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


     

Published online 9 August 2006
Published in J Environ Qual 35:1715-1730 (2006)
DOI: 10.2134/jeq2005.0124
© 2006 American Society of Agronomy, Crop Science Society of America, and 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)
Right arrow An erratum has been published
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 ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Phillips, D. H.
Right arrow Articles by Jardine, P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Phillips, D. H.
Right arrow Articles by Jardine, P. M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Phillips, D. H.
Right arrow Articles by Jardine, P. M.
Related Collections
Right arrow Structure and Properties
Right arrow Ground Water Quality
Right arrow Mixed Wastes
Right arrow Heavy Metals
Right arrow Soil Analysis

TECHNICAL REPORTS

Heavy Metals in the Environment

Distribution of Uranium Contamination in Weathered Fractured Saprolite/Shale and Ground Water

D. H. Phillipsa, D. B. Watsonb,*, Y. Rohc, T. L. Mehlhornb, J.-W. Moonb and P. M. Jardineb

a Environmental Engineering Research Centre, School of Planning, Architecture, and Civil Engineering, Queen's University of Belfast, Belfast BT9 5AG, Northern Ireland, UK
b Environmental Sciences Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN, USA 37831-6038
c Faculty of Earth Systems and Environmental Sciences, Chonnam National University, 300 Yongbong-dong, Buk-gu, Gwangju, 500-757, KR

* Corresponding author (watsondb{at}ornl.gov)

Received for publication April 8, 2005.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The objective of this study was to determine how structure, stratigraphy, and weathering influence fate and transport of contaminants (particularly U) in the ground water and geologic material at the Department of Energy (DOE) Environmental Remediation Sciences Department (ERSD) Field Research Center (FRC). Several cores were collected near four former unlined adjoining waste disposal ponds. The cores were collected, described, analyzed for U, and compared with ground water geochemistry from surrounding multilevel wells. At some locations, acidic U-contaminated ground water was found to preferentially flow in small remnant fractures weathering the surrounding shale (nitric acid extractable U [UNA] usually < 50 mg kg–1) into thin (<25 cm) Fe oxide-rich clayey seams that retain U (UNA 239 to 375 mg kg–1). However, greatest contaminant transport occurs in a 2 to 3 m thick more permeable stratigraphic transition zone located between two less permeable, and generally less contaminated zones consisting of (i) overlying unconsolidated saprolite (UNA < 0.01 to 200 mg kg–1) and (ii) underlying less-weathered bedrock (UNA generally < 0.01 to 7 mg kg–1). In this transition zone, acidic (pH < 4) U-enriched ground water (U of 38 mg L–1) has weathered away calcite veins resulting in greater porosity, higher hydraulic conductivity, and higher U contamination (UNA 106 to 745 mg kg–1) of the weathered interbedded shale and sandstone. These characteristics of the transition zone produce an interval with a high flux of contaminants that could be targeted for remediation.

Abbreviations: DO, dissolved oxygen • DOE, Department of Energy • Eh, redox potential • ERSD-FRC, Environmental Remediation Sciences Division-Field Research Center • HIV, hydroxyl-interlayered vermiculite • I, ionic strength • Kd, linear adsorption coefficient • KPA, kinetic phosphorescence analyzer • ORR, Oak Ridge Reservation • RCRA, Resource and Conservation Recovery Act • UNA, nitric acid extractable uranium • UTEVA, uranium tetravalent actinide • WI, weathering index • XRD, X-ray diffraction


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE Y-12 National Security Complex on the Oak Ridge Reservation (ORR), Oak Ridge, Tennessee, was constructed in 1945 during World War II as part of the Manhattan Project to process enriched U for nuclear weapon production. Releases of liquid wastes containing radionuclides into the soil and geological material at this facility have resulted in highly contaminated zones within the main plant area (Oak Ridge National Laboratory, 1983). From 1951 until 1983 U and other contaminants were disposed in the S-3 waste-disposal ponds consisting of four unlined ponds that are located on the fractured Nolichucky shale. Uranium in the pond sediments ranged from 280 µg g–1 in the northeast pond to 620 µg g–1 in the southeast pond. Uranium concentrations in the liquid phase of the S-3 ponds ranged from 26 mg L–1 where the pH was 4.9 and NO3– was 8 400 mg L–1 in the upper zone of the southwest pond to 60 mg L–1 where the pH was 0.54 and NO3– was 85 000 mg L–1 in the lower zone of the northeast pond (Oak Ridge National Laboratory, 1983). Waste-disposal practices in the S-3 ponds have caused extensive contamination of ground water, which discharges into Bear Creek (Watson et al., 2005).

The structure of geological material influences ground water flow and thus, contaminant transport which takes place three-dimensionally. For example, ground water that is transported to an area through fractures that cut across bedding planes may flow slower compared with ground water that flows preferentially along bedding planes (USDOE, 1997). Stratigraphy, or the layering of differing geological material, and certain characteristics of geological material (i.e. fractures) at a site can also encourage preferential flow of ground water. The bedrock and saprolite at the ORR are highly jointed and fractured. The fractures, which can be as dense as 100 to 200 per meter (Dreier et al., 1987), can be highly interconnected and surrounded by low-permeability but highly porous matrices. Rapid preferential flow in the fractures with storage of contaminants in the porous matrix is typical of this fractured shale and saprolite (Jardine et al., 2001). The matrices have been exposed to migrating contaminants for many decades and act as secondary sources of contaminants because they store a significant portion of the inventory of total waste in the subsurface (Jardine et al., 2000). Differential weathering can take place when two different rock materials are present that weather at different rates (Phillips et al., 1997), which also contribute to preferential flow. Heterogeneous and anisotropic subsurface hydrology and geochemistry may cause many pathways of contaminant migration and stabilization resulting in difficult predictions of the prevalent reactions (Mayes et al., 2000).

