Uranium Release from Acidic Weathered Hanford Sediments: Single-Pass Flow-Through and Column ExperimentsGuohui Wang,† Wooyong Um,*,†,‡ Zheming Wang,† Estela Reinoso-Maset,§ Nancy M. Washton,†

Karl T. Mueller,† Nicolas Perdrial,∥,⊥ Peggy A. O’Day,§ and Jon Chorover∥

†Pacific Northwest National Laboratory, Richland, Washington 99354, United States‡Pohang University of Science and Technology (POSTECH), Pohang, South Korea§Sierra Nevada Research Institute and School of Natural Sciences, University of California Merced, Merced, California 95343, UnitedStates∥Department of Soil, Water and Environmental Science, University of Arizona, Tucson, Arizona 85721, United States⊥Department of Geology, University of Vermont, Burlington, Vermont 05405, United States

*S Supporting Information

ABSTRACT: The reaction of acidic radioactive waste with sedimentscan induce mineral transformation reactions that, in turn, controlcontaminant fate. Here, sediment weathering by synthetic uranium-containing acid solutions was investigated using bench-scale experimentsto simulate waste disposal conditions at Hanford’s cribs (Hanford, WA).During acid weathering, the presence of phosphate exerted a stronginfluence over uranium mineralogy and a rapidly precipitated, crystallineuranium phosphate phase (meta-ankoleite [K(UO2)(PO4)·3H2O]) wasidentified using spectroscopic and diffraction-based techniques. Inphosphate-free system, uranium oxyhydroxide minerals such as K-compreignacite [K2(UO2)6O4(OH)6·7H2O] were formed. Single-passflow-through (SPFT) and column leaching experiments using synthetic Hanford pore water showed that uranium precipitated asmeta-ankoleite during acid weathering was strongly retained in the sediments, with an average release rate of 2.67 × 10−12 molg−1 s−1. In the absence of phosphate, uranium release was controlled by dissolution of uranium oxyhydroxide (compreignacite-type) mineral with a release rate of 1.05−2.42 × 10−10 mol g−1 s−1. The uranium mineralogy and release rates determined forboth systems in this study support the development of accurate U-release models for the prediction of contaminant transport.These results suggest that phosphate minerals may be a good candidate for uranium remediation approaches at contaminatedsites.

■ INTRODUCTION

Upon intentional or accidental introduction to the subsurface,radioactive waste interacts heterogeneously with host sedimentsby way of adsorption−desorption or precipitation−dissolutionreactions.1,2 Waste solutions, which are often highly acidic oralkaline, may be far from equilibrium with receiving sediments,thereby leading to mineral transformation reactions that canaffect the transport and fate of waste-derived contaminants.3,4

Considering the different geochemical properties betweenminerals with regard to their reactions with waste contami-nants, the resolution of molecular-scale interactions betweenwaste-derived contaminants and newly developed geomediaconstituents is needed to improve predictive modeling ofreactive contaminant transport.One pertinent example of this need pertains to uranium (U)

contamination of soil and groundwater at the Department ofEnergy’s Hanford site, (Hanford, WA), which is the largest andmost complex environmental cleanup site in the United Statesand a major concern for public health. At Hanford, hundreds ofmillions of liters of uranium-containing acidic aqueous waste,

generated during the biphosphate process for plutoniumprocessing during World War II and the Cold War, weredisposed of in a number of unlined underground, box-likeexcavation structures called cribs. Approximately 529 × 106 L ofacidic waste was discharged into two cribs (denoted 216-U-8and 216-U-12) at 200 Area of Hanford site from 1952 to 1988.The waste contained about 34 Mg U, 3400 Mg nitrate, andother fission products, e.g., Tc-99, H-3, Sr-90, and Cs-137.5

Furthermore, these waste streams were known to containvariable concentrations of phosphate, which could havesignificant implications for U fate. The large volume ofwastewater discharged to these two unlined cribs was sufficientto cause the contaminant-containing acidic aqueous waste topercolate into the sandy subsurface sediments and reach theunderlying groundwater.5 Thus, the vadose zone beneath the

Received: July 8, 2017Revised: September 7, 2017Accepted: September 8, 2017Published: September 8, 2017

Article

pubs.acs.org/est

© 2017 American Chemical Society 11011 DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

Dow

nloa

ded

via

UN

IV O

F A

RIZ

ON

A o

n N

ovem

ber

4, 2

019

at 1

8:28

:55

(UT

C).

See

http

s://p

ubs.

acs.

org/

shar

ingg

uide

lines

for

opt

ions

on

how

to le

gitim

atel

y sh

are

publ

ishe

d ar

ticle

s.

cribs is known to contain subsurface contamination byradioactive and nonradioactive contaminants from U process-ing.5

Uranium is hexavalent [U(VI)] and exists as the uranyl ion(UO2

2+) under oxic conditions. Interactions between U(VI)and mineral phases present in Hanford sediment, includingsorption and desorption, complexation, and reduction reac-tions, have been well-documented in the literature.1,2,6−12 Inaddition to these relatively direct U(VI)−mineral interactions,weathering of the underlying sediment minerals by the acidic oralkaline waste streams also affects the fate and transport ofradionuclides in the subsurface. Previous studies have shownthat leaking, hyperalkaline, caustic Hanford waste reacts withHanford sediments, resulting in primary mineral dissolutionand secondary mineral precipitation and significantly alteringsediment properties and radionuclide transport.3,4,13,14 How-ever, less is known about the geochemical reactions betweenacidic waste streams and the underlying sediments, and thesubsequent contaminant release from waste-weathered sedi-ment by natural pore water. Szecsody et al.15 investigated themineral dissolution of several Hanford sediments at acidicconditions and their impacts on U transport. They found thatmineral dissolutions led in rapid increases in aqueous carbonateand phosphate, resulting in significant shallow U mineraldissolution and deeper U precipitations (likely phosphates andcarbonates) with downward U migration. Gateman et al.16

examined the effects of low pH on uranium partitioning toHanford sediments. They reported that much greater Upartitioning was observed at conditions of pH 5 than at pH 2and pH 8, which was likely due to the precipitation of uranylphosphate in addition to adsorption mechanism. These studiesmainly simulated the U fate and transport behavior during theacidic waste reservoir operating time. But unfortunately, bothstudies did not use waste simulants (acidified deionized orgroundwater was used as the waste matrix), which could resultin geochemical reaction bias from the real legacy sedimentweathering history.Besides the above laboratory studies, which addressed the

