VZJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 20 November 2007
Published in Vadose Zone J 6:1042-1049 (2007)
DOI: 10.2136/vzj2006.0153
© 2007 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Evans, J. C.
Right arrow Articles by Farmer, O. T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Evans, J. C.
Right arrow Articles by Farmer, O. T., III
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Evans, J. C.
Right arrow Articles by Farmer, O. T.
Related Collections
Right arrow Isotopes
Right arrow Radionuclides
Right arrow Other Environmental Contamination

SPECIAL SECTION: HANFORD SITE

Inductively Coupled Plasma/Mass Spectrometric Isotopic Determination of Nuclear Wastes Sources Associated with Hanford Tank Leaks

John C. Evans*, P. Evan Dresel and Orville T. Farmer, III

Pacific Northwest National Lab., P.O. Box 999, Richland, WA 99352
* Corresponding author (john.evans{at}pnl.gov).

All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


Received 11 October 2006.



   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
The subsurface distribution of a nuclear waste tank leak on the USDOE's Hanford Site was sampled by slant drilling techniques to characterize the chemical and radiological characteristics of the leaked material and assess geochemical transport properties of hazardous constituents. Sediment core samples recovered from the borehole were subjected to distilled water and acid leaching procedures with the resulting leachates analyzed for isotopic and chemical signatures. Inductively coupled plasma–mass spectrometry (ICP–MS) techniques were used for determination of isotopic ratios for Cs, and Mo. Analysis of the isotopic patterns of Mo, combined with associated chemical data, showed evidence for at least two separate intrusions of nuclear waste into the subsurface. Isotopic data for Cs was inconclusive with respect to a source attribution signature.

Abbreviations: amu, atomic mass unit • bgs, below ground surface • ICP–MS, inductively coupled plasma–mass spectrometry • ICP–OES, inductively coupled plasma–optical emission spectrometry • NRC, Nuclear Regulatory Commission • PNNL, Pacific Northwest National Laboratory • %RSD, percent relative standard deviation, REDOX, reduction–oxidation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
Mass spectrometric methods for determination of stable and long-lived radiological isotope concentrations and concentration ratios have long been used for a wide range of geotechnical investigations, including such diverse applications as age dating, paleotemperature measurement, tracking of groundwater and seawater movement, mineral exploration, and chemical source attribution. Some hydrogeologic and contaminant isotopic studies relevant to the Hanford Site include Graham (1983), Hearn et al. (1989), Spane and Webber (1995), Christensen et al. (2004), DePaolo et al. (2004), Maher et al. (2003), Maher et al. (2006), Singleton et al. (2005a), Singleton et al. (2005b), Singleton et al. (2006), and Brown et al. (2006). This paper reports results from an investigation aimed at source attribution of highly radioactive materials associated with nuclear waste tank leaks at the USDOE's Hanford Site near Richland, Washington (Fig. 1 ). We supplemented the standard analysis of chemical and radiological constituents with analysis of stable and long-lived isotopes of Cs and Mo and used the isotopic ratios to characterize the fission component in the vadose zone beneath an underground tank that is known to have leaked. The sampling and analysis work was performed in early 2001 as part of a large-scale vadose zone characterization effort as reported earlier in the field investigation report (CH2M HILL, 2002).


Figure 1
View larger version (30K):
[in this window]
[in a new window]

  		FIG. 1. Location of the S-SX tank farms, Hanford Site in Washington State.

 
The occurrence of nuclear waste tank leaks at Hanford has been recognized since at least the late 1950s. Sixty-seven of the 149 single-shelled tanks used for storage of plutonium production waste are listed as "assumed leakers" (Naiknimbalkar, 2006). Over the past few years, an increased effort has been undertaken to improve the overall understanding of contaminant movement in the vadose zone and, in particular, the potential for groundwater impact from the more mobile constituents. The recognition of groundwater impacts at several tank farms (Johnson and Chou, 1998; Narbutovskih, 1998; Hodges, 1998; Hodges and Chou, 2000) has added a degree of urgency to this effort.

