A Review of a Field Study of Radionuclide Migration from an Underground Nuclear Explosion at the Nevada Test Site

Page 1

Los Alamos National Laboratory is operated by the University of California for the United Stales Department of Energy under contract W-7405-ENG-36. title: A Review of a Field Study of Radionuclide Migration From an Underground Nuclear Explosion at the Nevada Test Site submitted to: International Conference on Radioactive Waste Management International Atomic Energy Agency Seattle, WA, May 16-20, 1983 By acceptance of this article, the publisher recognizes that the U.S Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or to allow others to do so. for U.S. Government purposes The Los Alamos National Laboratory requests that the publisher identify this article as work performed under the auspices of the U S Department ot Energy Los Alamos National Laboratory

Page 2

INTERNATIONAL ATOMIC ENERGY AGENCY INTERNATIONAL CONFERENCE ON RADIOACTIVE Seattle, WA, USA, 16-20 May 1983 A REVIEW OF A FIELD STUDY OF RADIONUCLIDE MIGRATION FROM AN UNDERGROUND NUCLEAR EXPLOSION AT THE NEVADA TEST SITE* Los Alamos National Laboratory Los Alamos, New Mexico United States of America This is a preprint of a paper intended for presentation at a scientific meeting. Because of the provisional nature of its content and since changes of substance or detail may have to be made before publication, the preprint is made available on the understanding that it will not be cited in the literature or in any way be reproduced in its present form. The views expressed and the statements made remain the responsibility of the named author(s); the views do not necessarily reflect those of the govern ment of the designating Member State(s) or of the designating organization(s). tn particular, neither the IAEA nor any other organiiation or tody sponsoring this meeting can be held responsible lor sny materia! reproduced in this preprint.

Page 3

A REVIEW OF A FIELD STUDY OF RADIONUCLIDE MIGRATION FROM AN UNDERGROUND NUCLEAR EXPLOSION AT THE NEVADA TEST SITE* Los Alamos National Laboratory Los Alamos, New Mexico United States of America A REVIEW OF A FIELD STUDY OF RADIONULCIDE MIGRATION FROM AN UNDERGROUND NUCLEAR EXPLOSION AT THE NEVADA TEST SITE. Results from a long term (9 year) field study of the distribution of radionuclides around an underground nuclear explo sion cavity at the Nevada Test Site are reviewed. The goals of this Radionu clide Migration project are to examine the rates of migration underground in various media and to determine the potential for movement, both on and off the Nevada Test Site, of radioactivity from such explosions, with particular Interest in possible contamination of water supplies. Initial studies were undertaken near the site of the low-yield test Cambric, which was detonated 73 m beneath the water table in tuffaceous alluvium. Solid samples were obtained from just below ground surface to 50 m below the detonation point, and water was sampled from five different regions in the vicinity of the explosion. Ten years after the test, most of the radioactivity was found to be retained in the fused debris in the cavity region and no activity above background was found 50 m below. Only tritium and ••Sr were present in water in the cavity at levels greater than recommended concentration guides for water in uncontrolled areas. A satellite well is being used to remove water 91 m from the detonation point. During seven years (7 x 10‘ m*) of pumping, tritium, #,Kr, ,IC1, and 11*1 have been detected in the water. Approximately 40X of the total tritium from the cavity region has been removed by pumping at the satellite well, and the maximum in the tritium concentration is clearly defined. Use of sensitive analytical techniques has permitted measurement of the very low concentrations of ,€C1 and l,iI present in the water. The I#C1 peak precedes the trltlated water, possibly as a result of anion exclusion. Additional analyses are in progress to better define the shape of the ,lfI concentration curve. A field study of the distribution of radionuclides around an underground nuclear explosion cavity was initiated in 1974, about 9 years after detonation.1*2 The goals of this initial experiment in the Radionuclide Migration (RNM) project were to determine the rates of migration underground in various media at the Nevada Test Site (NTS) and the potential for movement, both on and off the NTS, of radioactivity from underground nuclear explosions. There was particular interest in possible contamination of water supplies. It was also envisioned that the study might provide data applicable to the underground *Work performed under the auspices of the United States Department of Energy.