vadose zone as a whole at Hanford site, extensive site-specific Uspeciation was investigated on the sediments below the inactiveNorth Process Pond at Hanford site 300 Area, which served asan acidic waste reservoir and neutralized with NaOH.Multiscale electron microscopic and spectroscopic analyses onthose samples focusing on U speciation and solid-phase hosthave been conducted by several researchers.17−21 They foundthat metatorbernite [Cu(UO2)2(PO4)2·8H2O] precipitated atthe intermediate depths in the vadose zone, likely during pondoperations when the U, Cu, and P present in this phase alloriginated from waste release, possibly at different times.However, the U(VI)−phosphates formation process discussedin these studies are speculative due to the lack of historical databecause the vadose zone pore water was not collected oranalyzed during the lifetime of the pond operation. In contrastto the 300 Area, less study on acidic waste contamination at the200 Area has been conducted. Recently, Kanematsu et al.22

synthesized acidic solutions according to waste disposaldocuments on cribs 216-U-8 and 216-U-12 and studieduranium speciation in the presence and the absence ofdissolved silica and phosphate under variable pH conditionsto assess thermodynamic and kinetic controls on theprecipitation of uranyl solid phases. Among other observations,they reported strong effect of phosphate on uranium speciationin the tested systems; in the presence of phosphate (3 mM),

meta-ankoleite initially precipitated as the primary phase at pH3, 5, or 7 regardless of the presence of 1 mM dissolved silica.Overall, investigations on uranium speciation and transport

in the Hanford vadose zone subjected to acidic weathering arenot yet well-documented. Specifically, the subsequent U releasefrom those legacy weathered sediments is rarely reported.Understanding how the acidic aqueous waste-sediment reactionaffects U speciation and coordination and how that, in turn,influences the subsequent U release is key to predicting theenvironmental impact of these legacy sediments. Therefore, theobjectives of this study were to (i) determine the U phasespresent in acid-waste weathered Hanford sediments below thewaste disposal cribs; (ii) understand how such sediments reactwith Hanford pore water; and (iii) estimate overall release ratesof the U-containing phases to predict U release and subsurfacetransport. Legacy sediments from 216-U-8 and 216-U-12 cribsites were simulated in bench-scale experiments in whichuncontaminated Hanford sediments were reacted withsynthetic acidic waste solutions containing U in the presenceand the absence of dissolved phosphate (PO4

3−). The Uspeciation and the nature of U-containing mineral phasespresent in the weathered sediments were characterized using acombination of spectroscopic, microscopic, and diffraction-based techniques. Uranium release from the weatheredsediments was investigated by leaching tests. Because of thestrong effect of phosphate on uranium immobilization,23−27 theweathering of sediment minerals by phosphate-containingacidic uranium waste would result in U sequestration indistinct solid-phase products with low aqueous solubility andslow release rates, thus limiting subsurface U transport.

■ MATERIALS AND METHODSAcidic Crib Waste Solution Simulants. Based on the type

and amount of waste discharged to the two cribs (216-U-8 and216-U-12), several acidic synthetic crib wastes (ASCW) havebeen proposed.22 Solution chemistry indicates that the presenceor absence of phosphate in the waste solution is a principalconstraint on reaction pathways.22 In this study, two acidicwaste simulants (pH 3) were adopted: ASCW3 containing noPO4

3− and ASCW6 containing PO43−. The two waste simulants

were prepared using NO3− and were identical in chemical

composition (U: 1 mM, NO3: ∼100 mM, Na: 0.75 mM, Cl:2.08 mM, and K: 100 mM) except PO4

3− content (0 and 3.00mM of PO4

3− for ASCW3 and ASCW6, respectively). InASCW6, rapid precipitation of a solid phase was observed aftersolution mixing; thus, a parallel ASCW6 solution was prepared,and the precipitate was filtered for solid characterization.

Hanford Fine Sands. Hanford Fine Sand (HS), collectedat near surface burial ground excavation at 218-E-12B locationand composed of the silty sand facies from the Hanfordformation in the Hanford 200 Area, was used to represent thesediments at the 216-U-8 and 216-U-12 crib sites. The HS,which contains 0.23 wt % gravel, 72.61 wt % sand, and 27.16 wt% silt and clay,28 was sieved to remove the size fraction greaterthan 2 mm prior to experiments. The HS is typical of Hanfordformation sediments and contains quartz, K/Na-feldspar andsmectite, along with illite and chlorite, as the dominant mineralphases with a low calcium carbonate content (1.75 wt %), asidentified by mineralogical analysis in Serne et al.28

Sediment Weathering with U-Containing Acidic-Waste Simulants. The HS was weathered in polyethylenebottles with a solid to solution ratio of 0.02 g/mL (6 L ofASCW3 or ASCW6 solution plus 120 g of dry HS) at room

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11012

temperature (20 °C). The bottles were shaken once a week andpH values of the supernatant solutions were monitored alongthe weathering process. After 3 or 15 months of reaction, thesupernatant solution was carefully removed, and the reactedsediment slurries were collected. Slurries were air-dried forfurther characterization and leaching experiments. A total offour weathered sediment samples were prepared and labeled asASCW3-3, ASCW3-15, ASCW6-3, and ASCW6-15, respec-tively, corresponding to the two solutions and the twoweathering times (see Table 1). The initial solution pH was

∼3.0 for ASCW3 and ASCW6 and was buffered to neutral(∼6.5−7.4) after being reacted with HS containing carbonateminerals. The evolution of the pH in both ASCW solutionsduring HS weathering as well as a sediment-free control isprovided in Figure S1.Characterization of Solid Samples. Detailed solids

characterization, including mineral identification and Uspeciation, was conducted on the weathered sediments, alongwith the precipitate formed homogeneously from the ASCW6solution, using X-ray diffraction (XRD), scanning electronmicroscopy (SEM) with a focused ion beam (FIB), trans-mission electron microscopy (TEM), energy-dispersive spec-troscopy (EDS) mapping, extended X-ray absorption finestructure (EXAFS) analysis, and time-resolved laser-inducedfluorescence (TRLIF) spectroscopy. In addition, U concen-trations in the weathered sediments were determined byinductively coupled plasma mass spectrometry (ICP-MS)analysis of HNO3, HF, and HCl microwave-assisted digests.The specific surface area of the weathered sediments wasmeasured using the Brunauer−Emmett−Teller (BET) method.Detailed characterization methods for XRD, SEM− and TEM−EDS, EXAFS, TRLIF, BET, and microwave digestion areprovided in the Supporting Information.Single-Pass Flow-Through and Column-Leaching