The geology beneath the 241-SX tank farm has been the subject of numerous reports (Johnson and Chou 1998, 1999; Johnson et al., 1999; Lindsey et al., 2000) and has been summarized in a series of reports by Serne et al. (2002a, b, c). The major stratigraphic units beneath the tank farm include (in descending order) the Hanford formation, the Cold Creek unit, and the Ringold Formation (Serne et al., 2002b) (Fig. 2 ). The upper 16 m (53 ft) of the Hanford formation was removed during construction of the tank farm (between 1953 and 1954), and the stockpiled sediments were later used as backfill around the underground storage tanks.


Figure 2
View larger version (35K):
[in this window]
[in a new window]

  		FIG. 2. Cross-section beneath the SX-108 high-level waste tank showing the depths of samples collected from the slant borehole.

 
The SX-108 slant borehole (borehole C3082) was emplaced, and samples were collected between June and July 2000 (Gardner and Reynolds 2000). This borehole was drilled approximately 12 m (40 ft) from the northeastern edge of tank 241-SX-108 at an angle of approximately 30 degrees from vertical. The borehole was intended to pass directly beneath the center of the tank to a vertical depth of 43.9 m (144 ft), with a total pipe run length of 52.2 m (171.2 ft) (Gardner and Reynolds 2000). Seventeen sampling attempts were made, and 16 samples were recovered at 1.5 and 3 m (5 and 10 ft) intervals throughout that portion of the borehole extending from the base of the backfill to the total depth of the borehole (Fig. 2). Borehole gyroscope surveys were conducted to verify the actual location of the drive tip and casing and to provide the precise position and depth of the sampling points (Serne et al., 2002b). Materials and data generated by that effort were used in planning and implementing the isotopic work described here.

Essentially all contaminants observed in the SX-108 core show concentration maxima at two or more depths. In this paper all depths are corrected for the drilling angle and reported as depth below ground surface. The multiple peaks could be interpreted as resulting from geologic effects (stratification, grain size), geochemical effects (sorption, precipitation, oxidation–reduction), or differences in chemical speciation (valence states, ligand complexation). Alternately, multiple leak events may be partially responsible for affecting the varied vertical distribution of contaminants. Separate leak events originating either from different tanks or separate fillings of the same tank should be detectable with isotopic signature analysis. Suitable systems that have been identified in this work include the fission derived isotopes of Cs and Mo. The ratios of various chemical species with little or no retardation following release from a tank into the vadose zone can also be useful as indicators of different source compositions (e.g., Johnson and Chou, 1998). Species that are used for that purpose here include NO3 and 99Tc. An understanding of the process of emplacement of the contamination is an important part of the site conceptual model and is needed for predicting contaminant fate and transport and for developing remedial alternatives. If multiple events are indicated, further evaluation may be needed to determine if these events are related to different sources and may have implications for estimates of contaminant inventory in the vadose zone.

The isotopic abundance distributions of elements produced through thermal neutron fission in nuclear reactors typically are different from natural abundances, and the magnitude of the effect on isotopic ratios is often large. For example, the radioactive isotopes of Cs considered in this paper, 135Cs and 137Cs, are essentially absent in natural Cs. Additionally, the isotopic ratio difference for 97Mo/98Mo is more than a factor of two between natural (0.396) and 235U fission (1.036) (Baum et al., 2002). Further, reactor conditions produce additional variation in isotopic ratios within the fission products. During reactor operation 239Pu ingrowth and fission shifts the fission yield from the initial 235U fission yield curve toward the 239Pu fission yield (Watrous and Wootan, 1997; Baum et al., 2002). Neutron capture may also shift the isotopic abundances due to differences in capture cross-section of the isotopes or their precursors.

In addition to variations caused by reactor conditions, chemical separations performed during fuel reprocessing may also alter the isotopic abundances of fission products. In particular, 95Mo has two radioactive precursors with moderately long half-lives, 95Zr (64.02 d) and 95Nb (34.97 d), that are likely to have survived the cool-off period before reduction–oxidation (REDOX) processing. Thus, some degree of chemical separation is likely to have perturbed the isotope ratio of 95Mo with respect to the other Mo fission isotopes. This effect potentially provides a means to discriminate between different sources because compositions will depend on variation in cool-off time and processing chemistry.


