VZJ sign up for etocs
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 20 November 2007
Published in Vadose Zone J 6:1018-1030 (2007)
DOI: 10.2136/vzj2006.0158
© 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 (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Christensen, J. N.
Right arrow Articles by Dresel, P. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Christensen, J. N.
Right arrow Articles by Dresel, P. E.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Christensen, J. N.
Right arrow Articles by Dresel, P. E.
Related Collections
Right arrow Isotopes
Right arrow Radionuclides
Right arrow Vadose Zone Processes and Chemical Transport

SPECIAL SECTION: HANFORD SITE

Isotopic Studies of Contaminant Transport at the Hanford Site, Washington

John N. Christensena,*, Mark E. Conrada, Donald J. DePaoloa,b and P. Evan Dreselc

a Center for Isotope Geochemistry, Earth Sciences Division, E.O. Lawrence Berkeley National Lab., 1 Cyclotron Rd., Berkeley, CA 94720
b Dep. of Earth and Planetary Science, Univ. of California, Berkeley, CA 94720
c Pacific Northwest National Lab., Richland, WA 99353
* Corresponding author (jnchristensen{at}lbl.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 1 November 2006.



   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Geology and Hydrology of...
 Review of Recent Isotopic...
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
Processes of fluid flow and chemical transport through the vadose zone can be characterized through the isotopic systematics of natural soils, minerals, pore fluids, and groundwater. In this contribution, we first review our research using measured isotopic variations, due both to natural and site-related processes, of the elements H, O, N, Sr, and U, to study the interconnection between vadose zone and groundwater contamination at the Hanford Site in south-central Washington State. We follow this brief review with a presentation of new data pertaining to vadose zone and groundwater contamination in the Waste Management Area (WMA) T-TX-TY vicinity. Uranium isotopic data for the C3832 core (WMA TX) indicate the involvement of processed natural U fuel and link the observed U contamination to releases near single-shelled tank TX-104. The data also preclude contamination from an early 1970s TX-107 leak. In the case of the C4104 core (WMA T), the U isotopic data indicate a mixture of processed natural and enriched U fuels consistent with the major leak from T-106 in 1973. Uranium and strontium isotopic data for the cores also provide direct evidence for chemical interaction between high-pH waste fluid and sediment. Isotopic data ({delta}15N and {delta}18O) for groundwater nitrate contamination in multidepth samples just to the northeast of WMA T are distinct from that seen in surrounding wells and suggest tank waste (possibly from the 1973 T-106 event) as a source of very high 99Tc concentrations recently observed at the northeast corner of WMA T.

Abbreviations: MC–ICPMS, multiple-collector inductively coupled plasma–mass spectrometer • WMA, Waste Management Area


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Geology and Hydrology of...
 Review of Recent Isotopic...
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
In 1942 Hanford, WA, was selected as the site for major plutonium production based on its relative isolation and its proximity to the Columbia River, which provided ample power from the Grand Coulee Dam and cooling water for nuclear reactors and Pu separation operations (Smyth, 1945). From the time of World War II–era production through the Cold War and the eventual end of Pu production in 1987, more than 100,000 metric tons of uranium fuel were processed through the reactors and chemical separation processes (USDOE, 1999). Highly radioactive waste resulting from these processes was stored in tank "farms" located in the 200 East and 200 West Areas on the Hanford plateau (Fig. 1 ). Discharges of radioactive wastes have left a legacy of vadose zone and groundwater contamination at the site. By far, the greatest contribution to groundwater contamination has been intentional discharges to cribs, trenches, and various other engineered structures, but tank leaks have also contributed to groundwater contamination. Some of this contamination has reached the Columbia River, although most of it is still resident in the 50- to 100-m thick vadose zone. There remains the potential for further contaminant migration from the vadose zone to groundwater and ultimately, to the river. Understanding the fate and transport of contaminants has been complicated by the presence of multiple potential sources within relatively small areas. We have developed and implemented a suite of isotopic techniques, using the elements H, O, N, Sr, and U, to study the interconnection between vadose zone and groundwater contamination at the Hanford Site. We demonstrate the use of isotopic measurements to establish sources of contamination and place constraints on the rates of transfer through the vadose zone to groundwater. Our multiple-isotopic system approach has proved to be a powerful means to identify sources of contaminants and, once the sources are identified, to understand the subsurface transport routes and mechanisms.


Figure 1
View larger version (58K):
[in this window]
[in a new window]

  		FIG. 1. Map of the 87Sr/86Sr of groundwater for the Hanford Site (after Singleton et al., 2006).

 
In this paper we review recent results from isotopic research concerning the Hanford Site conducted at the Center for Isotope Geochemistry (Lawrence Berkeley National Laboratory). New data are presented that relate to vadose zone contamination in WMA T and WMA TX-TY and the source of 99Tc groundwater contamination in the vicinity of WMA T.


   Geology and Hydrology of the Hanford Site
TOP
 ABSTRACT
 INTRODUCTION
 Geology and Hydrology of...
Review of Recent Isotopic...
 Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
The Hanford Site is situated on the Columbia Plateau within the Pasco Basin, which was formed by broad folding and faulting of the Miocene-aged Columbia River Basalt Group and overlying sediments. The Pasco Basin is filled with a series of sediments that unconformably overly the Miocene basalts (Hartman, 2000). Immediately above the unconformity are Pliocene fluvial and lacustrine deposits of the Ringold Formation, consisting of gravels, silts, and clays. At the top of the Ringold Formation is the Cold Creek unit, a zone of pedogenic carbonate that developed in response to arid climate conditions (Slate, 1996). During the Pleistocene epoch the informally designated Hanford formation was deposited by a series of episodic floods resulting from catastrophic failures of ice dams holding back Glacial Lake Missoula located 200 km to the northeast (Bretz, 1969; Waitt, 1984). The Hanford formation comprises unconsolidated sediments of generally granitic and basaltic provenance ranging in grain size from gravel through sand to silt. The Hanford formation is overlain by discontinuous Holocene eolian and fluvial deposits. To accommodate the tank farms, the top 10 to 15 m of the Hanford formation were excavated and backfilled once tank construction was complete.

The basalt ridges (Rattlesnake Hills, Yakima Ridge) along the southwest edge of the Hanford Site are the primary sources of natural groundwater recharge for the site along with diffuse infiltration through the vadose zone (Gee et al., 1992). Groundwater flows through the unconfined aquifer from the basaltic ridges generally eastward through the sediments of the Ringold Formation and Hanford formation and then enters the Columbia River along the eastern side of the site. In addition, there are a stacked series of confined aquifers within the Columbia River Basalt Group, which locally communicate with the overlying unconfined aquifer.