Experiments. Uranium release from the acid-waste weatheredsediments was investigated using the synthesized Hanford sitebackground pore water (BPW) with 0.3 mM HCO3

− accordingto Um et al.8 A single-pass flow-through (SPFT) approach29

was used to investigate the dissolution of the uranyl minerals inthe four weathered sediments, in which the BPW was contactedwith the acidic weathered sediments, resulting in the dissolutionof uranyl minerals and U release. In addition, two column-leaching tests using two weathered HS samples (ASCW3-3 andASCW6-3) were conducted to provide a more-realisticsimulation of the crib site leaching scenario. The column

experiments used high solid-to-solution ratios (3.6−4.0 g/mL;porosity of ∼0.38−0.4) and pore water flow rates (1 m/day;∼0.1 day solution resident time in the columns). The pH of theeffluent was monitored and the uranium concentrations weremeasured by ICP-MS. The details of the SPFT and columnexperimental setup are provided in the Supporting Information.The normalized apparent steady-state release rate (r) of U-

bearing minerals in weathered sediments under saturatedconditions was determined (in unit of mol g−1 s−1) as:

=

⎡⎣⎢

⎤⎦⎥

rt

mM

d

R

(1)

where md is the dissolved mass of U (mol) within samplingintervals, calculated through the multiplication of the steady-state element concentration in the effluent solutions by thecollected sample volumes; MR is the remaining mass (g) of U-minerals (meta-ankoleite or K-compreignacite) in the weath-ered sediments after each sampling event; and t is the length ofeach leaching interval in seconds. MR was calculated iterativelybased on mass balance by subtracting the released U massduring the n + 1 sampling step from the remained U mass at nstep (n = sampling times), and converted into the mass ofmeta-ankoleite or K-compreignacite based on the stoichiometryof U in each of the U-bearing minerals. It should be noted thatthe calculation of MR was simplified, such that only meta-ankoleite and K-compreignacite were assumed to be thedominant U-minerals, respectively, in ASCW6 and ASCW3solution-reacted sediment samples, without considering otherpotential minor U-minerals in the weathered sediments.

■ RESULTS AND DISCUSSION

Solid-Phase Characterization. Surface Area and UConcentration in Weathered Samples. The BET-determinedspecific surface area for air-dried weathered-HS varied between6.04 and 9.19 m2/g (Table 1). A BET surface area value of 13.7m2/g on the untreated HS was reported by Um et al.30 Lowermeasured specific surface area in this study on weathered HSsamples (ASCW solution pH 3) could be due to the dissolutionof silty and clay particles present in HS. The U concentrationsassociated with the weathered HS, as determined by microwavedigestion (Table 1), show a similar U loading (7.09−7.24 mg/g, corresponding to 0.71−0.72 wt %) among all the weatheredsamples, which minimizes the effect of differences in initial Uconcentration on U leaching among different samples.

X-ray Diffraction. The rapidly precipitated phase observedin ASCW6 was identified as meta-ankoleite by powder XRD(Figure 1a). The measured U concentration of 431 mg/g of theprecipitate from the ASCW6 solution (Table 1) is not far froma U concentration of 520 mg/g calculated for the theoreticalmeta-ankoleite composition, which further supports the XRDresult. The sharp peaks in the XRD (Figure 1a) indicate theformation of well-crystallized meta-ankoleite. This interpreta-tion is supported by SEM images in which clear and sharpmineral shapes were observed (Figure S2). Unreacted HS andHS samples weathered for 3 months in either ASCW6 orASCW3 solutions were also analyzed by XRD, but nosignificant difference was observed between the weatheredand unreacted samples (data not shown). Uranium was presentin the weathered samples, as shown by microwave digest results(Table 1), but the amount (0.71−0.72 wt %) was too low todetect a specific U-containing mineral phase by XRD.

Table 1. Summary of the Acidic Waste Solution, WeatheredHanford Fine Sands, U Loading, and BET-Specific SurfaceArea Measured on Sediments after Weathering

sample ID PO43−

ASCW HS weatheringtime (months)

U concentrationin weathered

samples (mg/g)

specificsurfacearea

(m2/g)

ASCW3-3 − 3 7.24 ± 0.01 6.04ASCW3-15

− 15 9.19

ASCW6-3 + 3 7.09 ± 0.04 8.37ASCW6-15

+ 15 8.58

meta-ankoleite

+ rapidly homogeneousprecipitate from

solution ASCW6 (withPO4

3−)

431 ± 10

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11013

FIB−SEM/TEM−EDS. The SEM−EDS analysis showed thatthe freshly precipitated meta-ankoleite from the ASCW6solution was present as thin platy micrometer-sized particles(Figure S2) with a P-to-U atomic ratio of ∼1. This ratio isfurther verified by wet chemistry analysis on ASCW6 solution(Table S1), where 0.99 mM U (99% of the U added) and 1.06mM P were removed from the ASCW6 solution through meta-ankoleite precipitation. In the ASCW6-HS weathering system,meta-ankoleite appeared to precipitate on sediment grainsshown in TEM images (Figure 1b), where the EDS mappinganalysis illustrates that U is clearly associated with P and K(indicating meta-ankoleite) but less with Si and Al, whichrepresent sediments in this study. Of particular interest,compared to the EDS spectra of the homogeneouslyprecipitated meta-ankoleite from the ASCW6 solution, inwhich no Ca signal was observed (Figure S2), and, considering,the later introduction of Ca into the weathering solution fromcalcite dissolution, the clear association of Ca with U, P, and Kin this system may imply a possible extra Ca-bearing layer“armoring” on the surface of the precipitated meta-ankoleite. Across-section of the sample was cut using the FIB, and theFIB−TEM−EDS analysis showed that an amorphous phase,consisting of Ca, P, and minor U as well as Si, Al, and Na, etc.,precipitated on the sediments (Figure S3). In this amorphouslayer, the possibility of Ca−P precipitates hosting minor Umight not be ruled out due to the high amount of P (3.0 mM/L) and Ca (up to 2.7 mM/L; see Figure S4) in this weatheringsystem. The significant differences of Ca aqueous concen-trations (ca. 0.5 mM/L) in the ASCW3-HS (without PO4

3−)and ACSW6-HS (with PO4

3−) during the sediments weathering(Figure S4) supported this assumption, in which the significant

Ca mass loss from the ASCW6-HS could be explained bycalcium phosphate precipitation. This finding is consistent withMehta et al.,31 who found that, when simultaneously mixing ofdissolved uranium, calcium, and phosphate at neutral pH, Uwas structurally incorporated into a newly formed amorphouscalcium phosphate solid. However, it should be pointed outthat the Ca−P phase is speculative, and further studies onreliable chemical compositions are needed.