   Materials and Methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
 Discussion
 Conclusions
 REFERENCES
 
Preliminary Characterization
All analyses were performed on either 1:1 sediment to water or 8 M nitric acid extracts of the sediment samples. Analyses routinely performed on the water and acid extracts included trace metals by inductively coupled plasma–optical emission spectroscopy (ICP–OES) and ICP–MS, and common anions by ion chromatography, as well as radiological measurements for 137Cs, 3H, 90Sr, 99Tc, and actinides. Details on the experimental protocols used and the results obtained are reported elsewhere (Serne et al., 2002a, 2002b, 2002c). Following completion of that work, the remaining extracts and associated blanks, each typically only a few milliliters, were transferred to our laboratory for more detailed mass spectrometric analysis.

An upgraded VG PlasmaQuad ICP–MS instrument (Thermo Fisher Scientific, Waltham, MA), in a radiologically controlled laboratory was dedicated to all subsequent ICP–MS analyses. Before detailed analysis, all samples were screened in the mass range from 91 to 139 and 200 to 243 atomic mass units (amu) to evaluate potential mass interferences and to determine required dilution factors.

Upon receiving the water and acid extract samples, all solutions were prepared by taking 0.1 mL of sample and adding 0.1 mL of a 1 µg/mL 115In internal standard and diluting to 10 mL with high-purity water. The samples were then analyzed for bulk impurity concentrations, and sample screening was performed. Each sample solution was nebulized into the instrument, and the concentration of each element/isotope was calculated on a semiquantitative basis using 115In as a surrogate standard. Using this data, each sample was then diluted to yield an analyte solution concentration such that each sample's solution contained an approximately constant amount of analyte. That procedure allowed for optimum precision and reduced memory effects. Concentration data were corrected back to porewater concentrations using the moisture contents reported in Serne et al. (2002b). It was not possible in the course of this study to fully characterize all possible sources of analytical error. The estimated analytical uncertainty is in the range of 1 to 5% or better for most analyses.

Cesium Analysis
Two analyses for Cs were performed on each sample. The first was a total 133Cs analysis using an appropriate dilution factor to yield a 133Cs concentration of approximately 100 pg/mL for each solution, where possible. Each solution was evaluated against a 133Cs calibration curve using 115In as the internal standard. The second analysis was performed using an on-line high pressure ion chromatography, ICP–MS procedure for the transient signal analysis of Cs isotopes. The chromatography separation used a Dionex CS3 cation column (4 x 250 mm; Dionex Corp., Sunnyvale, CA), employing a gradient elution of dilute nitric acid over a 10-min separation.

Molybdenum Analysis
For the analysis of Mo isotopic ratios, an appropriate amount of sample was diluted to yield a 100Mo concentration of approximately 0.1 µg/mL. One-hundred milligrams of Dowex-50x8 (50–100 mesh; Dow Chemical, Midland, MI) cation resin were then added to each sample solution. The resin was added to remove first-row transition elements that form molecular ion interferences with the plasma gas argon. The samples were then filtered and spiked with 115In to yield a solution having an 115In concentration of 500 pg/mL. Interferences were observed at masses 92 and 94 from samples containing high levels of Cr+6. For that reason, the 96 amu peak was used as a measure of naturally abundant Mo. A custom spreadsheet was developed to correct for minor residual zirconium interferences and to apply necessary mass bias corrections.

Technetium Analysis
Several possible plasma ion interferences can be present in the sample solution to interfere with the direct ICP–MS detection of 99Tc. These include the argides 59Co40Ar, 61Ni38Ar, and 63Ni36Ar along with the direct isobaric interference from natural 99Ru. Because of these potential interferences, each sample solution was pretreated with Dowex-50x8 cation resin. No attempt was made to remove any interference from 99Ru from the samples before analysis because ruthenium is a rare element not typically found at sensible levels in nature or detectable in these samples.

