Tailings and Mine Waste Conference

Uranium diffusion in soils and rocks Moore, Stephanie M. 2011

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
Moore_S_et_al_TMW_2011.pdf
Moore_S_et_al_TMW_2011.pdf
Moore_S_et_al_TMW_2011.pdf [ 223.44kB ]
Moore_S_et_al_TMW_2011.pdf
Metadata
JSON: 1.0107718.json
JSON-LD: 1.0107718+ld.json
RDF/XML (Pretty): 1.0107718.xml
RDF/JSON: 1.0107718+rdf.json
Turtle: 1.0107718+rdf-turtle.txt
N-Triples: 1.0107718+rdf-ntriples.txt
Citation
1.0107718.ris

Full Text

Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Uranium Diffusion in Soils and Rocks Stephanie M. Moore Civil and Environmental Engineering, Colorado State University, Fort Collins, Colorado, USA Charles D. Shackelford Civil and Environmental Engineering, Colorado State University, Fort Collins, Colorado, USA Abstract The results of a study undertaken to review the process of diffusion of uranium in soils and rocks are presented and discussed. After some background on the geochemistry of uranium and the different definitions for the diffusion coefficient in porous media, i.e. D*, De, Da, values of D*, De, and/or Da from the literature are summarized, analyzed and discussed. Diffusion of uranium in compacted bentonite is shown to increase semilog linearly with an increase in porosity (decrease in dry density), and appears to be significantly slower than that in other soils, probably due to the existence of semipermeable membrane behavior and interparticle/intraparticle diffusion as an attenuation mechanism. The effect of attenuation is shown to increase the variability in Da relative to that in either D* or De, and the variability in D*, De, and Da is shown to be greater in compacted bentonite than in unfractured biotitic granite, likely due to a more homogeneous pore structure and a lesser attenuation capacity for the biotitic granite. Overall, the typical preference for reporting only Da values is shown to severely restrict the usefulness of the literature on the subject. Introduction Despite the recent nuclear crisis at the Fukushima Dai-ichi nuclear power plant in Japan, interest in the USA in the use of nuclear power as a source of energy to reduce both global greenhouse gas emissions resulting from fossil-fuel generation of electricity and the dependence on the use of imported oil remains strong. As a result, there is renewed interest in uranium mining in the USA as a source of fuel for nuclear power plants as well as several other applications. Also, uranium mining activity remains strong in several other countries, including Canada and Australia, and the potential environmental impact resulting from disposal of uranium based mill tailings and waste rock is an ongoing issue. Uranium contamination of soil can occur due to natural processes acting on uranium inherently present in soil and rock, or from anthropogenic activities related to the extraction and processing of uranium from soil and rock for industrial applications, e.g. for use as fuel for nuclear power production and as a fissionable material in nuclear weapons. Examples of soil contamination resulting from natural processes include the deposition of uranium originally discharged into the atmosphere via wind erosion and volcanic activity or as a result of water erosion, dissolution, and precipitation. Examples of anthropogenic activities that may result in soil contamination include uranium mining and milling, uranium processing, phosphate mining, heavy metal mining, coal use and inappropriate waste disposal (Gavrilescu et al. 2009). Although soil contamination by uranium from natural processes far exceeds that from anthropogenic activities (Gavrilescu et al. 2009), such activities as mining and milling operations result in potential localized sources of uranium in the forms of waste rock and mill tailings, respectively, that can lead to significant environmental impacts if not contained properly. Accordingly, a study was undertaken to review the literature pertaining to the diffusion of uranium in soils and rocks. The primary purposes of the study were to evaluate the ranges of diffusion coefficients with respect to uranium diffusion in soils and rocks, to determine any trends in the values, and to identify any missing information for the purpose of directing future research efforts. The study was limited to include only liquid-phase diffusion of uranium through natural soils and rocks or compacted soils used as engineered barriers (e.g. compacted bentonites) and, therefore, does not include gas-phase Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 diffusion of radon or other decay products of radioactive uranium, nor diffusion coefficients resulting from stabilization/solidification of uranium with pozzolanic materials (e.g. cement). Geochemistry of Uranium Uranium (U) is a metallic chemical element in the actinide series of the periodic table, with an atomic number of 92. Uranium occurs naturally in low concentrations of a few parts per million in soil, rock and water, and is commercially extracted from uranium-bearing minerals such as uraninite. Although there are six unstable isotopes (i.e. radioisotopes) of uranium with between 141 and 146 neutrons, only uranium-238 (238U), uranium-235 (235U) and uranium-234 (234U), occur naturally, with 238U and 235U being the most abundant at 99.27% and 0.72%, respectively (Gavrilescu et al. 2009). Uranium decays slowly via emission of an alpha particle, such that the half lives of the 238U, 235U, and 234U are 4.47 billion, 245.5 thousand, and 704 million years, respectively (Gavrilescu et al. 2009). Uranium in aqueous solutions forms many chemical species as a result of hydrolysis and complexation reactions. These reactions are a function of the chemical conditions, especially the pH, oxidation– reduction (redox) potential (Eh), and the concentrations of the complexing ligands, such as [OH]−, [HCO3]−, [CO3]2−, [H2PO4]− [HPO4]2−, [PO4]3−, [SO4]2−. These reactions control the mobility of uranium in soils, primarily because of the different charges of the various chemical species. For example, uranium can exist in the +3, +4, +5 and +6 oxidation states, although the +4 (U(IV)) and +6 (U(VI)) oxidation states are the most important because only U(IV) and U(VI) are stable in aqueous solution. Thus, redox reactions that convert soluble U(VI) to insoluble U(IV) and vice versa are particularly important. Also, the formation of negatively charged uranium-ligand complexes, such as the uranium carbonate species (e.g. (UO2)2CO3(OH)3−, UO2(CO3)22−, UO2(CO3)34−, (UO2)3(CO3)66−), generally decreases the sorption of uranium to soils and, therefore, increases uranium mobility. Deposits of uranium generally contain relatively insoluble U(IV) in the form of the mineral uraninite, UO2(c).  However, upon exposure of these deposits to water and oxygen during mining activities and deposition of uranium bearing mine wastes, U(IV) is oxidized to U(VI) in the form of the aqueous soluble uranyl cation, UO22+ (Flury & Harsh 2000). Oxidation from U(IV) to U(VI) increases the solubility of the uranium and, therefore, the potential for release of uranium into the environment. The subsequent fate of U(VI) released from uraninite is governed by precipitation, the formation of solution complexes, and sorption/desorption to/from the solid phase due to surface complexation or ion exchange mechanisms of sorption (Bai et al. 2009). For example, if the Eh of the system decreases to a value below about 100 mV, precipitation of highly insoluble U(IV) minerals is possible (Zielinski et al. 1987). Also, uncomplexed UO22+ has a greater tendency to form complexes with organic substances, such as the fulvic and humic acids, than many other metals with a +2 valence (Gavrilescu et al. 2009). Finally, U(VI) is strongly sorbed to mineral constituents, such as iron oxides (e.g. hematite) and ferric oxyhydroxides (e.g. goethite), particularly for pH in the range 5 ≤ pH ≤ 8.5 (e.g. Langmuir 1978, Hsi & Langmuir 1985, Bruno et al. 1995, Duff & Amrhein 1996). However, carbonate complexing of U(VI) can appreciably reduce this adsorption (Gavrilescu et al. 2009). Diffusion Significance of Diffusion Diffusion of contaminants is a significant, if not dominant, contaminant transport process in scenarios where advection (i.e. hydraulically driven transport) is slow or negligible (Shackelford 1988). Such scenarios typically involve low-permeability (≤ 10-9 m/s), fine-grained soils (e.g. clay), and include analyses related to evaluation and design of engineered containment barriers for waste disposal (e.g. Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 landfills, surface impoundments, etc.), analyses related to the migration of contaminants in low permeability strata (i.e. aquitards and aquicludes), and analyses related to the remediation of existing contaminated soils (Shackelford & Lee 2005). In particular, the process of matrix diffusion, whereby contaminants diffuse from interconnected pores into the surrounding intact clay or rock matrix, may be an important attenuation mechanism when contaminant transport occurs through structured clay and rock formations (e.g. Parker et al. 1994). Finally, Bai et al. (2009) note the importance of intragranular diffusion of U(VI) in sediments from the DOE site at Hanford, WA, and the importance of interparticle diffusion for radionuclide waste disposal at Yucca Mountain, NV, and elsewhere. Thus, knowledge of diffusion of uranium in soils and rocks is important in mining and mine waste scenarios. Governing Equations Diffusion in saturated porous media is governed by Fick's first and second laws. Fick's first law for one-dimensional diffusion may be written as follows (Shackelford & Daniel 1991): * d cJ D x ∂ = −ε ∂  (1) where Jd is the diffusive mass flux of the chemical species perpendicular to the total cross-sectional area of the porous medium [ML-2T-1, where M = mass, L = length, and T = time],  ε is the porosity, D* is the effective diffusion coefficient [L2T-1], c is the concentration of the chemical species defined with respect to the volume of pore water [ML-3], and x is the distance in the direction of diffusion [L]. The effective diffusion coefficient given by Eq. 1 is related to the aqueous-phase or "free-solution" diffusion coefficient for the same chemical species, Do [L2T-1] as follows: * a oD D= τ  (2) where τa is the dimensionless apparent tortuosity factor (0 < τa < 1) as defined by Shackelford & Daniel (1991). Malusis & Shackelford (2002) further defined τa as the product of the actual matrix tortuosity factor, τm (0 < τm < 1), which accounts for the reduction in diffusion due to the geometric tortuosity of the network of connected pores that exist within the porous medium, and a restrictive tortuosity factor, τr (0 < τr ≤ 1), as follows: ( )1 2 1 N a m r m i m N i= τ = τ τ = τ τ = τ τ τ τ∏ L  (3) where N represents the number of all other factors, τi, that act to reduce or restrict the diffusive solute flux through the porous medium, such as ion exclusion in clays, constrictivity due to non-uniform cross-sectional areas of pores, dead-end or otherwise non interconnected pores, etc. Values of τa for a given porous medium generally cannot be determined a priori and, therefore, must be back calculated by dividing a measured value of D* for a given porous medium using a nonreactive tracer, typically an anion (e.g. Cl-, Br-) or radioisotope of water (e.g. tritium or deuterium), by the value of Do for the same tracer in accordance with Eq. 2. The τr appearing in Eq. 3 also may be considered as the factor by which ε is reduced to an effective or through-diffusion porosity, εeff. For example, substitution of Eqs. 2 and 3 into Eq. 1 results in the following alternative form of Fick's first law (e.g. Pearson 1999): ( )( )d a o m r o r m o eff pc c c cJ D D D D x x x x ∂ ∂ ∂ ∂ = −ετ = −ετ τ = − ετ τ = −ε ∂ ∂ ∂ ∂  (4) Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 where εeff is equal to the product τrε, such that εeff ≤ ε, and Dp (= τmDo) is referred to as the "pore diffusion coefficient" (Skagius and Neretnieks 1986). The distinction between Eqs. 1 and 4 is important for two reasons. First, when τr < 1, the value of D* (Eq. 2) will be less than the value of Dp (Eq. 4). Second, both εeff and Dp in Eq. 4 are unknown parameters, i.e. since both τm and τr generally are not known a priori, whereas only D* is unknown in Eq. 1, i.e. since ε generally is known a priori. This second reason is important because εeff often is assumed to be equal to ε, such that values of D* defined in accordance with Eq. 2 are reported even though Fick's first law has been defined in accordance with Eq. 4. As a result, Fick's first law as given by Eq. 1 was in this study. An alternative definition of the effective diffusion coefficient commonly encountered is one that includes the porosity, or: * e a oD D D= ετ = ε  (5) where De is the effective diffusion coefficient. When the effective diffusion coefficient is defined by Eq. 5, Fick's first law for diffusion in saturated porous media must be written as follows: d e cJ D x ∂ = − ∂  (6) In accordance with Eq. 5, since ε < 1 for porous media, D* > De. The important difference between Eqs. 1 and 6 is that D* is independent of ε whereas De is not. Fick's second law governing transient diffusion of a reactive chemical species subject to sorption and first-order linear decay in porous media may be written as follows: 2 2 2 2 2 2 * e a d Dc D c c c c c D c t R x x x ∂ ∂ ∂ ∂ = − λ = − λ = − λ ∂ α∂ ∂ ∂  (7) where Rd is the retardation factor, α is the rock capacity factor (i.e. Skagius & Neretnieks 1986), λ is the decay constant [T-1], and Da is the apparent diffusion coefficient. For reactive chemical species subject to linear, reversible, and instantaneous sorption, Rd > 1, whereas for non-reactive chemical species, Rd = 1. The rock capacity factor represents the volumetric capacity of the porous medium for the chemical species, and is related to the retardation factor as follows (Pearson 1999): dRα = ε  (8) Therefore, since ε < 1, α < Rd. The decay constant, λ [T-1], is inversely proportional to the half-life of a given radionuclide such that, for a non-decaying, non-reactive chemical species, λ = 0, Rd = 1, and α = ε. As previously noted, the half lives of the three dominant radionuclides of uranium, i.e. 238U, 235U, and 234U, are of such high magnitudes radioactive decay is negligible for most practical time frames. As is apparent from Eq. 7, the relationship among Da, D*, and De is as follows: * e a d DDD R = = α  (9) Thus, Da represents a lumped effective diffusion coefficient that includes the effect of attenuation via either Rd or α. For this reason, Da also has been referred to as the effective diffusion coefficient of a reactive chemical species (Shackelford & Daniel 1991). A comparison of Eqs. 2, 5, 8, and 9 reveals that the general order in the magnitudes of Da, D*, and De is D* > De > Da when Rd-1 < ε and D* > Da > Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 De when Rd-1 > ε. This latter relationship typically governs for diffusion through intact (unfractured) rock where sorption is minimal and the matrix porosity is low (e.g. see Parker et al. 1994). RESULTS AND DISCUSSION Diffusion of Uranium The diffusion coefficients for a variety of uranium chemical species diffusing in saturated, compacted specimens of different types of bentonites are summarized in Table 1. The focus of the majority of the studies upon which the results in Table 1 are based has been on the use of compacted bentonites as buffer barriers in containment of high level radioactive waste. The data in Table 1 (and subsequent tables) are