TRLIF Spectroscopy. TRLIF analysis was conducted on theprecipitated meta-ankoleite and the fluorescence spectrum wascompared with that of samples ASCW3-3 and ACSW6-3. Theemission fluorescence spectra for sample ASCW6-3 (Figure 2b)are the same as that for meta-ankoleite (Figure 2a) with fivecharacteristic peaks at 503, 525, 549, 575, and 602 nm. Thespectral similarity between the precipitated meta-ankoleite andASCW6-3 indicates that the U solid phase is dominated bymeta-ankoleite mixed with ASCW6-weathered sediments, inagreement with SEM− and TEM−EDS analysis. The emissionfluorescence spectrum for ASCW3-3 does not show the fivecharacteristic peaks for meta-ankoleite and instead shows threepeaks shifted to longer wavelengths (Figure 2c), suggesting thepresence of a different uranium phase. This can be explained bythe difference in solution composition as ASCW3 did notcontain added PO4

3−; thus, it is not expected that uranylphosphate phases, such as meta-ankoleite, would precipitate. Acomparison of the fluorescence spectrum for ASCW3−3 withthose for well-characterized standards32,33 shows that thespectrum is most similar to that of a becquerelite referencespectrum [Ca(UO2)6O4(OH)6·8(H2O)], which is structurallysimilar to K- and Na-compreignacite (Figure 2c).34,35 Theformation of compreignacite [K2(UO2)6O4(OH)6·7H2O] andbecquerelite is thermodynamically favorable at near-neutral pHand in the presence of sufficient K and Ca, respectively.36 Dueto the high concentration of K (100 mM) present in the ASCWsolutions, the formation of K-compreignacite was expected andwas observed by Kanematsu et al.22 to precipitate from similarsolutions at pH 5. Thus, additional EXAFS analyses wereundertaken to characterize the dominant uranium phases in thesample ASCW3-3.

EXAFS Analysis. EXAFS spectra were collected for thesample ASCW3−3 (HS weathered in ASCW3 for 3 months)(Figure 3). The best linear combination fit of the spectrum ofASCW3-3 (for uranium mineral EXAFS standards, see TableS2 and Figure S5) indicates that K-compreignacite is thedominant uranium phase (80%), but the inclusion of thespectrum of a natural compreignacite in the fit improves theresult and suggests substitution of other cations in the mineralstructure (Table S3). Given both the TRLIF and EXAFSspectra for ASCW3-3, U is present in a compreignacite-typephase that is mostly K but could contain some Ca (as indicatedby the TRLIF spectrum), which smears out the EXAFSfeatures.22,37 Substitution of different cations into theinterlayers between U(VI)−OH polyhedral sheets in com-preignacite-group minerals is common.37 Interestingly, despitethe different U-bearing mineral phases formed in solutions withor without PO4

3−, the amounts of U removed from the initialwaste solutions were similar (Table 1).

SPFT and Column-Leaching Experiments. SPFT Leach-ing Experiments. SPFT were performed with the fourweathered HS samples (ASCW3-3, ASCW3-15, ASCW6-3,and ASCW6-15), in which a constant effluent solution pH of∼7.4 (Figure S6) was observed in all the samples. HS samplesweathered with the same solution, but different weathering

Figure 1. (a) X-ray diffractogram of the homogeneous precipitatefiltered out of acidic waste solution ASCW6 compared with lines forreference meta-ankoleite; (b) TEM−EDS mapping of sampleASCW6-3 showing association of U with P, K, and Ca but less withSi and Al.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11014

times showed similar leaching behaviors, but ASCW3- andASCW6-weathered HS exhibited different leaching trends(Figure 4a). For the HS weathered in ASCW3 solution(without PO4

3−) for 3 or 15 months, a constant Uconcentration of ca. 47 μM/L (11420 μg/L) was measuredin the effluents after the first 20 mL (one cell solution volume).Thus, despite different weathering times, the steady-state rate ofU release was similar for both flow-through cells. The nearlyconstant U release from these two ASCW3-weatheredsediments are consistent with a leaching process that iscontrolled by steady-state dissolution of U-bearing solids(dominated by compreignacite in these samples). This behavioris contrast to previously reported U release from contaminatedHanford sediments, which was characterized by a rapid initialdecrease in U solution concentration followed by a gradualdecrease for the remainder of the experiment.24 The two-stageleaching behavior was consistent with leaching kinetics fromdiffusion-limited intragrain regions within the lithic frag-ments.1,38,39 The behavior observed in our experiments impliesthat U release from weathered HS is not diffusion-limited andsuggests that uranyl-bearing phases are widely distributedthroughout the HS rather than confined in lithic fragments.This likely resulted from the smaller HS grain radius (siltysand) and shorter weathering time (3−15 months) in this studycompared to decades-weathered field samples.

Figure 2. TRLIF fluorescence spectra of (a) meta-ankoleite, (b) ASCW6-3, and (c) ASCW3-3 with U-phosphate references.