Nuclear Modeling
As an aid to understanding isotopic ratio measurements, a simple nuclear model was developed to calculate expected ratios of fission produced Cs, Tc, and Mo. The methodology used is similar to that used in an earlier Pacific Northwest National Laboratory (PNNL) study performed for the Nuclear Regulatory Commission (NRC). Details of the FORTRAN-based computer code developed for the NRC work are described in two publications (Evans et al., 1983, 1988). A streamlined version of the calculational method used for the NRC study was adapted to an Excel spreadsheet. Input data included neutron flux and thermalization parameters for a production reactor, fission and neutron capture cross-sections and resonance integrals, fission yields from 235U and 239Pu, and 235U composition for typical production reactor fuel (0.71% 235U). The spreadsheet performed a series of neutron capture and fission event calculations on 0.1 d intervals for up to 150 d to simulate a typical production reactor irradiation during the REDOX process era at Hanford. Minor adjustments were made to reactor parameters to match the expected 239Pu/240Pu composition at the end of the irradiation period and to calculate reasonable values for 235U burnup and 239Pu accumulation. An average plutonium composition taken from the best basis inventory documents on tanks SX-108 and SX-109 was used to estimate the expected plutonium composition at the end of irradiation (4.5% 240Pu). ORIGEN2 calculations performed by Watrous and Wootan (1997) were also reconciled as a cross-check for accuracy.

Calculated fission product ratios for Mo and Tc proved to be relatively insensitive to reactor parameters and should thus be reasonably consistent. Table 1 lists the expected Cs, Tc, and Mo composition on a relative basis in the reactor fuel at the end of a 110-d irradiation producing plutonium containing 4.5% 240Pu. All results are normalized to 99Tc and assume no chemical fractionation effects. Relative 137Cs values shown in Table 1 were corrected for decay assuming a 40-yr age for the waste based on the tank leak history for tank SX-108 (WHC, 1992). The greatly reduced amount of 135Cs shown in Table 1 relative to other similar high-yield fission products is the result of burnup of its parent on the beta decay chain, 135Xe. 135Xe has an exceptionally high neutron capture cross-section, 2.65 million barns. The half-life of 135Xe, 9.1 h, is long enough that in a high flux reactor burnup, neutron capture is actually favored over decay. Because of this situation, the amount of 135Cs produced is expected to be extremely sensitive to reactor flux conditions. The measured values of 135Cs and 137Cs can thus be used to tightly constrain input parameters for the model. The sensitivity of the Cs isotope ratios to reactor conditions also suggests that some source-related variability may be expected.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Relative concentration of fission isotopes in irradiated Hanford reactor fuel from the nuclear model calculation.

 

   Results
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
Discussion
 Conclusions
 REFERENCES
 
Cesium Isotopes
The 137Cs distribution in the SX-108 slant borehole is shown in Fig. 3 . The plot is based on gamma counting data as reported by Serne et al. (2002b). As a point of comparison, the 99Tc concentrations measured in this study on water leach samples from the same core are plotted on the second axis. Both distributions show two peaks with the 99Tc penetrating much deeper and with more apparent vertical dispersion.


Figure 3
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 3. 137Cs sediment concentrations (Serne et al., 2002b) and 99Tc water extract concentrations (this study) in the SX-108 slant borehole.

 
Cesium isotopes were measured by ICP–MS on selected water and acid leach samples near the two peaks in the 137Cs vertical distribution. The results of those measurements are shown in Table 2. Only data with relative standard deviations of less than ~10%, as determined by replicate analysis, are included in Table 2. As expected, the acid extracts resulted in the dissolution of a large amount of native 133Cs from the mineral matrix of the host sediment. 133Cs results from the acid extracts are thus of little value; however, the ratio of fission product 137Cs/135Cs is of interest because the ratio can be determined in the acid extracts with very good precision. The ratio is relatively constant across both peaks. The water leach data show a small difference between the two peaks, but it is unclear if that is significant as it may be an artifact of the lower concentrations producing a larger measurement error for the water leach samples. The Cs data thus is inconclusive with respect to the issue of separate sources for the two peaks.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Cesium isotopic measurements in the SX-108 slant borehole vadose sediment leachates.