Page 4

disposal of radioactive waste. The RNM project is sponsored by the Nevada Operations Office of the Department of Energy with the participation of the Los Alamos National Laboratory, the Lawrence Livermore National Laboratory, the United States Geological Survey, the Desert Research Institute, and support The site of the 0.75-kt nuclear test Cambric, which was fired beneath the water table (294 m below the ground surface) in tuffaceous alluvium, was chosen for initial studies. It is anticipated that eventually tests in other geological media will be examined. Cambric was chosen for a number of reasons, including the following: The Cambric explosion cavity is within an NTS water-supply aquifer. It was predicted that sufficient time had elapsed so that the cavity and chimney had filled with groundwater to the preshot static water level, 73 m above the detonation point. If so, any radionuclides present in the water would constitute a potential source for transport. Sufficient tritium (SH or T) was present to provide an easily measurable tracer for water from the cavity region. The postshot debris also contained other radionuclides whose concentrations in the rubble and groundwater from the cavity and chimney regions could be measured and compared. (A summary of the radionuclide source term at the time of re entry is given in Table I.) The alluvium was believed to constitute a good medium for hydrologic studies because it is quite permeable and does not have large fissures or cracks through which the water might selectively flow. Finally, the CAMBRIC SOURCE TERM TEN YEARS AFTER DETONATION Half Life Activity

Page 5

small yield was expected to have had little effect on the The field study began with completion of a satellite well (RNM-2S) 91 m from the Cambric cavity, followed by drilling of a re-entry well (RNM-1) through the cavity itself (Fig. 1), thereby precluding the possibility of contaminating the satel lite well with radioactivity from RNM-1. Sidewall core samples were taken from RNM-1 from just below the ground surface to 50 m below the detonation point to measure the vertical distribution of radionuclides. The hole was cased and the casing perforated to permit pumping of water from different regions. In this way, variations in the radio nuclide content of water from different intervals were deter mined, and the extent of natural radionuclide migration during the 9 years after detonation was examined. Pumping of water from RNM-2S then induced sufficient articial gradient to draw water from the Cambric cavity to permit study of radionuclude migration. The appearance of tritiated water (HTO) in water from this satellite well signaled the forced breakthrough of water from the Cambric cavity region. Measurement of various radionuclides present in water pumped from RNM-2S provides a means of studying radionuclide transport under field conditions. ELEMENT DISTRIBUTION IN THE REGION OF THE CAMBRIC NUCLEAR EXPLOSION AFTER TEN YEARS Three types of samples were removed from the RNM-1 re entry hole into the Cambric cavity: sidewall cores, pumped water, and water with contained gases. As drilling progres sed, a total of 67 sidewall core samples was taken from RNM-1 at intervals ranging from 34 m below the surface to 37 m below the original detonation point. One core from each depth was placed immediately in a nitrogen-flushed, gas-tight, stain and for gamma-spectral analysis. Other core samples were sealed in water-tight plastic bags for later gamma-spectral and radiochemical analyses to determine the concentrations of the various radionuclides present. The cores were also examined to determine the lithology of the medium at different depths. Solid samples and water removed from the sidewall cores from the lower cavity region were analyzed radiochemcoefficients (ratio of the concentration in or on the solid to the concentration in the aqueous phase) were determined. These effective distribution coefficients are a measure of both retention in the fused material and sorption. The radio nuclides were found to be almost entirely incorporated in or on the solid material (Table II).

Page 6

EFFECTIVE DISTRIBUTION COEFFICIENTS IN THE LOWER CAVITY REGION After sidewall core sampling had been completed and the hole had been cleaned, casing with appropriately placed, inflat able, external packers was installed.2 The packers mini mized external water movement in the annular space between the casing and the wall. Beginning at the bottom, the water in five zones (ranging from below to above the cavity) was sampled successively by isolating the zones with internal packers and perforating the casing. Gas-tight water samples were taken at depth, and additional water samples were removed to the surface by a submersible pump. Representative activity levels of the radionuclides detected in water from each zone are given in Table III. Ten years after the test most of the radioactivity and the highest concentrations of all radio nuclides were still found in the region of the original explo sion cavity. No activity above background was found 50 m below the cavity. Measurements of HT and HTO removed from the cores and gas-tight water samples showed that more than 99.9% of the tritium was present as HTO. Although some ®5Kr and tritium were found in the collapsed zone above the explosion region, they were concentrated in the cavity region. The measured e5Kr to T ratios for water from the explosion cavity zone were consistent with the relative amounts resulting from the Cambric test; the t5Kr seemed to be dissolved in the water. No krypton was observed in water or solid material from cores taken above the water table. Water from the region of highest radioactivity at the bottom of the cavity contained only tritium and ,eSr at levels higher than the concentration By comparing the measured ratio of each nuclide detected in the water to the tritium in the water with the calculated ratio for the Cambric source term, an effective overall reten tion factor, E^, for each nuclide (ratio of the total activity in or on the solid to the total activity in the aqueous phase) was estimated. (This calculation assumes that all the radio nuclides are retained below the water table and that the fraction of a nulcide not in the water was in the solid.)