The fate and transport of radionuclides through the highly jointed and fractured bedrock and saprolite on the ORR are governed by hydraulic, geochemical and microbial processes (Jardine et al., 2001). Weathering byproducts in the form of phyllosilicates and metal (oxyhydr)oxides can coat and fill void spaces, thereby acting as sites for sorption of contaminants (i.e. heavy metals and radionuclides) (Lee and Tank, 1985; Jardine et al., 1999; Roh et al., 2000; Choi et al., 2005a; Choi et al., 2005b) and reducing ground water movement, respectively (Phillips et al., 1997; Phillips et al., 1998). The dissolution and immobilization of U is affected by ground water pH and Eh under aerobic and anaerobic conditions, which will change the speciation and/or oxidation state of U (Francis et al., 1991). Other components such as nitrate and sulfate influence oxidation–reduction reactions (Murray et al., 1995). In ground water, U (IV) complexes are much less soluble and mobile than U (VI) complexes. However, U (IV) readily oxidizes to U (VI) if O is available as gaseous or dissolved species (Abdelouas et al., 1999). Uranium (VI) forms several soluble complexes with carbonate (Langmuir, 1978) and at a pH > 4 in ground water, polymeric UO2CO30, UO2(CO3) 22–, and (UO2)2/CO3)4– complexes predominate. Uranium (VI) remains as uranyl ions [U(VI)O2]2– at a pH < 5 under aerobic conditions (Eh > 0.2 V). At an Eh of <0.2 V, U (VI) reduces to U (IV) and precipitates as uraninite [U(IV)O2]. At pH > 5, under aerobic conditions, U forms various uranyl carbonate species. The formation of insoluble uraninite in alkaline oxidations requires stricter anaerobic conditions (lower Eh). Naturally occurring amorphous and crystalline Fe oxides and certain phyllosilicates (i.e. illites) in geological material are efficient sorbers of U from ground water (Lee and Tank, 1985; Sato et al., 1997). The rate and direction of ground water flow and the extent to which minerals in host rocks will adsorb U will determine the direction and extent of the migration of U (Yanase et al., 1995; Sato et al., 1997).

A Field Research Center (FRC) on the Y-12 National Security Complex on the ORR in Tennessee has been set up by the USDOE-ERSD to conduct field research and obtain ground water and geological samples related to in situ bioremediation of metals and radionuclides (Istok et al., 2004). This site has been divided into three study areas where area 3 is closest to the source (about 20 m west), in the primary contaminant flow path, and is therefore the most contaminated. Area 2 (about 240 m south) is the farthest away and the least contaminated (Moon et al., 2006), whereas the levels of contamination in area 1 (about 40 m south) are between what is detected in areas 2 and 3. This study deals with the more highly contaminated ground water and weathered shale zones found in areas 3 and 1. The objective of this study was to examine the structure, stratigraphy, and weathering, and how they affect the fate and transport of U in the ground water and the geologic material at the FRC. This information will be used to assess effectiveness of potential remedial actions, such as the bioremediation of metals and radionuclides.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Study Area
The FRC is located in Bear Creek Valley on the Y-12 National Security Complex on the ORR in Oak Ridge, Tennessee (Fig. 1 ). This site is near four formerly unlined ponds, S-3 ponds, with a storage capacity of 40 million L, where high ionic strength nitric acid and U-containing wastes were disposed between 1951 and 1983 (Oak Ridge National Laboratory, 1983; Phillips et al., 2000; Phillips et al., 2003a). In 1984, the ponds were neutralized with limestone, denitrified, capped with a Resource and Conservation Recovery Act (RCRA) cover, and paved with asphalt to construct a parking lot. This waste contained U, technetium-99, nitrate, metals, and tetrachloroethylene (Gu et al., 2002).


Figure 1
View larger version (38K):
[in this window]
[in a new window]
  		Fig. 1. Location of the cores and wells in areas 1 and 3 at the NABIR FRC site around the S-3 ponds at the Y-12 Plant.

 
The study site is in the valley and ridge physiographic province and underlain by the Nolichucky shale member of the middle-to-late Cambrian Conasauga Group formation. The S-3 ponds are located in the upper reaches of Bear Creek in the Bear Creek Valley. The study area has a humid subtropical climate. Mean annual precipitation is 1360 mm for rain and 260 mm for snow. January is the coolest mo of the yr when daily temperatures average 6.3°C. July is the hottest month with temperatures that average 28°C (Moneymaker, 1981).

Core Sampling and Descriptions
Three cores (14 m deep continuous core by 15-cm diam.) (FWB103, 104, and 105) were collected by rotasonic drilling from area 3 in a transect going from up-gradient to down-gradient away from the former ponds. Two undisturbed set of cores (about 6.1 m deep by 3.8-cm diam.) (FWB19 and FWB21) were collected from area 1 using a pneumatic hammer-drive coring device. These cores were collected to examine the lithology, stratigraphy, and weathering of the geological material. Individual cores were taken from each area 1 sampling point by driving the corer containing a polyurethane tube (1.8 m long by 3.8-cm diam.) through a hollow barrel into the geological material, retrieving the core and repeating the procedure until total depth was reached. Core material was described according to the Soil Survey Division Staff (1993). The lithology and stratigraphy were plotted using RockWorks (2002). The bedrock at the study site shows different degrees of weathering and some of this weathered bedrock does not meet the criteria to be called saprolite. According to the glossary of soil science terms (Soil Science Society of America, 2004), the definition of saprolite is the "soft, friable isovolumetrically-weathered bedrock that retains the fabric and structure of the parent rock, exhibiting extensive inter-crystal and intra-crystal weathering." Even slight weathering of geological material can influence ground water flow and contaminant transport. To better specify the amount of weathering that has occurred in the cores, a weathering index (WI) was established based on Munsell color (indicates the oxidation–reduction states of the Fe oxides in the material) and consistency (indicates the resistance of the material to deformation) to delineate the degree of weathering of the stratigraphic and lithologic units of the cores. For the shale, the following index was applied: WI 1, unweathered shale that is firm and hard (material has to be broken with a hammer), WI 2, weathered shale with strong bedding that is firm but will easily break with moderate pressure with the hands, WI 3, unoxidized friable (material can be crushed between fingers) shale with strong bedding, and WI 4, oxidized and weathered friable shale with slight to medium bedding. Due to differential weathering, a separate weathering index was established for sandstone. The WI for sandstone was WI 1, unweathered sandstone, WI 2, weathered sandstone that is easily scrapped with a knife, WI 3, weathered friable sandstone, and WI 4, weathered, very friable (material easily crushes between fingers) to friable and loose (loose single-grained sand) sandstone. Subsamples of the cores were collected at changes in lithology in given stratigraphic zones or by degree of weathering. The samples were air-dried and ground to pass through a <80 mesh sieve (180-µm openings) for U analysis.