Sedimentary structure has a significant effect on the fluid transport of contaminants in the vadose zone and groundwater (Serne et al., 2004a). Clastic dikes potentially provide cross-cutting pathways for contaminant movement. In the sedimentary column within the vadose zone, local highs in moisture content are associated with boundary zones between layers of contrasting grain size. These capillary barriers can generate lateral movement of fluids, especially during times of high fluid flux through the vadose zone. Preferential pathways are also developed through coarser-grained units during high flow. Under low fluid flux conditions, the dominant pathways for fluid transmission are fine-grained layers where water content is highest.


   Review of Recent Isotopic Studies
TOP
 ABSTRACT
 INTRODUCTION
 Geology and Hydrology of...
 Review of Recent Isotopic...
Materials and Methods
 Results
 Discussion
 Conclusions
 REFERENCES
 
Groundwater Strontium Isotope Patterns
The Sr isotopic patterns in groundwater at the Hanford Site provide a good starting point for discussion. Using Sr isotopic analyses of 273 groundwater samples of the unconfined aquifer, Singleton et al. (2006) produced a highly detailed groundwater map of the variation of 87Sr/86Sr across the Hanford site (Fig. 1). These data provide a means of tracking groundwater sources, including the water in contaminant plumes, independent of the contaminants themselves. The Sr isotopic composition of groundwater is also sensitive to infiltration through the vadose zone, to weathering of the host sediments, and to communication with the deeper confined aquifers within the Columbia River Basalts. This large data set provides an assessment of the evolution of the undisturbed natural groundwater–vadose zone system, as well as the effects of perturbations to groundwater by industrial processes at the site and remediation activities (Singleton et al., 2006).

The groundwater 87Sr/86Sr map reflects known aspects of the hydrology and geology of the Hanford Site (Singleton et al., 2006) and provides new insights about the sources of the present groundwater and broad-scale diffuse infiltration rates. The basalt ridges (Rattlesnake Hills, Yakima Ridge) along the west and southwest edge of the Hanford Site are the primary sites of groundwater recharge, and due to the relatively low 87Sr/86Sr of basalts (0.704–0.707), the freshly infiltrated waters initially take on a low 87Sr/86Sr ratio. As groundwater moves to the east, it exchanges strontium with sediments that have higher 87Sr/86Sr (~0.712), while simultaneously obtaining additional diffuse recharge from the vadose zone, which also has relatively high 87Sr/86Sr. The resulting spatial pattern is a gradually increasing 87Sr/86Sr from west to east across the site along the lines of groundwater flow (Fig. 1). Superimposed on this general pattern are local effects of both natural and industrial origin. Two areas of low 87Sr/86Sr to the north and south of Gable Mountain reflect upwelling of deep groundwater with low 87Sr/86Sr from basalt-hosted confined aquifers. Infiltration of Columbia River water (87Sr/86Sr {approx} 0.714) along the shore of the 100-B/C, -K, and -N Areas produces increases in groundwater 87Sr/86Sr. In contrast, infiltration of Yakima River water (87Sr/86Sr {approx} 0.707) causes lowering of groundwater 87Sr/86Sr to the west of the Richland North Area. Surface application of Columbia River water to recharge ponds in the Richland North Area produces a strong local high in 87Sr/86Sr. In the 200 West and 200 East Areas, large volumes of process water were discharged to trenches, cribs, and ponds, producing major disturbances of water table elevation and pronounced local highs in groundwater 87Sr/86Sr (and Sr concentration) as a result of rinsing or stripping of high 87Sr/86Sr strontium from the vadose zone (Singleton et al., 2006; Maher et al., 2003).

Isotopic Constraints on Vadose Zone Processes
A major factor in the rate of contaminant migration within the vadose zone, and hence the timescale for eventual contamination of groundwater, is the rate of fluid infiltration through the vadose zone. Isotopic data derived from vadose zone fluids and sediments can be used to place constraints on this rate as well as on the rate of weathering of sediments. Maher et al. (2003, 2006) measured and modeled Sr and U isotopic profiles of pore fluids and sediments in an uncontaminated core (now well 299-W22-48) taken in the 200 West Area to the east of WMA S-SX. The Sr and U isotopic compositions of pore water vary systematically down core. In the case of Sr, decreasing 87Sr/86Sr with depth is a response to progressive weathering of the sediment that releases Sr of lower 87Sr/86Sr. For U, the 234U/238U atomic ratio (or activity ratio) generally increases down core, reflecting opposing effects of alpha-recoil loss of 234U, which tends to increase pore water 234U/238U, and chemical weathering that tends to lower the pore fluid isotope ratio toward the secular equilibrium value of the 234U/238U ratio (or an activity ratio of unity). For both Sr and U, the isotopic ratios are affected by the infiltration rate through the vadose zone.

Maher et al. (2003) modeled the Sr isotopic data to constrain the ratio of the infiltration rate to the weathering rate and thus, with an assumed range of weathering rates based on studies of other soils, estimated an infiltration flux of 7 ± 3 mm yr–1. Combining the constraints imposed by both the Sr and U data allows for simultaneous estimation of both the infiltration and weathering rates. Using this analysis, Maher et al. (2006) estimated a long-term (at least hundreds of years) infiltration rate of 5 ± 2 mm yr–1 and a long-term bulk weathering rate of between 10–15.7 to 10–16.5 mol m–2 s–1.

Information from the 87Sr/86Sr groundwater map described above can be use to provide model estimates for infiltration fluxes over wider areas. Singleton et al. (2006) used a one-dimensional model along a line extending ~20 km east–west through an undisturbed portion of the Hanford Site to estimate variation in the infiltration flux. The modeling results suggested, for a constrained value of the dissolution rate in the aquifer of 10–7.5 yr–1, infiltration fluxes of 0 to 1.4 mm yr–1 near the western end at Dry Creek Valley, with infiltration fluxes perhaps as high as 30 mm yr–1 in the central part of the site. These estimates are in general accord with those based on vegetation and soil type (Gee et al., 1992; Fayer and Walters, 1995).

Competition within the soil zone between evapotranspiration and infiltration imposes a profile in pore water {delta}18O that can be measured and modeled to elucidate these processes (DePaolo et al., 2004; Singleton et al., 2004). Vadose zone pore fluids deeper than ~2 m in an uncontaminated, relatively undisturbed core (299-W22-48) have essentially constant enrichments in {delta}18O of 2 to 4{per thousand} relative to winter precipitation. Pore fluids in the upper 2 m vary strongly in {delta}18O with both depth and season. Modeling suggests that the magnitude of the overall vadose zone {delta}18O shift varies inversely with infiltration rate, which in turn is highly dependent on soil type and vegetation. The results of these studies indicate that diffuse vadose zone drainage increases the {delta}18O of groundwater. Thus, natural vadose zone fluids (high {delta}18O) are distinguishable from process water derived from the Columbia River (low {delta}18O), so that fluids from leaking pipes and surface discharge can be distinguished from natural waters. In the 299-W22-48 core, there is evidence of lateral introduction of low-{delta}18O process water, elevated levels of tritium (DePaolo et al., 2004), and possibly 99Tc (Serne et al., 2002).

