 83415-3710
(208) 526-9208, (208) 526-5142 (fax)

Christopher M. Miller
University of Akron

Stephen W. Prewett
Orbit Technologies (Carlsbad, California)

A proof-of-concept experimental study was performed to investigate the use of Orbit Technologies' polysiloxane grouting material for encapsulation of U.S. Department of Energy mixed waste salts leading to a final waste form for disposal. Evaporator pond salt residues and other salt-like materials contaminated with both radioactive isotopes and hazardous components are ubiquitous in the Department of Energy complex and may exceed 250,000,000 kg. Current treatment involves mixing low waste percentages (less than 10% by mass salt) with cement or costly thermal treatment followed by cementation of the ash residue. The proposed technology involves simple mixing of the granular salt material (with relatively high waste loadingsCgreater than 50%) in a polysiloxane-based system that polymerizes to form a silicon-based polymer material. This paper presents the results of a mixing study to determine optimum waste loadings and compressive strengths of the resultant monoliths. Next, the results of durability testing such as wet/dry cycling and base and water immersion effects on compressive strength are presented. Also presented are results of leaching studies including the accelerated leach test and the toxicity characteristic leaching procedure applied to the resultant waste forms. In addition, results from Department of Transportation oxidizer testing on a high nitrate salt waste form are presented. Finally, the waste form was examined using a scanning electron microscope. Preliminary cost estimates for applying this technology to the Department of Energy complex mixed waste salt problem are also given.

A proof-of-concept experimental study was performed to investigate the use of Orbit Technologies' polysiloxane grouting material for encapsulation of U.S. Department of Energy (DOE) mixed waste salts leading to a final waste form for disposal. The DOE salts are basically evaporator pond material or other solid salt residuals that resulted from neutralizing acid-dissolving processes. The use of polysiloxane for encapsulating DOE salts is novel; however, it has been proposed that polysiloxane be used for helping fill void space in the sarcophagus at the Chernobly plant to eliminate subsidence. As part of this Chernobly work, polysiloxane has been shown to be highly radiation resistant (ref 1). More than 254,000,000 kg of mixed waste salt material is stored throughout the DOE complex as part of the weapons complex operations. The present best available technology for stabilizing this waste results in a large volume increase (mixing with cements at extremely low waste loadings [10% by weight maximum loadings]). Another option for the salt materials is thermal treatment. These thermal techniques involve expensive off-gas systems; therefore, application of a nonthermal, noncementatious technique has the potential to reduce cost and augment implementability.

The experimental study involved mixing the polysiloxane material with three different surrogate salt materials and performing a variety of leaching, compressive strength, and durability tests on the final surrogate waste forms. The surrogate salts were modeled after several DOE complex salt materials including Pad-A salts from the Idaho National Engineering and Environmental Laboratory's (INEEL) Subsurface Disposal Area at the INEEL's Radioactive Waste Management Complex and two generic surrogates suggested by the DOE Mixed Waste Focus Area (MWFA): one high nitrate and one high chloride with several heavy metals as contaminants.

The surrogates did not have radioactive components; however, heavy metal components were included. The Pad-A salt was approximately 88% nitrate salts with chromium (potassium dichromate at 1400 ppm and chromium trioxide at 800 ppm) as the main contaminant. The original source of the Pad-A salts was the Rocky Flats Plant. Other primary constituents were sodium sulfate at over 5% and sodium chloride at approximately 3%. The MWFA high chloride salt consisted of a mixture of salts, microCel E, Portland cement, water, and sodium chloride with heavy metal oxides at 1000 ppm (lead, chromium, mercury, cadmium, and nickel). The MWFA high nitrate salt consisted of 60% nitrates with similar cements and absorbing agents (microCel E) as with the high chloride salt. The identical heavy metal oxides were also used.

There was also 1000 ppm trichlorethylene added for both of the MWFA salts. For the exact constituents of the Pad-A salt surrogate and the MWFA salt surrogate, see Mixed Waste Salt Encapsulation Using PolysiloxaneCFinal Report (ref 2). Multiple versions of Orbit Technologies' polysiloxane material were tested; however, only results of the most promising version are reported in this paper. This material is referred to as Ceramic Silicon Foam #2 (CSF#2).