Uranium Analysis of Core Material
To determine the amount of U retention and its association with certain lithologies or stratigraphies and/or degree of weathering in the subsurface material, nitric acid-extractable U (UNA) was extracted from about 0.5 g of core material using 15 mL 2N HNO3 (three extractions). An aliquot of the acid-extracted sample was diluted to 0.01N HNO3 and analyzed with a laser-induced automatic kinetic phosphorescence analyzer (KPA-11) (ChemChek Instruments, Richland, WA) (instrument detection limit < 0.01 µg L–1). This data was interpolated across FWB103, 104, and 105 to examine the distribution of U in the geologic material using the inverse distance algorithm in RockWorks (2002).

To determine the adsorption of U on the geological material, bulk samples of the subsurface material were air-dried and sieved through a 2-mm screen. For each sample, about 0.1 g soil was placed in 30-mL polypropylene centrifuge tubes and agitated for 48 h on a reciprocal shaker with 15 mL aqueous U (VI) as UO2(NO3)2 in a CaCl2 matrix at 298 K. The solid and aqueous phases were separated via centrifuge at 1800 rpm for 15 min and the supernatant decanted and saved for analysis. The concentration of U (VI) varied from <0.01 µg L–1 to 20 mg L–1 and the ionic strength of the systems were maintained at I = 0.015 using 5 mM CaCl2. The loss of U (VI) via adsorption on the polypropylene tubes was negligible. Aqueous U (VI) was analyzed by KPA-11. A best linear fit to the slightly nonlinear experimental data provided an estimate of the partitioning coefficient Kd in mL g–1.

Ground Water Geochemistry Analysis
Ground water was sampled for analysis from unconsolidated material (saprolite and/or weathered bedrock) in multilevel well FW101 in area 3 near the S-3 ponds (Fig. 1) at vertical depths of 6.1 to 15 m at 1.5-m increments. Ground water wells are labeled FW, while borehole cores are labeled FWB. This data was collected in order to assess the amount of mixed waste, including U that was being transported through the geological material in the ground water. Multilevel ground water-monitoring well FW101 is located in area 3 between FWB103 and FWB104 (Fig. 1) and has seven depth intervals that can be monitored. Ground water was sampled from wells FW19 and 21 in area 1 from 3.1-m screens at the bottom of the wells. Ground water pH was measured inline (without exposing to the air) as a field parameter in the multilevel wells with a precalibrated multi-parameter probe (YSI XL6000M, Yellow Springs Instruments, CO). Dissolved oxygen (DO) was measured inline with a Hach HQ20 equipped with a luminescent probe. Sulfate (detection limit < 1 mg L–1) and nitrate (detection limit < 1 mg L–1) were analyzed with a Dionex 500 ion chromatograph (Dionex, Sunnyvale, CA). Aluminum, Fe, and Mn were analyzed using a Thermo Jarrell Ash inductively-coupled plasma atomic emission spectrometer (Franklin, MA).

Ground water samples from the wells were filtered with a 0.45-µm filter and acidified with HNO3 to a pH < 2 for U and metal analysis. U (VI) in the filtered ground water samples was measured with KPA-11 (ChemChek Instruments, Richland, WA). Because of the interference of elements such as Cl– and Ca2+ with the KPA-11 analysis, the samples were purified and preconcentrated using Eichrom U tetravalent actinide (UTEVA) resin (100 to 150 µm) in 2-mL extraction columns (Eichrom Technologies, Darten, IL) (Gu et al., 2002). Ground water samples (10 mL) were acidified to a concentration of 3 M to complex uranyl (UO2+) with NO3– (Horwitz et al., 1992). These uranyl-nitrate complexes were sorbed by the UTEVA resin, and dissociated from the column with 20 mL of 0.01 M HNO3 (Gu et al., 2002). U (VI) was measured from a solution of 1 mL of leachate and 1.5 mL of URAPLEX (ChemChek Instruments, Richland, WA) enhancing reagent with the KPA-11.