Isotopic Signatures of Contamination
One of the great ongoing challenges at the Hanford Site is the identification of the sources of subsurface contamination. Because there are typically many possible contaminant sources within a limited area, such as tank leaks, tank-related spills, cribs, and trenches, and because each source was active at different times and has different chemical characteristics, it is difficult to determine by what pathway, and at what rates, the contaminants were transported through the vadose zone to groundwater. An even greater challenge is to predict whether, when, and where additional contaminants may start arriving at the water table in the future.

An obvious target for contamination fingerprinting is processed uranium. Uranium from nuclear industrial activities has a wide range of 235U/238U and 236U/238U due to variable combinations of isotopic enrichment and transformations during the operation of uranium-fueled nuclear reactors. In contrast, natural background uranium has constant 235U/238U, virtually zero 236U/238U, but variable 234U/238U due to alpha-recoil effects. The contrasts in isotopic composition between natural and processed uranium, as well as the wide compositional range of processed uranium, provide the means to trace contaminant uranium in the environment and delineate the sources and history of contamination.

Christensen et al. (2004) used high-precision U isotopic measurements (234U/238U, 235U/238U, 236U/238U) to investigate the source of groundwater U contamination seen in the early 1990s in the vicinity of the 200 East Area WMA B-BX-BY. By comparing the U isotopic compositions of groundwater samples to that of pore water samples from two cores (299-E33-46 and 299-E33-45) through vadose zone U contamination, the source of the groundwater contamination was shown to be consistent with an event in 1951 during which tank BX-102 was overfilled. The association of the groundwater contamination with the BX-102 overfill would require that the U did not migrate just vertically downward 40 m through the vadose zone but instead had a component of lateral travel of as much as ~150 m before reaching the water table. The lateral transport is consistent with the large size of the spill and the local geological structure. Furthermore, the observations suggest that groundwater U contamination can appear decades (at least 40 yr in this case) after being released to the vadose zone, confirming the downward migration of contaminants in the presence of infiltrating fluids, and the difficulties of predicting transport paths in heterogeneous geologic media subject to extreme hydrological conditions.

Nitrate is a widespread groundwater contaminant at the Hanford Site, with groundwater concentrations locally reaching over 1000 mg L–1, and with 75 km2 of the unconfined aquifer above the USEPA drinking water limit of 45 mg L–1 (Hartman et al., 2006). Possible sources for this contamination include radioactive waste from leaking tanks or cribs, waste derived from site activities, and leaching of naturally occurring nitrate from soils and caliche layers within the vadose zone. Recent development of an isotopic analysis technique involving a bacterially mediated process for preparation of nitrate samples has allowed the rapid analysis of the {delta}15N and {delta}18O of small amounts of nitrate (Sigman et al., 2001; Casciotti et al., 2002). Singleton et al. (2005) applied an adapted version of this technique to groundwater and vadose zone pore water samples for the analysis of both {delta}15N and {delta}18O of dissolved nitrate. Nitrate associated with tank waste is characterized with high {delta}15N (>10{per thousand}) and normal {delta}18O (~5{per thousand}). Low-activity waste synthetic nitrate has high {delta}18O (>10{per thousand}) with low-to-normal {delta}15N (<5{per thousand}), while naturally occurring vadose zone nitrate has low {delta}18O (<3{per thousand}) and low {delta}15N. These results provide a signature for nitrate source identification where concentrations alone would not be a distinguishing factor. This nitrate isotopic tracing technique is particularly valuable for identifying potential tank-related waste plumes as nitrate is a nearly conservative tracer of water movement in the vadose zone and groundwater and can be used to identify possible problem areas where no radioactive contamination (e.g., U) may yet be evident due to retardation.


   Materials and Methods
TOP
 ABSTRACT
 INTRODUCTION
 Geology and Hydrology of...
 Review of Recent Isotopic...
 Materials and Methods
Results
 Discussion
 Conclusions
 REFERENCES
 
Sample Collection and Background
The WMA T and WMA TX-TY tank farms are located within the 200 West Area in the central portion of the Hanford Site. WMA TX-TY is located due south of WMA T and consists of the contiguous TX and TY single-shell tank farms. Three boreholes, C3830, C3831, and C3832 (Fig. 2 ), were drilled in 2003 to evaluate possible leaks from single-shell tanks TX-105, TX-107. and TX-104 (Serne et al., 2004a). We analyzed a set of vadose zone samples from C3832 near tank TX-104 for the U and Sr isotopic compositions of pore fluids.


Figure 2
View larger version (27K):
[in this window]
[in a new window]

  		FIG. 2. Map of the T-TX-TY Waste Management Areas (WMAs) in the 200 West Area, Hanford, WA. Shown are locations of vadose zone boreholes including the studied borehole cores C4104 and C3832. Locations of groundwater sampling wells, and the location of the multilevel well W11-25B are also shown. For brevity, the 299 prefix has been dropped from well names. Contours of 99Tc groundwater concentration are from Hartman et al. (2006).

 
Two vadose zone cores were taken in the WMA T in 2003 (Fig. 2): C4104 just to the southeast of the single shell tank T-106, and a second, C4105, just to the southwest (Serne et al., 2004b). Tank T-106 was involved in a major event in 1973 in which 435,000 L of high-level waste leaked to the vadose zone over a 7-wk period (Routson et al., 1979). Tank T-103 is suspected to have leaked, while T-101 was reported to have been overfilled in the 1960s and to have had a leaky inlet port in 1969 (Jones et al., 2000). We analyzed pore water from vadose zone samples from the C4104 core for U and Sr isotopes. This core has been identified as having the higher U pore water concentrations of the two cores (Serne et al., 2004b).

Groundwater samples analyzed for Sr and nitrate isotopic composition come from monitoring wells in the vicinity of WMA T (Fig. 2, Table 1). Sampling was conducted in conjunction with the Hanford Groundwater Monitoring Project and occurred in May–June 2005. These samples supplement samples analyzed by Singleton et al. (2005, 2006). Groundwater under WMA T is contaminated with carbon tetrachloride, chromium, nitrate, 99Tc, and tritium, representing local and dispersed contamination from disposal cribs and trenches, as well as leaks from single-shell tanks and transfer lines (Hartman et al., 2006). In particular, increasing 99Tc concentration in monitoring well 299-W11-39 (to 27,000 pCi L–1 in 2005) near the northeast corner of the WMA T area prompted the emplacement of a new well, 299-W11-25B (Fig. 2), during which depth discrete samples were collected to 51 m below the groundwater surface. The depth profile from 299-W11-25B revealed high 99Tc concentrations peaking at 182,000 pCi L–1 at about 10 m below the water table (Hartman et al., 2006). We obtained aliquots for isotopic analysis of nine groundwater samples produced by purge-and-pump sampling during the drilling of W11-25B.