limited in several ways. First, space limitations prevent detailed descriptions of the various bentonites and the methods used to measure the diffusion coefficients, both of which can be found in the references. Second, in some cases, relevant information such as the chemical species, type of bentonite, and/or porosity were not specified in the reference and, therefore, are unknown. Finally, in some cases, values of De and/or D* are missing from Table 1. These cases generally occur when only Da or both Da and De values have been reported, but the associated values of ε and Rd (or α) have not been reported, such that determination of De and/or D* is not possible. This relatively common practice of reporting only Da limits the usefulness of the data, since the diffusion and attenuation aspects cannot be evaluated independently (Shackelford & Daniel 1991). Also, values of Da are not appropriate for use in scenarios involving flux controlled boundary conditions, since diffusive mass flux depends on either D* or De ( Eqs. 1 or 6), not Da. Thus, the practice of reporting only Da values should be avoided. Given the aforementioned limitations, the general ranges of diffusion coefficients are 1.1x10-13 ≤ D* ≤ 2.8 x 10-10 m2/s, 4.3x10-14 ≤ De ≤ 1.8x10-10 m2/s, and 3.1x10-14 ≤ Da ≤ 4.8x10-12 m2/s. Thus, D*, De, and Da vary over 3.4, 3.6, and 2.2 orders of magnitude, respectively. Also, the general trend of D* > De > Da holds true, but the lower limit on Da is only marginally lower than that on De (i.e. 3.1x10-14 m2/s vs. 4.3x10-14 m2/s), whereas the upper limit on Da is 1.6 and 1.8 orders of magnitude lower than that on De and D*, respectively. Since the difference between Da and D* or De primarily reflects the effect of attenuation, the marginal differences between Da and D* or De values reported in Table 1 likely reflect, in part, the anionic or neutral forms of the chemical species, such that sorption via cation exchange to the bentonite was limited. Finally, the upper limit on D* of 2.8 x 10-10 m2/s is only slightly lower than the Do value of 3.9-4.9x10-10 m2/s for U(VI) reported by Bai et al. (2009), which is required since D* must be less than Do if τa < 1 in accordance with Eq. 2. In terms of a specific type of bentonite, the results from Muurinen (1990) for saturated specimens of compacted MX-80 bentonite are shown in Figure 1a. As indicated, both D* and De tend to increase semilog linearly with increasing ε (or decreasing dry density, ρd), albeit at different slopes, and D* > De for a given ε. Of course, in the limit as ε approaches unity, De must approach D*, and both De and D* must approach Do. The limited number of Da values precludes a similar conclusion with respect to the trend in Da with ε (or ρd) Similar relationships as shown in Figure 1a are expected for the other types of bentonites, but the lack of D* and De values from the relevant studies precludes such an evaluation. Values of τa calculated in accordance with Eq. 2 based on the D* and Do values shown in Figure 1a are plotted as a function of ε in Figure 1b. Parker et al. (1994) note that τa typically is empirically correlated with ε via a power relationship as follows: b a aτ = ε  (10) where a = 1 and b typically ranges from 1.3 to 5.4, depending on the porous medium. However, as shown in Figure 1b, reasonable estimates of τa for the results based on Muurinen (1990) are obtained Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 using Eq. 10 with values of a ranging from 0.150 to 0.188 and a value for b of 8.18. These values for a and b result in significantly lower values of τa than suggested by Parker et al. (1994), i.e., for the range of ε considered, reflecting the significantly increased apparent tortuosity afforded by highly compacted Table 1: Diffusion coefficients for uranium (U) chemical species in compacted bentonites(1) Diffusion Coefficients (m2/s) Chemical Species Type of Bentonite ε ρd (g/cm3) D* De Da Reference UO2(NO3)2 MX-80 0.41(2) 1.6 1.1x10-13 4.3x10-14 NS UO2(NO3)2 MX-80 0.59(2) 1.1 9.4x10-13 4.9x10-13 NS UO2(NO3)2 MX-80 0.67(2) 0.9 3.6x10-12 2.1x10-12 8.4x10-13 UO2(NO3)2 MX-80 0.78(2) 0.6 1.3x10-11 8.9x10-12 4.8x10-12 UO2Cl2 MX-80 0.56(2) 1.2 1.4x10-13 6.0x10-14 NS UO2Cl2 MX-80 0.78(2) 0.6 1.1x10-11 7.7x10-12 3.0x10-12 Dissolved from UO2 MX-80 0.67(2) 0.9 2.6x10-12 1.5x10-12 2.6x10-12 Dissolved from UO2 MX-80 0.78(2) 0.6 1.2x10-11 8.3x10-12 2.6x10-12 Muurinen (1990) NS NS NS 1.15 NS 5.7x10-11 7.5x10-14 NS NS NS 1.15 NS 5.8x10-12 6.2x10-13 NS NS NS 0.76 NS 1.9x10-12 1.5x10-13 NS NS NS 0.7 NS 1.9x10-12 6.3x10-13 NS Kunigel V1 NS 2 NS NS 4.5x10-13 NS Kunigel V1 NS 1.8 NS NS 4.2x10-13 NS Kunigel V1 NS 1.6 NS NS 8.6x10-13 NS Kunigel V1 NS 1.4 NS NS 1.3x10-12 NS Kunigel V1 NS 1.2 NS NS 1.2x10-12 NS Kunigel V1 NS 1 NS NS 2.2x10-12 NS Kunigel V1 NS 2 NS NS 3.4x10-14 NS Kunigel V1 NS 1.8 NS NS 4.8x10-14 NS Kunigel V1 NS 1.6 NS NS 7.3x10-14 NS Kunigel V1 NS 1.4 NS NS 2.1x10-13 NS Kunigel V1 NS 1.2 NS NS 2.5x10-13 NS Kunipia F NS 1.8 NS NS 3.1x10-14 NS Kunipia F NS 1.6 NS NS 4.4x10-14 NS Kunipia F NS 1.4 NS NS 8.7x10-14 NS Kunipia F NS 1.2 NS NS 3.3x10-13 NS Kunipia F NS 1 NS NS 6.9x10-13 Yu & Neretnieks (1997) Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011  Table 1 (continued): Diffusion coefficients for uranium (U) chemical species in compacted bentonites(1) Diffusion Coefficients (m2/s) Chemical Species Type of Bentonite ε ρd (g/cm3) D* De Da Reference UO2(OH)3- MX-80 0.62 0.997 1.4x10-10 8.5x10-11 1.9x10-12 UO2(OH)3- MX-80 0.62 0.996 1.8x10-10 1.1x10-10 1.9x10-12 UO2(OH)3- MX-80 0.63 0.978 9.5x10-11 6.0x10-11 1.4x10-12 UO2(OH)3- MX-80 0.63 0.958 1.2x10-10 7.8x10-11 1.2x10-12 UO2(OH)3- MX-80 0.64 0.953 2.6x10-10 1.6x10-10 1.9x10-12 UO2(OH)3- MX-80 0.64 0.951 2.8x10-10 1.8x10-10 2.3x10-12 UO2(OH)3- MX-80 0.64 0.946 1.3x10-10 8.1x10-11 1.5x10-12 Wang et al. (2005) (1)  NS = not specified; ε = porosity; ρd = dry density; (2) Based on ρs = 2.70 g/cm3 bentonites relative to other porous media. Such low τa values are likely due, in part, to the existence of semipermeable membrane behavior (ion exclusion) in bentonites, which can significantly reduce the magnitude of D* in bentonites depending on the porosity of the bentonite and the concentration of the chemical species (Malusis & Shackelford 2002). 10-14 10-13 10-12 10-11 10-10 10-9 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 D* D e D a 0.60.811.21.41.6 Di ffu si o n  Co ef fic ie nt  (m 2 /s ) Porosity, ε D* = 1.96x10 -16exp(14.1ε) or D* = 2.63x10 -10exp(-5.23ρ d ) D e  = 3.86x10 -17exp(15.7ε) or D e  = 2.62x10 -10exp(-5.83ρ d ) (a) Dry Density, ρ d  (g/cm3)  (r2 = 0.987)  (r2 = 0.989) D o  (U6+) = 3.9-4.9 x 10 -10 m2/s  0.0001 0.001 0.01 0.1 1 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 (b) Ap pa re n t T o rtu os ity  Fa ct or ,  τ a Porosity, ε τ a  = 0.188ε8.18 (r2 = 0.989) τ a  = 0.150ε8.18 (r2 = 0.989) τ a  = ε 1.3 τ a  = ε 5.4  Figure 1: Uranium diffusion through compacted MX-80 bentonite: (a) diffusion coefficient versus porosity; (b) apparent tortuosity factor versus porosity (data from Muurinen 1990) NS Kunipia F NS 0.8 NS NS 3.7x10-12 NS Kunipia F NS 1.8 NS NS 4.3x10-14 NS Kunipia F NS 1.6 NS NS 4.5x10-14 NS Kunipia F NS 1.4 NS NS 6.5x10-14 NS Kunipia F NS 1.2 NS NS 8.8x10-14 NS Kunipia F  NS 1 NS NS 1.2x10-13 NS Kunipia F NS 0.8 NS NS 2.2x10-12 Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 The results from Muurinen (1990) for MX-80 bentonite are compared with those reported by Wang et al. (2005) for MX-80 bentonite in Figure 2. The comparison reveals that the values of Da from Wang et al. (2005) are on the same order of magnitude as those from Muurinen (1990). However, the D* and De values from Wang et al. (2005) are significantly higher than those from Muurinen (1990), and the values D* and De values from Wang et al. (2005) tend to be only slightly lower than the upper limit given by Do. Aside from the differences in the techniques used to measure the diffusion coefficients (i.e. Muurinen (1990) used a through-diffusion technique whereas Wang et al. (2005) used a capillary method), the higher D* and De values from Wang et al. (2005) can be attributed, in part, to the anionic complex, UO2(OH)3-, evaluated by Wang et al. (2005) versus the neutral species of U(VI) evaluated by Muurinen (1990), since UO2(OH)3- presumably is repelled by the negatively charged surfaces of the bentonites particles, thereby facilitating diffusion through the compacted bentonite. Since this explanation excludes cation exchange as an attenuation mechanism, the apparent attenuation of UO2(OH)3- as reflected by the significantly lower Da values was attributed by Wang et al. (2005) to diffusion of UO2(OH)3- into the interlamellar space within the mineral structure of the smectite comprising the bentonite particles. This observation is consistent with the study by Jo et al. (2006) indicating attenuation in granular bentonites due to both interparticle and intraparticle diffusion. Thus, the comparison shown in Figure 2 illustrates the potential importance on uranium diffusion of both chemical speciation and interparticle/intraparticle diffusion as an attenuation mechanism. 10-14 10-13 10-12 10-11 10-10 10 -9 10 -8 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 Di ffu si o n  Co ef fic ie nt  (m 2 /s ) Porosity, ε Circle = D* Square = D e Diamond = D a UO 2 (OH) 3 - (a) D* = 1.96x10 -16exp(14.1ε) D e  = 3.86x10 -17exp(15.7ε) (r2=0.989) Closed = Muurinen (1990) Open = Wang et al. (2005) (r2=0.987)  10-14 10-13 10-12 10-11 10-10 10 -9 10 -8 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 (b) D iff us io n Co ef fic ie nt  (m 2 /s ) Dry Density, ρ d  (g/cm3) Circle = D* Square = D e Diamond = D a UO 2 (OH) 3 - Closed = Muurinen (1990) Open = Wang et al. (2005) D* = 2.63x10 -10exp(-5.23ρ d ) D e  = 2.62x10 -10exp(-5.83ρ d )  (r2 = 0.989)  (r2 = 0.987)  Figure 2: Effect of chemical speciation on diffusion of uranium through compacted MX-80 bentonite as a function of (a) porosity and (b) dry density. The diffusion coefficients for uranium diffusing in saturated specimens of porous media other than solely compacted bentonite are summarized in Table 2. The specimens for the study by Alonso et al. (2003) consisted of sheets of granite adjacent to either a solution containing uranium or FEBEX bentonite to evaluate the effect of the bentonite on uranium diffusion in the granite. They found that the presence of the bentonite reduced Da about two orders of