Page 7

REPRESENTATIVE ACTIVITY LEVELS IN CAMBRIC WATER SAMPLES TEN YEARS AFTER DETONATION ^bg - no activity detectable above background levels. uncontrolled areas. Values for tritium and ••Sr are public drinking water standards established In Ref. 4. where K is the distribution coefficient. were all found to have high retention factors, indicating they

Page 8

are either retained in the fused debris or highly sorbed on the solid material, or both (Table IV). Additional information describing early work at RNM-1 is given in Refs. 1 and 2. TRANSPORT OF RADIONUCLIDES TO THE SATELLITE WELL The satellite well RNM-2S is located 91 m from the Cambric explosion cavity. Pumping was begun in October 1975 at a rate of about 1 m3/min; in October 1977 the rate was increased to about 2.3 m3/min. Significant amounts of tritiated water, signaling arrival of water from the Cambric cavity region, were detected after a total of 1.44 x 106 m3 of water had been pumped from RNM-2S. The tritium concentration in water from RNM-2S reached a maximum just below the maximum permissible3 concentration (3nCi/mt) for drinking water in controlled areas and is now decreasing (Fig. 2). Pumping of 7.09 x 10s m3 of water from RNM-2S (by early October 1982) removed 42% of the tritium available from Cambric. Figure 3 shows a comparison of the transport of tritiated water to the satellite well with the calculations of Sauty3 for instantaneous tracer injection in a radial, converging flow field, similar to the RNM experiment. The dimensionless time is given on the upper abscissa* A dimensionless, time T , of one corresponds to the time re quired to pump the volume of water contained in a cylinder whose radius is the distance from the satellite well to the tration relative to the maximum concentration defined by the data. This treatment invokes symmetry and treats the source as an annular ring. The shape of the elution curve depends on the Peclet number, which is inversely proportional to the dispersivity. An excellent fit to the data was obtained with a Peclet number of 10, corresponding to a dispersivity of 9.1 m. This result can be compared with the data presented by Borg et al.,6 who reported longitudinal dispersivity values that were estimated by calibrating mathematical models for transport against observed transport in various field studies. Values ranged from 11.6 to 91 m for a wide variety of lithologies; the value was 21.3 m for a sand or gravel deposit — the lithology that probably most closely resembles the tuffaceous alluvium of the Chlorine-36 produced by the Cambric event has been de tected in water samples from RNM-2S using accelerator-based mass spectrometry. The technique is described in Ref. 7. A comparison of the concentrations of tritiated water and 3*C1 is given in Fig. 4. As for tritium, the two-dimensional solu tions of Sauty5 were used to fit the 36C1 data and an excel lent fit was obtained with Peclet number 10 (Fig. 5).

Page 9

RETENTION FACTORS FROM CAMBRIC WATER SAMPLES abg ■ no activity detectable above background levels, The breakthrough of 36 Cl appears to occur earlier than that of tritium, and the maximum in the chloride concentration clearly occurs before that of tritium. The phenomenon whereby anions are eluted earlier than cations or neutral species from packed columns has been observed in soil chemistry studies0' 9 and column chromatography and is called anion exclusion. The effect has been shown to occur in soils containing smectites, and it correlates fairly well with the cation exchange capacity of the soil. Anion exclusion is due to the negative surface charge on aluminosilicates which repels anions, thereby effectively constricting the pores and making the intragranular porosity of the mineral unavailable to nega tively charged species. A method has been devised for determining the concentra tion of 129I in RNM waters by neutron activation and counting of the 1301 product.10 Although the concentrations are well below the concentration guide3 for water in an uncontrolled significant because X2*I is considered a hazardous radio nuclide in the storage of nuclear waste.11 The high ranking of l23I is due to its biological activity and lack of sorption on most minerals. The concentration of l29I was determined in samples taken from RNM-1 before pumping started on RNM-2S. The concentra tion of 12SI in water removed from the lower cavity region is measured in the RNM-2S samples was 7.6 x 10x0 atoms/liter. The iodine and tritium concentrations for RNM-2S water are plotted in Fig. 6. The …