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

  		TABLE 1. Isotopic and concentration data for strontium and nitrate in groundwater samples from the WMA T vicinity.

 
Analytical Methods
Two sets of groundwater samples were collected; one set for U and Sr in high-density polyethylene bottles acidified with Ultrex nitric acid (J.T. Baker, Phillipsburg, NJ), and a second filtered into brown glass bottles and unacidified for nitrate isotopic ({delta}15N and {delta}18O) analysis. Pore water extracts from the vadose zone core samples were obtained by a 1:1 (w/w) deionized water rinse with centrifugation and filtration. Separate aliquots of the pore water rinses were then used for the U and Sr isotopic analyses. The depths of the analyzed samples from cores C4104 (near T-106) and C3832 (near TX-104) are presented in Tables 2 and 3.


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

  		TABLE 2. Isotopic and concentration data for strontium and uranium in pore-water from vadose zone samples from borehole C3832.

 

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

  		TABLE 3. Isotopic and concentration data for strontium and uranium in pore-water from vadose zone samples from borehole C4104.

 
For isotopic analysis, the uranium was chemically separated from the allotted sample aliquots using TRU-SPEC (Eichrom Industries, Darien, IL) in small volume Teflon columns scaled down from the procedure of Luo et al. (1997). This separation scheme provides column U yields of ≥95%.

The uranium isotopic compositions (234U/238U, 235U/238U, and 236U/238U) were measured on an IsoProbe (GV Instruments, Manchester, UK; multiple-collector inductively coupled plasma–mass spectrometer, MC–ICPMS) at Lawrence Berkeley National Laboratory. Uranium isotopes 235 and 238 were measured simultaneously on separate Faraday cups, while 234 and 236 were measured on a Daly ion counting system situated behind a wide-angle retardation potential lens. Static simultaneous measurement routines were used, one for 235U/238U and 234U/238U and a second for 236U/238U. Corrections for mass fractionation, Daly–Faraday intercalibration, and for any peak-tail under mass 236 were calculated from bracketed analyses of an in-house secular equilibrium natural uranium standard (20 µg L–1 solution of U ore from the Schwartzwalder Mine, CO, provided by W. Sharp, Berkeley Geochronology Center, Berkeley, CA). Isotopic compositions were normalized to the natural 238U/235U ratio (= 137.88 by convention; Steiger and Jäger, 1977) of the standard solution using an exponential mass fractionation law. Sample solutions were introduced to the MC–ICPMS via a desolvation system (Aridus; CETAC Inc., Omaha, NE) equipped with a low uptake microconcentric nebulizer. Typical precision for 235U/238U is ± 0.05% 2{sigma} or better; for 234U/238U and 236U/238U, it is ± 0.15% 2{sigma}. 236U/238U can be measured down to and below the 10–7 range where precision degrades by about a factor of 10 with a minimum measurable ratio of ~2 x 10–8.

Strontium isotopic analyses of vadose zone pore water and groundwater samples were conducted using thermal ionization mass spectrometry. Strontium was separated for analysis using Sr specific resin (Eichrom Industries, Darien, IL), loading and eluting the sample with 3N HNO3, and then stripping the separated Sr with clean H2O. The sample was then dried down with a drop of HClO4 and a drop of concentrated HNO3 in preparation for loading on an outgassed Re filament with 1% H3PO4 and a TaCl5 emitter solution. Strontium isotopic analyses on a VG 54 multicollector mass spectrometer (GV Instruments, Manchester, UK) were performed using a multidynamic analysis routine, with normalization to 86Sr/88Sr = 0.1194. The average 87Sr/88Sr measured for NBS 987 over the period of analysis was 0.710281 ± 0.000014 (2{sigma}).

The {delta}15N and {delta}18O of nitrate in the groundwater samples were determined using a denitrifying bacterial species to generate N2O from dissolved sample NO3– (Singleton et al., 2005, adapted from Sigman et al., 2001; and Casciotti et al., 2002). Sample nitrite/nitrate was low enough that the nitrite contribution to the generated N2O was negligible. Isotopic analysis of the N2O was then conducted with a continuous flow mass spectrometer (Micromass JA Series IsoPrime, GV Instruments, Manchester, UK), and isotopic values were corrected to KNO3 standards (USGS32, USGS34, USGS35, and IAEA-N3). Analytical precisions were 0.5 per mil for both the {delta}15N and {delta}18O of nitrate. This technique allows analysis of low sample volume (<4 mL) and low concentration samples down to 0.5 mg L–1 nitrate (Singleton et al., 2005).


   Results
TOP
 ABSTRACT
 INTRODUCTION
 Geology and Hydrology of...
 Review of Recent Isotopic...
 Materials and Methods
 Results
Discussion
 Conclusions
 REFERENCES
 
Results from Vadose Zone Borehole C3832 (WMA TX)
The pore water U concentrations in core C3832 (near TX-104) outline a vadose zone contamination plume (Serne et al., 2004a), with U concentrations up to 28.5 mg L–1, peaking just at the contact between the upper and lower Cold Creek units (Fig. 3a ). Fourteen pore water samples with sufficient U were analyzed for U isotopic composition. The measured 236U/238U ratios of the 10 samples within the depth range of 18.96 to 33.6 m fall within a narrow range of 90.12 ± 0.07 (x 10–6) to 91.37 ± 0.16 (x 10–6), despite pore water U concentrations varying by a factor of 10 (Fig. 3a). Likewise, the 235U/238U ratios of these samples also fall in a narrow range of 0.0066701 ± 27 to 0.0066877 ± 18 (errors given are 2{sigma} and apply to the last two digits). The pore water U concentrations, 2.7 to 28.5 mg L–1, are much higher than background values of 0.015 to 0.15 mg L–1. The near constant U isotopic composition across the central nine samples of the plume (Fig. 3a; Table 2) indicate that these U samples represent essentially the U isotopic composition of the pure contaminant, little affected by mixing with background U. The uniformity of the isotopic ratios also suggests a single relatively well-mixed contamination source. The plume boundaries as defined by 236U/238U (or 235U/238U, not shown) are relatively steep, falling to near background values over at most 5 m on the up-core margin, while on the down-core margin falling to ~1% contaminant U over an interval of 1.2 m before rising again to ~20% contamination in the next 0.2-m interval. Serne et al. (2004a) noted that the lowest extent of any sort of contamination was probably not reached in this core.


Figure 3
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 3. Profiles for the (A) C3832 cores and (B) C4104 cores of Uranium concentration and 236U/238U. Uranium concentrations are from Serne et al. (2004a, 2004b).