magnitude, from ~10-13 m2/s to ~10-15 m2/s. The Na-boltwoodite evaluated by Liu et al. (2006) is a uranyl silicate precipitate that was used as a surrogate for precipitated U(VI) resulting from the discharge in 1951 of 350 m3 of radioactive waste solution containing 7000 kg of U(VI) at the Hanford site in Washington. The primary purpose of the study by Patra et al. (2011) was to evaluate the influence of distilled water versus 0.1 N NaNO3 as solvents on the leaching or uranium from a variety of metallurgical wastes and ore samples. The general characteristics of the other materials noted in Table 2 should be self evident. The data in Table 2 are noteworthy only in terms of the paucity of ε, D* and De values that have been Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 reported, which significantly limits any attempt to assess trends in the data.  Essentially, the only observation that can be drawn from the data in Table 2 is that the Da values fall within the range 1.0x10-15 ≤ Da ≤ 2.8x10-10 m2/s, which covers 5.4 orders of magnitude. However, the upper limit Da value of 2.8x10-10 m2/s for UO22+ is so close to the aforementioned Do value for U(VI) of 3.9-4.9x10-10 m2/s as to warrant suspicion. Thus, the Da values in Table 2 bentonite-rock and non-bentonite materials are more variable than those in Table 1 for compacted bentonites. Diffusion of Radionuclides of Uranium The diffusion coefficients for 233U in the form of several chemical species diffusing in saturated specimens of MX-80 bentonite, sand-bentonite mixtures, and biotitic granite are summarized in Table 3. Several observations from the data in Table 3 are readily apparent. First, the values of ε and ρd for     Table 2: Diffusion coefficients for U chemical species in bentonite-rock and non-bentonite materials(1) Diffusion Coefficients (m2/s) Reference Chemical Species Material ε ρd (g/cm3) D* De Da UO2(NO3)2 FEBEX bent. & granite NS NS NS NS 1.0x10 -15 Alonso et al. (2003) UO22+ Na-boltwoodite NS NS NS NS 2.6x10-12 UO22+ Na-boltwoodite NS NS NS NS 2.8x10-10 Liu et al. (2006) Ca2UO2(CO3)3 Silt/clay  0.45 1.48 1.6x10-10 7.2x10-11 NA Bai et al. (2009) NS Rock NS 1.37 NS NS 2.6x10-14 NS Rock NS 1.37 NS NS 6.0x10-14 NS U tailings NS 1.43 NS NS 6.6x10-15 NS U tailings NS 1.43 NS NS 4.2x10-15 NS Cu kinker ash NS 1.33 NS NS 2.4x10-15 NS Cu kinker ash NS 1.33 NS NS 6.6x10-15 NS Cu tailings NS 1.87 NS NS 2.6x10-14 NS Cu tailings NS 1.87 NS NS 1.1x10-13 Patra et al. (2011) (1)  NS = not specified; ε = porosity; ρd = dry density each of the three types of materials are the same, such that the effect of ε or ρd on the measured diffusion coefficients cannot be evaluated. Second, several values of D* and De for studies involving compacted MX-80 bentonite could not be ascertained due to the lack of information given in the publications. Third, the D* value of 1.3x10-9 m2/s extracted from the information given in the study by Torstenfelt et al. (1983) is greater than the Do value for U(VI) of 3.9-4.9x10-10 m2/s reported by Bai et Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 al. (2009) and the Do value for UO2(CO3)34- of 7.2x10-10 m2/s reported Yamaguchi & Nakayama (1998) and, therefore, is likely in error. Finally, the Da values reported by Yamaguchi & Nakayama (1998) for biotitic granite are greater than the De values (i.e. Da > De), which as previously noted is possible when ε < Rd-1. Otherwise, the diffusion coefficients shown in Table 3 provide an indication of the apparent variability that can exist in the measurement of the diffusion coefficients of 233U in specific porous media at the given values of ε (or ρd). For example, for compacted MX-80 bentonite (at ε = 0.35), excluding the aforementioned D* value of 1.3x10-9 m2/s and the associated De value of 4.5 x10-10 m2/s, D* and De vary by a factor of about 3 (1.0x10-10 ≤ D* ≤ 3.1 x 10-10 m2/s, 3.7x10-11 ≤ De ≤ 1.1x10-10 m2/s), whereas Da varies by a factor of about 48 (1.2x10-14 ≤ De ≤ 5.8x10-13 m2/s). In the case of biotic granite (at ε = 0.007) for the species UO2(CO3)34-, D* and De vary by a factor of only about 1.2 (1.8x10-1 ≤ D* ≤ 2.2 x 10-11 m2/s, 1.3x10-13 ≤ De ≤ 1.6x10-13 m2/s), whereas Da varies by a factor of only about 1.7 (3.5x10-13 ≤ Da ≤ 5.9x10-13 m2/s). Thus, the variability in the measured diffusion coefficients is greater in the MX- 80 bentonite than in the biotitic granite, and the variability in Da is greater than that in D* or De. The greater variability in Da relative to either D* or De can be attributed to the complicating effect of attenuation included in the values of Da, and likely reasons for the lesser variability in the diffusion coefficients for the biotitic granite relative to those for the MX-80 bentonite include a more homogeneous pore structure and the lesser attenuation capacity for the biotitic granite. Table 3: Diffusion coefficients for 233U in a variety of porous materials(1) Diffusion Coefficients (m2/s) Species Material ε ρd (g/cm3) D* De Da Reference NS MX-80 B 0.35 2.0 3.1x10-10 1.1x10-10 5.8x10-13 NS MX-80 B 0.35 2.0 1.0x10-10 3.5x10-11 1.9x10-13 NS MX-80 B 0.35 2.0 3.0x10-10 1.1x10-10 5.7x10-13 NS MX-80 B 0.35 2.0 1.1x10-10 3.7x10-11 2.0x10-13 NS MX-80 B 0.35 2.0 1.3x10-9 4.5x10-10 4.5x10-14 