HNF-3233

Analysis of SX Farm Leak Histories --Historical Leak Model (HLM) August 1998

V. Description of HLM Spreadsheets

Appendix A shows the plots of the fill, thermal, and heat load histories of these tanks as well as the spreadsheets that were used to generate the plots. Also shown in each plot is the calculated volume of each tank in the absence of any leak. The criteria used to define each leak basically depended on the comparing of unaccounted volume loss rates and adjusting the size and duration of each leak in order to reduce those unaccounted volume losses.

Figures A1, A3, A5, A7, and A10 shows reported tank volumes, calculated tank volumes with leak, calculated tank volume without leak, calculated heat load of the tank, and reported tank temperature. All of this information is shown on a month-by-month basis in the HLM although much of the historical information that we now have is tabulated quarterly and sometimes only semi-annually.

Figures A2, A4, A6, A8, A9, and A11 all show the volume rates for evaporation, leak, and unaccounted gain or loss. These plots illustrate the means of calibrating the leaks using unaccounted volume loss information.

VI. Leak inventories

Given the leak volumes, starting times, and ending times, we calculate the composition leaked in terms of R1, R2, RSltCk, and other HDW types. Given the composition of each of these wastes, we then derive leak inventories for each chemical and radionuclide. As tank wastes were concentrated and blended, the compositions changed somewhat from the original wastes that were placed in each tank. These changes have been calculated by the SMM (Supernatant Mixing Model), which assumes an initial sludge/supernatant partitioning and propagates partitions of each waste type expressed in kgal’s of original waste.

We also calculate the amount of Cs-137 placed into each waste tank directly from fuel element processing records. This provides a parallel estimate of the Cs-137 that leaked from each of these tanks independent of the blending approximations inherent within the HDW model. These estimates are shown in Table 5 in total MCi (1 MCi = 1e6 Ci = 3.7e4 Tbq). The difference in the HLM and HDW model’s Cs-137 totals is due to the approximations of the HDW model. The HDW model defines wastes that are blended composites of entire campaigns whereas the HLM has placed each batch of fuel into tanks on a month-by-month basis. Therefore, the soluble radionuclides should all be corrected with the factors shown in Table 5.

For example, Tc-99 for tank SX-108 is 1.07e-4 Ci/L, corresponding to a leak inventory of 203 kgal x 3785 L/kgal x 1.07e-4 = 82 Ci Tc-99 by the HDW estimate. For the HLM estimate, multiply by 1.6 to get 131 Ci Tc-99. This assumes that Tc-99 behaves like Cs-137 in the tank liquors.

Table 4. Chemical and Physical Characteristics of HLM Leaks.

Table 5. Radionuclide Concentrations for HLM Leaks.*

VII. Uses and Limitations:

Clearly, there are limitations to the HLM for leak determination and there are also limitations in every approach that has been used in the past for leak determination. The HLM is only appropriate for aging waste tanks where water evaporation is cooling the tank waste. This limits its use to SX, A, AX, and possibly S and BY farms as well. Second, the HLM is only able to discern leak rates on the order of 0.5 kgal/mo or greater and it really is meant to record trends of volume lost over many months of operation.

The HLM has assumed a constant conductive heat loss for each tank that is scaled with tank temperature relative to seasonal average ground and air temperatures. Therefore, seasonal variations in temperature are not taken into account and these are appreciable for heat loss to air. Seasonal variation of ground temperatures is much less and therefore not expected to be a substantial source of HLM variation. Likewise, diurnal variations in temperature are not expected to have an impact on tank heat loss variation. The waste tanks are very large and therefore simply do not respond to temperature variations on the time scale of one day.

These four tanks were chosen for the HLM analysis because they had already demonstrated a substantial amount of contamination to the soil column by their associated drywell activities. The real question for these tanks was not if they leaked, but rather how much they leaked. The HLM confirms that these tanks have unaccounted volume losses and uses those unaccounted losses to extend the tank leaks over periods of operation where leak determination was very difficult by any other means.

The partitioning of radionuclides among tanks by subsequent transfers of waste supernatant is modeled only very crudely by the HLM. We have not evaluated the effect of the partitioning assumption on the leak estimates.

Acronyms, Abbreviations, and Definitions

References:

WSTRS is Agnew, S.F.; Corbin, R.A.; Duran, T.B.; Jurgensen, K.A.; Ortiz, T.P.; Young, B.L. "Waste Status and Transaction Record Summary (WSTRS Rev. 4)," LA-UR-97-311, April 1997. Also see WHC-SD-WM-TI-614, 615, -669, -689, Rev. 2, September 1995.

Agnew, S. F. "Hanford Defined Wastes: Chemical and Radionuclide Compositions," LA-UR-94-2657, Rev. 2, September 1995.

Agnew, S.F.; Boyer, J.; Corbin, R.A.; Duran, T.B.; FitzPatrick, J.R.; Jurgensen, K.A.; Ortiz, T.P.; Young, B.L. "Hanford Tank Chemical and Radionuclide Inventories: HDW Model Rev. 4," LA-UR-96-3860, January 1997.

Allen, G. K. "Estimated Inventories of Chemicals Added to Underground Waste Tanks, 1944 through 1977," ARH-CD-6108, March 1976.

Anderson, J. D. "A History of the 200 Area Tank Farms," WHC-MR-0132, June 1990.

Brevick, C.H.; Gaddis, L.A.; Williams, J.W. "Historical Vadose Zone Contamination of S and SX Tanks Farms," WHC-SD-WM-ER-560, Rev. 0, November 1996.

Hanlon, B. M. "Tank Farm Surveillance and Waste Status and Summary Report for November 1993, "WHC-EP-0182-68, February 1994, published monthly.

(a) Jungfleisch, F. M. "Hanford High-Level Defense Waste Characterization—A Status Report," RH-CD-1019, July 1980. (b) Jungfleisch, F. M. "Supplementary Information for the Preliminary Estimation of Waste Tank Inventories in Hanford Tanks through 1980," SD-WM-TI-058, June 1983. (c) Jungfleisch, F. M. "Preliminary Estimation of Waste Tank Inventories in Hanford Tanks through 1980," SD-WM-TI-057, March 1984.

(a) no author "Redox Technical Manual," HW-18700, July 1951. (b) Crawley, D. T.; Harmon, M. K. "Redox Chemical Flowsheet HW-No.6," October 1960, HW-66203. (c) Isaacson, R. E. "Redox Chemical Flowsheets HW No.7 and HW No.8," RL-SEP-243, January 1965. (d) Jenkins, C. E.; Foster, C. B. "Synopsis of Redox Plant Operations," RHO-CD-505-RD-DEL, July 1978, declassified with deletions.

There are a whole series of documents pertaining to the evaporator campaigns. See for example (a) Bendixsen, R. B. "Dilute Customer Waste Concentration, First Pass 242-A Evaporator-Crystallizer Campaign 80-1," RHO-CD-80-1045, July 1980. (b) Starr, J. C. "242-A Evaporator/Crystallizer Fiscal Year 1986 Campaign Run 86-5 Post-Run Document," SD-WM-PE-032, June 1987. (c) Reynolds, D. A. "Double-Shell Slurry Campaign," RHO-CD-1268, December 1981.

no author, "Tank Farm Process Engineering Evaporator Monthly Reports December 1976 to December 1978.

Watrous, R.A.; Wootan, D.W. "Activity of Fuel Batches Processed Through Hanford Separations Plants, 1944 Through 1989," HNF-SD-WM-794, Rev. 0, July 1997.

Continue to next section