 
The U isotopic data for C3832 pore water are presented in plots of 236U/238U versus 235U/238U (Fig. 4a ) and 236U/238U versus 234U/238U (Fig. 4b), which are useful for deducing the character of the contaminant sources. In Fig. 4a natural background U plots on the x axis (where 236U/238U = 0) at a value of 235U/238U of 0.0072527 ± 21 (Cowan and Adler, 1976; Steiger and Jäger, 1977; Cheng et al., 2000). In Fig. 4b natural background U also plots on the x axis, but within a wide potential range of 234U/238U. In natural pore water–rock systems, such as in the vadose zone or the saturated zone, the alpha-recoil effect, through the decay of 238U, leads to the unsupported build-up of 234U in interstitial porefluid compared to solids (Kigoshi, 1971). This results in the common observation in groundwaters (Osmond and Cowart, 1992; Porcelli and Swarzenski, 2003, and references therein) of 234U/238U higher than the secular equilibrium value of 54.89 (x 10–6) (Cheng et al., 2000). In contrast, after timescales of 106 years, the bulk solid remains close to secular equilibrium, while the outer 10s of nanometers of grains and small (<40 microns) grains tend to become strongly depleted in 234U and so have 234U/238U distinctly less than 54.89 (x 10–6) (Maher et al., 2006; DePaolo et al., 2006).


Figure 4
View larger version (29K):
[in this window]
[in a new window]

  		FIG. 4. Plots of uranium isotopic data. (A) 236U/238U vs. 235U/238U; (B) 236U/238U vs. 234U/238U. Black circles are data for pore water extracts from C4104 (Waste Management Area [WMA] T); pink squares are data for pore water extracts from C3832 (WMA TX). Errors for data are approximately the size of the symbols or smaller. Red triangles represent estimates by Jones et al. (2000) of the U isotopic compositions of suspected or known tank leaks in WMA T-TX-TY. The blue square and green diamond represent the U isotopic compositions of the BX102 and B110 tank leaks established by Christensen et al. (2004). In A, data for a 1999 sample of W10-24 is from Dresel et al. (2002). Inset in B at expanded scale shows the best-fit line through the top four samples and its relationship to model compositions for processed enriched U fuels from Watrous and Wootan (1997). Numbers along best-fit line represent percentage of processed enriched fuel in the mixture.

 
In Fig. 4a the data for the 10 central samples of the C3832 U plume form a tight cluster on the line representing the array of Hanford processed natural U fuels (Dresel et al., 2002; Christensen et al., 2004). The remaining four samples (at depths of 13.7, 18.5, 34.9, and 35.1 m) fall along this line toward the composition of natural U, indicating dilution by admixed background U. In Fig. 4b the 10 central samples also form a cluster on the array for processed natural U fuels (Christensen et al., 2004). While the shallowest analyzed sample (13.66 m) has a small excess of 234U (234U/238U activity ratio = 1.051 ± 0.001), the remaining three samples fall very close to the natural U fuels array.

For the C3832 core, 19 pore water samples were analyzed for Sr isotopic composition. The pore water Sr concentration in the C3832 core ranges from 0.1 to 4 mg L–1, sharply increasing after a depth of 29 m correlating with an increase in CaCO3 (Serne et al., 2004a). The measured 87Sr/86Sr ratios are between 0.7136 and 0.7168, showing no obvious relationship to pore water Sr concentration.

Results from Vadose Zone Borehole C4104 (WMA T)
The pore water U concentrations in the C4104 core are as high as 17.8 mg L–1, 60% of the highest concentration in the C3832 core (Serne et al., 2004b). The zone of elevated U concentration extends from 14.32 to 28.23 m depth where the range of 1.15 to 17.8 mg L–1 U is 15 to >200 times greater than the background pore water U concentrations (Fig. 3b). Measured 236U/238U ratios in samples within the elevated U concentration plume are between 100.65 ± 0.11 (x 10–6) and 184.08 ± 0.30 (x 10–6), reaching values twice as high as in the C3832 core (Fig. 3). The overall trend down-core, with one reversal at 28.2 m, is of decreasing 236U/238U. This is in contrast to the relatively uniform isotopic composition across the U plume in the C3832 core. The two deepest analyzed pore water samples in C4104, at 30.89 and 32.74 m, have low U concentrations but still measurable 236U/238U (0.115 mg L–1, 38.027 ± 0.093 [x 10–6] and 0.053 mg L–1, 0.941 ± 0.012 [x 10–6], respectively).

The U isotopic data for C4104 are also plotted in Fig. 4a (236U/238U vs. 235U/238U) and Fig. 4b (236U/238U vs. 234U/238U). In Fig. 4b all the samples plot off the line representing processed natural U fuels. The four shallowest analyzed samples form a short array between Hanford processed enriched fuels and the array of natural U fuels. The positions of these data points along a best-fit line imply that the contaminant in C4104 consisted of no more than 16% processed enriched U fuel. The spread of the data for these four samples suggests that the proportion of processed enriched U decreases by 2 to 3% down core over the depth interval 14.3 to 23.2 m. The composition of the processed natural U end member is indicated by the intersection of the best-fit line with the array of natural U fuels and has a 236U/238U of approximately 120 x 10–6 and a 234U/238U of ~53 x 10–6 (Fig. 4b). In Fig. 4a (236U/238U vs. 235U/238U) the remaining five samples (depth interval 24.7–32.7 m) must fall along a fanning set of mixing lines projecting from the singular composition of natural background U and ending at various points along the line through the samples from the 14.3- to 23.2-m interval. The segment thus subtended by the fan of mixing lines indicates a greater range in the isotopic composition of the U contaminant than is observed in the four samples from the 14.3- to 23.2-m depth interval, extending it down to a minimum of 9% processed enriched U fuel.

In Fig. 4b, the distribution of the five samples from the 24.7- to 32.7-m interval constrains the range of 234U/238U of the natural background U that diluted/mixed with the contaminant U and so is suggestive of the source (e.g., pore fluid vs. sediment) of the natural U. Using sample C4104-9a (23.2-m depth) as representing the contaminant end member, a line including it and samples C4104-10a (24.7-m depth) and C4104-11a (26.6-m depth) intersects (at 236U/238U = 0) the 234U/238U axis at 54.74 ± 0.19 (x 10–6) (± 95% confidence), indicating that the natural U end member had a secular equilibrium value for 234U/238U. For samples C4104-9a, C4104-12a (28.2 m), and C4104-18a (32.7 m), the best-fit line indicates a different natural U 234U/238U of 56.76 ± 0.09 (x10–6), somewhat above the secular equilibrium value. The extreme example in the core is sample C4104-16a (30.9-m depth); its position in Fig. 4b requires that the natural U end member 234U/238U was significantly below the secular equilibrium value, as low as 50.86 ± 0.07 (x 10–6) (± 95% confidence) if sample C4104-9a is used as the contaminant end member. Even using the minimum of 9% processed enriched fuel suggested above for the contaminant in C4104 (Fig. 4a) does not affect the requirement for sample C4104-16a for a component of natural U significantly below secular equilibrium. However, for samples C4104-10a and C4104-11a, it would allow values above but close to secular equilibrium. Still, these values are below the range for vadose pore water observed in the B-BX WMA (Christensen et al., 2004) and a background clean core (Maher et al., 2006) (ranges shown in Fig. 4b).