Figure 3. Linear combination fits (blue dashed lines) to the U LIII-EXAFS spectra (black solid lines) of the ASCW3-3 sample anddeconvolution of the contribution from reference compound spectra(pointed and dashed−pointed lines). Numerical values are given inTable S3.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11015

Different U release behavior was observed for HS weatheredin ASCW6 solution (with PO4

3−) for 3 or 15 months (sampleASCW6-3 and ASCW6-15) (Figure 4a). After a sharp decreasein U concentration during the initial 20 mL of effluent, Uconcentrations gradually increased over the course of theexperiment until ∼227 mL of leaching volume was reached.This was followed by a slow release to nearly constantconcentration for the remainder of the experiment (Figure 4a).Given the constant temperature, pH, and flow conditions, agradual increase in U leaching would only occur as a function ofspecific geochemical processes affecting the properties of theweathered sediments. Shi et al.39 reported a similar leachingtrend in their column experiments to investigate the inhibitoryeffect of secondary phosphate precipitation on U release fromHanford sediments. By comparing U leaching from Hanford300 Area sediment samples treated with 1 and 50 mM PO4

3−

solutions, they concluded that the inhibitory effect of PO43−

decreased during U leaching if insufficient PO43− was provided.

Similarly, uranium release from the ASCW6-weathered HS mayhave been inhibited by the observed amorphous layer (see theFIB−SEM/TEM−EDS section). In addition, considering thehigh amount of phosphate (3 mM) applied in this weatheringsystem, phosphate adsorption on the solid phase mightoccur,40,41 which can also contribute to the inhibitory effect.The rapidly P decrease in the early stage of the columnexperiments (Figure 4c) might mainly result from thephosphate desorption, but the dissolution of the possibleamorphous Ca−P could not be ruled out. Thus, U leachingfrom meta-ankoleite would be strongly impacted by the aboveinhibition effects, resulting in a low initial U concentration inthe effluent. Following the infiltration of PO4

3−-free BPWleachate (pH 7.28), it is likely that the inhibitory effects fromthe armoring layer and phosphate-adsorption graduallydecreased during leaching through dissolution and desorptionprocesses; eventually, a constant U concentration in the effluent

from the dissolution of meta-ankoleite would be expected in thelater stages of the leaching experiment. Although theconcentration of U in the effluent did not reach steady stateduring the leaching experiment, the decrease in slope of the Urelease after 227 mL of cumulative leached volume suggeststhat the system is approaching an apparent steady state. Incontrast to the ASCW3-weathered samples, HS weatheringtime in ASCW6 solutions affected U release behavior. Uraniumeffluent concentrations from the HS sample weathered for 3months (ASCW6−3) were 2−3 times higher than effluentconcentrations for the HS sample weathered for 15 months(ASCW6−15). The difference in leaching behavior may be dueto an increase in particle size and crystallinity (i.e., Ostwaldripening) of the precipitated meta-ankoleite with increasedweathering time.Effluent U(VI) concentration from HS weathered in ASCW6

solution (with PO43−) were significantly lower (2−3 orders of

magnitude) than those from HS weathered in ASCW3 solution(without PO4

3−) (Figure 4a). Cumulatively, more than 99% ofthe initial U mass remained in the sample ASCW6-3 aftergreater than 400 mL of pore-water leaching (∼21 porevolumes), whereas almost 25% of the U mass was removedfrom sample ASCW3-3 (Figure 4b). Clearly, the differentleaching behavior exhibited by these two weathered HS samplesresulted from the sequestration of U in distinct mineral phasesthat were formed during HS weathering by acidic wastesolutions with or without dissolved phosphate. The presence ofphosphate exerts a strong control over uranium mineralogy; therapid precipitation of meta-ankoleite, whereas in the phosphate-free system, the main U-bearing mineral phases werecompreignacite-type oxyhydroxides. The insoluble U phos-phate-phase meta-ankoleite strongly retained U in the HS. Incontrast, the rate of U release was much higher from thedissolution of more soluble U-bearing oxyhydroxide minerals inthe phosphate-free system. The application of polyphosphate

Figure 4. SPFT (top: panels a and b) and column (bottom: panels c and d) leaching test results on the ASCW solution-weathered sediments. (left)Effluent uranium concentrations as a function of cumulative leached fluid volume and (right) remaining uranium mass as a function of leached fluidvolume. The black circle in panel c indicates P concentrations in the ASCW6-3-column effluent.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11016

for the remediation of U has been tested for uranium-contaminated Hanford sediments and results showed that thelong-term decrease in U leaching rate and mass were significant(∼54% to 1 year).24 Similar to the findings in this study,phosphate treatment decreases U leaching by forming Ca−Pprecipitates coated on uranium mineral surface phases, Uadsorption to precipitates, or the formation of uranium−phosphate precipitates.24,26,42

The apparent steady-state U release rate was calculated usingthe effluent concentration data through eq 1. For sampleASCW3-3, the average U concentration when U release hadplateaued, after a cumulative flow-through volume of ca. 48 mL(see Figure 4a), were selected to represent the steady-state Uconcentration. The rate of U leaching dominated by K-compreignacite dissolution at steady state was 1.05 × 10−10 molg−1 s−1 using BPW (HCO3

− = 0.3 mM). For sample ASCW6-3,U release plateaued later after the inhibitory effect (see Figure4a), so the average U concentration after ca. 227 mL leachingvolume was used to estimate the release rate. Uranium releaserate from the meta-ankoleite dominated ASCW6-3 sample was2.67 × 10−12 mol g−1 s−1. It should be noted that this is apseudo release rate because U leaching had not reached steady-state during the ASCW6-3 leaching experiment. Above all, it isclear that the U release rate for weathered samples dominatedby meta-ankoleite are about 2 orders of magnitude lower thanthose dominated by compreignacite. As far as we know, there isno reported U release rate for meta-ankoleite (K-autunite),although the dissolution rates (in terms of mol m−2 s−1) forsynthesized Na- and Ca-autunite has been reported underdifferent geochemical conditions.43−46 Unfortunately, it is hardto determine the correct dissolution rates in terms of mol m−2

s−1 in this study and compare them with those literature values.Because the surface area contributions of U minerals in the bulkweathered sediments were not easy to be determined due topossibly the dissolution of silt and clay particles in the acidicASCW solutions. As described before, lower BET-specificsurface areas were observed in the ASCW-weathered HSsamples compared to the untreated HS, even with neo-precipitated U phases.Column-Leaching Test. Similar to the SPFT results, much