Torstenfelt et al. (1983) UO22+ MX-80 B 0.35 2.0 NS NS 3.4x10-12 UO22+ MX-80 B 0.35 2.0 NS NS 6.4x10-13 UO22+ MX-80 B 0.35 2.0 NS NS 2.7x10-13 UO22+ MX-80 B 0.35 2.0 NS NS 3.3x10-11 UO22+ MX-80 B 0.35 2.0 NS NS 1.6x10-13 UO22+ MX-80 B 0.35 2.0 NS NS 8.2x10-12 UO22+ MX-80 B 0.35 2.0 NS NS 3.1x10-13 UO22+ MX-80 B 0.35 2.0 NS NS 1.0x10-13 UO22+ MX-80 B 0.35 2.0 NS NS 8.5x10-12 UO22+ MX-80 B 0.35 2.0 NS NS 2.3x10-13 UO22+ MX-80 B 0.35 2.0 NS NS 8.3x10-14 UO22+ MX-80 B 0.35 2.0 NS NS 6.7x10-12 Torstenfelt (1986) Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 UO22+ MX-80 B 0.35 2.0 NS NS 6.8x10-13 UO22+ MX-80 B 0.35 2.0 NS NS 2.3x10-13 UO22+ 90% S-10% B 0.13 2.0 1.7x10-10 2.3x10-11 1.5x10-12 UO22+ 90% S-10% B 0.13 2.0 1.7x10-10 2.3x10-11 1.5x10-12 Albinsson & Engkvist (1989) U(IV) MX-80 B 0.35(2) 2.0 NS NS 1.2x10-14 U(IV) MX-80 B 0.35(2) 2.0 NS NS 2.8x10-14 U(IV) MX-80 B 0.35(2) 2.0 NS NS 1.4x10-13 Albinsson et al. (1991) Biotitic granite 0.007 2.6 2.2x10-11 1.6x10-13 5.9x10-13 Biotitic granite 0.007 2.6 2.0x10-11 1.4x10-13 4.4x10-13 UO2(CO3)34- Biotitic granite 0.007 2.6 1.8x10-11 1.3x10-13 3.5x10-13 UO22+ Biotitic granite 0.007 2.6 5.1x10-12 3.6x10-14 3.3x10-12 Yamaguchi & Nakayama (1998) (1)  NS = not specified; B = bentonite; S = sand; ε = porosity; ρd = dry density; (2) Assumed value based on values given by Torstenfelt et al. (1983) and Torstenfelt (1986) for the same bentonite. Diffusion coefficients for uranium radionuclides other than 233U also have been reported, but both the number of studies and the results contained in the studies are limited. For example, Sims et al. (2008) report diffusion coefficients ranging from 3 x10-14 m2/s to 1.2 x10-8 m2/s for 238U diffusion in an unsaturated prairie soil, but the definition of the diffusion coefficient is not specified. Also, several of the values appear to be exceedingly high so as to approach or exceed a representative Do value, which is physically impossible. Sims et al. (2008) also report a single diffusion coefficient for 235U of 7x10-10 m2/s for the same soil and conditions and with the same limitations. Finally, Rameback et al. (1998) report Da values for 236U diffusing in saturated specimens of compacted MX-80 bentonite (ρd = 2.00 g/cm3) ranging from 3.2x10-15 m2/s to 2.2x10-13 m2/s. Conclusions A review of the literature on the diffusion of uranium in soils and rocks reveals several conclusions. First, the common practice of reporting only Da values as opposed to both Da and D* and/or De values severely restricts the usefulness of the data and should be avoided in the future. Second, diffusion of uranium in compacted MX-80 bentonite is shown to increase semilog linearly with increasing porosity (or decreasing dry density), and to be potentially significantly affected by chemical speciation. Third, the potential existence of semipermeable membrane behavior in bentonites is a likely reason for the significantly slower rate of uranium diffusion in compacted MX-80 bentonite than would be predicted on the basis of other soils. Fourth, interparticle/intraparticle diffusion can significantly restrict the diffusion of uranium in compacted MX-80 bentonite. Finally, the effect of attenuation is shown to increase the variability in Da relative to that in either D* or De, and the variability in D*, De, and Da is shown to be greater in compacted bentonite than in unfractured biotitic granite, likely due to a more homogeneous pore structure and the lesser attenuation capacity for the biotitic granite. Acknowledgment This study was supported by the U. S. Department of Energy, under Cooperative Agreement Number DE-FC01-06EW07053 entitled, "The Consortium for Risk Evaluation with Stakeholder Participation III" awarded to Vanderbilt University.  The opinions, findings, conclusions, or recommendations expressed herein are those of the authors and do not necessarily represent the views of the Department Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 of Energy or Vanderbilt University. References Abdelouas, A., Lutze, W. and Nuttal, E., 1998. Chemical reactions in ground water at a mill tailings site. Journal of Contaminant Hydrology, 34 (4), 343-361. Albinsson, Y. and Engkvist, I., 1989. Diffusion of Am, Pu, U, Np, Cs, I and Tc in compacted sand-bentonite mixture. Swedish Nuclear Fuel and Waste Management Co., SKB TR 89-22. Albinsson, Y., Christiansen-Satmark, B., Engkvist, I., and Johansson, W., 1991. Transport of actinides and Tc through a bentonite backfill containing small quantities of iron or copper. Radiochimica Acta, 52/53 (1), 283- 286. Alonso, U., Missana, T., Patelli, A., Ravagnan, J., and Rigato, V., 2003. RBS and µPIXE analysis of uranium diffusion from bentonite to the rock matrix in a deep geological waste repository. Nuclear Instruments and Methods in Physics Research B, 207 (2), 195-204. Ames, L.L., McGarrah, J.E., and Walker, B.A., 1983. Sorption of trace constituents from aqueous solutions onto secondary minerals. I. Uranium. Clays and Clay Minerals, 31 (5), 321-334. Bai, J., Chongxuan, L., and Ball, W.P.,  2009. Study of sorption-retarded U(VI) diffusion in Hanford silt/clay material. Environmental Science & Technology, 43 (20), 7706-7711. Bruno, J., de Pablo, J., Duro, L, and