Mineralogical Analysis
Untreated samples of the clay size fraction (<2 µm) from selected clay coatings, clay zones, and clay-rich highly weathered saprolite were obtained by wet sedimentation. The mineralogy of the samples, particularly clay mineralogy, was determined to examine the degree of weathering and to determine if U retention in the geological material was associated with certain clay minerals. Iron and Mn oxides were removed from the clay fraction in half of the samples using sodium citrate bicarbonate dithionite (CBD) (Loeppert and Inskeep, 1996). The other half of the samples was untreated to detect crystalline Fe oxides. The samples were then split and saturated with either 1 M KCl or 0.1 M MgCl2 (Jackson, 1975). The untreated and treated samples were oriented onto glass slides using the filter-membrane peel technique (Drever, 1973) for identification of phyllosilicate minerals. Air-dried Mg-saturated clay samples were analyzed by XRD at room temperature and after a 10% glycerol treatment to detect 2:1 expandable clays (Jackson, 1975). Air-dried K-saturated clay samples were analyzed by XRD after dried at 25°C and heat-treated at 300°C and 550°C to identify clay minerals (Jackson, 1975). Additionally, selected core samples were ground very fine, then packed into boxed mounts to obtain random orientation of the grounded materials and then analyzed by XRD to determine the mineralogy of the whole samples. The XRD analysis was performed on a Scintag XDS-2000 X-ray diffractometer (Sunnyvale, CA) using Fe-filtered Co-K{alpha} radiation operated at 40 kV and 35 mA using a continuous scan speed of 2°{theta} min–1. Cobalt K-{alpha} radiation ({lambda} = 0.179026 nm) was used for the XRD analysis to minimize fluorescence of Fe-rich minerals by Cu radiation.

Optical Microscopy
Thin sections were studied to examine the different geological material components, the extent of weathering, and the presence and distribution of weathering byproducts. Selected undisturbed samples of the core materials from areas 3 and 1 were impregnated with Epotek 301 resin (Epotek, MA). After hardening, thin sections (2 cm by 3 cm) were cut from these impregnated samples using isopropanol and ethanol to cool the saw. The thin sections were polished using 0.3-µm alumina (Materials and Chemical Laboratory, Oak Ridge, TN) and described according to FitzPatrick (1993) with a Nikon OPTIPHOT-POL microscope (Nikon, Merietta, GA) using polarized light.


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Field Characteristics of the Geological Material
Cores were collected from area 3 to a depth of 15 m. The stratigraphy of cores from area 3 is separated into three major units (i) fill, (ii) weathered interbedded shale and sandstone, and (iii) less weathered interbedded limestone, shale and sandstone. The lithology, or individual units, are identified as fill, shale, interbedded shale (predominant), and sandstone, interbedded shale and sandstone (predominant), and interbedded limestone, sandstone, and shale (Fig. 2 , Table 1). The cores from area 3 are less weathered with depth. About 1 to 3 m (average 2 m across the site) of calcareous fill (zone 1), which is composed of a mixture of limestone, soil, and saprolite from past construction activities overlies the weathered interbedded shale and sandstone stratigraphic unit. The fill is located above the ground water table in the unsaturated zone. The weathered interbedded shale and sandstone is differentially weathered (Fig. 3a ).


Figure 2
View larger version (54K):
[in this window]
[in a new window]
  		Fig. 2. Stratigraphy, lithology and the hydraulic conductivity of the cores from area 3 at the study site (mbgl = meters below ground level; V, very; Ex, extremely) (hydraulic conductivity expressed as Ki/Kavg where Kavg = 0.0005 cm s–1).

 

View this table:
[in this window]
[in a new window]
  		Table 1. Field characteristics, mineralogy, and UNA concentrations in the geological material.

 

Figure 3
View larger version (161K):
[in this window]
[in a new window]
  		Fig. 3. Photomicrographs of the weathered core material from areas 1 and 3 at the study site. a) Differential weathering of the interbedded shale and sandstone at 4.9 m from FWB24 in area 3 (Note shale is showing greater weathering compared with sandstone). b) Shale from zone 5 of FWB100 at 12 m showing calcite veins. c) Iron-rich sandstone from zone 5 of FWB100 at 12 m. d) Clayey zone from FWB21 from area 1 at 5.8 m. e) Highly weathered saprolite showing Fe oxides and clay adjacent to the clayey zone in (D) at 5.9 m.

 
The weathered interbedded shale and sandstone (unit 2) is separated into four zones (zones 2 to 5) based on color and consistence (firmness) of the shale and sandstone. Most of the upper section of the weathered interbedded shale and sandstone (zone 2) is in the vadose zone. A great part of the upper portion of zone 2 is mixed with the fill material and only about 61 cm remains undisturbed. The undisturbed portion of zone 2 is oxidized and highly weathered saprolite with very friable to friable strong brown shale and very friable to firm sandstone. Beds of loose sand grains are also present as a result of weathering of the sandstone. In zone 2, the weathering index is WI 4 for shale and WI 2 to 4 for sandstone. As this saprolite weathers, reddish brown to yellowish red (5YR 4-5/4-6), reddish brown to dusky red (2.5YR 3/1-3) Fe and Mn oxides are leached and deposited in between shale bedding and around sandstone, particularly on the tops of fine grained sandstone layers. Thin light gray (10YR 7/1-2) and gray to light brownish gray (2.5YR 6/1-3) gleyed areas are present between some bedding and around sandstone. Zone 3 is similar to zone 2, except that the shale saprolite is light olive brown (2.5Y 5/4) in color in places. Zone 3 marks the beginning of the saturated zone and has a weathering index of WI 3 for shale and WI 2 to 3 for sandstone.

Zones 4 and 5 are noticeably less weathered compared with zones 2 and 3. In zone 4, the light olive brown (2.5Y 5/4) shale is friable and the yellowish brown (10YR 5/4) sandstone is friable to very firm. In zone 4, the weathering index is WI 2 to 3 for shale and WI 2 to 3 for sandstone. Zone 5 is the transition zone between the interbedded shale and sandstone and underlying predominately limestone bedrock in zone 6. In zone 5, the weathering index is WI 1 to 3 for shale and WI 1 to 2 for sandstone. The shale is calcareous where there is less weathering (Fig. 3b) and grayish brown (2.5Y 5/2) to light olive brown (2.5Y 5/4) and leached of calcite in zones where there is more weathering. At the interface of sandstone layers and overlying shale lithologies in zone 5, the thin upper portion of the sandstone is friable and black (2.5YR 2.5/1) to dark reddish brown (5YR 3/3) due to the accumulation of reduced Fe oxides (Fig. 3c) leached from the overlying adjacent gleyed, very friable Fe-bearing shale. These black Fe oxide zones are absent from the firm underside of the sandstone layers. Below the interbedded shale and sandstone stratigraphic layer, interbedded limestone, shale, and sandstone bedrock (zone 6) with small sections of gray to light brownish gray (2.5Y 6/1-2) and grayish brown to light brown shale (2.5Y 5/2-3, 4/3) saprolite are present.