less U was released from the meta-ankoleite-dominated sample(ASCW6-3 column) compared to the compreignacite-domi-nated sample (ASCW3-3 column). Cumulatively, more than99% of the initial U mass remained in the ASCW6-3 columnafter ca.176 mL of pore water leaching (263 pore volumes),whereas almost 13% of the U mass had been leached from theASCW3-3 column after ca.163 mL (256 pore volume) (Figure4d). A steady-state U effluent concentration of ca. 58 μM/L(13843 μg/L) was observed for ASCW3-3 column after 51 mLof leaching (Figure 4c). Despite the different reaction times (>5days versus ∼0.1 day of solution residence time) in the columnand SPFT systems, U release from the ASCW3-weathered HScolumns reached steady state rapidly. The estimated rates forthe two systems were similar, with the rate for the ASCW3-3column (2.42 × 10−10 mol g−1 s−1) within a factor of ca. 2 ofthe SPFT rate. However, in ASCW6-3-column, the Uconcentration continued to increase gradually during the entiretest (ca. 176 mL of BPW), and thus, no reliable release rate canbe determined; U leaching was far from steady-state, possiblydue to less-effective removal of the armoring layer (comparedto SPFT, to which more vigorous leaching (ca. 431 mL) wasapplied). Interestingly, release of U from ASCW6-3 reactedsediments in SPFT or column gives similar calculated release

rates. Using average U concentrations for BPW effluentvolumes of 60−176 mL, pseudo release rates of 4.29 × 10−13

or 6.7 × 10−13 mol g−1 s−1 were obtained for SPFT or columnexperiments, respectively.

Environmental Implications. Results of this study show thaturanium mineralogy below the two studied cribs could becontrolled by the presence of phosphate through rapidprecipitation of the U-bearing phosphate mineral meta-ankoleite. When the U- and phosphate-containing acidicaqueous waste was historically disposed in the cribs andsubsequently released into the subsurface, the waste streamcould have reacted and precipitated mainly as meta-ankoleite orsimilar phosphate phases in these legacy contaminated Hanfordsediments. Future U speciation study on field sediments shouldpay attention on uranium phosphate phases based on thissimulation results. The meta-ankoleite that formed undersimulated field conditions was well crystalline, and uranylphosphate phases in general have low thermodynamic solubilitycompared with uranyl oxyhydroxides and silicate phases,22

which could limit dissolved U(VI) concentrations in ground-water. The U mineralogy and release rates determined in thisstudy are important for the development of accurate U-releasemodels at these sites. Due to the slow U release rate of meta-ankoleite and low meteoric water infiltration at the Hanfordsite, limited U solubilization and downward migration isexpected beneath the 216-U-8 and 216-U-12 crib sites.However, the experimental effluent U concentrations ofhundreds of micrograms per liter in this study is still abovethe Environmental Protection Agency regulatory standard (30μg/L; National Primary Drinking Water Regulations, 2000).46

In addition, the phosphate concentrations in the historicallydisposed waste streams are not well-known; thus, somedischarges may have contained high PO4

3−, while othersmight have been very low. In the absence of dissolvedphosphate, U precipitates in uranyl oxyhydroxide minerals withrelatively high solubility and fast release rates.37 Thus, thereality in terms of U retention would be somewhere betweenthe two end-members explored in this study. The weatheredsediments below the two studied cribs may still represent a U-source term and require remediation to prevent contaminationof the underlying groundwater.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.est.7b03475.

Additional details on the buffering of solution pH, thecharacterization of weathered sediments, and SPFT andcolumn-leaching experiments. Figures showing measuredsolution pH, SEM and EDS results, Ca concentrations,uranium LIII-EXAFS spectra, and effluent pH. Tablesshowing calculation of the P-to-U ratio; chemicalformula, source, and spectra source; and numericalresults for linear combination fits. (PDF)

■ AUTHOR INFORMATIONCorresponding Author*Phone: (509) 372-6227; fax: (509) 371-7249; e-mail:wooyong.um@pnnl.gov.ORCIDWooyong Um: 0000-0002-3826-0030Nancy M. Washton: 0000-0002-9643-6794

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11017

Karl T. Mueller: 0000-0001-9609-9516Present Address#Division of Advanced Nuclear Engineering (DANE) andDivision of Environmental Science and Engineering (DESE),Pohang University of Science and Technology (POSTECH),Pohang, South Korea. E-mail: wooyongum@postech.ac.kr.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis research was funded by the U.S. Department of Energy(DOE) through the Subsurface Biogeochemical Research(SBR) program under grant no. SBR-DE-SC0006781. Portionsof this research were carried out at the Stanford SynchrotronRadiation Lightsource, a national user facility operated byStanford University on behalf of the U.S. Department ofEnergy, Office of Basic Energy Sciences. A portion of thisresearch was supported by the Radioactive Waste ManagementTechnology Program of the Korea Institute of EnergyTechnology Evaluation and Planning (KETEP)-grantedfinancial resource from the Ministry of Trade, Industry &Energy, Republic of Korea (no. 20141720100610). The PacificNorthwest National Laboratory (PNNL) is operated for theDOE by Battelle Memorial Institute under contract no. DE-AC05-76RL01830. The authors appreciate Dr. Carolyn I.Pearce for technical review and discussion and are grateful toDr. Michele A. Conroy and Bruce Arey for the SEM, TEM, andEDS measurements. The XRD, SEM, TEM, EDS, and TRLIFanalyses were performed in the Environmental MolecularSciences Laboratory (EMSL). EMSL located at PNNL is anational scientific user facility sponsored by the U.S. Depart-ment of Energy’s Office of Biological and EnvironmentalResearch.