Figuerola, E., 1995. Experimental study and modeling of the U(VI)- Fe(OH)3 surface precipitation/coprecipitation equilibria. Geochimica et Cosmochimica Acta, 59 (20), 4113-4123. Duff, M.C. and Amrhein, C., 1996. Uranium(VI) adsorption on goethite and soil in carbonate solution. Soil Science Society of America, Journal, 60 (6), 1393-1400. Flury, M. and Harsh, J.B., 2000. Remediation of Uranium Contaminated Mine Waste. State of Washington Water Research Center Report WRR-04, U.S. Geological Survey Grant No. 1434-HQ96GR02704, Department of Crop and Soil Science, Washington State University, Pullman, WA. Gavrilescu, M., Pavel, L.V., and Cretescu, I., 2009. Characterization and remediation of soils contaminated with uranium. Journal of Hazardous Materials, 163 (2-3), 475-510. Ho, C.H. and Doern, D.C., 1985. The sorption of uranyl species on a hematite on a hematite sol. Canadian Journal of Chemistry, 63 (5), 1100-1104. Hsi, C.-K. D. and Langmuir, D., 1985. Adsorption of uranyl onto ferric oxyhydroxides: application of surface complexation site-binding model. Geochimica et Cosmochimica Acta, 49 (9), 1931-1941. Jo, H.Y., Benson, C.H., and Edil, T.B., 2006. Rate-limited cation exchange in thin bentonitic barrier layers. Canadian Geotechnical Journal, 43 (4), 370-391. Kohler, M., Curtis, G.P., Kent, D.B., and Davis, J.A., 1996. Experimental investigation and modeling of uranium(VI) transport under variable chemical conditions. Water Resources Research, 32 (12), 3539-3551. Langmuir, D., 1978. Uranium solution-mineral equilibria at low temperatures with application to sedimentary ore deposits. Geochimica et Cosmochimica Acta, 42 (6, Part 1), 547-569. Liu, C., Zachara, J.M., Yantasee, W., Majors, P.D., and McKinley, J.P., 2006. Microscopic reactive diffusion of uranium in the contaminated sediments at Hanford, United States. Water Resources Research, 42, W12420, 15 p. Malusis, M.A. and Shackelford, C.D., 2002. Coupling effects during steady-state solute diffusion through a semipermeable clay membrane. Environmental Science and Technology, 36 (6), 1312-1319. Muurinen, A., 1990. Diffusion of uranium in compacted sodium bentonite. Engineering Geology, 28 (3-4), 359- 367. Proceedings Tailings and Mine Waste 2011 Vancouver, BC, November 6 to 9, 2011 Parker, B.L., Gillham, R.W., and Cherry, J.A., 1994. Diffusive disappearance of immiscible-phase organic liquids in fractured geologic media. Ground Water, 32 (5), 805-820. Patra, A.C., Sumes, C.G., Mohapatra, S., Sahoo, S.K., Tripathi, R.M., and Puranik, V.D., 2011. Long-term leaching of uranium from different waste matrices. Journal of Environmental Management, 92 (3), 919-925. Pearson, F.J., 1999. What is the porosity of a mudrock? In: Aplin, A.C., Fleet, A.J., and Macquaker, J.H.S., eds. Muds and Mudstones: Physical and Fluid Flow Properties. Geological Society, London, Special Publications, 158, 9-21. Rameback, H., Skalberg, M., Eklund, U.B., Kjellberg, L., and Werme, L., 1998. Mobility of U, Np, Pu, Am and Cm from spent nuclear fuel into bentonite clay. Radiochimica Acta, 82 (1), 167-171. Shackelford, C.D., 1988. Diffusion as a transport process in fine-grained barrier materials. Geotechnical News, Bi-Tech Publishers, Vancouver, B.C., 6 (2), 24-27. Shackelford, C.D. and Daniel, D.E., 1991. Diffusion in saturated soil. I: Background. Journal of Geotechnical Engineering, 117 (3), 467-484. Shackelford, C.D. and Lee, J.-M., 2005. Analyzing diffusion by analogy with consolidation. Journal of Geotechnical and Geoenvironmental Engineering, 131 (11), 1345-1359. Sims, D.J., Andrews, W.S., and Creber, K.A.M., 2008. Diffusion coefficients for uranium, cesium and strontium in unsaturated prairie soil. Journal of Radioanalytical and Nuclear Chemistry, 277 (1), 143-147. Skagius, K. and Neretnieks, I., 1986. Porosities and diffusivities of some nonsorbing species in crystalline rocks. Water Resources Research, 22 (3), 389-398. Torstenfelt, B., 1986. Migration of the actinides thorium, protactinium, uranium, neptunium, plutonium and americium in clay. Radiochimica Acta, 39, 105-112. Torstenfelt, B., Allard, B., Andersson, K., Kipatsi, H., Eliasson, L., Olofsson, U., and Persson, H., 1983. Radionuclide diffusion and mobilities in compacted bentonite. SKB Technical Report, 83-34, 31 p. Wang, X.K., Chen, C.L., Zhou, X., Tan, X.L., and Hu, W.P., 2005. Diffusion and sorption of U(VI) in compacted bentonite studied by a capillary method. Radiochimica Acta, 93 (5), 273-278. Yamaguchi, T. and Nakayama, S., 1998. Diffusivity of U, Pu, and Am carbonate complexes in a granite from Inada, Ibaraki, Japan studied by through diffusion. Journal of Contaminant Hydrology, 35 (1-3), 55-65. Yu, J.-W. and Neretnieks, I., 1997. Diffusion and sorption properties of radionuclides in compacted bentonite. SKB Technical Report, 97-12, 105 p. Zielinski, R.A., Otton, J.K., Wanty, R.B., and Pierson, C.T., 1987. The geochemistry of water near a surficial organic-rich uranium deposit, northeastern Washington State, U.S.A., Chemical Geology, 62 (3-4), 263-289.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 5 4
Japan 2 0
France 2 0
Mexico 1 0
City Views Downloads
Unknown 4 0
Sunnyvale 2 0
Tokyo 2 0
Columbus 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items