 
Molybdenum Isotopes
Molybdenum has seven stable isotopes amenable to mass spectrometric analysis (92, 94, 95, 96, 97, 98, and 100). Four of those isotopes may also be formed as fission products resulting from beta decay of neutron rich high-yield parent isotopes (95, 97, 98, and 100). The other three can be produced in very low yields as direct fission products, but because all three are blocked from the beta chain by stable isotopes of zirconium, cumulative fission yields are negligible. Molybdenum isotopes at masses 92, 94, and 96 can thus be regarded as essentially invariant in composition and representative of naturally occurring Mo. These nonfission-derived isotopes were used to correct for the natural Mo component present in each core segment leachate and to calculate a composition for the purely fission derived components. The measured 92Mo and, to a lesser extent, 94Mo abundances are distorted by Cr–Ar interferences, so only 96Mo was used in this study for calculating the contribution due to naturally occurring Mo.

Table 3 lists the measured Mo isotopic compositions of the water leach samples. Measurements were also performed on the acid leach samples; however, it was clear from the screening results that the large quantity of naturally occurring Mo leached out of the sediment obscuring the fission product signal for the acid leachate. By contrast, the Mo in the water leaches was dominated by a fission-derived component accounting for up to 96% of the total water leachable Mo at the maximum concentration. Table 3 shows the composition of natural and purely fission derived Mo compositions across the top two rows for comparison.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Molybdenum isotopic abundance in water extracts from SX-108 slant borehole.

 
Table 4 shows the calculated isotopic ratios for the fission-only component for the three peaks with the contribution from naturally abundant Mo stripped out. Only the middle peak was wide enough to calculate a relative error directly on multiple samples. Molybdenum isotope ratios are similar with the exception of 95Mo/98Mo, where the 24.8–27.4 m bgs peak is 22% greater than the 20.8 m bgs peak and 17% greater than the 35.4 m bgs peak.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Fission molybdenum isotope ratios for maximum concentrations in the SX-108 slant borehole.

 
Figure 4 shows two different representations of the distribution of Mo in the SX-108 borehole. The plot compares the total Mo concentration measured by ICP–OES (Serne et al., 2002b) with the mass spectrometric determination of fission-derived 100Mo expressed as a percentage associated with fission. The separate peaks are more readily discernable using the fission-derived isotopes. Three fission containing peaks are evident with maxima at 20.8, 27.4, and 35.4 m. Because the shallowest and deepest peaks contain much less total fission Mo, the percent fission component is somewhat less; however, the fission component still represents more than 80% of the total Mo even in the smaller peaks. The fission contribution is a very large effect not significantly impacted by analytical uncertainty associated with the use of a quadrupole ICP–MS for the isotopic measurements.


Figure 4
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 4. Comparison of total Mo measured by inductively coupled plasma–optical emission spectrometry (bottom axis) in water extracts from SX-108 slant borehole (Serne et al., 2002b) with this study's fission derived 100Mo measured by inductively coupled plasma–mass spectrometry (top axis). The other three fission derived isotopes of Mo as 94, 97, and 98 amu show a similar pattern. (amu, atomic mass unit; BGS, below ground surface.)

 
Molybdenum is almost certainly present in tank supernatant as the molybdate oxyanion. Molybdenum falls beneath Cr in the periodic table and may function in a geochemical system as a surrogate for Cr(VI). Hexavalent chromium is of particular interest in tank leak studies because of its copious use in the REDOX process as an oxidizing agent combined with its high toxicity and high environmental mobility. Figure 5 shows a comparison between Cr and Mo measured in the core water extracts by ICP–OES (Serne et al., 2002b). Both species show evidence of three separate peaks, with the Mo peaks at somewhat shallower depths relative to the corresponding Cr peaks.


Figure 5
View larger version (19K):
[in this window]
[in a new window]

  		FIG. 5. Chromium and Mo distribution in water extracts from SX-108 slant borehole (Serne et al. 2002b). (BGS, below ground surface.)

 
Chromium itself appears to be significantly retarded in its vertical transport relative to Tc (Fig. 6 ). The retardation of both Cr and Mo appears to be associated with either sorption or precipitation on a solid phase. Zachara et al. (2004) show that Cr is sequestered in the solids as both Cr(III) and Cr(VI). Figure 7 shows the comparison between water and acid extracts for Cr. A substantial fraction of the total Cr is unavailable for extraction by water. While it was not possible to perform the same comparison directly for contaminant Mo because of the substantial natural component in the acid extracts, it is reasonable to infer that an even larger fraction of the Mo is tied up in a solid phase based on the greatly reduced peak height for the lower fission peak at 35.4-m depth. Except for differences in peak heights and minor differences in retardation, the Cr and Mo appear to exhibit similar profiles.