■ REFERENCES(1) Qafoku, N. P.; Zachara, J. M.; Liu, C.; Gassman, P. L.; Qafoku, O.S.; Smith, S. C. Kinetic desorption and sorption of U(VI) duringreactive transport in a contaminated Hanford sediment. Environ. Sci.Technol. 2005, 39, 3157−3165.(2) Liu, C.; Zachara, J. M.; Qafoku, O.; McKinley, J. P.; Heald, S. M.;Wang, Z. Dissolution of uranyl microprecipitates from subsurfacesediments at Hanford Site, USA. Geochim. Cosmochim. Acta 2004, 68,4519−4537.(3) Chorover, J.; Choi, S.; Rotenberg, P.; Serne, R. J.; Rivera, N.;Strepka, C.; Thompson, A.; Muller, K. T.; O’Day, P. A. Silicon controlof strontium and cesium partitioning in hydroxide-weathered sedi-ments. Geochim. Cosmochim. Acta 2008, 72, 2024−2047.(4) Wang, G.; Um, W. Mineral dissolution and secondaryprecipitation on quartz sand in simulated Hanford tank solutionaffecting subsurface porosity. J. Hydrol. 2012, 472−473, 159−168.(5) Faulk, D. A.; Hedges, J. A. 216-U-8 crib and 216-U-12 crib vadosezone characterization sampling and analysis plan. 2010. DOE/RL-2009−94, revision 0.(6) Qafoku, N. P.; Icenhower, J. P. Interaction of aqueous U(VI)with soil minerals in slightly alkaline natural systems. Rev. Rev. Environ.Sci. Bio/Technol. 2008, 7, 355−380.(7) Qafoku, N. P.; Kukkadapu, R. K.; Mckinley, J. P.; Arey, B. W.;Kelly, S. D.; Wang, C.; Resch, C. T.; Long, P. E. Uranium inframboidal pyrite from a naturally bioreduced alluvial sediment.Environ. Sci. Technol. 2009, 43, 8528−8534.(8) Um, W.; Serne, R. J.; Krupka, K. M. Surface complexationmodeling of U(VI) sorption to Hanford sediment with varyinggeochemical conditions. Environ. Sci. Technol. 2007, 41, 3587−3592.(9) Um, W.; Wang, Z.; Serne, R. J.; Williams, B. D.; Brown, C. F.;Dodge, C. J.; Francis, A. J. Uranium phases in contaminated sediments

below Hanford’s U tank farm. Environ. Sci. Technol. 2009, 43, 4280−4286.(10) Ilton, E. S.; Heald, S. M.; Smith, S. C.; Elbert, D.; Liu, C.Reduction of uranyl in the interlayer region of low iron micas underanoxic and aerobic conditions. Environ. Sci. Technol. 2006, 40, 5003−5009.(11) Ilton, E. S.; Qafoku, N. P.; Liu, C.; Moore, D. A.; Zachara, J. M.Advective removal of intraparticle uranium from contaminated vadosesediments, Hanford, U.S. Environ. Sci. Technol. 2008, 42, 1565−1571.(12) Ilton, E. S.; Wang, Z.; Boily, J. F.; Qafoku, O.; Rosso, K. M.;Smith, S. C. The effect of pH and time on the extractability andspeciation of uranium (VI) sorbed to SiO2. Environ. Sci. Technol. 2012,46, 6604−6611.(13) Bickmore, B. R.; Nagy, K. L.; Young, J. S.; Drexler, J. W. Nitrate-cancrinite precipitation on quartz sand in simulated Hanford tanksolutions. Environ. Sci. Technol. 2001, 35, 4481−4486.(14) Rod, K. A.; Um, W.; Flury, M. Transport of strontium andcesium in simulated Hanford tank waste leachate through quartz sandunder saturated and unsaturated flow. Environ. Sci. Technol. 2010, 44,8089−8094.(15) Szecsody, J. E.; Truex, M. J.; Qafoku, N. P.; Wellman, D. M.;Resch, T.; Zhong, L. Influence of acidic and alkaline waste solutionproperties on uranium migration in subsurface sediments. J. Contam.Hydrol. 2013, 151, 155−175.(16) Gartman, B. N.; Qafoku, N. P.; Szecsody, J. E.; Kukkadapu, R.K.; Wang, Z.; Wellman, D. M.; Truex, M. J. Uranium fate in Hanfordsediment altered by simulated acid waste solutions. Appl. Geochem.2015, 63, 1−9.(17) Catalano, J. G.; McKinley, J. P.; Zachara, J. M.; Heald, S. M.;Smith, S. C.; Brown, G. E. Changes in uranium speciation through adepth sequence of contaminated Hanford sediments. Environ. Sci.Technol. 2006, 40, 2517−2524.(18) Stubbs, J. E.; Veblen, L. A.; Elbert, D. C.; Zachara, J. M.; Davis,J. A.; Veblen, D. R. Newly recognized hosts for uranium in theHanford Site vadose zone. Geochim. Cosmochim. Acta 2009, 73, 1563−1576.(19) Singer, D. M.; Zachara, J. M.; Brown, G. E., Jr Uraniumspeciation as a function of depth in contaminated Hanford sediments− A Micro-XRF, Micro-XRD, and Micro- and Bulk-XAFS study.Environ. Sci. Technol. 2009, 43, 630−636.(20) Arai, Y.; Marcus, M. A.; Tamura, N.; Davis, J. A.; Zachara, J. M.Spectroscopic evidence for uranium bearing precipitates in vadosezone sediments at the Hanford 300-Area Site. Environ. Sci. Technol.2007, 41, 4633−4639.(21) McKinley, J. P.; Zachara, J. M.; Wan, J.; McCready, D. E.; Heald,S. M. Geochemical controls on contaminant uranium in vadose zoneHanford formation sediments at the 200 Area and 300 Area, HanfordSite, Washington. Vadose Zone J. 2007, 6, 1004−1017.(22) Kanematsu, M.; Perdrial, N.; Um, W.; Chorover, J.; O’Day, P. A.Influence of phosphate and silica on U(VI) precipitation from acidicand neutralized waste waters. Environ. Sci. Technol. 2014, 48, 6097−6106.(23) Mehta, V. S.; Maillot, F.; Wang, Z.; Catalano, J. G.; Giammar, D.E. Effect of co-solutes on the products and solubility of uranium (VI)precipitated with phosphate. Chem. Geol. 2014, 364, 66−75.(24) Szecsody, J. E., Zhong, L., Oostrom, M., Vermeul, V. R.,Fruchter, J. S., Williams, M. D. Use of Polyphosphate to DecreaseUranium Leaching in Hanford 300 Area Smear Zone Sediment; PNNL-21733; Pacific Northwest National Laboratory; Richland, WA, 2012.(25) Wellman, D. M.; Glovack, J. N.; Parker, K.; Richards, E. L.;Pierce, E. M. Sequestration and retention of uranium (VI) in thepresence of hydroxylapatite under dynamic geochemical conditions.Envir. Chem. 2008, 5, 40−50.(26) Wellman, D. M.; Fruchter, J. S.; Vermeul, V. R.; Richards, E.;Jansik, D. P.; Edge, E. Evaluation of the efficacy of polyphosphateremediation technology: Direct and indirect remediation of uraniumunder alkaline conditions. Technol. Innovation 2011, 13, 151−164.(27) Bovaird, C. C., Rod, K., Wellman, D. M., Strandquist, S. C.Supplemental Laboratory Development of Polyphosphate Remediation