Cores FWB19 and 21 were collected from area 1 to a depth of 6.1 m (Fig. 1). Area 1 cores have a similar stratigraphy between each core, which consists of weathered Nolichucky shale that underlies about 3.1 m of calcareous fill material (limestone gravel, soil, and saprolite). This weathered shale is composed of Fe-rich dusky red (2.5YR 3/3) claystone that is very firm, but highly fractured and jointed and does not meet the criteria of saprolite (Fig. 4 ). Seams (2.5 to 25 cm) of Fe oxide-rich strong brown (7.5YR 5/6) clay that retain faint rock structure and have gleyed light gray (10YR 7/1-2) streaks (Fig. 3d, 4) are generally present between thin zones (2 to 5 cm) of highly weathered olive (5Y 5/4) and olive brown (2.5Y 3-4/4) saprolite that retains rock structure. The saprolite has dusky red (2.5YR 4/4) and dark brown (7.5YR 3/2) Fe and black (10YR 2/1) Mn coatings on the fracture faces (Fig. 3e).


Figure 4
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 4. Stratigraphy and lithology of the cores from area 1 at the study site (mbgl = meters below ground level).

 
Ground Water Geochemistry
Uranium, sulfate, and Al concentrations have similar trends with depth in three area 3 multilevel wells (seven ports each), with the highest concentrations occurring in stratigraphic zone 5 where ground water flow is also the greatest (Fig. 2, 5) . There appears to be an inverse relationship with U, Al, and sulfate concentrations to the pH of the ground water throughout the depth interval of the multilevel wells, and there is excellent direct linear correlation between U and Al concentrations in groundwater with an r2 of 0.98 (Fig. 6 ). Uranium, sulfate, nitrate, Al, Mn, and Fe concentrations, and pH measurements of ground water remain fairly constant through zone 3 for FWB101 (Fig. 5). Zone 4 core material was wet at the depths of 9.1 m in FWB104 near FW101 and 8.8 m in FWB103 where there are underlying firm sandstone layers. This is where hydraulic conductivity, and U, sulfate, nitrate, Fe, Mn, and Al concentrations increase and pH decreases. In well FW101, U concentrations increased from 0.4 mg L–1 at 6.1 m to 3.1 mg L–1 at 11 m. Uranium and SO42– concentrations in area 3 ground water increase, whereas pH decreases between 11 and 14 m in the interbedded calcareous shale and sandstone (zone 5) in FW101. Uranium is highest at 38 mg L–1 and SO42– is highest at 3042 mg L–1 concentrations in FW101 at 12 m. In the limestone bedrock (zone 6), pH increased to 5.5 in FW101 at 15 m. Nitrate increases from 1100 mg L–1 at 6.1 m to 28 000 mg L–1 at 15 m in FW101. These values are comparable to concentrations of U at about 7 mg L–1, NO3– at about 12 000 mg L–1 and a pH of about 3.3 to 6.7 in area 1 wells FW19 and 21.


Figure 5
View larger version (23K):
[in this window]
[in a new window]
  		Fig. 5. Distribution of pH, U (IV), NO3–, SO42–, Fe, Al, and Mn in ground water with depth in multilevel well FW101.

 

Figure 6
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 6. The relationship between U (IV) and Al over a pH range of 3 to 7 in ground water from FW100.

 
Dissolved oxygen (DO) measured in a cluster of wells about 9 m away from FW101 shows a decreasing trend with depth. The results were 0.53 mg L–1 at 9.7 m, 0.24 mg L–1 at 12 m, and 0.17 mg L–1 at 14 m for the three ports in multilevel well FW113; however, FW114, which is only 6 m deep had a DO of 4.5 mg L–1.

Byproducts of Weathering
In areas 3 and 1, shale interbeds are observed in thin sections as thin horizontal laminations of mica-derived minerals and quartz that form claystones and siltstones. The mica is mostly illite, fine-grained mica (mainly biotite), that is weathering to phyllosilicate minerals including kaolinite, vermiculite, hydroxyl-interlayered vermiculite (HIV), Fe oxides (goethite) as observed in thin sections (Fig. 3c and 3d), and X-ray diffractograms (Fig. 7 ).


Figure 7
View larger version (50K):
[in this window]
[in a new window]
  		Fig. 7. X-ray diffractograms of clay fractions from a) FWB 19 at 2.5 to 2.6 m, the vadose zone, b) FWB 19 at 5.3 to 5.4 m, c) FWB 21 at 5.7 to 5.8 m, and d) FWB24 at 6.7 m in the saturated zone (HIV, hydroxy-interlayered vermiculite; I, illite; K, kaolinite; Q, quartz).

 
The highly weathered shale in core FWB24, located near FWB103 in area 3, has an illite > > kaolinite > vermiculite > goethite > quartz clay mineralogy at 6.7 m in the upper section of the saturated zone (zone 3) (Fig. 7b). Clay mineralogy was not determined on lower portions of zone 3 to zone 6 in area 3 cores because of a lack of clay minerals due to less weathering. A similar suite of minerals was detected by XRD in whole-sample fractions of sandstone (quartz > Ca feldspar > vermiculite > mica > goethite) and shale (vermiculite > mica > HIV > Ca feldspar > quartz) from zone 5, however the amount of these minerals varied between the two different rock types (Fig. 8a ).