Eighteen pore water samples from core C4104 were analyzed for 87Sr/86Sr ratio. Pore water concentrations of Sr ranged from <0.1 to 20 mg L–1, with the peak in Sr concentration coinciding with peaks in nitrate and 99Tc concentrations (Fig. 5 ; (Serne et al., 2004b) (see discussion below). The 87Sr/86Sr of pore water ranges from 0.7112 to 0.7148, and except for the shallowest sample, all are lower than measured in core C3832 pore water samples.


Figure 5
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 5. Profile through core C4104 of the pore water concentrations of Sr (µg L–1), nitrate (mg L–1), U (µg L–1), and 99Tc (nCi L–1). Data from Serne et al. (2004b).

 
Groundwater Data for the WMA T
Strontium isotopic and concentration data for groundwater monitoring wells in the vicinity of WMA T and WMA TX-TY are presented in Table 1 and in Fig. 6 , along with previously published data from Singleton et al. (2006). The resulting map of groundwater 87Sr/86Sr highlights two areas near WMA T of enhanced recharge from the vadose zone indicated by 87Sr/86Sr > 0.710 (Fig. 6a). One area appears associated with the T-36, T-7, and T-32 cribs, where ~1.4 x 105 kL of low-activity waste, including nitrate, was disposed to the vadose zone. Monitoring wells reveal very high groundwater nitrate concentrations, up to 3000 mg L–1 near these cribs as well as high Sr concentrations up to 2400 µg L–1 (~10 x normal Sr groundwater concentrations). A second area of very high 87Sr/86Sr is located just north of WMA T near the beginning of the 216-T-4-2 ditch that drained dilute process water. Groundwater in this area contains moderately high nitrate (250 mg L–1) and 99Tc concentrations (140 pCi L–1). Near the northeast corner of WMA T, there is groundwater with moderately high 87Sr/86Sr (>0.7095 and <0.7100). These wells are also associated with very high 99Tc concentrations. The set of discrete depth samples from 299-W11-25B (location shown in Fig. 2) provides a vertical section of the unconfined aquifer in the region of highest 99Tc concentration (Fig. 6b). The samples describe a zone of vertical mixing within the aquifer between Sr with high 87Sr/86Sr flushed from the vadose zone and Sr with lower 87Sr/86Sr representative of the broader 87Sr/86Sr trend for groundwater in the area. This parallels the contrast seen in pairs of adjacent wells drilled to different depths, where groundwater from the deep well has lower 87Sr/86Sr than groundwater from the shallow well (Fig. 6a).


Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 6. (A) Map of groundwater 87Sr/86Sr in WMA T-TX-TY and vicinity. Data from Table 1 and Singleton et al. (2006). (B) Profile below water table in W11-25B (see Fig. 2 for location) of 87Sr/86Sr.

 
Nitrate isotopic data ({delta}15N and {delta}18O; Table 1) for shallow groundwater wells surrounding WMA T and groundwater samples from the well 299-W11-25B depth profile are presented in Fig. 7 . Most of the WMA T groundwater samples form an array separate from the array described by the nine discrete depth samples from 299-W11-25B. The two exceptions are samples from wells 299-W11-39 and 299-W10-8, located near the northeast corner of WMA T and 299-W11-25B. The WMA T array traverses the region in Fig. 7 between the field representing the composition of synthetic nitrate and the field representing the isotopic composition of naturally occurring nitrate in the vadose zone. The high {delta}18O end of the WMA T array is represented by samples with very high (>900 mg L–1) nitrate concentrations, including well 299-W10-4 (2420 mg L–1 nitrate), adjacent to the T-36 crib. The low {delta}18O end of the array has the lowest nitrate concentrations. These observations suggest that this array represents mixing between synthetic nitrate and natural nitrate flushed from the vadose zone. The data for discrete depths in well 299-W11-25B describe a separate trend (Fig. 7) from synthetic nitrate toward the field representing the isotopic compositions of nitrate from vadose zone samples contaminated by tank related waste (Singleton et al., 2005). The deepest sample in 299-W11-25B falls at the upper end of the array, closest to the field for synthetic nitrate. This sample also has the lowest 99Tc concentration. The shallowest two samples, including the sample with the highest 99Tc concentration (151,000 pCi L–1) fall at the low {delta}18O, high {delta}15N end of the array toward the composition of nitrate from tank-related waste. This indicates that nitrate in 299-W11-25A represents variable mixing between two sources and implies that the source of the very high levels of 99Tc seen in the upper part of 299-W11-25B and in the northeast corner of WMA T is from tank-related waste.


Figure 7
View larger version (26K):
[in this window]
[in a new window]

  		FIG. 7. Plot of {delta}15N versus {delta}18O for nitrate in groundwater samples from Waste Management Area (WMA) T and for the sub-water table depth discrete samples from W11-25B. Additional data, the field for background cores, and data for tank-related waste from contaminated cores are from Singleton et al. (2005). The field for synthetic nitrate (representing low-activity waste) is from Kendall (1998) and Amberger and Schmidt (1987). Lines connect samples from the same well taken at different times, with the more recent sample indicated by a white dot.

 
Uranium concentrations in the analyzed WMA T and 299-W11-25B samples are low, less than 2 µg L–1. Their U isotopic compositions are presented in Table 4. The measured 235U/238U ratios of the samples are indistinguishable or very close to the natural ratio. The 236U/238U of these samples are all very low, less than 5 x 10–7. Both these factors indicate a very low level of contamination by processed U. The contaminant contribution, <1%, is too minor to identify accurately the source of the U contamination. The 234U/238U of these six samples are all relatively high (74.1 x 10–6 to 96.5 x 10–6) compared with groundwater in the vicinity of WMA B-BX-BY in the 200 East Area (Christensen et al., 2004) and at the location of the 299-W22-48 borehole in the southern portion of the 200 West Area (Maher et al., 2006) but similar to the value, 98.4 x 10–6, for a clean well (299-W11-6) situated ~650 m to the southeast (Fig. 6).


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

  		TABLE 4. Isotopic and concentration data for uranium in groundwater samples from the WMA-T vicinity.