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11018

Technology for In Situ Treatment of Uranium Contamination in theVadose Zone and Periodically Re-wetted Zone; report PNNL-SA-76114;Pacific Northwest National Laboratory: Richland, WA, 2010.(28) Serne, R. J., Bjornstad, B. N., Schaef, H. T., Williams, B. A.,Lanigan, D. C., Horton, D. G., Clayton, R. E., Mitroshkov, A. V.,LeGore, V. L., O’Hara, M. J., Brown, C. F., Parker, K. E., Kutnyyakov,I. V., Serne, J. N., Last, G. V., Smith, S. C., Lindenmeier, C. W.,Zachara, J. M., Burke, D. S. Characterization of Vadose Zone Sediment:Uncontaminated RCRA Borehole Core Samples and Composite Samples;report PNNL-13757−1, revision 1; Pacific Northwest NationalLaboratory; Richland, WA, 2008.(29) Cantrell, K. J.; Um, W.; Williams, B. D.; Bowden, M. E.;Gartman, B.; Lukens, W. W.; Buck, E. C.; MAusolf, E. J. Chemicalstabilization of Hanford Tank residual waste. J. Nucl. Mater. 2014, 446,246−256.(30) Um, W.; Serne, R. J.; Brown, C. F.; Rod, K. A. Uranium (VI)sorption on iron oxides in Hanford Site sediment: Application of asurface complexation model. Appl. Geochem. 2008, 23, 2649−2657.(31) Mehta, V. S.; Maillot, F.; Wang, Z.; Catalano, J. G.; Giammar, D.E. Effects of reaction pathway on the extent and mechanism ofuranium (VI) immobilization with calcium and phosphate. Environ. Sci.Technol. 2016, 50, 3128−3136.(32) Wang, Z.; Zachara, J. M.; Gassman, P. L.; Liu, C.; Qafoku, O.;Yantasee, W.; Catalano, J. G. Fluorescence spectroscopy of U(VI)-silicates and U(VI)-contaminated Hanford sediment. Geochim.Cosmochim. Acta 2005, 69, 1391−1403.(33) Wang, Z.; Zachara, J. M.; Liu, C.; Gassman, P. L.; Felmy, A. R.;Clark, S. B. A cryogenic fluorescence spectroscopic study of uranylcarbonate, phosphate and oxyhydroxide minerals. Radiochim. Acta2008, 96, 591−598.(34) Burns, P. C. The structure of compreignacite,K2[(UO2)3O2(OH)3]2(H2O)7. Can. Miner. 1998, 36, 1061−1067.(35) Burns, P. C.; Li, Y. The structures of becquerelite and Sr-exchanged becquerelite. Am. Mineral. 2002, 87, 550−557.(36) Gorman-Lewis, D.; Burns, P. C.; Fein, J. B. Review of uranylmineral solubility measurements. J. Chem. Thermodyn. 2008, 40, 335−352.(37) Reinoso-Maset, E.; Steefel, C. I.; Um, W.; Chorover, J.; O’Day,P. A. Rates and mechanisms of uranyl oxyhydroxide mineraldissolution. Geochim. Cosmochim. Acta 2017, 207, 298−321.(38) Liu, C.; Zachara, J. M.; Yantasee, W.; Majors, P. D.; McKinley, J.P. Microscopic reactive diffusion of uranium in the contaminatedsediments at Hanford, United States. Water Resour. Res. 2006, 42, 1−15.(39) Shi, Z.; Liu, C.; Zachara, J. M.; Wang, Z.; Deng, B. Inhibitioneffect of secondary phosphate mineral precipitation on uranium releasefrom contaminated sediments. Environ. Sci. Technol. 2009, 43, 8344−8349.(40) Stollenwerk, K. G. Simulation of phosphate transport in sewage-contaminated groundwater, Cape Cod, Massachusetts. Appl. Geochem.1996, 11, 317−324.(41) Szecsody, J. E., Rockhold, M. L., Oostrom, M., Moore, R. C.,Burns, C. A., Williams, M. D., Zhong, L., Fruchter, J. S., McKinley, J.P., Vermeul, V. R., Covert, M. T., Wietsma, T. W., Breshears, A. T.,Garcia, B. J. Sequestration of Sr-90 subsurface contamination in theHanford 100-N Area by surface infiltration of a Ca-citrate-phosphatesolution; report PNNL-18303; Pacific Northwest National Laboratory;Richland, WA, 2009.(42) Troyer, L. D.; Maillot, F.; Wang, Z.; Wang, Z.; Mehta, V. S.;Giammar, D. E.; Catalano, J. G. Effect of phosphate on U(VI) sorptionmontmorillonite: Ternary complexation and precipitation barriers.Geochim. Cosmochim. Acta 2016, 175, 86−99.(43) Wellman, D. M.; Icenhower, J. P.; Gamerdinger, A. P.; Forrester,S. W. Effects of pH, temperature and aqueous organic material on thedissolution kinetics of metaautunite minerals, (Na, Ca)2−1[(UO2)-(PO4)]2 · 3H2O. Am. Mineral. 2006, 91, 143.(44) Wellman, D. M.; Gunderson, K. M.; Icenhower, J. P.; Forrester,S. W. Dissolution kinetics of synthetic and natural meta-autunite

minerals, X3−n(n)+ [(UO2)(PO4)]2•xH2O, under acidic conditions.

Geochem. Geophys. 2007, 8, 16.(45) Gudavalli, R. K. P.; Katsenovich, Y. P.; Wellman, D. W.;Idarraga, M.; Lagos, L. E.; Tansel, B. Comparison of kinetic rate lawparameters for the dissolution of natural and synthetic autunite in thepresence of aqueous biocarbonate ions. Chem. Geol. 2013, 351, 299−309.(46) U.S. Environmental Protection Agency. National PrimaryDrinking Water Regulations; Radionuclides, Final Rule. Fed. Regist.65; U.S. Environmental Protection Agency: Washington, D.C., 2000; p76707.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.7b03475Environ. Sci. Technol. 2017, 51, 11011−11019

11019