The study involved the following tests and analyses of the waste forms: (1) a mixing study to optimize the waste loading (salt mass to polysiloxane mass) relative to basic compressive strength of the monolith; (2) laboratory hydraulic conductivity of the sample; (3) durability testing including effects on compressive strength of wet/dry cycling, base immersion, and water immersion; (3) leachability (toxicity characteristic leaching procedure [TCLP]), (4) Department of Transportation oxidizer testing, and (5) evaluation of volatile organic content (zero head space testing). Finally, the samples were subjected to a scanning electron microscope and evaluated for encapsulating properties.

Results for the Pad-A surrogate and the MWFA high nitrate and high chloride are presented.

The Orbit Technologies polysiloxane mixture CSF#2 with Pad-A salts resulted in a cohesive monolith with compressive strength well in excess of the proposed 500 psi Nuclear Regulatory Commission (NRC) requirement for cemented waste for up to 50% waste loadings. The polymerization process (silicon chains) is essentially nonexothermic. In addition, even with waste loadings as high as 65% by mass, the current NRC limit of 60 psi was achieved. For both the high chloride and high nitrate MWFA-suggested surrogate material, monoliths with compressive strength higher than 637 psi could be formed at 30% by mass when using a mixture of CSF#2 polysiloxane and salt. In addition, when formed at 50% by mass, the monoliths had compressive strengths greater than 420 psi. These monoliths, which could easily be poured into waste form shapes for disposal in shallow land burial, were elastic in nature and therefore not prone to brittle cracking in the event of ground movement such as earthquakes. In addition, the base polysiloxane material has a high enough viscosity that particulate waste material such as granular salt remains suspended prior to polymerization (curing).

The basic stabilized waste form consists of polysiloxane, polymer catalyst, and surrogate salt (Pad-A or MWFA waste). Depending on the waste loading, the appropriate amount of polysiloxane, catalyst, and waste was measured to the nearest 0.1 g. Mixing of all samples took place in 1-gal plastic buckets using a hand-held drill with a stainless steel mixing paddle. The order of addition was polysiloxane, salt waste, and catalyst. The mixture was thoroughly stirred and then transferred to plastic test cylinders (usually 2 in. in diameter by 4 in. tall). The mixture was allowed to cure for at least 24 hours before compressive strength testing. A waste form's mechanical integrity and ability to withstand loading pressures in a disposal environment are directly related to compressive strength. Compressive strength is also used as a gauge of the resistance of samples that have undergone durability tests. For each of the surrogates, five replicate samples at 30% and 50% waste loadings were tested for compressive strength. The Geotest Instrument load cell apparatus used for this study has a maximum load capacity of 2000 lb, which results in a maximum compressive strength of 637 psi for a 2 in. diameter by 4 in. tall sample. The American Society for Testing and Materials (ASTM) D-695 technique was applied.

A laboratory-scale study of the final waste form showed that the Pad-A surrogate salt waste mixed with the polysiloxane system at 50% waste loading was an order of magnitude less than the NRC hydraulic conductivity requirement of 1e-7 cm/s. A hydraulic conductivity test using ASTM D-5084 resulted in an average hydraulic conductivity for the CSF#2-based waste form of 6.4 e-8 cm/s. Hydraulic conductivity is a direct measurement of the potential ability of a waste form to leach out contaminants. Test samples (50% waste loading) were prepared using 3 in. diameter by 6 in. tall cylinders. According to ASTM D-5084, test samples were placed in a flexible wall permeameter, sealed with a latex membrane, and filled with de-aired water at a cell pressure of 5 psi. Once the cells were stabilized, the cell pressure was set at 96 psi, the lower cap at 93 psi, and the upper cap at 90 psi. A hydraulic gradient of 50 in./in. was used throughout the tests. Using a falling head constant tailwater method, the flow of water through the samples was recorded versus time using a graduated pipette, and the hydraulic conductivity calculated.

Durability testing showed that the waste form created from the CSF#2 Orbit formulation mixed with the Pad-A surrogate high nitrate waste material was unaffected by wet/dry cycling and 30-day water and base immersion. A variety of durability testing was performed to investigate the effects on compressive strength (using ASTM D-695) due to water immersion, wet/dry cycling, and base immersion. For the water immersion testing, CSF#2 material remained above the upper limit of the testing apparatus at 637 psi. For the wet/dry cycling testing, the waste form created with CSF#2 material showed only a 1.22% mass loss and again a compressive strength value above the maximum 637 psi for the apparatus. When subjected to base resistance (pH 12.5 for 30 days), the waste form created using CSF#2 material showed again a compressive strength above the maximum value obtainable of 637 psi.