 

   Discussion
TOP
 ABSTRACT
 INTRODUCTION
 Geology and Hydrology of...
 Review of Recent Isotopic...
 Materials and Methods
 Results
 Discussion
Conclusions
 REFERENCES
 
Sources and Histories of Contamination for Boreholes C3832 and C4104 Cores
In the C3832 core, collected from WMA TX-TY, the isotopic composition of the contaminant is consistent with processed natural U fuel with a 236U/238U as high as 91 x 10–6. No evidence for a component of processed enriched U is seen. These observations place constraints that can be used to infer the source of the contaminant U based on the history of fuel type and usage at the Hanford Site. A perusal of a model history of processed natural U (Wootan and Finfrock, 2002) indicates that the first time that T-plant processed fuel reached a 236U/238U of ~91 x 10–6 was in 1951. Enriched U fuel was not processed until 1958, thus suggesting that the contamination in C3832 came either from a pre-1958 source or from a source/waste stream that never received processed enriched U. Although it is not listed in site documents as having leaked, single-shell tank TX-104, located just south of C3832 (Fig. 2) matches well the above constraints. A possible alternative is the set of trenches arrayed 125 m to the west of WMA TX. These retention trenches received more than 6000 kL of first cycle supernatant via tanks TX-109, TX-110, and TX-111 to provide needed tank volume for processed fuel separations (Maxfield, 1979). However, the first decontamination cycle waste had considerably lower uranium concentrations than the "metal waste" that contained nearly all the uranium from the extraction (Anderson, 1990). Thus, the T-21 to T-25 retention trenches are not a likely source of the contamination observed in C3832. Estimate of the isotopic composition of U associated with TX-107 leaks during the period 1975 to 1977 (Jones et al., 2000) are not compatible (Fig. 4) with the apparent lack of processed enriched U in C3832. Myers (2005) presents a scenario where the metal waste loss from tank TX-104 or other tanks took place during a process to sluice sludge from the tank for recovery of the uranium. Thus, the chemistry and mobility of the waste may have been modified during the sluicing process. The uranium isotopic data is consistent with this hypothesis.

The U contamination in C4104 in the WMA T, as described above, represents a mixture of processed natural U and processed enriched U, in roughly the proportions of 85 and 15%, respectively. In detail the proportion of processed enriched U appears to decrease down hole from 16 to 9%, suggesting that contamination came from a source that was not perfectly well mixed, or that the composition of the source changed somewhat with time during the leak event. The inferred compositions of the two contaminant end members place some constraints on the time frame of the contamination. From the line through the shallowest four samples, the processed natural U component had a 236U/238U of ~120 x 10–6, while the processed enriched U component had a 236U/238U of approximately 525 x 10–6. These values are both consistent with processed enriched U in the late 1960s (Wooton and Finfrock, 2002). No natural U fuel was processed in 1970, while the batches run in 1971–1972, the last years of the use of natural U fuel, never reached a 236U/238U ratio greater than 95 x 10–6. The material leaked in 1973 from T-106 originated from fuel rods processed as much as 5 yr before the event (Routson et al., 1979, Appendix G) and so is compatible with the broad time frame indicated by the U isotopic observations in C4104. Mixing within the vadose zone of the T106 leak fluids with fluids from a nearly contemporaneous leak at T-103 is a possible explanation of the range seen in C4104, but although the estimated U concentrations are similar for the two leaks (Jones et al., 2000), the leak volume for T-106 and hence the mass of U involved was 40 times greater. Alternatively, the leak composition may have changed with time. The 1973 T-106 incident occurred during the transfer of supernatant to T-106 via cascade flow by pumping the supernatant first into tank T-105 (Routson et al., 1979), probably causing some time dependent mixing among the contents of T-105, T-106, and the newly added supernatant. This would have produced a nonconstant U isotopic composition to the contaminant fluid that percolated downward at the location of C4104, producing the isotopic range observed down core over the 14.3- to 23.2-m depth interval.

Isotopic Evidence for Sediment–Contaminant Fluid Interaction
The U and Sr isotopic compositions of vadose zone pore water in cores C4104 and C3832 presented above provide evidence of chemical interaction of contaminant fluids with sediment. In Fig. 8 , for C4104 236U/238U shows a correlation with pH, indicating that as the extreme pH of the contaminant fluid was reduced or neutralized, the contaminant U was diluted by natural composition U with 236U/238U = 0. Wan et al. (2004a, b) showed that high pH (~14) tank supernatant will react with Hanford sediment silicates in a chemical process that reduces pH in the main plume, while pushing ahead a front of high Ca, Mg, and so on, pore fluid concentrations resulting from cation exchange due to high Na+ contaminant fluid (Serne et al., 2004b). This front would likely be accompanied by mobile contaminants such as nitrate and 99Tc. As described above, C4104 samples (10a, 11a, and 16a) in the lower portion of the core likely represent mixing between the U contaminant and background natural U with 234U/238U ratio just near to well below, in the case of 16a, the secular equilibrium value. With time the alpha-recoil effect imparts distinctive 234U/238U signatures to sedimentary material and pore water and groundwater that can be used to distinguish between simple mixing between contaminant fluids and pore fluids versus chemical interaction between the contaminant fluid and the host sediment. For C4104 samples 10a and 11a, the near secular equilibrium 234U/238U ratio of the natural U end member is consistent with wholesale release of U from the bulk sediment as a result of the reaction between the high pH fluid and sediment. Sample 16a presents an extreme value of 234U/238U for the natural U component, significantly below the secular equilibrium value. This is consistent with release of U from the outer edges of grains or from fine-grained material (Maher et al., 2006). Interestingly, sample 16a consists of a sandy mud (Serne et al., 2004b), providing a fine-grained source material for low 234U/238U.


Figure 8
View larger version (11K):
[in this window]
[in a new window]

  		FIG. 8. Plot of 236U/238U versus pH for core C4104 in WMA T. Data for pH from Serne et al. (2004b).

 
Although no high pH was observed in C3832 (Serne et al., 2004a), samples 121b and 61b fall on or just to the left of the natural U fuel line in Fig. 4b, suggesting some interaction between contaminant fluid and sediment. This suggests the initial occurrence of high pH in the contaminant plume seen in C3832 and would thus emphasize the unreliability of pH as an indicator of the involvement of tank waste since pH can be altered (Wan et al., 2004a, b; Serne et al., 2004b).

Further evidence for contaminant–sediment interaction is provided by the Sr isotopic profiles of C4104 and C3832. In Fig. 9 they are compared to the 87Sr/86Sr profile of the 299-W22-48 core (Maher et al., 2003), which showed minimal impact from site contamination. Both the C4104 and C3832 profiles appear shifted in 87Sr/86Sr toward the range of bulk sediment and away from the clean core profile, with the shift for C4104 being the greatest. This suggests the release of relatively low 87Sr/86Sr from the whole rock or feldspars (Fig. 9) due to interaction with the contaminant fluid. In detail, the greatest apparent 87Sr/86Sr shift in C4104 occurs within the zone of elevated pH (> ~9 between 14- and 28-m depth). Near the peak of pore water 99Tc and Sr concentrations (Sr = 20 mg L–1 at 35.4-m depth, Fig. 5), there is a local high in 87Sr/86Sr (= 0.7123, Fig. 9), perhaps reflecting Sr released from clays in the ion exchange process (Serne et al., 2004b) that produced the high Ca concentration front.