Figure 6
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 6. 99Tc (this study) and Cr (Serne et al., 2002b) in distributions in water extracts from SX-108 slant borehole. (BGS, below ground surface.)

 

Figure 7
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 7. Chromium distribution in water and acid extracts from SX-108 slant borehole (Serne et al., 2002b). (BGS, below ground surface.)

 
Nitrate and Technetium
A number of species are believed to migrate in the vadose zone in an almost purely unretarded fashion. Concentration ratios of species with similar geochemical behavior, NO3 and 99Tc, were examined for their applicability in identifying contaminant sources. The NO3 data discussed here was taken from the work of Serne et al. (2002b), while our 99Tc measurements were made by ICP–MS on samples from the same leachates. Figure 8 shows the distribution of NO3 and 99Tc in the SX-108 slant borehole core. In the upper peak, the NO3 appears at a shallower depth than the 99Tc and at a lower relative concentration compared with the lower peak. There is no strong indication of a third peak, as was observed in the Mo data. Chloride versus NO3 (data not shown) exhibited a similar profile to 99Tc vs. NO3.


Figure 8
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 8. NO3 (Serne et al., 2002b) and 99Tc (this study) distributions in water extracts from SX-108 slant borehole. (BGS, below ground surface.)

 

   Discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
Conclusions
 REFERENCES
 
Multiple chemical and isotopic species in water extracts of vadose zone sediments show two or three peaks with depth below ground surface in the SX-108 slant borehole. Although the borehole was not drilled perpendicular to the surface, and thus, there is a lateral offset between the peaks, the fact that the peaks appear at different vertical depths for different species indicates that differential retardation in transport in the vertical direction is occurring. It is difficult to envision preferential lateral transport of different species in different layers or another mechanism to explain the variation with depth. The data indicate that the order of retardation for the species considered is Cs >> Mo > Cr, NO3 > Tc.

The SX-108 tank leak history is summarized in WHC (1992). Tank SX-108 was placed in service in late 1955 for disposal of REDOX process high level waste. The tank began self-boiling in mid-1956 and was completely filled by early 1959. The first evidence of unaccounted volume loss and other indications of a potential leak were noted in late 1962, although it appears that the tank was not considered to be leaking until 1964. The tank was emptied in 1962. Before refilling, NaNO3 salt cake from SX-108 was recovered by leaching with water. It is possible that the leak that had self-healed reopened as a result of dissolution of the salt plug at that time. SX-108 was then filled with another round of fresh REDOX waste from mid-1963 through the end of 1964. Data presented in WHC (1992) indicate that further leakage took place after mid-1965. Estimates of the volume of leaked material vary widely. The best estimate for leakage up to 1966 based on sediment 137Cs concentrations as a function of depth from several boreholes drilled around the tank is approximately 2400 gallons (Raymond and Shdo, 1966), with an upper bound in the range of 35,000 gallons. There is thus evidence from the historical record that a significant volume of material possibly representing two different fills leaked over a relatively long period of time at that location. More recent estimates of contamination leak volumes at tank farms by Field and Jones (2006) identify two periods of leakage, the leak noted in 1964 and a second leak believed to have begun in 1966. However, it is also possible that other unidentified leak events occurred. Measurements conducted on recovered sediment from the SX-108 slant borehole, drilled through the heart of the leak plume, show evidence to support multiple events.

The high molybdenum concentration peaks in the water extracts were identified as being dominated by a fission component through the isotopic abundances. With the exception of the 95Mo/98Mo, the isotopic ratios for the fission component were similar throughout the depth interval sampled. Variability in 95Mo/98Mo is attributed to chemical separation before complete decay of precursor isotopes. The data indicate three peaks of different isotopic ratio, which can be explained by source variability. Although the potential for two leak events is presented above, three separate sources or events are suggested by the Mo data from the SX-108 slant borehole. It is unclear why Mo and Cr show three apparent peaks while 99Tc and NO3 show only two.