The waste form created by mixing polysiloxane and the Pad-A salt surrogate showed excellent resistance during TCLP testing. At 30% waste loadings, the waste form involving CSF#2 material actually meets the current and proposed Universal Treatment Standard (UTS) TCLP values of 5 ppm and 0.86 ppm, respectively. The Pad-A surrogate material had approximately 1045 ppm of the highly soluble Cr⁺⁶ (1400 ppm potassium dichromate and 800 ppm chromium trioxide), and, when subjected to TCLP testing at the University of Akron, the results were 1.36 ppm for the 30% waste loadings and 5.6 ppm for the 50% waste loadings.

Additionally, samples of waste forms based on CSF#2 at both 30% and 50% loadings were evaluated by an independent laboratory (Ecology and Environment, Inc.) at 2.4 ppm and 5.9 ppm, respectively (basically the same values as the University of Akron). At 30% waste loadings, the waste form meets current TCLP requirements, and, when considering that the source term for Cr⁺⁶ in the Pad-A salts was 180 ppm (approximately 5.8 times that found in the actual waste), it is almost certain that the waste form will also meet the proposed UTS. It is speculative at this time whether the proposed UTS can be met for the 50% waste loadings; however, the fact that the source term is higher for the surrogate than the actual Pad-A salts suggests that the current standards can also be met (5 ppm). However, metal scavengers may be required at the higher waste loadings (greater than 30%) to meet the proposed UTS. Recent studies have shown that the scavengers can be introduced without affecting the matrix and that the TCLP results are greatly improved.

The waste form generated by mixing the suggested MWFA surrogate and CSF#2 also passed TCLP testing for a variety of metals and were all below proposed universal treatment standards. Samples of CSF#2 mixed with the MWFA high nitrate and high chloride salts were sent to Ecology and Environment, Inc. (Lancaster, New York) for TCLP testing (USEPA SW-486) of cadmium, chromium, mercury, and lead. Table I shows the results for CSF#2 with 30% and 50% waste loadings. All samples showed levels under current Resource Conservation and Recovery Act (RCRA) regulatory limits.

For the MWFA salt material, all hazardous metals were below current and proposed guidelines, except for one test reported in Table I. The exception is that for 30% waste loading, the chromium TCLP value was 0.36 ppm for the MWFA high nitrate waste loading, which is below the proposed UTS of 0.86 ppm. However, for 50% waste loading, the value was 1.3 ppm, which is slightly above the proposed UTS. This is confusing in that the high chloride waste material was below the proposed UTS for chromium for both 30% and 50% waste loadings, and the physical characteristics of the salt material were essentially the same. It is possible that the trichloroethylene loading in the surrogate material affected the high nitrate salt material differently than the high chloride material or alternatively that the chromium behaves differently in the presence of nitrate rather than chloride salts. It is particularly puzzling in that, for the high nitrate salt-based Pad-A surrogate with 1045 ppm Cr⁺⁶ (highly soluble), the leaching for TCLP was on the same order of magnitude as for the much less soluble chromium oxide used in the MWFA surrogate for the same waste loading.

A "zero head space" evaluation for trichloroethylene was made of the waste form created with the MWFA surrogate high chloride salt and CSF#2. The result was nondetect for trichloroethylene even though the source term in the surrogate was 1000 ppm.

A mixture of the Pad-A waste and polysiloxane material CSF#2 passed the Department of Transportation oxidizer test in that a mixture of sawdust and the waste form material did not burn after 30 minutes compared with the surrogate salt material, which burned under similar conditions in 30 seconds.

The basis of this test is to compare a test substance and three reference substances with regard to their ability to increase the burning rate or burning intensity of a combustible solid. According to 49 CFR 173.127, samples must be able to pass a sieve size of 0.3 mm and are mixed with sawdust at a 1:1 mass ratio. Testing was carried out by Hark Laboratories, Inc. (Barberton, Ohio). Each mixture was arranged in a conical pile and a wire loop placed inside the pile. The wire was then heated to 1000° C until the first sign of combustion or until it was clear that the mixture would not ignite. The time for combustion was then recorded for each substance. Based on favorable compressive strength and leaching results, only CSF#2 was tested. At 30% waste loading, the CSF#2 material did not burn after approximately 30 minutes, while the surrogate salt waste had a burn time of approximately 30 seconds. Based on these results, the salt waste would be placed in Packing Group I, while the stabilized CSF#2 waste form would not have a specific packing requirement as an oxidizer, i.e., it did not burn.

Scanning electron microscope studies show qualitatively that the simple mixing of polysiloxane material, catalyst, and salt causes a general encapsulation of the salt particles. There is some change (reduction) in the overall size of the salt particles following the encapsulation process, suggesting some unknown chemical recombination as part of the polymerization process. The average size of the untreated salt particles is approximately 150 microns, and, when mixed with the polysiloxane, the size appears to be on the order of 50-100 microns with some particles in the original 150 micron range. It is not clear whether the mixing action breaks up the original salt size or whether there is a dissolution of the salt and incorporation into the polysiloxane matrix.

A preliminary cost estimate shows that the Orbit Technologies polysiloxane system has a cost savings over concrete of at least 3:1, with the added benefit of a superior resultant waste form. Details of this cost savings basis are found in Mixed Waste Salt Encapsulation Using PolysiloxaneCFinal Report (ref 2). Basically, the cost of material is outweighed in the cost equation by the volume reduction. Therefore, disposal cost reduction is achieved by using the polysiloxane system. Once there is a market for the base polysiloxane material, this cost savings could increase.

Overall, the process of mixing the salts with Orbit Technologies' polysiloxane-based systems results in stable nonleachable waste forms suitable for disposal. The process of mixing uses only off-the-shelf equipment and has an easily controllable curing process. The final waste form can be poured into suitable shapes for shallow land burial, and the base polysiloxane material is high enough in viscosity that waste particles remain suspended prior to curing. The final product is elastic, such that when poured into waste forms for shallow land burial sudden earth movement caused by earthquakes or other shifting would not tend to cause brittle cracking. When considering the relatively high waste loadings (durability and leach testing were performed up to 50% by mass), the process appears cost effective even though the cost of the base material may be as high as $8/lbm.

Examination of outside laboratory TCLP results for the waste form created by CSF#2 and the Pad-A salt surrogate shows that the waste form passes current limits for chromium of 5 ppm unconditionally for 30% waste loadings and most likely for 50% waste loadings when considering that the source term for the surrogate was a factor of 5.8 higher than the actual Pad-A waste (the surrogate had 1045 ppm Cr⁺⁶ [1400 ppm potassium dichromate and 800 ppm chromium trioxide] and the actual Pad-A salts only had a source term of 180 ppm Cr⁺⁶). It is most likely that the 30% waste loading case in the actual Pad-A salts will also pass the proposed UTS of 0.86 ppm chromium, because the proposed UTS is only a factor of 2.7 lower than the value achieved for the surrogate. Yet, the surrogate had a factor of 5.8 higher Cr⁺⁶ loadings than the actual waste. Nevertheless, for the higher waste loadings (greater than 30%), application of polysiloxane may require a metal scavenger like iron oxide or phosphates in the matrix to meet the UTS, which has been shown to not affect the overall curing process.

It is recommended that the retrieved Pad-A salts presently stored at the INEEL Waste Reduction Operations Complex be sent to the University of Akron for encapsulation, testing, and eventual disposal at a permitted facility as a "hot" demonstration of the whole process.

It is further recommended that the technology be investigated for application beyond mixed waste salts. Specifically, this technology could be applied to stored transuranic waste, alpha mixed low-level waste, or retrieved buried transuranic waste. A waste form suitable for shallow land burial could be created by encapsulating shredded waste material and placing the material in the same 4 by 4 by 8-ft polyethylene boxes for interim storage or disposal.

1. I.K. SHVETSOV et al., "Determination of the Structural Peculiarities of the Material and the Components Distribution," Cerametalic Silicone Foam Rubber Research Programme, Kurchatov Institute, (April 1994).
2. G.G. LOOMIS et al., "Mixed Waste Salt Encapsulation Using PolysiloxaneCFinal Report," INEEL/EXT-97-01234, (November 1997).

BACK