Figure 9
View larger version (34K):
[in this window]
[in a new window]

  		FIG. 9. Plot of depth below ground surface vs. 87Sr/86Sr comparing vadose zone pore waters from the contaminated cores C4104 (black circles) and C3832 (red squares) with vadose zone pore water from an uncontaminated core (blue diamonds, W22-48 Maher et al. 2003). Green stars represent bulk sediment analyses for W22-48 (Maher et al. 2003). Dashed lines connect corresponding stratigraphic depths for the cores. Peach shaded and purple shaded areas represent zones of U and 99Tc contamination.

 
Isotopic Constraints on the Source of 99Tc Contamination in the Vicinity of WMA T
The isotopic data for nitrate in groundwater samples from 299-W11-25B suggest that it is derived from a mixture of high {delta}15N tank-related waste and waste with isotopic composition in the range of synthetic nitrate discharged to disposal cribs. The most likely source of that crib waste is a plume of groundwater nitrate extending from the southwest corner of the WMA T area. Simpson et al. (2001) have estimated that 3.6 x 106 kg of nitrate was discharged to the T-7 and T-32 cribs located in that area (Fig. 2). Groundwater samples from the closest monitoring wells to those cribs, 299-W10-4 and 299-W10-28, consistently have the highest nitrate concentrations in the WMA T area. In addition, the {delta}18O values of the nitrate in all these samples are high (up to 15.4{per thousand}), indicating a high proportion of synthetic nitrate.

Given the relatively low 99Tc concentrations in these samples (<1000 pCi L–1), most of the 99Tc in 299-W11-25B was derived from the source of the tank-related waste nitrate. The T-106 spill observed in the C4104 core is a likely candidate for the source of the tank waste contamination, but as documented by Rohay (2007), there are alternate conceptual models for sources. These other possible sources include other tanks that leaked and nearby trenches, but the T-106 leak was the largest. The T-101 tank (which is closer to the northeast corner of WMA T) also leaked, but that leak was much smaller volume (<10%), and the concentration of 99Tc in the leak fluids was much lower (~7% of T-106). The T-14, T-15, T-16, and T-17 trenches received first decontamination cycle supernatant from the T tank farm and are also a potential source. However, data from electrical resistivity surveys across the trenches, although not conclusive, do not show the vadose zone contaminant plumes extending to groundwater (Rucker and Levitt, 2006).

The net contributions of fluid from the T-106 leak that would be necessary to account for the 99Tc concentrations observed in 299-W11-25B is relatively low. The highest 99Tc concentrations in the C4104 core (between 35- and 39-m depth on Fig. 5) range from 4 x 106 to 25 x 106 pCi L–1, averaging 8 x 106 pCi L–1. At that level, only a 2% contribution to groundwater from the pore fluids from C4104 would be necessary to produce the highest 99Tc concentration measured in 299-W11-25B. The nitrate concentrations in the peak interval in C4104 range from about 4500 to 10,500 mg L–1 averaging about 6000 mg L–1. Two percent of the average composition would be 120 mg L–1, or only about 21% of the nitrate concentration observed in the C4104 sample with the peak 99Tc concentration. That implies that the remaining 79% of the nitrate in the peak sample is derived from the low-activity groundwater nitrate plume. The amount of T-106 fluids necessary to account for the other samples from 299-W11-25B will all be less (<0.3% for the deepest sample).

The alternate conceptual models proposed for the 99Tc by Rohay (2007) are being investigated at this time through drilling and characterization. Additional samples from the other possible sources that are planned to be collected and analyzed may further constrain the source. The suggestion that the T-106 spill as the source of contamination may also be tested if a sufficient fraction of contaminant U appears in groundwater samples associated with high 99Tc. Chemical retardation of U relative to 99Tc may be delaying the arrival of associated contaminant U, if any.


   Conclusions
TOP
 ABSTRACT
 INTRODUCTION
 Geology and Hydrology of...
 Review of Recent Isotopic...
 Materials and Methods
 Results
 Discussion
 Conclusions
REFERENCES
 
Isotopic studies add an extra dimension to the characterization of subsurface contamination, leading to insights into contaminant source and behavior as well as the processes that affect contaminant transport and mobility. Natural strontium isotopic composition (87Sr/86Sr) is particularly sensitive to water–rock interaction and show areas of enhanced recharge through elevated 87Sr/86Sr in groundwater. The nitrogen and oxygen isotopic compositions of nitrate show distinct contrasts between the attributed tank-related and low-activity waste sources investigated so far. Uranium isotopic measurements can also be useful for understanding water–rock interactions. In addition for the Hanford Site, because the U isotopic ratios of potential sources changed through time as a result of varying fuel enrichment and reactor exposure, U isotopic measurements can also be used to identify the source and timing of contamination. Combining multiple isotopic measurements with conventional chemical data constrains the possible contaminant sources and the transport path and rate in the vadose zone and groundwater.

Our recent studies implicate tank-sourced waste, quite possibly associated with the T-106 leak in 1973, in the shallow 99Tc and nitrate groundwater contamination near the northeast corner of WMA T. Ongoing site characterization activities are expected to further constrain contaminant sources. The uranium and strontium isotopic data for the C3832 and C4104 cores, and chemical data from previous studies, show varying amounts of interaction between the infiltrating waste fluids and vadose zone sediments. The high pH, high ionic strength tank fluids greatly enhance the reaction with sediments over that of background infiltrating recharge. Even the relatively low levels of uranium contamination seen in the WMA T and WMA TX-TY vadose zone cores are sufficient to provide considerable constraints on the source and timing of contamination.


   ACKNOWLEDGMENTS
 
Funding for this study was provided by the U.S. Department of Energy under Contract DE-AC02-05CH11231 to Lawrence Berkeley National Laboratory (LBNL) through the Hanford Science and Technology Program, and by the Assistant Secretary of the Office of Environmental Management, Office of Science and Technology, under the Environmental Management Science Program (now Environmental Remediation Sciences Program) of the U.S. Department of Energy under Contract DE-AC02-05CH11231 (LBNL) and Contract DE-AC06 = 76RL01830 (Pacific Northwest National Laboratory). The authors gratefully thank the Hanford Groundwater Monitoring Program for providing groundwater samples, as well as Jeff Serne and Clark Lindenmeier for providing core samples produced by the CH2M HILL Hanford Group.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Geology and Hydrology of...
 Review of Recent Isotopic...
 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]

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 (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Christensen, J. N.
Right arrow Articles by Dresel, P. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Christensen, J. N.
Right arrow Articles by Dresel, P. E.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Christensen, J. N.
Right arrow Articles by Dresel, P. E.
Related Collections
Right arrow Isotopes
Right arrow Radionuclides
Right arrow Vadose Zone Processes and Chemical Transport

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