Molybdenum isotopes also look promising as tracers of tank leak material into groundwater, where such impacts are known or suspected. Because of the high background of naturally occurring stable Mo typically found in groundwater, high-precision mass spectrometric techniques would be needed. However, it appears that the Mo is retarded with respect to other constituents of interest, Cr and 99Tc.

The Cs isotopes showed relatively constant isotopic composition in water and HNO3 electrolyte leaches. The presence of separate leak sources inferred from several other lines of evidence was not discernable from the Cs data despite an expected variability in 135Cs concentration predicted from nuclear modeling considerations. A more extensive series of Cs measurements on other tanks is needed to better define the range of variability, and higher-precision data would be valuable.


   Conclusions
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
REFERENCES
 
The data analyzed in this study support a model where the vadose zone beneath tank SX-108 is impacted by two or three distinct leak events. These could be leaks from separate tanks or multiple leak events from tank SX-108. Multiple leak events of fluids of varying composition are consistent with the usage of the tank but other tanks in close proximity are also believed to have released contaminants to the subsurface.

The chemical and isotopic data show multiple concentration peaks with depth in the borehole. The differences in peak depth for different constituents are interpreted to result from differences in retardation, even among constituents generally considered to be highly mobile in the vadose zone. These minor differences in peak depths can lead to major variations in constituent ratios, complicating identification of different source signatures when using ratios of different chemical species.

Isotopic measurements of Mo and Cs complement the chemical measurements and can provide signatures for different sources. Isotopic ratios among fission products vary due to differences in reactor operations and chemical processing, and this variability is carried through waste disposal, with the exception of radioactive decay. Isotopic fractionation for a given element by natural processes such as sorption, precipitation, and diffusion in the mass range of these elements is believed to be insignificant at the level of the precision of our measurements. Fission is the dominant source of Mo in the SX-108 vadose zone water extracts. The three Mo concentration peaks show varying isotopic ratios of 95Mo/98Mo. The fission Cs isotopes, 135Cs and 137Cs, showed little variability in the samples but from theoretical considerations are worth pursuing during characterization of additional sources.


   ACKNOWLEDGMENTS
 
This study was supported by the U.S. Department of Energy's Richland Operations Office through the Groundwater/Vadose Zone Science and Technology Project. We are grateful for the support and insight provided by John Zachara, who oversaw the task for Pacific Northwest National Laboratory. Samples were graciously provided by CH2M HILL Hanford Group Single Shell Tank Vadose Zone Characterization Project funded by the U.S. Department of Energy's Office of River Protection. A special thanks also to Dr. Thomas Jones for providing valuable information and suggestions. Sample extracts and baseline analyses were provided by Ray Clayton, Virginia LeGore, Robert Orr, Mathew O'Hara, Christopher Brown, Todd Schaeff, and Theresa Wilson of Pacific Northwest National Laboratory. The insights and advice of Chris Brown, Clark Lindenmeier, and Jeff Serne were invaluable. This work was performed for the U.S. Department of Energy under Contract DE-AC06-76RLO 1830.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 




This article has been cited by other articles:


Home page
 Vadose Zone JHome page
G. W. Gee, M. Oostrom, M. D. Freshley, M. L. Rockhold, and J. M. Zachara
Hanford Site Vadose Zone Studies: An Overview
Vadose Zone J., November 20, 2007; 6(4): 899 - 905.
[Abstract] [Full Text] [PDF]

Home page
 Vadose Zone JHome page
J. M. Zachara, J. Serne, M. Freshley, F. Mann, F. Anderson, M. Wood, T. Jones, and D. Myers
Geochemical Processes Controlling Migration of Tank Wastes in Hanford's Vadose Zone
Vadose Zone J., November 20, 2007; 6(4): 985 - 1003.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Evans, J. C.
Right arrow Articles by Farmer, O. T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Evans, J. C.
Right arrow Articles by Farmer, O. T., III
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Evans, J. C.
Right arrow Articles by Farmer, O. T.
Related Collections
Right arrow Isotopes
Right arrow Radionuclides
Right arrow Other Environmental Contamination

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome