Alternative High-Level Waste Treatments at the Idaho National Engineering and Environmental Laboratory (1999)

Chapter 4 briefly discussed the proposed treatment options for sodium-bearing waste (SBW), and noted how classification and disposal restrictions were important considerations affecting the treatment method and final waste form. Guided by these ideas, the committee presents in this chapter its views of what should be done with the SBW, in a discussion similar to that presented in Chapter 11 for the HLW calcine.

Because of the potential for tank leakage, a study of options should be made promptly to identify ways to stabilize the liquid SBW into solid waste forms that are suitable for disposal or, if necessary, interim on-site storage pending disposal or further processing. In this effort, the committee's preferred general approach for treating SBW is to convert it to a solid, low-level, remote-handled transuranic (RH-TRU) waste to be shipped to a TRU repository designed to handle it [e.g., the Waste Isolation Pilot Plant (WIPP)¹] or, if necessary, stored on-site. Treatment methods suitable for this purpose may also generate a non-TRU low-level waste (LLW) product that can be shipped to a LLW disposal site or, if necessary, stored or disposed on-site.

Five candidate SBW treatment methods are described below and are consistent with Department of Energy (DOE) program plans to empty liquid from all tanks by 2012. These methods are low-temperature (< 150 °C) and involve evaporation, simple chemical adjustments such as pH adjustment or addition of precipitants, and in some cases, solid-liquid separation. These methods do not involve calcination, vitrification, or complex multistage separation processes. The first four candidate methods involve evaporation and hydroxide precipitation; the fifth accomplishes TRU (and also much of the Sr) precipitation by sequential lanthanum fluoride precipitation.

In contrast, a sixth solidification treatment method, calcination in a repermitted calciner, is a thermal process producing a calcine waste form. If SBW is made into calcine, this calcine should not be mixed with the existing HLW calcine, principally because this changes its waste classification and thereby restricts eventual disposal options. While the calcination of

SBW and addition of the resulting calcine to bins of HLW calcine may be straightforward technically, this procedure creates additional inventory of HLW by mixing. HLW is the waste category for which final disposal options are most limited and restrictive.

Six examples of possible processes are given below. They vary as to (1) whether they develop two separate waste fractions, one a TRU waste and the other a non-TRU LLW; (2) how this separation is done, if at all; and (3) the relative amounts of the two fractions. DOE research and development efforts over the next few years should focus on developing sufficient information (much of which already exists at other DOE sites in the form of treatment options that can be adapted for use on the SBW) for a future decision to select a satisfactory flowsheet to process the SBW.

This option is conceptually the simplest. The SBW would be retrieved and evaporated to dryness or near-dryness (about 120 °C) such that the slurry concentrate would solidify when cooled. Water and nitric acid could be recovered from the overheads. The concentrate, containing all the radioactivity and salts in the SBW, would be solidified in drams or other appropriate containers and either shipped to a suitable TRU repository as RH-TRU waste (preferred) or stored on site. The evaporation step would probably be a two-stage operation, using a thermo-syphon first stage to concentrate the SBW and a wiped-film evaporator to carry it to a concentrated slurry product that would solidify on cooling.

The free-water content can be controlled by adjusting the temperature of evaporation (i.e., the boiling point). The slurry should be evaporated to the point that it contains less water than will be tied up as water of crystallization in the various salts present after cooling. As a result, there would be no free water. The slurry product would essentially be a salt cake and could be made into various physical forms, such as a monolithic salt cake or a powder.

The primary technical problem with this approach is probably evaporator and storage container corrosion with the acid system containing fluoride. If this is serious, one of the other options listed below should be used. Another disadvantage of this option is that all salts in the SBW go to the final waste, thereby generating a large volume of RH-TRU solid waste. The major advantage is that the only unit operation is evaporation.

The excess acid in SBW would be destroyed by addition of formic acid during an evaporation-concentration step with a conventional evaporator. The concentrate, containing little excess acid, would be neutralized with NaOH to a pH of about 8 to 10, precipitating most of the polyvalent metals (mostly Al) present. The slurry would then be evaporated and solidified for disposition as in Option 1. There has been considerable study of this method as well as other approaches at the Oak Ridge National Laboratory (ORNL) in managing neutralized waste reasonably similar in composition (McNeese et al., 1998).

The possible corrosion problem in Option 1 is due to evaporation of acidic fluoride solutions and would be largely circumvented in Option 2. Both of these options have the disadvantage that all the salts in the waste are converted to RH-TRU waste and must by disposed of in a suitable way.

As in Option 2, addition of formic acid, evaporation, and neutralization with NaOH would be done, producing a slurry. The slurry, after neutralization, would be treated for metals regulated by the Resource Conservation and Recovery Act (RCRA) using best available technology (probably sulfide addition), and then would go to a solid-liquid separation (SLS) step. The TRU, Al, other insoluble metal hydroxides, along with RCRA metals and much of the Sr would go to the solids fraction. The bulk of the salts (Cs, Na, K, and nitrates) would remain in solution. The solids fraction would be dehydrated, convened (e.g., by grouting) to a monolithic solid, if required to meet transportation or repository acceptance criteria, and shipped to a suitable TRU repository as RH-TRU waste or stored on-site. The liquid fraction would be evaporated, probably in two stages as in Option 1, with the wiped-film evaporator product being a solid non-TRU LLW that would be stored on-site or shipped to a suitable LLW repository as a Class C waste. There is considerable evidence from work at other sites with similar wastes that the supernate indeed would be non-TRU.

The primary disadvantage of Option 3 is the use of a somewhat difficult SLS process to recover a fairly large amount of precipitate. The advantage over Options 1 and 2 is that less TRU waste is generated.

This is similar to Option 3 except that more NaOH is added to the SBW to increase the hydroxide concentration enough to redissolve the Al. In this case, the SLS step is simpler because the amount of solids is smaller. Again, the solids contain the TRU and insoluble metal hydroxides including RCRA metals, and much of the Sr. The solution contains the soluble salts plus Al, and would be non-TRU LLW. The solids fraction would be evaporated and solidified to a form suitable for disposal in a TRU repository as RH-TRU waste, as in Option 3, but the quantity of solids would be smaller. The supernate would also be evaporated and solidified for storage or disposal, as in Option 3, but the quantity would be larger because of the Al and added NaOH.

The disadvantage here is that more NaOH is added, thereby increasing the amount of LLW. The advantage is that the TRU waste volume is smaller than for Option 3. Option 4 is analogous to the "enhanced sludge leaching" process at Hartford. The overall process is also quite similar to the method planned for disposition of a similar waste at ORNL under a recent privatization contract (Brass, 1998; DOE, 1998c).

The SBW would be processed in fluoride solution by addition of lanthanum. Multiple LaF₃ scavenges would precipitate a TRU fraction as an insoluble fluoride. The precipitate would be a high-density solid containing the rare earth elements and a large fraction of the Sr, and could be further processed, if necessary, into a more durable waste form. In this process, excess fluoride can be removed from the supernatant solution by adding sodium borate and boiling to drive off the excess fluoride as BF₃, if other fluoride complexants (e.g., Zr and Ca) do not interfere.² The lanthanide and actinide solubility product is so small that the SBW

could be converted to a non-TRU product with only a few LaF₃ scavenge cycles, assuming adequate mixing and precipitation at 85 °C and fast settling of the dense fluoride precipitate.

If the calciner can be modified to satisfy current regulatory requirements, another solidification option is to calcine the SBW (with additives such as aluminum nitrate [AI(NO₃)₃], and store the resulting solid calcine in separate, RCRA-approved storage facilities. This option does not mix the new calcine with the HLW calcine. One disadvantage is that it requires operation of the calciner, and hence development work to repermit it to satisfy applicable U.S. Environmental Protection Agency regulations. Another disadvantage is that, compared to the other options, the solid form produced here is relatively harder to redissolve if future liquid-phase separations are needed.

A more comprehensive examination of options such as 1 through 6 above is clearly required, based on a review of the extensive literature and relevant experimental work. At that point, a comparative evaluation could be used to rationally select the best approach. The choice would likely be influenced strongly by the availability of repositories that would accept the different waste products. If a suitable TRU repository is available, then options such as 4 and 5 above have appeal in that they provide relatively small RH-TRU fractions for disposal in this repository, albeit with added complexity in the processing steps. If a suitable TRU repository is not available for the TRU fraction of solids developed from the SBW, then it appears that at least the TRU fraction must remain on site. Nevertheless, the committee's position is to convert the liquid SBW into a solid form, preferably one that is very likely to be suitable for shipment to a future suitable TRU repository, if and when one becomes available.

This report recommends separate treatment of SBW and HLW calcine waste streams. This approach could be challenged. For example, if vitrification or cementation is to be performed, mixing the SBW (containing the sodium-fluxing agent) and HLW calcine (containing refractory elements) would serve to reduce the final waste volumes compared to the separate vitrification or cementation of the two waste streams. In response to this challenge, Chapters 4, 9, and 12 of this report note potential advantages (i.e., an opportunity to use non-HLW disposal options and more simplified treatment options) if the SBW is segregated from HLW and treated separately. These are benefits that can be assessed and quantified in the near term.

If the approach of this chapter is adopted, the final waste form for SBW need not be a vitrified one, and the SBW need not go to a HLW repository. Combining the SBW and HLW calcine would force the SBW to become classified as HLW via mixing, thereby restricting disposal options.³ The committee's proposal to remove the SBW from the HLW stream achieves greater HLW volume reduction than the specific proposal in the preceding

paragraph.⁴ Moreover, if the SBW were solidified as a saltcake and stored on site, this waste could be later added to the HLW calcine if this is ever deemed desirable.