Figure 8
View larger version (35K):
[in this window]
[in a new window]
  		Fig. 8. X-ray diffractograms of the finely ground whole saprolite material from a) FWB24, and b) FWB19 (HIV, hydroxy-interlayered vermiculite; I, illite; K, kaolinite; G, goethite; Q, quartz; V, vermiculite).

 
The XRD analysis of FWB19 from area 1 indicates that the clay band at 2.5 to 2.6 m has a clay mineralogy of kaolinite > illite >> vermiculite > HIV > goethite > quartz in the vadose zone (Fig. 7a). The clay bands in FWB19 at 5.3 to 5.4 m and FWB21 at 5.7 to 5.8 m had a greater amount of HIV and vermiculite compared with the clay band in FWB19 at 2.5 to 2.6 m (Fig. 7a, 7b, 7c) in the saturated zone. Iron oxide-segregated areas are observed in thin sections throughout the clay bands. Sharper peaks for goethite from area 1 are observed in the X-ray diffractograms than for goethite from area 3 (Fig. 7). This indicates that the goethite from area 1 is more crystalline. Core FWB24 from area 3 had more crystalline primary minerals present, especially quartz, as indicated by sharper XRD peaks and less Fe oxides with a lower crystallinity compared with the mineralogy of cores from Area 1 (Fig. 8a, 8b).

Uranium Distribution in the Geological Material
The UNA concentration in the fill material from area 3 ranges from <0.01 to 100 mg kg–1, but most fill has <50 mg kg–1 UNA (Fig. 9 ). The UNA concentration ranged from <0.01 to 50 mg kg–1 in zone 2, 10 to 200 mg kg–1 in zone 3, and was mainly 50 to 100 mg kg–1 in zone 4. In zone 5, there is a notable increase in UNA concentrations ranging from 106 mg kg–1 in core 104 to 745 mg kg–1 in FWB103, where the highest UNA concentrations are associated with the thin black zones. In FWB105, which has the thickest zone 5 compared with FWB103 and 104 (Fig. 4), there are overall lower amounts of UNA ranging from 1 to 386 mg kg–1. Zone 6 generally has between 0.2 and 7 mg kg–1 UNA; however, FWB104 has UNA concentrations from 300 to 400 mg kg–1 at the top of some sandstone layers in this zone. This is where contaminated ground water has moved into zone 6 through fissures created by the weathering out of CaCO3 veins in the shale, thus depositing U on sandstone layers.


Figure 9
View larger version (51K):
[in this window]
[in a new window]
  		Fig. 9. Contouring of the distribution of the UNA in the geological material in area 3. Note that the greatest amount of UNA occurs in stratigraphic zone 5. The thin horizontal zones of UNA concentrations show how UNA is controlled by the lithological units.

 
In area 1 cores, UNA concentrations range from 18 to 38 mg kg–1 in the fill material and about 50 mg kg–1 in the weathered shale lithology. The clay bands have the highest amount of UNA (239 to 375 mg kg–1), compared with other parts of the cores from area 1. The saprolite adjacent to the clay bands has lower UNA concentrations of 67 to 110 mg kg–1.

Figure 10 shows the U adsorption isotherms for three samples collected from area 3 boreholes. The linear adsorption coefficient (Kd) at a pH of 4 for three samples collected at various depths from FWB100 and FWB101 ranged between 30 and 42 mL g–1.


Figure 10
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 10. Isotherms of U adsorption on geological material from boreholes FWB100 and 101 (I = 0.015 using 5 mM CaCl2).

 

   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Effects of Structure and Stratigraphy on Preferential Flow and Diffusion of Contaminated Ground Water
The structure of the geological material plays a major role in directing the buffered acidic ground water containing U and other contaminants away from the ponds and across the study area. At the FRC site, ground water is preferentially migrating along bedding plane fractures and towards the nearby tributaries of Bear Creek and discharging through seeps in the creek bed (Phillips et al., 2000; Phillips et al., 2003a; Gu et al., 2002). The underlying Nolichucky Shale bedrock and saprolite at the site dips approximately 45° to the southeast and has a strike of N55E which is parallel to Bear Creek Valley (Solomon et al., 1992) (Fig. 11 ). Cores from area 3 (about 20 m), which are closer to the former ponds relative to area 1 cores (about 40 m), were collected from the dominant flow path that follows along the strike and dip of fractured bedding planes (USDOE, 1997) where the acidic contaminated ground water (pH < 4) flow is greatest (Fig. 1), which has caused U contamination and enhanced weathering of the core material. Area 1 is U-contaminated and weathered, but tangential to the primary flow path where contaminated ground water was transported to this area through fractures that cut across bedding planes and are generally less permeable and interconnected than the bedding plane fractures (USDOE, 1997). Therefore, since the geological material has not been exposed to the low pH, high U, nitrate, and other contaminants to the same extent as area 3 geologic materials, it is less weathered and contaminated compared with cores from area 3.


Figure 11
View larger version (82K):
[in this window]
[in a new window]
  		Fig. 11. Generalized diagram of the flow of the contaminants along the strike and dip of the saprolite and bedrock bedding planes at the NABIR-FRC site.

 
Stratigraphy and other geological material characteristics (i.e. fractures) also influence preferential flow of acidic U-contaminated ground water through the study areas. Primary contaminant transport occurs in a more permeable stratigraphic transition zone (zone 5) in area 3 at 11 to 13 m between overlying unconsolidated saprolite (zones 2 to 4) and an underlying, less permeable, and less weathered bedrock (zone 6). The overlying saprolite has fractures and void space plugged with clays and oxides making this material less permeable. However, diffusion of contaminants may take place along small fractures and into the silt, clay, and/or shale pores, whereas preferential flow occurs through wider fractures (Fig. 11). Due to a combination of high fracture density and lower clay content, the transition zone (zone 5) has higher permeability and contaminant transport compared with the overlying saprolite and underlying bedrock. Hydraulic conductivity vectors have a high degree of anisotropy with the greatest value oriented parallel to bedding and the least oriented perpendicular to bedding (Hatcher et al., 1992), as observed at the study site. Hydraulic conductivities as high as 1 x 10–3 to 1 x 10–2 cm s–1 (Solomon et al., 1992) and 0.002 to 0.006 cm s–1 (Fienen et al., 2004) are reported for this transition zone, while hydraulic conductivities between 1 x 10–4 and 1 x 10–7 cm s–1, usually occur in the overlying clay-rich saprolite (<0.001 cm s–1) (Solomon et al., 1992), underlying limestone bedrock unit, and overlying clayey saprolite (Fienen et al., 2004). The hydraulic conductivity increases by a factor of 16 in FW26 in this zone where U concentrations in core material are highest, associated with less soluble sandstone layers, and where the ground water flows on the surface of these layers. In less contaminated area 1 shale which has bedding that is tightly bound and compact, acid U-contaminated ground water diffuses along small fractures and into pores, whereas preferential flow occurs in wider fractures weathering the surrounding shale into thin (<25 cm) Fe oxide-rich clayey seams that retain U and are surrounded by thin layers of saprolite. These bands have redoximorphic features in the form of Fe-rich concretions, segregated zones, and streaks of gleying, which indicate the presences of water moving through the clay bands. These clay bands appear to be pathways for localized contaminant transport but not as significant to regional transport as the high Kd zones found in zone 5 and in area 3.

Effects of Weathering on Preferential Flow
The cores from area 3 are both naturally and anthropogenically weathered. These cores are also less weathered with depth because natural processes that weather the soil and saprolite from the surface downward have been occurring at the site for hundreds of thousands of years. Water percolates through joints and fractures of Nolichucky shale, causing oxidation and hydrolysis reactions to occur at considerable depths (Lietzke et al., 1988).

Saprolite (3 to 10 m thick) in area 3 consists of weathered interbedded shale and sandstone and underlies about 2 m of fill. The saprolite retains some of the original structural fabric of the rock including remnant bedding planes and fractures (Hatcher et al., 1992). The weathered interbedded shale and sandstone in area 3 is differentially weathered where the sandstone is less weathered than the shale (Fig. 3a), which weathers to clayey material. This encourages differential ground water flow through the material and impacts contaminant transport (Jardine et al., 2000). Zones 2 through 4 in area 3 contain saprolite and show greater natural weathering compared with underlying zones 5 and 6. However, close to half of the weathered materials in zone 4 do not meet the criteria for saprolite because they are too firm (Table 1). The permeability of the silt and clay saprolite matrix is low, due to weathering which resulted in infillings and coatings of clay and oxides in the pores and fractures, thereby reducing the permeability and flow of ground water. Nevertheless, the overall permeability can be higher due to the presence of fracturing along remnant bedding planes. Uranium contamination indicates that acidic contaminated groundwater has also diffused through this material and therefore plays some role in weathering.

In area 3, most of the weathered shale and sandstone in the transition zone (zone 5) have a firm consistency so they do not meet the criteria to be called saprolite. A greater amount of anthropogenic weathering has occurred in zone 5 compared with overlying zones 2 through 4. Greater void space (fracture aperture) in zone 5 is partially due to dissolution of CaCO3 veins from the shale by the rapid flow of acidic ground water with higher U contamination, which has resulted in higher amounts of UNA in the solid phase. In FWB105, which has the thickest zone 5 compared with FWB103 and FWB104 (Fig. 2), there are lower overall amounts of UNA because it is farther away from the source compared with other cores in area 3, therefore contamination becomes more diffuse.

In area 1, saprolite and clayey bands in the shale are possibly partially the result of weathering from the acidic ground water from the S-3 ponds that has preferentially flowed through fractures. The clay bands are a result of both migrating clay and clay that has formed in situ from the weathering of the shale into saprolite, then into clay as small shale saprolite fragments are observed in the clay bands. There were signs of clay migration in the soil thin sections where some, but not all, of the clay appeared highly oriented under cross-polarized light.

Effects of Weathering Byproducts on Uranium Distribution
Weathering of the Nolichucky shale bedrock at the FRC site has produced clay minerals which contain phyllosilicates and metal oxyhydroxides that coat the surfaces of void spaces in the geological material, and react with the contaminants, especially U. Bertsch and Seaman (1999) report the surface-reactive phases of soils and aquifers containing phyllosylicates and metal oxyhydroxides minerals play a critical role in contaminant fate and transport.

The geological material is rich in biotite and Ca feldspars which weather to the phyllosilicates in the clay fraction. The weathering of biotite from areas 3 and 1 generally follows the following pathway with increasing alteration of biotite to illite to vermiculite to kaolinite by the increasing release of K, Mg, and Fe from the structure. In the area 1 clay fraction, hydroxyl-interlayered vermiculite is also present as a result of the weathering of vermiculite (Barnhisel and Bertsch, 1989). The Ca feldspars weather directly to kaolinite by releasing Ca (Arocena and Sanborn, 1999).

In zone 2 of FWB24 from area 3 there was a greater abundance of kaolinite and an absence of HIV due to the large amount of weathering of the saprolite compared with FWB19 and 21 in area 1. Similar to the clay mineralogy of FWB24, there were increases in kaolinite in highly weathered soil and saprolite formed from Knox formation on the ORR (Arnseth and Turner, 1988). The Knox formation is mainly dolostone (Harris, 1971; Millici, 1973), that contain up to 50% chert, quartz, feldspars, pyrite, and micas (Miller, 1972), with occasional beds of limestone, shale, and sandstone (Harris, 1971; Millici, 1973). This mineralogy is similar to the mineralogy of area 3 sandstone beds.

In area 1 the increase in vermiculite and HIV in the saturated zone, compared with the vadose zone could have been produced by dissolution. Clay mineralogy of the clay bands in the saturated zone is similar to a poorly drained soil and saprolite formed in interbedded Maryville limestone and Nolichucky at the ORR (Arnseth and Turner, 1988). Similar clay mineralogy to the clay band in the vadose zone in area 1 was observed for a well-drained soil and saprolite formed in interbedded Maryville limestone and Nolichucky shale on the ORR (Arnseth and Turner, 1988). The absence of smectite from the core samples in our study is perhaps due to the low pH (pH < 4). However, higher pH values of 4.5 to 6 are reported for fractured, weathered shales at the ORR (Jardine et al., 2001). The presence of illite, HIV, and vermiculite may help retain U in the clay bands. Uranium (VI) adsorption to clay minerals varies due to the type of clay, with illites being greater sorbers of U compared with kaolinite (Lee and Tank, 1985).

Goethite, a byproduct of the weathering, is present in the form of coatings on the surfaces of minerals, void spaces, and as segregations and/or concretions in core material from areas 3 and 1 (Fig. 3c, 3d, 3e). Iron oxides are reported to heavily coat minerals in interbedded shale-limestone saprolite on the ORR (Lee et al., 1990; Phillips et al., 1997; Phillips et al., 1998; Barnett et al., 2000; Phillips et al., 2001). The UNA is highest in the thin black zones on the surface of the sandstone layers where Fe oxides were leached down from the overlying reduced shale in zone 5 of area 3 cores. Metals (Fe, Al, Mn) released from the geological material and in ground water increase along with U in this zone (Fig. 5) and are also deposited on the sandstone layers. However, there is no statistical correlation between UNA and oxides in zone 5 and in all of the core material from area 3 (Phillips et al., 2003b). The black color of the Fe accumulation layers indicates that Fe is in a reduced state in an anoxic environment which is supported by DO concentrations in nearby wells. The underlying alkaline carbonate-rich zone (zone 6) induces the precipitation of the Fe oxides. Also, after drying, core samples from this zone redden due to oxidation of the Fe oxides. At the study site, the U is in the U (VI) state (Kelly et al., 2005) and in the form of the uranyl ion which is relatively mobile (Langmuir, 1978), and less conducive to being adsorbed in the acidic anaerobic environment. Uranium (VI) adsorption on samples from area 3 borehole FWB104 are highly dependent on pH (Wu et al., 2006). Peak sorption occurs at pH 6 to 6.5, decreasing above and below that range, consistent with prior studies for similar soils (Barnett et al., 2002). Cores FWB19 and 21 from area 1 had a strong correlation between Fe oxides and UNA (Phillips et al., 2003b). The adsorption of U (VI) to ferrihydrite (amorphous iron hydroxide) and goethite (crystalline Fe oxyhydroxide) is well known (McKinley et al., 1995; Turner et al., 1996; Prikryl et al., 2001). Also, goethite is reported to adsorb more U (VI) compared with phyllosilicate minerals (Jong et al., 1999). More research is needed to determine the adsorption of U on different Fe oxides at the study site.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Complex interactions between the solid and liquid phase over the past 54 years at the FRC site has resulted in the contaminant distribution currently observed in ground water and in the solid phase. This study shows that structure, stratigraphy, and weathering of geological material have impacted the fate and transport of U and other contaminants across the study areas at the ERSD FRC. Structure of the geological material has played an important role in directing the buffered acidic (pH < 4) groundwater containing U and other contaminants across the study areas. Ground water flows preferentially along the strike and dip of fractured bedding planes, which has resulted in greater flow in area 3, compared with less permeable, less interconnected, impeded ground water flow through fractures that cut across bedding planes in area 1. Therefore, there is greater U contamination and enhanced weathering of the core material in area 3, since weathering happens even without contaminated ground water, compared with area 1 core material. Stratigraphy and other characteristics of the geological material (i.e. fractures) influenced preferential flow of acidic U-contaminated ground water through the study areas, resulting in enhanced weathering and the production of weathering byproducts (i.e. illite and goethite) that may retain U in certain places in the subsurface. Greatest contaminant transport occurs in a more permeable stratigraphic transition zone in area 3 at 11 to 13 m where acidic U-contaminated ground water has differentially weathered away calcite veins, resulting in greater porosity, higher hydraulic conductivity, and higher U contamination of the weathered interbedded shale and sandstone. In the less contaminated area 1, ground water diffuses along small fractures and into pores in the compact shale, while preferential flow occurs in wider fractures weathering the shale surfaces into thin Fe oxide-rich clayey seams that retain U. More research is underway to study the adsorption of U on different Fe oxides at the site which can be used to determine the availability of the U for bioremediation. This study provides detailed information on some of the major geological processes that affect the fate and transport of U and other contaminants across the ERSD FRC study site. These findings suggest that more limited remediation strategies focused on reducing contaminant migration in the relatively thin high contamination, high flow zone 5 could significantly reduce the flux of contaminants leaving the site.


   ACKNOWLEDGMENTS
 
We thank Sharon Graves for her help with the U adsorption analysis. We also gratefully acknowledge support from the U.S. Department of Energy, Office of Science, Biological and Environmental Research Program. This project was supported under the DOE Environmental Remediation Sciences Program. ORNL is managed by UT-Battelle, LLC, for the U.S. Department of Energy under contract DE-AC05-00OR22725.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The submitted manuscript has been authored by a contractor of the U.S. government under contract no. DE-AC05-00OR22725. Accordingly, the U.S. government retains a non-exclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. government purposes.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES