a

NUREG/CR-6627PNNL-12185

The Role of OrganicComplexants andColloids in theTransport of Radionuclidesby Groundwater

Final Report

Pacific Northwest National Laboratory

U.S. Nuclear Regulatory CommissionOffice of Nuclear Regulatory Research +Washington, DC 20555-0001 RJ9

AVAILABILITY NOTICE

Availability of Reference Materials Cited in NRC Publications

NRC publications in the NUREG series, NRC regu-lations, and Title 10, Energy, of the Code of FederalRegulations, may be purchased from one of the fol-lowing sources:

1. The Superintendent of DocumentsU.S. Government Printing OfficePO. Box 37082Washington, DC 20402-9328<http://www.access.gpo.gov/su-docs>202-512-1800

2. The National Technical Information ServiceSpringfield, VA 22161-0002<http://www.ntis.gov>1-800-553-6847 or locally 703-605-6000

The NUREG series comprises (1) brochures(NUREG/BR-XXX)Q, (2) proceedings of confer-ences (NUREG/CP-)XXXX, (3) reports resultingfrom international agreements (NUREG/IA-XX)00,(4) technical and administrative reports and books[(NUREG-XXXX) or (NUREG/CR-)OOOQ], and (5)compilations of legal decisions and orders of theCommission and Atomic and Safety UcensingBoards and of Office Directors' decisions underSection 2.206 of NRC's regulations (NUREG-x)009.

A single copy of each NRC draft report for com-ment is available free, to the extent of supply, uponwritten request as follows:

Address: Office of the Chief Information OfficerReproduction and Distribution

Services SectionU.S. Nuclear Regulatory CommissionWashington, DC 20555-0001

E-mail: <DISTRIBUTION@nrc.gov>Facsimile: 301-415-2289

A portion of NRC regulatory and technical informa-tion is available at NRC's World Wide Web site:

<http:l/www.nrc.gov>

After January 1,2000, the public may electronicallyaccess NUREG-series publications and other NRCrecords in NRC's Agencywide Document Accessand Management System (ADAMS), through thePublic Electronic Reading Room (PERR), link<http:llwww.nrc.gov/NRC/ADAMS/index.html>.

Publicly released documents include, to name afew, NUREG-series reports; Federal Register no-tices; applicant, licensee, and vendor documentsand correspondence; NRC correspondence andinternal memoranda; bulletins and information no-tices; inspection and investigation reports; licens-ee event reports; and Commission papers andtheir attachments.

Documents available from public and special tech-nical libraries include all open literature items, suchas books, journal articles, and transactions, Feder-al Register notices, Federal and State legislation,and congressional reports. Such documents astheses, dissertations, foreign reports and transla-tions, and non-NRC conference proceedings maybe purchased from their sponsoring organization.

Copies of industry codes and standards used in asubstantive manner in the NRC regulatory processare maintained at the NRC Ubrary, Two White FlintNorth, 11545 Rockville Pike, Rockville, MD20852-2738. These standards are available in thelibrary for reference use by the public. Codes andstandards are usually copyrighted and may bepurchased from the originating organization or, ifthey are American National Standards, from--

American National Standards Institute11 West 42nd StreetNew York, NY 10036-8002<http://www.ansi.org>212-642-4900

DISCLAIMIERThis report was prepared as an account of work sponsored byan agency of the United States Government. Neither the UnitedStates Government nor any agency thereof, nor any of their em-ployees, makes any warranty, expressed or Implied, or assumes

any legal liability or responsibilityfor any third party's use, or theresults of such use, of any information, apparatus, product, orprocess disclosed in this report, or represents that its use bysuch third party would not infringe privately owned rights.

NUREG/CR-6627PNNL-12185

The Role of OrganicComplexants andColloids in theTransport of Radionuclidesby Groundwater

Final ReportManuscript Completed: April 1999Date Published: January 2000

Prepared byD. E. Robertson, C. W. Thomas, S. L. PrattaK. H. Abel, E. A. Lepel, A. J. SchilkaE. L. Cooper, F. Caron, I( J. King, R. W. D. Killey, P. G. HartwigbJ. F. Mattie, M. K. Haas, E. Romaniszyn, M. Benz, W. G. EvendenbS. 0. Link0

P. Vilksd

'Pacific Northwest National Laboratory

Richland, WA 99352

Subcontractors:bChalk River Laboratories

Chalk River, Ontario, Canada KOJIJO

'Washington State University at Tri-Cities

Riclhland, WA 99352

dWhiteshell Laboratories

Whiteshell, Manitoba, Canada

E. O'Donnell, NRC Project Manager

Prepared forDivision of Risk Analysis and ApplicationsOffice of Nuclear Regulatory ResearchU.S. Nuclear Regulatory CommissionWashington, DC 20555-0001NRC Job Code L1935

Disclaimer

NUREG/CR-6627 is not a substitute for NRC regulations and compliance is not required. Theapproaches and/or methods described in this NUREG/CR are provided for information only.Publication of this report does not necessarily constitute NRC approval or agreement with theinformation contained herein.

NUREG/CR-6627 iii

Abstract

This final report for project JCN L1935,"Role of Organic Complexation and Colloidsin the Transport of Radionuclides byGroundwater", describes the results of fieldand laboratory work conducted during thelast two years of the project. It also containssummaries of important observationsreported in earlier progress reportsNUREG/CR-6429 and NUTREG/CR-6587, andother open literature publications.

The field studies were conducted at two low-level radioactive waste (LLW) managementareas within the boundaries of the ChalkRiver Laboratories, Chalk River, Ontario,Canada, under a cooperative arrangementwith the Atomic Energy of Canada, Ltd. Atthese locations, several well-defined, slightlyradioactive groundwater contaminantplumes have developed over the past 25 to40 years, providing excellent field sites forconducting studies of radionuclide transportby groundwater. This work was conductedto provide NRC, licensees, and otherinterested parties with information to betterunderstand and predict the groundwatertransport of radionuclides from LLWdisposal facilities.

These studies have primarily addressed: 1)characterization of the sub-surfacegeochemical environment near the disposalfacilities, 2) identification and quantificationof the migrating radionuclides and theirchemical speciation in groundwater, 3) thesorption behavior (Kd measurements) of themobile radionuclide species (cationic andanionic) onto site soils, 4) identification ofcolloidal radionuclides, and 5) theenvironmental dynamics of 14C in thevicinity of a solid LLW disposal facility(published earlier as NUJREG/CR-6587).

The results of these studies indicate that themobile radionuclides in groundwater at theChalk River LLW disposal sites areprimarily in solution as dissolved anionicorgano-radionuclide complexes. Thesecomplexes appear to be quite stable overlong time periods in the groundwater

environment. Ion chromatographicseparations of the soluble anionicradionuclide species from both ultra-filteredand 0.45 ffiltered groundwater from theChemical Pit contaminant plume havedemonstrated the presence of multiplechemical species for each radionuclide thatare anionic in nature and consist of organo-radionuclide species. There is some colloidaltransport, but it is minor compared to thedissolved radionuclide species.

Soil Kd measurements using the naturalanionic species as radio-tracers gave muchlower Kd values compared to measurementsmade using commercially availableuncomplexed radio-tracers. Theseexperiments confirm that the naturalanionic organo-radionuclide species ingroundwater are much more mobile thanwould be predicted from transport modelingusing classical literature Kd data bases.

Carbon-14 migrating in groundwater fromthe Area-C LLW disposal site emerges in awetland area 300 m downgradient from thewaste trenches and enters the ambientvegetation. Several field studies showedthat the 14C almost exclusively enters thevegetation by photosynthetic uptake of '4CO2released from the ground surface, whiletranspiration uptake directly from thecontaminated groundwater is essentiallynegligible.

These studies will provide performanceassessment modelers and regulators withadditional information to better assess therates and mechanisms of radionuclidetransport from LLW disposal facilities thatbecome infiltrated with water.

111 NUREG/CR-6627

-N

Contents

Abstract ............................................................................................................................................... iiiFigures ................................................................................................................................................ viiTables .................................................................................................................................................. ixExecutive Sum m ary ........................................................................................................................... xiForeword ............................................................................................................................................. xvAcknowledgem ents ............................................................................................................................. xix1 Introduction .................................................................................................................................. 1

1.1 Background Inform ation ..................................................................................................... 21.2 Description of Experim ental Study Sites at CRL .............................................................. 3

1.2.1 Chem ical Pit Liquid W aste Disposal Site ................................................................ 91.2.2 Area C Solid W aste Disposal Site .............................................................................. 141.2.3 Previous Related Chemical and Radiological Characterization .............................. 20

2 Experim ental ................................................................................................................................ 232.1 Observation W ells ................................................................................................................ 232.2 Groundwater Sam pling and Analyses ............................................................................... 23

2.2.1 Sam pling ................................................................................................................... 232.2.2 Radiom etric Analyses ............................................................................................... 24

2.3 Ultra-filtration Experim ents ............................................................................................... 242.3.1 Colloid Sam pling and Analysis ................................................................................ 242.3.2 Anion Exchange Chromatography of Ultra-filtered Groundwater ...................... 25

2.4 Laboratory Chromatography Experiments .......................................................................................... 282.4.1 Field Sam pling ......................................................................................................... 282.4.2 Chrom atographic Equipm ent ................................................................................. 282.4.3 Preparation of AG M P-1 Colum ns .......................................................................... 302.4.4 Procedure for Loading and Eluting AG MP-1 Columns ........................................ 302.4.5 Radionuclide Analysis ....................................................................................................... 302.4.6 M easurem ent of UV Spectra ........................................................................................... 30

2.5 Laboratory Kd Experiments with Anionic and Cationic Radionuclide Species ............... 302.5.1 Soil and Groundwater Sampling and Preparation for Kd Measurements .......... 312.5.2 Isolation of Anionic and Cationic Radionuclide Fractions from Contaminated

Groundwater ........................................................................................................... 322.5.3 Blanks, "Conventional Batch" Kd Measurements, and "Reverse Osmosis"

Sam ples ..................................................................................................................... 322.5.4 Speciation Checks for Soluble Radionuclides ......................................................... 332.5.5 Analysis of Soluble Groundwater Constituents ..................................................... 372.5.6 Addition of Spikes for Kd M easurem ents .............................................................. 37

3 Results and Discussion ................................................................................................................ 403.1 Concentrations and Chemical Speciation in Groundwater ............................................... 403.2 Radionuclide Association with Colloids in Groundwater ................................................. 46

3.2.1 Evolution of Colloid Speciation Along the Flowline ............................................... 463.2.2 Evolution of Chemical Speciation Along the Flowline .......................................... 50

3.3 Speciation of Radionuclides and Organics in Groundwater ............................................ 613.4 Adsorption of Anionic and Cationic Radionuclide Species onto Soil ................................ 67

3.4.1 Groundwater Characterization .............................................................................. 673.4.2 Kd Measurements Using Cationic and Anionic Radionuclide Species ................ 753.4.3 Speciation Checks During Kd M easurem ents ........................................................ 80

4 Sum m ary and Conclusions .......................................................................................................... 875 References ................................................................................................................................... 89

V V NUREG/CR-6627

6 A ppendices ............................................................ e....................................................................... 93Appendix 1. Instructions for Preparation of Chromatographic Columns, 96-02 ........ 93Appendix 2. Procedure for Chromatographic Separation, 96-05 ................................ 94

NUREG/CR-6627 vvi

Figures

1.1 Chalk River Laboratories site location map ....................................................................... 51.2 Liquid Dispersal Area and Waste Management Area A ..................................................... 61.3 Borehole and cross section locations in the vicinity of the Chemical Pit ........................... 71.4 Stratigraphic section through the Chemical Pit and East Swamp ................................... 81.5 Water table contours in the vicinity of the Chemical Pit .................................................... 111.6 Annual average concentrations of total beta-gamma activity and 9oSr at the

East Swamp Stream weir (1965-1993) .................................................................................. 121.7 Annual average concentrations of 6OCo and 1O6Ru at the East Swamp Stream

w eir (1974-1993)612 ............................................................................................................. 131.8 Borehole and cross-section locations in the vicinity of Waste Management Area C .......... 161.9 Stratigraphic section through Area C and the Duke Swamp groundwater

discharge ar ea ........................................................................................................................ 171.10 Surface hydrology, water table contours, and areal extent of groundwater

contamination by 3H (bifurcating, medium-gray pattern) in the vicinity of Area C ........ 181.11 Annual fluxes of 3H and 6oCo through the Duke Stream weir .......................................... 192.1 W ater Filtration Schem e ................................................................................................................ 262.2 Chrom atography equipm ent .................................................................................................. 292.3 Simplified flowchart of sample manipulation and analysis ................................................. 342.4 Flowchart showing the procedure for blanks (with uncontaminated groundwater)

and for speciation checks ...................................................................................................... 363.1 Cobalt-60 concentrations and chemical forms at sampling well ES-16 at the CRL

Chemical Pit groundwater plum e ........................................................................................ 413.2 Ruthenium-106 concentrations and chemical forms at sampling well ES-16 at the

CRL Chemical Pit groundwater plume ................................................................................. 423.3 Antimony-125 concentrations and chemical forms at sampling well ES-16 at the

CRL Chemical Pit groundwater plume ................................................................................. 433.4 Plutonium-239/240 concentrations and chemical forms in groundwater at well ES

-16 at the CRL Chemical Pit groundwater plume ............................................................... 443.5 Cobalt-60 concentrations and chemical forms in the CRL Chemical Pit groundwater

plume as a function of distance from the pit ....................................................................... 453.6 Cobalt-60 chromatograms for samples from the three wells .............................................. 523.7 Ruthenium-106 chromatograms for samples from the three wells .................................... 533.8 Uranium-238 chromatograms for the samples from the three wells ................................. 543.9 Neptunium-237 chromatograms for the samples from the three wells ............................. 553.10 Neptunium-237 chromatograms for samples from the three wells .................................... 563.11 Plutonium-239+240 chromatograms for samples from the three wells ............................. 573.12 Plutonium-239+240 chromatograms for samples from the three wells ............................. 583.13 Americium-241 chromatograms for samples from the three wells ................................... 593.14 Curium-244 chromatograms for samples from the three wells .......................................... 603.15 Cobalt-60 chromatogram for ES16 1996 April 22 sample .................................................. 633.16 Cobalt-60 chromatogram for ES16 1993 August 17 sample ............................................... 643.17 UV absorbance chromatogram for ES16 April 22 1996 sample ......................................... 653.18 UV absorbance chromatogram for ES9 May 27 1996 sample ............................................. 663.19 UV absorbance chromatograms for successive runs ........................................................... 693.20 UV absorbance chromatograms for successive runs (continued) ....................................... 703.21 Cobalt-60 chromatograms for successive runs .................................................................... 713.22 Cs-137 and Co-60 chromatograms for well ES16 1996 May 5 sample ............................... 723.23 Speciation of nuclides and tracers in contaminated groundwater (from well ES16)

Before and after pretreatment through a cation exchange resin ...................................... 74

Vii NUREG/CR-6627

3.24 Speciation of nuclides and tracers in contaminated groundwater (from well ES-16)Before and after pre-treatment through a anion exchange resin column. The 57Co,

94Nb, and 134Cs were the commercial tracers ..................................................................... 823.25 Nuclide speciation (tracers only) in the samples fpr the conventional "batch" Kd

determinations before and after equilibration with CRL sand. Note: "Am-241-g" refersto the results obtained with gamma spectrometry, and "Am-241-a" were obtained withalpha spectrom etry ............................................................................................................... 83

3.26 Nucide speciation (tracers only) in the samples for "batch" conventional Kddeterminations, after nanofiltration of the uncontaminated groundwater to removeorganics and colloidal particles ............................................................................................. 84

3.27 Nuclide speciation stability in aqueous solutions during the experiment(E S-16 w ith tracers) .............................................................................................................. 85

3.28 Nucide speciation stability on aqueous solutions during the experiment(ES16 uncontaminated groundwater, with tracers) ............................................................. 86

NU.REGICR-6627 viViii

Tables

1.1 Major ion chemistry in groundwater from the Chemical Pit vicinity ..................... 151.2 Major ion chemistry in groundwater from the Area C vicinity ...................... 211.3 Anthropogenic organics from the Area C groundwater plume in units of Ag/L ................ 221.4 Carbon and 14C concentrations in Area C waters ................................ 222.1 Experimental Conditions for AG MP-1 separations ............................................................ 272.2 Levels of radionuclides present in contaminated and uncontaminated groundwaters

(measured), and levels of artificial tracers added to each solution (calculated) ................ 352.3 Concentration of major ions in ES-16 contaminated groundwater at various

stages of sam ple processing .................................................................................................. 382.4 Concentrations of major ions in ES-9 uncontaminated groundwater at various

stages of sam ple processing .................................................................................................. 393.1 Gravim etric particle size analysis ........................................................................................ 473.2 Size fractionation of gamma-emitters .................................................................................. 483.3 Size fractionation of actinides .............................................................................................. 493.4 Summary of counts in the peaks of chromatograms from ultra-filtered samples .............. 513.5 Levels of radionuclides present in contaminated and uncontaminated groundwaters

(measured), and levels of artificial tracers added to each solution (calculated) ............... 733.6 Concentrations of major ions in ES-16 contaminated groundwater at various

stages of sam ple processing .................................................................................................. 773.7 Concentrations of major ions in ES-9 uncontaminated groundwater at various

stages of sam ple processing .................................................................................................. 783.8 Kd values obtained for the ambient radionuclides and the tracers for the different

sam ple m anipulations ......................................................................................................... 793.9 Comparison of values obtained in this work with ES-16 groundwater (pre-treated

with cation exchange resin to isolate the anionic species) and with artificial tracersvs oth er w orks ........................................................................................................................ 80

ix NUREG/CR-6627

,11-11

Executive Summary

This final report describes the results ofradiological and geochemical investigations ofthe mechanisms of radionuclide transport ingroundwater at two low-level waste (LLW)disposal sites within the Chalk RiverLaboratories (CRL), Ontario, Canada. Thesesites, the Chemical Pit liquid dispersal facilityand the Waste Management Area C solid LLWdisposal site, have provided valuable 30- to 40-year old field locations for characterizing themigration of radionuclides and evaluating anumber of recent site performance objectivesfor LLW disposal facilities. This informationwill aid the NRC and other federal, state, andlocal regulators, as well as LLW disposal sitedevelopers and waste generators, inmaximizing the effectiveness of existing orprojected LLW disposal facilities for isolatingradionuclides from the general public andthereby improving the health and safetyaspects of LLW disposal.

These studies have focused on identifying thephysico-chemical species of mobileradionuclides in groundwater at the field sitesand characterizing their behavior in the sub-surface environment. Following theidentification and characterization of themobile radionucide species, these solubleanionic chemical forms were isolated fromcontaminated groundwater and used astracers for conducting Kd measurements onsite soils. These special Kd measurements canthen be used to provide more realisticestimates of the transport rates of the mobileradionuclide species in groundwater. Inaddition, the role of colloids in transportingradionuclides in groundwater has beeninvestigated at this field site.

Field and laboratory studies, using large-volume groundwater sampling techniques toconcentrate and separate the soluble charge-forms of the radionuclides, have shown that anumber of mobile radionuclide species,including 55Fe, 6OCo, 106Ru, 125Sb, and 23 9 ,2 4 0pu,

are generally anionic in nature. Supportiveevidence suggests that the radionucides arebeing sequestered by naturally occurringand/or man-made complexing materials,including fulvic and humic substances. These

complexed organo-radionuclide species appearto be quite stable in the groundwater overtime. Once formed, these species change verylittle in their physicochemical propertiesduring transport in the groundwater.

Ion chromatographic separations of themobile radionuclides and dissolved organicconstituents concentrated fromcontaminated groundwater onto anionexchange resin were conducted to furthercharacterize these organo-radionuclidespecies. These separations have identifiedmultiple individual chemical species of60Co, 106Ru, and 2 3 9 ,2 4 0 Pu, indicating thepresence of a complex mixture of thesesequestered radionuclides beingtransported in the groundwater.Chromatograms for 6OCo from samplescollected in 1993 and 1996 showed that theconcentrations declined and the chemicalspeciation slightly changed since use of thepit was discontinued. Chromatograms ofsoluble organic constituents in thegroundwater were produced frommeasurements of UV absorbance at 254 nmwith a UV detector. Speciation of organicsin samples collected inside and outside theplume showed many similarities both inthe shapes of the chromatograms and inthe UV absorbance spectra of separatedfractions. This indicated that many of theorganic species in the plume were naturalin origin. Repeated separations fromportions of the same sample showed thatthe technique gave very consistent resultsand could be used to measure smallchanges in chemical speciation. Severalsoluble anionic species of 137Cs were alsoobserved, which confirms that organiccomplexation is also sequestering some ofthe 137Cs. These results have provided abetter understanding of the chemicalnature and behavior of radionuclides andorganics in groundwater and theirinteractions within the contaminant plume.These studies have shown, through the use ofmultiple state-of-the-art characterizationmethodologies, the consistent relationshipbetween migrating radionuclides andnaturally occurring humic and fulvic

xi xi NUREG/CR-6627

substances in CRL groundwater. Theubiquitous presence of humic and fulvicmaterials in sampled waters, their substantialcapacity for complexation of cationicradionuclides under the existing geochemicalconditions, and their general lack of attractionfor most mineral surfaces in the aquifermedium suggest that these organic specieswere primarily responsible for the facilitatedtransport of radionucides at this location. Asthe CRL site serves as a viable model for LLWdisposal sites in humid environments, theresults of this investigation have far-reachingimplications in the fields of facility design,maintenance, and remediation.

The evolution of the physico-chemicalspeciation of radionuclides in a contaminantplume was studied by sampling three wells ona flowline from the Chemical Pit disposal siteto a nearby wetland. Ultra-filtration was usedto separate the samples into fractions: >450nm, 10 to 450 rm, 1 to 10 nm and <1 nmnominal sizes. The results showed that theradionucide species were predominantly inthe < 1 nm size range and could be classifiedas "soluble" species. This was likely due toorganic complexation, since previous studiesshowed much evidence of complexed species inthe plume.

However, small but significant quantities ofradionuclides were associated with colloidalsized particles in the 1-10 rm and the 10-450nm size ranges. This suggests that colloidaltransport of radionuclides in this groundwatersystem could be occurring to a small butmeasureable degree. Colloidal transport is,nevertheless, small compared to the transportof soluble complexed organo-radionucidespecies.

The "soluble" species in the < 1 nm fractionwere further studied by pre-concentratingthem on AG MP-1 anion exchange resin andthen chromatographically separating them byeluting with various concentration gradientsof KC1 and HCl. The chromatograms for 60Co,106Ru, and actinides separated from the threesamples showed small changes in theintensities of peaks with distance from the pit.These changes indicated that the chemicalspeciation was slightly evolving as theradionucides move through the plume.

Radionuclides present in contaminatedaquifers have varying degrees ofinteraction with geological material, whichcn be determined experimentally using astandard liquid-solid partition coefficient(Kd). This coefficient is an empiricalparameter that lumps together severalsurface-driven processes at the surface ofthe solids, as well as differences in aqueousspeciation of the radionuclides.

Previous studies at the Chalk RiverLaboratories (CRL) have revealed thatsome radionuclides in contaminatedgroundwater are much more mobile thanexpected, based on predictions using Kdvalues from laboratory batchmeasurements. This increased mobilitywas linked to the presence of stable anionicspecies of the radionuclides which have arelatively low affinity for sorbing to soil. Inthe current work, aqueous species of theradionuclides present in the contaminatedaquifer (6OCo, lO6Ru, 137Cs, 238Pu, 239/24 0Pu,241Am and 244 Cm), were isolated in their

anionic and cationic forms to separatelydetermine the Kd's for these two groups ofcharge-forms. These results werecompared with batch Kd's measured afterspiking groundwater with differentradioisotopes of the same elements thathad no prior opportunity to formcomplexes.

Batch Kd values obtained with isolated anionicforms of 6 0Co, 137Cs, Pu isotopes and 24 1Amfound in the aquifer, were lower by one to twoorders of magnitude compared to the Kdvalues of 57Co, 134Cs, 239/24OPu and 24

lAin, addedas tracers in laboratory short-termconventional measurements. These Kdvalues, measured by using the actual anionicradionucide species that have exhibitedenhanced mobility in groundwater, will allowtransport modelers to make more accuratepredictions of the "first arrival" times ofradionuclides migrating from LLW facilities.

With the exception of Ru, all of theradioelements considered here would beexpected to occur as cationic species in theaquifer. Mixes of negative and positivespecies of the aquifer radionucides wereshown to be stable over the duration of the

NUREG/CR-6627 xii

experiment, whereas the added tracersremained predominantly positively charged,except for 94Nb. This confirmed that theaqueous speciation observed during theseexperiments was as close as possible to the insitu conditions, and had a strong andapparently lasting influence on sorption.

A special study of the dynamics of 14C in theDuke Swamp wetland area near the Area-Csolid LLW disposal site was conducted duringthis investigation. This study has recentlybeen published as a separate report(NUREG/CR-6587), but a summary of theimportant observations are given here.

An understanding of the pathways for transferof 14CO2 from low-level radioactive wastemanagement sites to the environment isneeded for risk assessment analyses. At theChalk River Laboratories (CRL), AtomicEnergy of Canada (AECL), 14C is beingreleased from a LLW management area and itis being transported in groundwater to awetlands discharge area some 300 metersfrom the waste trenches. This area providesthe opportunity to quantify the distributionand movement of 14C in the environment.Observations of 14C in ground water, surfacewater, air, plants and peat were made. Inaddition, it was possible to determine themajor pathway for 14C uptake by vegetationgrowing at this site, i.e., photosyntheticuptake through the leaf system versustranspirational uptake through the rootsystem.

Concentrations of 14C in the groundwaterbetween the LLW disposal facility andDuke Swamp remained constant untildischarge into the swamp surface. Carbon-14 dissipated very quickly at the surface,with 270,000 pCi-g"' carbon in thegroundwater changing to 1890 pCi-g-1 atthe first ground-level passive air sampler.Surface waters leaving the swamp had noenhanced 14C specific activity.

An experimental study area(approximately 10m x 10m square) locatedaround the highest concentrations of 14C

observed in vegetation growing in DukeSwamp was fenced, and studies of theenvironmental dynamics of the 14C were

conducted within this fenced area. Airsamplers placed within this experimentalsite near the contaminant plume dischargepoint in Duke Swamp, showed little changein 14C specific activity with horizontaldistance or time within the 270 M2 fencedarea. At a height of 6 meters, the 14Cspecific activity was 297 pCi g-1 which was adecrease of approximately 6-fold fromconcentrations on the ground.

A vegetation survey within the fenceshowed little heterogenity with distance.The 14C specific activity was significantlygreater in the moss and grass than in forbs,ferns, or fir. The moss and grass sampleshad greater specific activity because theywere collected closer to the ground than theforbs, ferns and fir. There was anexponential decrease in 14C specific activityin vegetation with height. Cedar leavescollected at the same heights as the lichensshowed 14C specific activity similar to thelichens.

The specific activity in peat was 1890pCi g-1 C near the surface, dropping tobackground levels at a depth of 5.5 cm.Using 30 years as the exposure time in theecosystem, it was estimated that the peataccumulation rate was 1.8 mm yr 1.

The significance of the transpirationalpathway for transport of 14C ingroundwater through roots to the shootswas tested in two ways. The first methodcompared the ratio of specific activity oflichens to specific activity of vascularplants in two sites, one without 14C in thegroundwater and one with 14C in thegroundwater. The specific activity ratioswere the same. The second methodcompared the specific activity in shoots ofvascular plants transplanted into theexperimental area with 14C in the groundwater. Pairs of each species were eitherbagged or left open (for root uptake). Forthe time period of the experiment, noincrease in 14C was noted in the plants thatwere left open compared with the plantswith the bagged root systems. By thesetests, it was concluded that there was nosignificant uptake of 14CO2 in the

xiii NUREG/CR-6627

transpirational stream within the limits of themeasured precision.

Carbon-14 concentration ratios for air tovegetation ranged from 1.6 to 5.9. Thesevalues, being greater than one, suggestsignificant gradients in air and vegetationlevels near the ground. Because there was no

transport of 14C from groundwater directlyinto vegetation, the concentration ratio fromgroundwater-to-vegetation must be close tozero. Background levels of 14C were found inthe peat below 5.5 cm and it could be assumedthat the concentration ratio fromgroundwater-to-peat was close to zero.

NUREG/CR-6627 vxiv

Foreword

Any candidate site intended for use as afacility for the shallow burial of low-levelradioactive waste must satisfy therequirements of 10 CFR 61 (LicensingRequirements for Land Disposal ofRadioactive Waste), subpart D, section 61.5(2).This regulation states that any proposed sitemust be capable of being characterized,modeled, analyzed, and monitored. Whensuch facilities are licensed, the primaryconcern is the means (natural and/orenhanced) by which radionuclides migratebeyond site boundaries. A thoroughunderstanding of these means could lead tothe prevention or minimization of suchtransport, thereby sparing the general publicfrom undue exposure. Furthermore, shallowland burial has been routinely used for thedisposal of low-level radioactive wastes duringwhich imperfect disposal practices,precipitation infiltration, and groundwaterleaching of various waste forms have led tothe contamination of many surface andsubsurface waters (Riley and Zachara 1992).Despite current waste-management programs,which are intended to modify and/or reducewaste streams and to contain pollutants frompast disposals, the active migration ofradionuclides from many of theaforementioned sites is an almost ubiquitousphenomenon (e.g., McCarthy 1988; Olsen et al.1986; and references therein). Whether one isconcerned with projected or existingradioactive waste repositories, a clear anddetailed understanding of the mechanisms ofradionuclide transport (both natural andfacilitated) is imperative and critical to anyrealistic performance assessment.

To ensure the compliance of proposed sitesand to perform environmental assessments ofexisting ones, most investigators usepredictive models of radionuclide transportthat employ laboratory-derived values for aparameter referred to as the radionuclidedistribution coefficient (Kd). This parameter isessentially a ratio of the radionuclide activityper unit mass of soil

.(aquifer medium) to the radionuclide activityper unit volume of groundwater. Themagnitude of this ratio is directly related tothe extent to which the radionuclide inquestion is retarded (or retained by thestationary phase) with respect to themigrating fluid. A number of recent studieshave identified significant discrepanciesbetween actual migration rates and thosepredicted by models for some radionuclides innatural settings (Robertson et al. 1989; andreferences cited). It is generally acknowledgedthat transport models based onthermodynamic calculations of radionuclidespeciation' (Rai and Serne 1978) andlaboratory analyses (e.g., studies involving"ideal" radiotracers that may be in some formother than those of the actual mobile speciesin natural groundwaters) often fail to producereliable results. This is undoubtedly becausethese procedures disregard the potential forradionuclide complexation by natural andman-made organic species as well asincorporation of radionuclides inmicroparticulate or dissolved colloidalmaterial (Bondietti 1982; Champ et al. 1988;McCarthy 1988; McCarthy and Zachara 1989;Petit 1990; Warwick et al. 1991).Consequently, the current study supports thecontention that to understand themechanisms that determine the extent ofradionuclide migration or retardation in anexisting or proposed setting, it is imperativethat species exhibiting facilitated transport beexamined in actual field situations andsubsequently compared with results fromlaboratory studies and calculatedthermodynamic data. Fundamentalinformation of this sort can then beincorporated into models for more accuratepredictions of radionuclide movement fromrepositories and thereby enhance strategiesfor site planning or remediation by artificiallymanipulating retention or mobilization ofthese species.

IThis term refers to the individual physico-chemical forms of anelement, which together make up its total concentration in asample (Florence 1982).

XV NUREG/CR-6627

Several mature low-level waste managementand experimental dispersal sites exist atChalk River Laboratories, Ontario, Canada,and are excellent analogues for shallow buridfacilities in humid environments. These sitesprovide a unique and invaluable opportunityto study the long-term behavior and transportof a number of critical radionuclides (virtuallyall of those addressed in 10 CFR 61), whichhave resided in a shallow groundwater flowsystem for nearly 40 years. The current studywas initiated to sample groundwaters fromthis natural setting, use multifarioustechniques to elucidate the speciation of theradionuclides migrating in this environment,and investigate the roles of organiccomplexation and microparticulates in thefacilitation of this phenomenon.

Results of this study point to the existence ofcomplex, heterogeneous, and highly variable

chemical and redox environments that arecontrolled, at least in part, by the presence ofnatural organic material in the Chalk Rivergroundwaters. Anionic speciation of mostradionuclides predominates in the areasstudied, and particle-size and spectroscopicanalyses indicate a correlation between theseradionuclides and large organicmacromolecules, lending support to thepremise that naturally occurring humic andfulvic materials are facilitating the transportof radioactive contaminants in the subsurface.Additional evidence seems to suggest thatmultiple organo-radionuclide complexes are inthese waters, and considerable conversionmay be occurring between migrating species,both spatially and temporally. This hassignificant implications with regard to thepotential for controlling radionuclidetransport to either ensure contaminantretention or facilitate the extraction thereof.

Foreword References

Bondietti, E-A. 1982. "Mobile Species of Pu,Am, Cm, Np, and Tc in the Environment."In Environmental Migration of Long-LivedRadionuclides, IAEA-SM-257/42,International Atomic Energy Agency,Vienna, Austria.

Champ, D.R., J.L. Young, D.E. Robertson,and K.H. Abel. 1985. "Chemical Speciationof Long-Lived Radionuclides in a ShallowGroundwater Flow System." Water Poll.Res. J. Canada 19(2).

McCarthy, J.F. 1988. Role of ColloidalParticles in the Subsurface Transport ofContaminants. DOE/ER-0384, U.S.Department of Energy, Washington, D.C.

McCarthy, J.F., and J.M. Zachara. 1989."Subsurface Transport of Contaminants."Environ. Sci. Technol. 23:496-502.

Olsen C.R., P.D. Lowry, S.Y Lee, I.L.Larsen, and N.H. Cutshall. 1986."Geochemical and EnvironmentalProcesses Affecting Radionucide Migrationfrom a Formerly Used Seepage Trench."Geochim. Cosmochim. Acta 44/45:165-170.

Petit, J.C. 1990. "Migration ofRadionuclides in the Geosphere: What canwe learn from Natural Analogues?"Radiochim. Acta 51:181-188.

Rai, D., and R.J. Serne. 1978. Solid Phasesand Solution Species of Different Elementsin Geologic Environments. PNL-2651,Pacific Northwest Laboratory, Richland,Washington.

Riley, R.G., and J.M. Zachara. 1992.Chemical Contaminants on DOE Landsand Selection of Contaminant Mixtures forSubsurface Science Research. DOE/ER-0547T, U.S. Department of Energy,Washington, D.C.

Robertson, D.E., M.P. Bergeron, D.Holford, K.H. Abel, C.W. Thomas, D.A.Myers, D.R. Champ, R.W.D. Killey, D.L.Moltyaner, J.L. Young, and T. Ohnuki.1989. Demonstration of PerformanceModeling of a Low-Level Waste Shallow-Land Burial Site. NUREG/CR-4879 vol. 2,U.S. Nuclear Regulatory Commission,Washington, D.C.

Warwick, P., A. Hall, P. Shaw, J.J.W.Higgo, G.M. Williams, B. Smith, D. Haigh,

NUREG/CR-6627 vxvi

and D. Noy. 1991. "The Influence ofOrganics in Field Migration Experiments

(Part 2)." Radiochim. Acta 52/53:465-471.

xvii xvii NUREG/CR-6627

Acknowledgements

Financial support for this work was providedby the U. S. Nuclear Regulatory Commissionthrough a contract with Pacific NorthwestNational Laboratory.

The authors wish to express their appreciationto the NRC Project Manager, Dr. EdwardO'Donnell, of the Radiation Protection,Environmental Risk, and Waste ManagementBranch, Division of Risk Analysis andApplications, Office of Nuclear RegulatoryResearch, for helpful guidance andsuggestions in conducting this work. Theauthors also acknowledge and expressappreciation to Mr. John W. Craig, Director,Division of Regulatory Applications and Dr.Sher Bahadur, Chief, Waste ManagementBranch at NRC for their support for this work.

We express appreciation to the Atomic Energyof Canada, Ltd., and the Chalk RiverLaboratories for sharing their facilities, staff,and expertise in making this collaborativeresearch venture a resounding success.

Without their cooperation and assistance, theimportant research observations andregulatory implications resulting from thiswork, which will improve the health andsafety aspects of low-level radioactive wastemanagement and disposal, would not havebeen possible.

We acknowledge the assistance of E.T. McGeeand P.J. Durepeau in the collection ofgroundwater samples. G. Fisher, D.P. Batisteand E.F. Romaniszyn assisted in thepreparation of sources for actinide analysis,while J.M. Cox counted the samples on thealpha spectrometers.

We also wish to thank D.S. Hartwig, C.Lafontaine, A. Billo, and especially T.L. Eve,for their help in sample analysis, and R.W.D.Killey for providing electronic copies of themaps used for several of the figures. Hiscomments have helped improve thismanuscript. M. Scrimgeour has kindlyperformed the sand size distribution.

xix XIX NUR.EG/CR-6627

1 Introduction

A major concern regarding the operation ofexisting and future low-level radioactivewaste (LLW) disposal facilities is theability to adequately model and predict theperformance of the site with respect tooffsite migration of radionuclides from thedisposal facility. Unfortunately, thephysical and chemical processes thatcontrol the transport of radionuclides byinfiltrating groundwater are not wellunderstood.

The migration potential of radionuclides isstrongly influenced by the physical andchemical speciation of the radionuclides.For example, radionuclides existing incationic chemical forms are more readilyadsorbed onto soil components, therebyretarding the transport of the radionuclidesby groundwater. Conversely, radionuclideswhich exist in anionic chemical formsmigrate at much faster rates because of thepoor adsorption onto soil (Cooper andMcHugh, 1983; Killey, et al., 1984; Champand Robertson, 1986; Cooper and Mattie,1993; Robertson, et al., 1995; Schilk, et al.,1996). Earlier studies have also shownthat in the natural environment somecationic species reacted with organiccomplexants in the groundwater to formanionic organo-radionuclide complexeswhich exhibit significantly lower Kd's onsoil than predicted by geochemicalmodeling (Fruchter, et al., 1985). Inaddition, colloidal species have been shownto transport radionuclides in groundwaterover considerable distances (Buddemeierand Hunt, 1988; Penrose, et al., 1990;Robertson, et al., 1995, Schilk, et al., 1996;Kersting, et al., 1999). If the physico-chemical speciation of these radionuclidesis modified by organic complexation andlorcolloid formation (either natural or man-made), the rate of migration may bedramatically enhanced.

To more fully understand the mechanismsthat determine the extent of radionuclide

migration or retardation in existing orfuture LLW disposal facilities, it isimperative that those species exhibitingtransport be examined in actual fieldsituations. Field studies are importantbecause of the occurrence of natural andman-made complexing agents present ingroundwater which alter the speciation ofradionuclides. Field observations can thenbe compared with the results of laboratoryand thermodynamic modeling predictionsto better understand the nature of thecomplexed species. Empirical data canthen be incorporated into transport modelsfor more accurate predictions ofradionuclide migration rates from LLWdisposal facilities, thereby enhancing thehealth and safety aspects of radioactivewaste disposal.

To assist in the characterization of theprocesses affecting transport, sampling ofgroundwater plumes contaminated withradionuclides was conducted at ChalkRiver Laboratories (CRL) in Ontario,Canada, to identify and quantify themobile radionuclide species originatingfrom two individual LLW disposal sites.Sampling was conducted at two locations:1) the Chemical Pit within the Chalk RiverLiquid Dispersal Area, which has receivedaqueous wastes containing variousradioisotopes, acids, alkalis, complexingagents, and salts since 1956, and 2) WasteManagement Area C, a 30-year old solidLLW disposal facility consisting of a seriesof trenches containing LLW from CRL andvarious other facilities throughout Canada.These mature LLW management sites areexcellent analogues for prospective shallowland burial facilities that are located inenvironments where precipitation exceedsevapotranspiration. They provide aunique and invaluable opportunity to studythe long-term behavior and transport of anumber of important radionuclides, manyof which have resided in this shallowgroundwater flow system for nearly 40years.

I NUREG/CR-6627

The current study was initiated to: 1)sample groundwater from this naturalsetting, 2) use multiple analyticaltechniques to determine thephysicochemical speciation of the mobileradionuclides, 3) investigate the role oforganic complexation and colloids in thefacilitation of radionuclide transport, and4) determine the soil adsorption propertiesof the mobile anionic complelxedradionuclide species isolated from thecontaminated groundwater.

1.1 Background Information

A candidate site intended for use as afacility for the shallow burial of low-levelradioactive waste must satisfy therequirements of 10 CFR 61 (LicensingRequirements for Land Disposal ofRadioactive Waste), subpart D, section61.5(2). This regulation states that anyproposed site must be capable of beingcharacterized, modeled, analyzed, andmonitored. When such facilities arelicensed, the primary concern is the means(natural and/or enhanced) by whichradionuclides migrate beyond siteboundaries. A thorough understanding ofthese mechanisms could lead to theprevention or minimization of suchtransport, thereby sparing the generalpublic from undue exposure.

Furthermore, shallow land burial has beenroutinely used for the disposal of low-levelradioactive wastes during which imperfectdisposal practices, precipitationinfiltration, and groundwater leaching ofvarious waste forms have led to thecontamination of many surface andsubsurface waters (Riley and Zachara1992). Despite current waste-managementprograms, which are intended to modifyand/or reduce waste streams and to containpollutants from past disposals, the activemigration of radionuclides from many ofthe aforementioned sites is an almostubiquitous phenomenon (e.g., McCarthy1988; Olsen et al. 1986; and referencestherein). Whether one is concerned withprojected or existing radioactive waste

repositories, a clear and detailedunderstanding of the mechanisms ofradionuclide transport (both natural andfacilitated) is imperative and critical to anyrealistic performance assessment.

To ensure the compliance of proposed sitesand to conduct performance assessments ofexisting ones, most investigators usepredictive models of radionuclide transportthat employ laboratory-derived values for aparameter referred to as the radionuclidedistribution coefficient (Kd). Thisparameter is essentially a ratio of theradionuclide activity per unit mass of soil(aquifer medium) to the radionuclideactivity per unit volume of groundwater.The magnitude of this ratio is directlyrelated to the extent to which theradionuclide in question is retarded (orretained by the stationary phase) withrespect to the migrating fluid. A number ofrecent studies have identified significantdiscrepancies between actual migrationrates and those predicted by models forsome radionuclides in natural settings(Fruchter, et al., 1985; Robertson et al.1989; and references cited). It is generallyacknowledged that transport models basedon thermodynamic calculations ofradionuclide speciation (Rai and Serne1978) and laboratory analyses (e.g., studiesinvolving "ideal" radiotracers that may bein some form other than those of the actualmobile species in natural groundwaters)often fail to produce reliable results. Thiscan result in underestimation ofradionuclide transport because theseprocedures disregard the potential forradionuclide complexation by natural andman-made organic species as well asincorporation of radionuclides in colloidalmaterial (Bondietti 1982; Champ et al.1988; McCarthy 1988; McCarthy andZachara 1989; Petit 1990; Warwick et al.1991).

Consequently, the current study supportsthe contention that to understand themechanisms that determine the extent ofradionuclide migration or retardation in anexisting or proposed setting, it isimperative that species exhibitingfacilitated transport be examined in actual

NUREG/CR-6627 2

field situations. The field observations canthen be compared with results fromlaboratory studies and/or calculatedthermodynamic modeling to reconciledifferences or add further confidencebetween empirical versus calculatedresults.

Fundamental information of this sort canthen be incorporated into models for moreaccurate predictions of radionuclidemovement from repositories and therebyenhance strategies for site planning orremediation by artificially manipulatingretention or mobilization of these species.Results of studies conducted at the slightlycontaminated field sites at Chalk RiverLaboratories' LLW disposal facilities pointto the existence of complex, heterogeneous,and highly variable chemical and redoxenvironments that are controlled, at leastin part, by the presence of natural organicmaterial in the Chalk River groundwater.Anionic speciation of most radionuclidespredominates in the areas studied, andparticle-size and spectroscopic analysesindicate a correlation between theseradionucides and large organicmacromolecules, lending support to thepremise that naturally occurring humicand fulvic materials are facilitating thetransport of radioactive contaminants inthe subsurface. Additional evidence seemsto suggest that multipleorgano-radionuclide complexes of eachradionuclide are in these waters, andconsiderable conversion may be occurringbetween migrating species, both spatiallyand temporally. This has significantimplications with regard to the potentialfor controlling radionucide transport toeither ensure contaminant retention orfacilitate the extraction thereof.

1.2 Description ofExperimental Study Sites atCRL

The Chemical Pit is located in the LiquidDispersal Area near the northern boundaryof the Lower Perch Lake Basin (Figures 1.1and 1.2). The Liquid Dispersal Area has aseries of gravel-lined infiltration pits in a

sandy, unsaturated horizon. This area isbounded to the southeast by a steep slopeand the East Swamp, and to the southwestby a shallower grade that overlooksanother wetland (South Swamp). TheChemical Pit, situated within this area, iscomposed of a pair of 5-m-deep,gravel-filled infiltration pits (Figure 1.3)that have accepted low-level liquid wastessince the mid-1950s. These wastewatersincluded a host of radionuclides (activationand fission products) and occasionaldischarges of small quantities of acids,alkalis, salts, and complexing agentsoriginating from various research anddevelopment operations at CRL (Killey andMunch 1984). The disposal ofanthropogenic complexing agents at thissite has been discouraged since 1961.The local geology in the vicinity of theChemical Pit (Figure 1.4) is made up of avariety of unconsolidated Quaternarysediments overlying a topographicallyirregular and locally fractured bedrock ofPrecambrian granitic to monzonitic gneiss,typical of much of the Canadian Shield.The bedrock is overlain by a sandy till ofvarying thickness that exhibits a low tomoderate silt content with abundantgravel, cobbles, and boulders (oftenexceeding 3 m in diameter). Above the tillis a thin ( 1 m) layer of clayey silt, which is,in turn, overlain by silty tomedium-grained sands with interbeddedlayers of clayey silt. These sand layers,generally less than 30 m in thickness, arecomposed largely of quartz (30-40%) andaluminosilicate minerals (30-40%) withassociated alteration products(horneblende, biotite-vermiculite,magnetite, garnet), and trace amounts ofsulfides, carbonates, organic matter, andhydrated oxides of Fe and Mn (Champ etal. 1985; Killey and Munch 1984; Killey etal. 1984). The lower portion of these sandsis probably composed of fluvial deposits,whereas the shallower materials areconsidered to be aeolian sheet depositscreated from the reworking of the fluvialsands.

Precipitation, totaling approximately 75 cmper annum, is distributed evenlythroughout the year and recharges a water

3 3 NUREG/CR-6627

table normally less than 5 m from theground surface. This has led to theformation of swampy areas that containbog deposits of peat and other organicmatter (generally less than 0.5 m inthickness throughout the region. Existingboreholes in this area were used todetermine hydraulic head distributions. Asubsurface flowpath was found to extendfrom the Chemical Pit to the East Swamp

(Figure 1.5), with a groundwater residencetime of 6 months to 2 years and a meanflow rate of 5 to 25 cm/day. Consequently,t.hree wells along this gradient wereselected for sampling near-field (CP-4),far-field (ES-16), and intermediate (ES-39)groundwater in the hopes of observing anytemporal changes associated with themigrating radionuclide species (refer toFigures 1.3 and 1.4 for actual welllocations).

NUTREGICR-66274 4

Figure 1.1 Chalk River Laboratories site location map

Waste Management Area

Paved Road

Gravel Road

A StreamWetland

Lake o 50o 1000 1500 m

5 5 NUREG/CR-6627

Figure 1.2 Liquid Dispersal Area and Waste Management Area A

0 l100

NUREG/CR-66276 6

Figure 1.3 Location of boreholes and cross-section in the vicinity of the Chemical Pit

7 7 NUREG/CR-6627

N0%0~

175

0)N

0

N

170

105

00

.......................... .............. .... .. ..

... .. .. .. .Ko

C+

0

0

o 0

Fi13DlI~ i ne* Rnto Msedhzti.Fl Sands

Sity Vary Fine Sand

~ZGmvad Backfd

- OeSM

z2GladalNWater.T.ble

1.2.1 Chemical Pit Liquid WasteDisposal Site

The field study site was adjacent to theChemical Pit infiltration trench in dune sands,located on CRL property approximately 1 kmSW of the main laboratory (see Figs. 1.2-1.4).The disposal site consisted of a circular pit,24 m dia. u 4 m deep, connected to atrapezoidal pit, both excavated in dune sandand backfilled with coarse aggregate.

The Chemical Pit overlies an aquifer thatcarries water that has infiltrated through thedune deposits, as well as the wastewaterspumped to the pit, southeast to a groundwaterdischarge area in a perennial wetland (theEast Swamp). The base of the gravel backfillof the infiltration pit is about 2 m above thewater table. The groundwater flowpath to thedischarge area is about 80 to 120 m long.Groundwater residence time in the permeablefine-medium sand is relatively low; estimatesof water (and non-reactive contaminant)subsurface travel times ranged from about 180to 670 d (Killey and Munch, 1984).

Aqueous solutions containing variousradioisotopes, acids, alkalis, and/or salts havebeen discharged to this pit since 1956,although available data indicate that theinputs of non-radiological contaminants weregenerally very low. Nearly 331,000 m 3 ofwaste water, containing approximately 230TBq (TBq = 1012 Bq) or 6200 Curies of betaemitters (liquid + solids) and 0.31 TBq or 840Curies of alpha emitters has been dischargedto this pit (Welch and Killey, in prep.) Avariety of activation and fission products, anumber of which were gamma emitters, werepresent in the waste water. The pit was usedregularly for waste water discharges until1992, and a few intermittent releases wererecorded during the period of 1993-1995. Aseries of interception wells, installeddownstream from the pit in 1994, now supplywater to a treatment plant whereradionuclides are removed.

The mineralogy consists of quartz andfeldspars (each making up 30-40%), withminor or accessory muscovite and

ferromagnesian minerals (hornblende, biotite-vermiculite, magnetite, garnet).

The concentrations of radionucides in theaquifer have decreased in recent yearsbecause of radioactive decay and thediscontinuation of the routine use of theinfiltration pit as a waste managementfacility. Nevertheless, a wide variety ofradioisotopes remain detectable in the aquifer.Multiple chemical species of a number ofradionuclides (including 60Co, 106Ru, and2391240 Pu), first identified in the early 1980's(Cooper and McHugh, 1983; Killey et al., 1984,Champ et al., 1984; Champ and Robertson,1986) have also remained present in thecontaminant plume (Cooper and Mattie, 1993;Robertson et al., 1995; Schilk et al., 1996).Some of these (e.g., ruthenium) would beexpected to form inorganic anionic complexes,but the anionic forms of other radioisotopes(including cobalt and plutonium) have beenattributed to the formation of complexes withdissolved organic compounds, most probably ofnatural origin. The persistence and stabilityof the organo-radionucide complexes, and theinference that the organics are of naturalorigin, and hence, likely to be commonlyencountered in other flow systems, indicatethat they may often influence the migration ofradioisotopes. It is also probable that theireffects have not been present in many of thelaboratory experiments that have been used indeveloping Kd databases. This study wasundertaken to investigate the effects of theseaquifer complexes on sorption coefficients withgeological material.

A previous sampling effort in July of 1980(Killey and Munch 1984) indicated that themajor ion loading at the Chemical Pit wasrather low, and total dissolved solids in thecontaminant plume were not substantiallydifferent from local uncontaminatedgroundwater (Table 1.1). The keydistinguishing features of the plume, apartfrom the presence of radionucides, wereelevated Na and chloride concentrations,elevated Eh, and lower dissolved iron (Fe2+)concentrations compared to nearbygroundwaters. It was also noted during thisstudy that sands within the Chemical Pitplume had significantly heavier surface

9 9 NUREGICR-6627

coatings of iron oxyhydroxides than didequivalent sands outside of the plumeboundaries. These coatings were interpretedto be a direct result of the more oxidizingconditions associated with the plumeenvironment relative to the surrounding,uncontaminated waters, which led to theformation of insoluble ferric (Fe3+)compounds.

Weekly records of streamflows and totalbeta-gamma-emitter concentrations have beenmaintained for the East Swamp Stream. Thisstream provides a potential discharge path forthe plume contaminants, since 1965.Radiochemical analyses and gamma-rayspectrometry were performed on quarterlycomposites of the stream water from 1974 to1992. Currently, these are done on a monthlybasis, as are tritium and gross-alpha analyses.Since 1989, the average annual pH has rangedfrom 5.5 to 6, and the conductivity has rangedbetween 300 and 600 S/cm.

Perhaps the most significant event in theoperational history of the Chemical Pitoccurred in 1961 when a solution containingappreciable quantities of citrate (and, possibly,a number of other complexing agents) wasdischarged to the pit. This event was followed3 months later by a marked increase in 9°Srconcentration and the first appearance of 60Coin the East Swamp Stream. The presence ofcitrate and other complexing agents wasapparently responsible for these phenomena

and may have dissolved the previouslyadsorbed cationic species and facilitated theirtransport toward the discharge zone.

Another large increase in East Swamp Streamradionuclide concentration occurred in 1990(Figure 1. 6). The total beta-gamma and 9°Srconcentrations were observed to risedramatically throughout the year, resulting inaverage annual concentrations that wereroughly an order of magnitude greater thanthose observed during the previous year.These were accompanied somewhat later byrelatively smaller increases in theconcentrations of 60Co and 10°Ru (Figure 1.7).The reasons for this rise in radionuclideconcentrations in the East Swamp Streamhave not been determined, and samplingresults since 1989 have provided noindications of appreciable changes in theconductivity or pH of the dischargedwastewater. Although these results do notinclude analyses for complexing agents,groundwater sampling during the summer of1991 failed to detect any complexed forms of9°Sr. The presence of complexing agents,major shifts in groundwater pH, or stableelements that might compete withradionuclides for sorption sites on the aquifersands are expected to be observed at the EastSwamp Stream within a few months of thedischarge event, as indicated above. However,no such precursors to the enhancedcontaminant concentrations have beenidentified.

NUREG/CR-6627 110

Figure 1.5 Water table contours in the vicinity of the Chemical Pit

16.737.000 N

I1I 11 NTJREG/CR-6627

C)

0\0\

-J

250

200

C

Cu

I..

U

CU

'.4

U,

150

100

10,000

8,000

0

4.0004

2,000 •"

i0

10

W1

50

00

MOO

50

01965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993

70

60

I.bJ

0)

0

0U

CU

50

40

30

20

C-04SRu4106

CD-a-J

I1~

-aCDCo,1

CD

0C)

CD'1

e.g.a-.0

CA

0a

0

e.g.

e.g.

CD

L~jCAeg.cn

.1U)

I0~

10

01974 1976 1978 1980 1982 1984 1986 1988 1990 1992

It may be that the increases in radionuclideconcentrations at the East Swamp Streamare the result of some much earlierperturbation in the Chemical Pit influent.The effects of such changes might not beevident in the discharge zone for manyyears if radionuclide migration rates wereretarded by reversible or partiallyreversible sorption reactions. Records fromthe analyses of disposed waters show thatthe release ofSr to the Chemical Pit in 1978 wasapproximately three times larger thanreleases in previous and subsequent years.A comparison between this 12-year delay(from the time of the 90Sr release to theobserved increase at the East SwampStream) and the 3-month delay observed inthe 1961 event leads to a 9°Sr Kd of around12 mL/g. This is in reasonable agreementwith 9°Sr radioactivity distributioncoefficients observed in other CRL aquifers(e.g., Robertson et al. 1987), making ittenable to propose that the observedincrease in East Swamp Streamradiostrontium in 1990 is a direct result ofthe discharges in 1978. This hypothesisdoes not address the accompanyingincreases in 60Co and 1°6Ru, yet theseenhancements fall within the ranges ofvariability observed for these isotopes overthe past decade. According to Champ et al.(1985), the total inventory of 60Co on theaquifer matrix downgradient of theChemical Pit was 10 Ci, while the quantityof alpha emitters disposed of at this sitewas estimated to be 5 Ci.

Earlier investigations (Champ et al. 1985;Killey et al. 1984) identified radioisotopesof Co, Zr, Ru, Sb, Cs, Ce, Eu, Fe, Sr, Ni, I,Tc, Pu, Am, and Cm in downgradient watersamples. Nonetheless, routine monitoringof Perch Lake, which receives drainagefrom this waste pit and other nearbywaste-management areas, indicates thatradionuclide concentrations do not exceedlevels permissible for drinking water(Killey et al. 1984). The only exception is90Sr, which is present at Perch Creek Weirat an average concentration of 160 pCi/t,compared to the present drinking waterstandard of 135 pCile. (Cooper, 1999)

1.2.2 Area C Solid WasteDisposal Site

Waste Management Area C (Figures 1.1and 1.8-1.10) includes a series of unlinedtrenches excavated from a large sand ridgethat forms the southern margin of a nearbyintermittent lake (Lake 233). This facilityis still in operation and has receivedlow-level radioactive wastes (paper, plastic,wood, metal, glass, liquid scintillationcocktails, organic and inorganic chemicals,etc.) from CRL, industry, hospitals, anduniversities throughout Canada since 1963.Until 1983, wastes were placed in trenchesapproximately 5 m deep (above the watertable), 7 m wide, and 60 m long, orientedparallel to the short axis of the rectangularsite. After being filled with low-levelradioactive waste, the trenches werecovered with about 1 m of local sand andsandy topsoil. Since 1983, wastes havebeen placed within a large trench andcovered with sand as the face of the filladvances. The southern portion of the sitehas been covered with a high-densitypolyethylene membrane and a tile drainagesystem to minimize precipitationinfiltration (Figure 1.8).

Terrain to the northeast of Lake 233 isprimarily bedrock with a thin, sandy tillcover (see Figure 1.9). There isintermittent overland runoff from this areainto Lake 233, which may contain up to 1m of water in spring, but is normallyalmost dry in late summer. This lake hasno surface drainage channel; thereforerunoff that enters the lake from thenortheast (as well as direct precipitationthat exceeds evapotranspiration losses)drains by infiltration into the bed of thelake.

Apart from a basal wedge of bouldery till,the unconsolidated sediments in Area Cconsist of fine and medium-fine sands offluvial origin overlain by a unit ofinterstratified, very-fine to fine sands andsandy silts. This interstratified unit isoverlain by fine and medium-fine sandsthat make up the dune ridge hosting AreaC. A unit of very extensive, continuous,

NUREG/CR-6627 14

laminated clayey silts (generally less than30 cm in thickness) is present along themargin of and beneath the wetland locatedabout 250 m soujhwest of thewaste-management facility. This wetland,known as Duke Swamp, contains up to 3 m

of peaty organics. The southwest boundaryof this wetland, as well as that of thegroundwater flow system itself, is abedrock ridge that trends roughlynorthwest to southeast.

Table 1.1 Major Ion Chemistry in the Chemical Pit Vicinity in 1984 (from Killey andMunch, 1984). Refer to Figure 1.3 for Well Locations. Major ion loading was

rather low and total dissolved solids in the contaminant plume were not

substantially different from local uncontaminated groundwater.

WeU NoJ pH Eh Ca Mg Na K Fe CI" S0 42 - HCO3 " F1 N0 3"

bepth (mV)

Concentration (mg/L

Es-911.o 5.98 130 3.7 0.85 2.2 1.34 6.3 0.8 5.7 15.4 0.21 0.0

ES-9/2.o 5.78 260 6.3 1.66 2.5 1.20 1.6 0.8 18.5 10.4 0.11 0.5

ES-913.0 5.85 100 5.5 1.87 2.6 1.26 2.5 0.7 8.8 21.2 0.13 0.0

ES-12/1.0 5.81 270 7.9 4.10 4.4 1.27 0.8 1.5 14.6 28.6 0.21 0.4

ES-12/1.5 6.40 160 21.2 6.70 4.5 2.23 1.1 1.0 11.3 84.8 0.17 0.3

ES-12/2.0 6.61 100 10.0 6.00 3.9 1.85 21.0 1.0 0.1 116.0 0.17 0.0

ES•131.0 5.82 80 7.4 3.40 4.3 1.10 4.2 2.5 16.2 23.6 0.23 0.0

E•-13/2.0 5.87 70 12.5 4.80 4.6 1.60 6.2 2.9 11.0 41.7 0.29 0.0

ES-51o.6 5.15 660 2.4 0.70 3.0 1.15 0.1 2.3 9.2 10.0 0.25 1.0

ES-511.8 4.60 470 1.5 0.79 6.3 1.00 0.1 1.5 14.7 41.0 0.10 4.4

ES-5/2.4 5.70 330 3.7 2.30 3.8 1.65 3.0 0.5 12.5 21.8 0.20 0.0

ES-1411.0 7.06 210 17.1 4.90 4.1 2.60 0.3 9.6 12.8 80.3 0.18 0.7

ES.•14a.0 7.32 230 31.2 9.00 3.0 1.00 0.1 13.5 28.0 97.5 0.13 4.0

ES-15/1.0 5.40 310 0.9 0.29 8.0 0.80 0.4 7.7 6.3 14.3 0.09 0.0

ES-15/2.0 5.50 310 4.4 0.81 11.1 1.90 0.3 9.3 5.8 24.9 0.25 0.0

ES•-52.5 5.15 280 5.1 1.07 12.0 1.98 1.3 12.3 7.0 28.8 0.07 0.0

ES-1/1.06 5.40 650 0.7 0.52 13.0 0.86 0.3 12.8 10.7 14.5 0.15 3.0

ES-16/1.5 5.65 640 0.4 0.47 16.8 0.55 0.3 16.0 8.1 19.9 0.00 5.8

ES-16/2.o 6.28 540 1.0 4.40 16.2 0.49 0.5 16.4 11.4 83.4 0.39 0.0

5.16/2.5 5.50 560 2.0 0.55 16.6 0.84 0.1 12.5 9.2 40.8 0.15 0.0

ES-17/1.0 4.75 360 1.7 0.49 7.3 1.04 0.1 0.6 12.5 10.0 0.12 2.5

ES-17P2.0 5.10 300 2.4 0.67 5.4 1.14 0.9 0.6 14.7 6.1 0.05 0.0

As mentioned previously, much of therecharge from Lake 233 that passesbeneath Waste Management Area C flowssouthwest (see Figure 1.10) through theunconsolidated sediments, and directprecipitation infiltration through Area Cand downgradient sands is added to theaquifer en route. Groundwater from thisflowpath discharges to Duke Swamp and

subsequently drains by overland flowthrough Duke Stream to Maskinonge Lake.Groundwater flowrates range from 15 to 30cm/day, leading to a total residence time ofapproximately 2 to 4 years between Area Cand the wetlanid (Killey et al. 1993).Additionally, the groundwater flowdowngradient of Area C bifurcates (Figure1.10), such that those waters passing

15 NUREG/CR-6627

beneath the south end of the facility arediverted almost due south, flowing beneaththe site highway and discharging toanother wetland (which in turn drainsthrough Bulk Storage Stream). This

bifurcation is caused by a partially buriedbedrock ridge that trends north-northwestand outcrops just east of the southern endof Duke Swamp.

Figure 1.8 Borehole and cross-section locations near Waste Management Area C

10

IN &2

K1 1069 227

105. 22

*163

164

1,027.000 1.029,000

I9, C-Series plezometer nest Waste Mgmt Area C, Sand Cover

Waste Mgmt Area C, Poly Cover8 A TL-Serles piezometer nest

- Cross Section

I I I I I I II I I I I I ¢

0 300 m

NUREGICR-6627 116

LZ99

-"H

OM

UM

R

LLI! E

leva

tion

(m

asl

)-J U

'

UL~

I ~I[I

ii' ~II

'Si'

Ii I K I

'~1 iii

a C C

I11

-91

eantj

ui

g-q

ola

j-

-ra

9 8jx~

qos~

pj~

jw~

puno

zi2

drun

mS

931

n(Ia

q put,

0 ý

axt~

qnoa

zq

uoio

s ox

~qdr

2IV

.ItuS

6-

1 m

~

Figure 1.10 Surface hydrology, water table contours, and areal extent of groundwatercontamination by 3H (bifurcating, medium-gray pattern) in the vicinity ofArea C. Dots indicate piezometer nest locations (refer to Figure 1.8 for nestidentifiers).

NUJREGICR-6627 118

Figure 1.11 Annual flux of 3H and GOCo through the Duke Stream weir

60

:so

10

0

160

140

• 120

1¶00

080•Q .

-"60

~40

20

0

1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991

_ _ _~~ ~~ -. - . - --------. -

1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991

19 19 NUREG/CR-6627

1.2.3 Previous Related Chemicaland Radiological Characterization

Some groundwater outside of the Area Cplume is affected by salt used for roadde-icing, but is otherwise uncontaminated.Table 1. 2 summarizes the major ionchemistry for these surrounding groundwaterand for samples collected from the Area Cplume. Boreholes C-112 and C-212 (refer toFigure 1.8 for locations) were sampled foranthropogenic organic compounds in 1991,and Table 1.3 lists the compounds that weredetected.

Tritium is the primary radionuclide of concernat this site in terms of quantity, although 14Cand 60Co have also been detected indowngradient water and vegetation. Area Cbegan releasing 3 H to the aquifer before 1973when routine monitoring of local surfacewaters was initiated (see Figure 1.11).Tritium in water downgradient of the coveredportion of Area C has virtually disappeared,however, since the fall of 1985 (Killey andMunch 1993), and this can be attributed to theimpermeable polyethylene cover installed in1983. Although this barrier appears to haveeliminated the infiltration of precipitation andconsequent release of tritiated water from thislocale, 14C apparently is still migrating fromthe area. Tritium releases continue from thatportion of the site that has no cover, butanalyses from the streams that drain thegroundwater discharge area that receive theArea C water has shown a declining trend in3H releases since 1983.

When regular measurements of Duke Streamwere initiated in 1973, low concentrations of6 0 Co were detected (Figure 1.11). The 60Coflux peaked at 3.6 mCi per annum in 1977 andhas decreased in a roughly exponentialfashion since that time. At present, thisradionuclide is only marginally detectable inDuke Stream, exhibiting an annual flux of lessthan 81 gCi. No program currently exists for

mapping 6 0 Co distribution in thecontaminated aquifer or the groundwaterdischar: e area.

Surveys in 1982 and 1983 detected elevated

14C in vegetation growing in and adjacent to

the Area C compound with specific activitiesapproaching 1.1 nCi/gC. In 1990, a study ofsoil gas and groundwater around Area Cfound specific activities up to 1.3 Ci/gC insoil C02 above the waste trenches and 14C

concentrations in downgradient waters of upto 34 nCi/L (Killey and Mattie 1993). Furtherstudies from 1991 to 1993 addressed thedistribution of 14C in dissolved inorganic andorganic forms (Killey et al. 1993). Dissolvedorganic carbon (DOC) was generally found toaccount for less than 10% of the dissolved 14C

in these samples (Table 1.4). This is becausethe DOC in the ground and surface water isless abundant than the dissolved inorganiccarbon (DIC), and the specific activity of theformer is generally lower than that of thelatter. The variability in the distribution of14C between the two classes of compounds isattributed to variability in the source area.Although these data suggest that most of the14C inventory in Area C is in inorganic forms,a portion of the dissolved inorganic 14C most

likely originates from C02 produced duringthe bacterial degradation of radiocarbon-bearing organic materials. The DIC and DOCconcentrations in the Area C plume watersrange from background levels to about 3 and 8times background, respectively.

Recent studies of 14C in the environs of DukeSwamp have characterized the transport andbiological uptake from groundwater-to-air-to-vegetation (Link, et al., 1998; Milton, et al.,1997). These studies have shown that thepredominant pathway for 14C uptake byvegetation growing in Duke Swamp is via thegroundwater-to-air-to-leaf route, and uptakeof 14C by transpiration through the rootsystem is essentially negligible.

NUREG/CR-6627 20

Table 1.2 Major ion chemistry within and adjacent to the Area C contaminatedgroundwater plume. Refer to Figure 1.8 for well locations.

Well pH Eh Ca Mg Na K Fe SiO2 HC0 3" C1" S042" F"

No./ (mV)Depth

Concentration (ra/m

1/10.4 6.80 164 16.3 5.3 3.8 1.1 29.0 33.8 93.8 47.0 0.9

1/5.4 5.70 235 5.5 2.2 21.2 1.3 6.0 12.8 25.9 54.0 5.5 --

2/9.2 6.35 82 14.2 5.0 3.7 1.1 6.5 28.9 35.0 5.6 16.3 --

2/2.5 5.80 284 6.6 2.5 7.0 2.2 0.1 13.1 18.1 9.2 9.9 --

3/11.0 7.10 44 15.3 6.7 3.1 1.3 8.6 30.4 11.6 3.4 --..

3/4.1 5.50 254 8.2 2.1 4.1 1.3 0.3 13.5 24.5 7.0 10.6 --

4/9.4 6.65 124 8.0 3.2 7.4 1.0 19.1 14.6 44.0 13.4 18.5 0.1

7/15.6 6.50 119 8.9 2.8 5.9 2.3 12.2 20.3 48.1 9.6 12.4 --

7/11.8 6.20 176 10.8 3.2 10.1 1.8 11.8 11.1 30.8 34.0 10.3 0.1

8/14.5 6.84 90 8.3 2.8 5.6 1.6 17.9 21.0 79.3 7.3 11.0 --

8/14.5 6.41 154 9.2 2.8 7.2 1.2 15.8 17.6 46.3 17.4 11.9 0.1

8/10.3 6.32 265 12.6 3.6 5.9 1.1 0.3 14.8 35.8 11.3 12.8 0.0

8/8.1 6.60 233 5.0 1.3 5.7 1.9 0.1 15.4 26.3 2.0 10.6 --

8/8.1 6.30 294 11.0 2.7 12.8 1.3 0.1 10.7 44.9 11.1 12.4 0.0

9/14.1 6.30 239 17.9 7.1 16.7 1.7 0.7 12.0 93.0 15.5 20.0 --

9/11.1 6.00 374 32.1 9.1 37.1 2.9 0.5 19.1 77.1 47.0 41.4

9/11.1 6.04 334 26.0 7.6 19.3 2.3 0.8 11.1 93.0 36.0 30.0 --

9/7.2 5.90 334 3.8 1.3 0.8 1.3 0.1 12.4 10.9 0.3 10.1 --

9/7.2 6.14 327 20.4 6.0 15.3 2.0 0.8 11.8 9.1 47.5 33.7 --

10/4.8 6.70 114 8.5 2.7 5.7 1.0 17.2 21.4 62.1 12.5 5.4 0.2

10/2.6 6.32 274 3.6 1.3 3.4 1.4 0.7 13.1 16.3 2.8 9.0 -

14/9.5 6.30 219 8.0 2.2 8.1 1.5 3.7 11.1 24.9 18.7 13.2 0.1

14/7.1 6.15 243 9.5 2.7 9.0 1.6 1.8 10.9 24.7 17.6 18.2 0.1

14/10.5 6.25 214 9.5 2.8 8.4 1.4 3.2 13.3 28.6 16.2 13.2 0.0

15/13.3 6.75 120 13.0 4.1 5.0 1.6 6.5 27.9 64.8 11.8 12.5 0.1

15/8.8 7.25 116 11.8 3.2 9.3 9.4 0.4 24.8 72.1 7.5 14.7 --

15/8.8 7.05 74 7.8 1.5 3.6 1.6 0.8 22.3 40.8 0.8 7.5 0.1

16/14.1 6.70 134 15.0 4.0 5.1 1.4 9.0 34.7 66.7 9.5 12.8 0.1

23/11.6 6.99 74 4.9 2.8 5.8 1.3 20.9 12.8 51.2 11.9 5.1 0.1

24/12.8 6.60 123 7.6 3.5 6.3 1.4 14.6 13.7 35.8 20.3 31.1 0.1

24/10.5 6.55 139 6.6 3.5 6.6 1.7 8.4 13.9 29.9 16.2 13.7 0.1

24/7.8 7.00 86 7.8 2.9 6.6 3.7 2.0 14.1 29.9 13.0 12.2 0.1

33/8.8 5.90 266 18.0 5.8 9.5 1.9 10.2 17.1 27.2 59.0 16.9 0.1

33/4.0 6.95 15 10.4 3.5 2.7 2.0 7.6 19.6 34.5 5.7 1.3 0.1

34/9.5 6.59 191 6.5 2.1 7.3 1.3 6.0 18.4 32.2 16.8 11.3 0.2

34/6.6 6.28 194 4.4 1.7 2.4 0.8 6.9 18.2 32.2 5.7 1.3 0.1

36/9.0 7.70 -104 10.2 3.4 9.4 2.3 2.9 10.1 66.2 6.9 4.5 0.7

37/11.0 6.93 169 7.6 3.3 10.6 1.9 3.4 37.2 51.7 5.5 3.4 0.3

Lake 233 -- -- 4.2 1.6 21.3 0.6 0.2 -- 3.9 41.2 3.1 0.0

21 NUREG/CR-6627

Table 1.3. Anthropogenic organics from Area C plume in units of Irg/L

Organic Compound C-112 (11 m to 12.5 m) C-212 (10 m)

1,1-Dichloroethane 32.0 55.9

1,2-Dichloroethane 4.0 7.4

Carbon Tetrachloride 4.6 1.2

Chloroform 39.4 6.0

Tetrachloroethene 4.4 1.2

Trichloroethene 409.0 182.0

trans- 1,2-Dichloroethene 1.4 257.0

Benzene 1.7

0-Xylene _ 1.6

Naphthalene 16.5

Bis(2-ethylhexyl)phthalate 3.5 9.0

Table 1.4 Carbon and 1 4 C concentrations in Area C waters. Refer to Figure 1.8 for welllocations.

Well Carbon Concentration 14 C Concentration Specific Activity

No.

DIC (mgfL) DOC (mg/L) DIC DOC (nCi/L) DIC DOC(nCifL) (nCi/gC) (nCi/gC)

104 13.82 0.83 1.1 0.1 78.9 147.2

105 12.10 2.45 2.8 0.7 232.5 294.1

112 46.65 19.76 7.0 1.0 150.3 50.5

114 44.47 7.84 4.6 0.9 103.1 111.5

211 37.78 24.7 5.5 3.3 5390 134.7

212 38.10 7.25 2.9 0.0 76.7 6.1

213 52.96 9.12 28.6 0.7 540.5 78.0

214 29.86 7.78 8.7 0.8 290.2 97.6

216 23.22 5.29 8.3 0.1 358.5 22.9

221 30.58 5.58 1.6 0.1 52.5 14.1

222 55.30 21.6 1.9 0.3 34.6 11.4

Duke 7.66 3.57 0.1 0.0 17.9 4.5

Stream

Bulk Storage 7.77 1.97 0.1 0.3 9.3 159.1

Stream I

NUR.EG/CR-6627 222

2 Experimental

The general intent of the experimental phasewas to identify and quantify as many of themobile radionuclide species as practical. Thisprocess is inherently difficult due to theinvariably low radionuclide concentrations,the potential instability of the complexedspecies, the nature of their kinematics,changing physico-chemical conditions in thespatial and temporal domains, etc. Towardthis end, a number of diverse procedures andmethodologies were used to study thespeciation of migrating radionuclides: 1) fieldsampling of groundwaters and subsurficialsands, 2) radioanalytical techniques, 3) wetchemistry, 4) ion exchange, 5) hollow-fiber andmembrane ultrafiltration, 6) ion-exchangechromatography, and 7) liquidchromatography.

2.1 Observation Wells

Wells were drilled at Area C and the ChemicalPit using a rotary drilling rig. When soilsampling and multi-level piezometerinstallation were the objectives of theborehole, 9.5-cm-ID hollow-stem augers wereused. When installing the groundwatersampling wells, 15.2-cm-ID hollow-stemaugers were employed. Portable equipmentwas deployed in the wetland down-gradient ofthe Chemical Pit, and a gas-poweredjackhammer for driving the flush-joint casingwas used. While drilling, the casings wererinsed with water at a flow rate ofapproximately 50 L/min.

Cores of soil were collected with a fixed-pistonsampler fitted with thin-wall, 5-cm-IDaluminum tubing. Core lengths were either0.76 m or 1.52 m, and the soil samples werecollected ahead of the advancing borehole.Cores were checked for beta and gammaactivities as they were retrieved, logged forvisual features, and subsampled. Thesesubsamples (except for soils collected fromwell CP-4B) were subsequently oven dried andarchived. Following a second screen for betaand gamma activities in the dried

samples, composites from the contaminated portionsof CP-4, ES-39, and ES-16 were prepared.Individual samples were subdividedusing a mechanical splitter, and appropriatequantities of these subsamples were combined.

Sampling wells used to collect groundwaters forspeciation analyses were installed at boreholelocations CP-4, ES-39, C-112, and C-213 at depthsselected to correspond with the maximumconcentrations of radionuclides as defined byprevious soil cores, borehole gamma scans, andgroundwater samples from multilevel piezometers.The wells are constructed of 10-cm-ID, 1-m-long,stainless-steel well screens (0.15-mm slot width)attached to 10-cm-ID stainless-steel pipe thatextends about 1 m above the water table. Theremainder of the well casing consists of 10-cm-IDSchedule 40 PVC pipe. The local sands, which haveno cohesive strength below the water table, slumpedaround the well screens as soon as the augers werewithdrawn. Above the water table, the annularspace around the well casing was backfilled withclean local sand, and a 0.6-m seal of powderedbentonite was installed just below grade. The well atES-16 consists of a 7.6-cm-ID, 0.91-m-long stainless-steel screen with a 7.6-cm-ID Schedule 40 PVCcasing to the surface.

2.2 Groundwater Sampling and Analyses

2.2.1 Sampling

All of the sampling wells were developed afterinstallation. Initially, the wells were pumped todryness repeatedly (with the exception of C-112,which sustained pumping at a rate of 40 L/min) andthen pumped for a minimum of 1 hour at asustainable rate (>12 L/min) within 5 days of samplecollection. A 4.6-cm-OD stainless-steel, variable-speed, submersible pump or a bellows-type meteringpump with polypropylene wetted parts was used fordrawing groundwater samples.

All water samples were collected after removing atleast three column volumes of water or afterpumping the well to dryness twice. Sample flowrates were maintained as low as possible to precluderesuspension of any microparticulates retained bythe aquifer medium (generally less than 100mL/min). During this process, groundwater

23 23 NUREG/CR-6627

temperature, pH and Eh, dissolved oxygen,alkalinity, and Fe2 ÷/Fe3÷ ratios weredetermined. Aliquots were also drawn foranalyses of trace metals (inductively-coupledplasma/atomic-emission spectrometry), anions(ion chromatography), and organic carboncontent (persulfate/ultraviolet-oxidationanalysis). Samples were pressure filteredthrough sterile 0.2gm or 0.45-gm membranesat the wellhead or suction-filtered through0.45-11m membranes in a CRL laboratorywithin 4 hours of sample collection. Thepurpose of this prefiltration step was to retainany undissolved or particulate matter. Waterswere maintained on ice during transit andstorage to hinder kinematic processes andmitigate against any physico-chemicalchanges to the dissolved species.

A number of in-line prefiltered groundwaterswere passed through a large-volume watersampler (Robertson and Perkins 1975) tofacilitate the removal of charged anduncharged soluble species. Generally, about200 L of groundwater were sampled at eachwell through this system. This modular unitis composed of six stacked resin or sorbentchambers (2.5 cm thick by 20 cm in diameter)separated by glass-fiber filter sheets. Waterenters the base of the unit and is directed tothe first of two cation-exchange resin beds(Na+ form, 200-400 mesh), the redundant bedbeing present in case the first becomescompromised or saturated with ions. Thisresin was originally in the H÷ form, but wasconverted to the Na÷ form to eliminate thepotentially destructive or denaturing effectson existing radionuclide species that mighthave been caused by the resulting acidicsolution. Following the cation beds areduplicate layers of anion-exchange resin (Cl-form, 200-400 mesh) for the retention ofnegatively charged species. Finally, a pair ofactivated aluminum-oxide sorbent beds areused to remove those radionuclides that existin soluble, non-ionic forms.

2.2.2 Radiometric Analyses

Following this ion-exchange groundwater-sampling procedure, the individual resin andsorbent beds (as well as the prefilters) wererinsed with about 1 L of clean groundwater,de-watered, disassembled, and packaged for

shipment and radiological characterization.Gamma-ray spectrometry was performed by placingthe prefilters (or their component membranes),resins, and sorbent powders in standardizedgeometries and counting them with lithium-driftedor intrinsic germanium detectors for suitable periodsof time. Aliquots of the above materials were thenleached with appropriate acid mixtures, and theresulting solutions were subjected to variousradiochemical separation procedures for quantifyingradioisotopes that were not amenable to directgamma-ray analyses.

2.3 Ultra-filtration Experiments

A series of ultra-filtration separations wereconducted on groundwater samples collected down-gradient from the Chemical Pit to determine the sizedistribution and the radionuclide association ofcolloidal-sized particles present in the groundwaterat this site.

2.3.1 Colloid Sampling and Analysis

Samples of groundwater were collected from CP4,ES39 and ES16 (Figures 1.3 and 1.4) during theperiod August 9-11, 1993. The wells were locatedalong a flowline from the pit to the nearby wetland.Well CP4 was just outside the fence around the pit,while ES16 was down in the wetland and ES39 wasin between. Low sampling flow rates (0.25 to 1L/min) were used to minimize resuspension ofsediments in the boreholes during sampling.

In order to collect enough material for colloidcharacterization, Millipore tangential-flow ultra-filtration systems were used to produce particleconcentrates (2 to 6 L) from 100 L groundwatersamples. These concentrates contained enoughmaterial for colloid concentration and size analysis.Chemical and radiochemical analysis of particleconcentrates and fitered water provided informationon the association of various elements andradionuclides with colloids and suspended particles.The groundwater was not pre-filtered before ultra-filtration to avoid loss of colloids and suspendedparticles on the pre-filter. The filtration scheme isshown in Figure 2.1. In the first step particles largerthan about 10 nm were concentrated by filteringgroundwater through a Pellicon system equippedwith 100,000 nominal molecular weight limit(NMWL) polysulfone membrane packets, separatedby retentate

NUREG/CR-6627 24

screens. The concentrate (retentate) producedby the 100,000 NMWL filter was filtered with aMinitan tnagential flow filtration system(Millipore) equipped with a 450-nm cutoffDurapore membrane. This produced a filtratecontaining colloids between 10 and 450 nm, anda concentrate enriched in suspended particles>450 nm. About twenty liters of filtrateproduced by the initial 100,000-NMVL filtrationwere filtered through another Pellicon system,equipped with a 10,000-NMWL polysulfonefilter cassette. This produced a sample offitered water containing only dissolved species,and a concentrate enriched in colloids with asize range between 10,000 NMVL (-1 nm) and100,000 NMWL (- 10 rm).

Particle concentrations and size distributionswere measured by pressure-fitering 25 mL ofparticle concentrates through a series of 25-mmNuclepore polycarbonate filters with cutoff sizesof 10,000, 5000, 1000, 400, 100, 50 and 15 nm.Particle concentrations were determined by theweight of material deposited on the fiters. Thepresence of colloidal material in the 1- to 10-nmsize range was determined by comparing thechemical composition of colloid concentrateswith that of fitered water. Particle sizeanalysis was also carried out with an UltrafineParticle Size-Analyzer (UPA-Leeds andNorthrup), which used backscattered laser lightto measure the Brownian motion of particleswith sizes between 6 nm and 2.5 mm. Brownianmotion was used to calculate particle sizedistributions.

A portion of fitrate with size < 10,000 NMWLfrom each sample was retained for speciation byanion exchange. These were kept covered in acoldroom until the separations could beperformed.

2.3.2 Anion Exchange Chromatographyof Ultra-Filtered Groundwater

To further characterize the ultra-filteredgroundwater, anion exchange chromatographicseparations were conducted on the 10,000NMWL fraction to help identify thephysicochemical speciation of the radionuclides.

An Alcott model 760 piump was used to load thesamples and chromatographically separate thespecies. A Scanivalve rotary valve (modelKEM5 12) was used to select the various eluLntsolutions, which were stored in 4 L polyethylenebottles. Salt gradients were generated using aMilton-Roy model 196-0066-003 piston pumpand an acrylic mixing chamber on a magneticstirrer. The conductivity of the eluent solutionswas monitored downstream of the Alcott pumpwith a conductivity flow cell and meter salvagedfrom a Dionex ion chromatograph. The outputsignal from the conductivity meter was recordedwith a UNEAR model 0142-0000 strip chartrecorder. A 10 m length of Teflon capillarytubing (0.020 inch I.D.) between the Alcottpump and the conductivity cell provided back-pressure for the pump.

Chromatographic columns were constructedfrom 0.25 inch I.D. polyethylene tubing and 3/8inch stainless steel Swagelok fittings, whichwere adapted to chromatography fittings. Thecolumns were slurry packed and flowconditioned at 10 mL/minute prior to use.Further details of the columns and eluents aregiven in Table 2.1. Gamma-emittingradionuclides in the effluent from the analyticalcolumn were monitored online with a custom-built flow cell and a 7.6 cm by 7.6 cm well-typeNaI detector (Harshaw). The gamma-rayspectra from the detector were recorded with aDavidson model 1056C multichannel analyzer(MCA) and the data from regions of interest inthe spectra were transferred to a personalcomputer and printed out. The on-line flow cellprovided qualitative results which were onlyused to monitor the elution of 6 0Co during theruns. Fractions of effluent (nominally 6 mL)were collected with an LKB model 2070ULTRORAC II programmable fraction collector.A signal from the fraction collector was used tocontrol the MCA which read out the data andstarted a new accumulation at each change to anew fraction. Teflon tubing was used for all ofthe interconnections in the system.

25 NUREG/CR-6627

Figure 2.1 Water filtration scheme for ultra-filtration and chromatography studies

WATER FILTRATION SCHEME

1F21

DissolvedSpecies(<1 rm)

R2 F R3 I

Enriched inColloids

(1-10 frm)

Enriched inColloids

(10-450 nm)

Enriched inParticles

(>450 mn)

NUREG/CR-6627 26

Table 2.1 Experimental conditions for AG MP-1 anion exchange chromatography separations of ultra-filtered groundwater

Resin Mesh Size

Pre-Column Length

Analytical Column Length

Nominal Flow Rate

Sample Volume

Eluents:

pH 4.5 HCl

0-0.5 M KCI gradient

0.5 M KCI

0.1 M HCI

200 - 400

7 cm

25 cm

6 mLimin

3 to -6 L

30 mL

500 mL

375 mL

200 mL

The general procedure for the separationsusing AG MP-1 resin was to transfer aportion of sample to an opaque container,weigh it and load it onto the columnsovernight. The sample container was re-weighed after loadingto determine the amount loaded and the flowrate. The elutions with pH 4.5 HC1 (i.e.

double distilled water adjusted to pH 4.5with HU1), 0 to 0.5 M KM1 gradient, 0.5 MKC1 and 0.1 M HC1 were carried out the nextday. Each elution took most of a workingday. The samples and eluents were de-gassed before each run by sparging with Hegas for about 5 minutes.

2.3.3 Radionuclide Analysis of Ultra-Filtered Groundwater

Gamma-ray spectrometry was used toidentify the radionuclides present in theeffluent fractions from the anionchromatographic separations.

Quantitative results for 60Co and 106Ru wereobtained by counting the fractions in 7 mLscintillation vials with a Packard MINAXI&automatic gamma counter. Selected

samples were analysed on a gamma-rayspectrometer consisting of a 30% EG&G Gedetector and ND 6700 analyzer. Largerliquid samples were either counted in a 400g geometry in a polyethylene bottle or driedon polyethylene film and pressed into plasticpiliboxes. Alpha spectrometry wasperformed with Nucleus model 5300 orEG&G Ortec model 676 alpha spectrometersconnected to an EG&G model 920-16multichannel buffer. The counters andspectrometers were calibrated withstandards traceable to NIST.

Actinide isotopes present in selectedfractions were radiochemically separated asa group. The fractions were transferred toglass beakers and 10 mL of nitric acid and 1mL of 5% NaHSO4 solution were added. Thefractions were then evaporated to drynessand taken up in 1 M HC1 (10-15 mL). Thesolutions were transferred to polypropylenecentrifuge tubes. One mL of Nd carriersolution (100 jig of Nd/mL) was added priorto reducing the actinides with TiClb (30%solution). Then the actinides were co-precipitated on NdF3 by adding concentratedHF and sources were prepared by filteringthrough NdF3 beds supported on 0.1 Ammembrane filters (Gelman Suporg). The

27 NUREG/CR-6627

filters were dried prior to analysis by alphaspectrometry.

Total actinide sources were also preparedfrom the filtrates and retentates obtainedafter ultra-filtration. Approximately 500 mLof each sample were acidified with about 50mL of concentrated nitric acid. One mL of5% NaHSO4 solution was added to each andthen they were boiled down to about 20 mL.Each sample was then diluted to about 200mL and 10 mg of Fe+3 carrier was added.Ferric hydroxide was precipitated by raisingthe pH to about 9 with NH4OH. The ferrichydroxide was then dissolved in a minimumvolume of concentrated HC1 and the volumewas made up to 10 to 15 mL with 1 M HCl.Actinide sources were then prepared byco-precipitation on NdF3 in the same manneras described above for the fractions.

2.4 LaboratoryChromatographyExperiments

A series of laboratory anion chromatographyexperiments using 0.45 fln filteredgroundwater from well ES16 were conductedto separate and identify individualphysicochemical forms of the migratingradionuclides present in anionic forms.

2.4.1 Field SamplingThe samples were pumped from well ES16(see Fig. 1.3) with a peristaltic pump andfiltered off-line through a 0.45 pim membranefilter. The fitered groundwater wascollected in 20-L polyethylene carboys, whichwere covered to exclude light and stored in acoldroom until the experiments wereperformed. Generally, the first experimentwith a given sample was started on the dayof collection or on the following day.Subsequent experiments with each samplewere performed over periods ranging up toseveral weeks.

2.4.2 Chromatographic Equipment

The chromatographic equipment andmethodology for ion exchangechromatography is similar to that describedin Section 2.3.2 and is shown schematicallyin Figure 2.2. The Alcott model 760 micro-processor controlled dual piston pump wasprogrammed to pump at 3 mL/min. AScanivalve rotary valve (model KEM5 12)was used to connect the inlet of the pump tovarious eluent solutions, which were storedin 4-L polyethylene bottles, and to thegradient programmer. Salt gradients weregenerated with an ISCO model 2360gradient programmer.

This programmer can generate ternarygradients and it was used to produce bothKC1 and HCl gradients. The conductivity ofthe eluent solutions was monitoreddownstream of the Alcott pump with aconductivity flow cell and meter from aDionex ion chromatograph. The outputsignal from the conductivity meter wasconnected to one of the detector inputs on aDionex Advanced Computer Interface, whichwas used to record data and to control theScanivalve and gradient programmer. A 10-m length of Teflonco capillary tubing (0.020inch I.D.) between the Alcott pump and theconductivity cell provided back-pressure forthe pump. Online UV measurements at 254nm were made with a Shimadzu model SPD-10A UV detector. The signal from the UVdetector was connected to the seconddetector input on the Advanced ComputerInterface.

NUREG/CR-6627 228

Figure 2.2 Chromatography equipment used for anion separations of organo-radionuclide species

The Advanced Computer Interface wasconnected to a PC and controlled withDionex software (AI-450). Fractions ofeffluent (6 mL) were collected with an LKBmodel 2070 ULTRORAC II programmablefraction collector, which was programmed tocollect fractions for 2 minutes.

Chromatographic columns were constructedfrom 0.25 inch I.D. polyethylene tubing andchromatography fittings from spent Dionexcolumns. Porous polyethylene frits wereused to retain the resin at the ends of thecolumns. Teflon tubing was used for all ofthe interconnections in the system.

29 29 NUREGICR-6627

2.4.3 Preparation of AG MP-1 Columns

The procedure used to prepare the guard (7cm) and separation (25 cm) columns is givenin Appendix 1. After preparation, thecolumnswere connected to the rest of the system andcovered with aluminum foil to exclude light.The columns were kept dark becauseprevious studies (Killey et al., 1984) hadshown that some complexes were lightsensitive.

2.4.4 Procedure for Loading andEluting AG MP-1 Columns

The procedure used to load the samples ontothe columns and elute the species is given inAppendix 2. Two separate elution runs wereconducted, an initial run on May 7, 1996,and a second run on May 15, 1996. A slightmodification of this procedure was used forthe run on May 7, when the 0.1 M HC1elution was terminated after only 50minutes.

2.4.5 Radionuclide Analysis

Cobalt-60 in the column effluent fractionswas measured in 7 mL scintillation vialswith a Packard MINAXTy automatic gammacounter. The 60Co results were normalizedto the weight of groundwater loaded onto thecolumns. Selected samples from the May 15run were analysed on a gamma-rayspectrometer consisting of a 20% CanberraGe detector connected to a Canberra Series10 analyzer. Since quantitative results werenot needed for the chromatographs, nodetector calibrations were applied. Theanalyses on the Ge detector provided datafor 137Cs and better counting statistics for60Co.

2.4.6 Measurement of UV Spectra

The UV absorbance spectra of selectedfractions from the May 7 run were obtainedoff-line with the UV detector. Fractionswere selected on the basis of peaks observedat 254 rn in the UV chromatograms.

Samples from the fractions were manuallyloaded into the UV cell with a syringe. Thedetector was then set to scan mode and theabsorbance was recorded by the PC throughthe Advanced Computer Interface.

2.5 Laboratory Kd Experimentswith Anionic and CationicRadionuclide Species

Subsurface contaminant transport fromburied wastes is an environmental problemthat faces many sites. In the nuclearindustry, the primary contaminants ofconcern are radioisotopes, and due to thenature of the contaminants and the soils,some are quite mobile (i.e., tritiated water),but many exhibit various degrees ofretention in soils. Contaminant transportmodels used to determine the fate of theradionuclides in the environment, whenlicensing is sought for nuclear wastemanagement facilities, frequently simulatethe retardation of contaminants in soilsusing the solid-liquid partition coefficient Kd.It is generally defined as:

Kd = solidC liquid

(1)

where:

Kd = solid-liquid partition coefficient(volume/weight)

Csolid = concentration of contaminant sorbedon the solid at equilibrium(amount/weight of solid)

Cliquid = concentration of contaminant left insolution at equilibrium(amount/volume of solution).

Solid-liquid partition coefficients are lumpedparameters that may include severalprocesses, such as sorption onto a variety ofbinding sites, or the influences of aqueousspeciation, into one single value. Many ofthe Kd's in the databases that are frequentlyreferenced in tranport simulation models(e.g., Sheppard and Thibault, 1990; Caron,1995), are derived from laboratoryexperiments that may not completely

NUREG/CR-6627 330

represent conditions and reactions that takeplace in field settings. Contaminant plumesthat have been present in actual flowsystems have long been of interest as casestudies, where the adequacy of the KI modelcan be tested. The simple groundwater flowsystem adjacent to the Chemical Pit used forlow-level aqueous radioactive wastes at theChalk River Laboratories (CRL) has been avaluable field site for conductinggeochemical investigations.

Studies of this site (Cooper and McHugh,1983; Killey et al., 1984; Champ andRobertson, 1986; Cooper and Mattie, 1993;Robertson, et al., 1995; Schilk et al., 1996 ),have found that several radionuclides thatwould be expected to exist as cationic speciessorbing strongly to aquifer sediments haveformed stable, mobile complexes. Similarphenomena, leading to transport of actinidesover distances of several kilometres, werereported by Penrose et al. [1990]. Suchobservations raise obvious questions aboutthe validity of the laboratory-derived Kd'sthat are used in predictive models.

The objectives of this task were to: 1)determine Kd values of radionuclides in theaquifer down-gradient of the previouslymentioned infiltration pit at CRL as afunction of their field chemical speciation,and 2) to compare these measurements withthose obtained by the conventionallaboratory approach of "batch" Kddetermination using a tracer of the sameelement. In this manner, more realistic Kdvalues for the mobile anionic species ofseveral important radionuclides weredetermined. These values could then beused by transport modelers to moreaccurately predict "first arrival times" of themigrating radionucides.

2.5.1 Soil and Groundwater Samplingand Preparation for Kd Measurements

Samples of contaminated (ES-16) anduncontaminated (ES-9) groundwaters werecollected at two locations in the vicinity ofthe Chemical Pit (Figures 1.3 and 1.4). TheASTM Method D 4319-83 (ASTM, 1984) wasused to determine the solid-liquid partition

coefficient K&. The solution-to-solid ratiowas 4:1 (volume:weight basis). Theradionuclides of interest were present in thecontaminated ES16 groundwater (60Co,106Ru, 137Cs, Pu isotopes, 24'Am, 244Cm), andthese were used for the Kd measurementsfor the field chemical species. Commerciallyavailable tracers in known chemical formswere added to the solutions at specific pointsin the experimental sequence, as describedin the following sections.

The calculation of the Kd using the aboveradionuclides was given by the relationship:

Kd =CO - Ceq XVceq M

Where:

&d = Liquid-solid partition coefficient (mL/g)

C = Concentration of element (Bq/L orequivalent); subscripts 0 and eqcorrespond to initial and equilibrated,respectively

V = Volume of solution (mL)

M = Mass of solid in contact with thesolution (g).

Four sets of sorption experiments wereconducted. In the first two, samples of ES16contaminated groundwater were passedthrough columns of ion exchange resins toisolate the cationic and anionic species ofradionuclides already present in solution. Inthe other two sets, one aliquot of untreatedES9 uncontaminated groundwater was usedwith added tracers to determine Kd valuesaccording to conventional methods. A secondaliquot was processed by reverse osmosis toremove colloidal material, including largehumic/fulvic acid (HFA) molecules, and somedissolved inorganic species, and then used tocarry out the same type of batch Kddetermination.Uncontaminated sand collected fromboreholes located outside the contaminatedarea (ES41, ES42, ES47, ES48; Figure 1.3),were used to make a composite soil sample.Approximately 4.2 kg of material (total) was

31 31 NUREGICR-6627

pre-conditioned in a 4-L plastic containerwith a sample of uncontaminatedgroundwater (from well ES9), for -2 weeksat room temperature. After conditioning,the sand was air-dried on a filter paper.Four sub-samples were taken for particle-size distribution during the homogenizationstep, to ensure the homogeneity of thesample prior to subsampling for sorptionstudies. The size distribution was verysimilar between the replicates.

In June 1996, approximatively 20 L and 40 Lof groundwater were collected from ES-9 andES-16, respectively, using a peristaltic pumpand existing piezometers. Dark plasticgarbage bags were used to cover the samplecontainers to avoid any possible speciationchange induced by photolysis (Cooper andMcHugh, 1983). The samples were filteredthrough 0.45 gm pore size polysulfone filterdiscs (Gelman, 15 cm dia.) within one weekof collection. Throughout the program,waters were stored in the dark at 4 oC, andall sample manipulations were conductedunder subdued light.

2.5.2 Isolation of Anionic and CationicRadionuclide Fractions fromContaminated Groundwaters

Figure 2.3 shows the flowchart of samplemanipulations to isolate the native anionicspecies of the radionuclides present in ES16groundwater. Immediately after 0.45 gmfiltration, a 1-L aliquot of the groundwaterwas reserved for total radionuclide contentand for major ion analysis. Approximately20 L of the filtered contaminatedgroundwater was passed through a bed of800 mL of AG50-X8 cation exchange resin(100-200 mesh, Bio-Rad; previouslyconverted to its Na÷ form and rinsed withdeionized water), at a rate of 28 mL/min toremove the cationic species of nuclides,leaving only the anionic species in solution.The amount of resin was calculated to exceedthe expected cationic content by a factor of10-20 times. A -1-L aliquot of the eluatewas used for radionucide analysis, and therest of the solution was kept in the dark (at 40C) until the Kd measurements were

performed. The radionuclides left in solutionwere operationally defined henceforth as"anionic species".

A second 20-L aliquot of ES16 groundwaterwas passed through a bed of 500 mL of AGI-X8 anion exchange resin (100-200 mesh,obtained from Bio-Rad in its Cl- form; theresin was rinsed with deionized water beforeusing) to remove the anionic species andisolate the positively charged native cationicspecies present in the ES16 contaminatedgroundwater. The amount of resin was 10-20 times in excess of the expected anionexchange capacity, and a -1-L aliquot of theeluate was also taken for radionuclideanalysis. This fraction was henceforth called"cationic species", although it was recognizedthat neutral species would pass through bothresins, and that there might be very smallamounts of chemical retention of somecomplexes on either or both resins.

After elution through each resin, and after aresting period, each solution was split intotwo aliquots. The first aliquot was used fornon-radiogenic analysis (major ions, etc.),whereas the second aliquot was spiked withtracers which are not present in the originalsamples (03 4Cs, 57Co, 94Nb; see Table 2.1).Except for the addition of radionuclides, bothaliquots were treated identically, includingthe equilibration times. The volume ofsolution used was 800 mL along with 200 gof sand.

2.5.3 Blanks, "Conventional Batch" KdMeasurements, and "Reverse Osmosis"Samples

In parallel with the above, a sample ofuncontaminated groundwater from well ES9was taken and split into three portionswhich were used for blanks, conventional"batch" Kd determinations, and "nanofilteredbatch" Kd determinations.

Blanks were prepared by fitering thegroundwater (Figure 2.4). A sub-sample wastaken for total radionuclide analysis, and theremainder of the sample was split into twoaliquots. One aliquot was passed through acolumn of cation exchange resin, while theother one was passed through a column of

NUREG/CR-6627 32

anion exchange resin. After elution througheach column, water samples for major ionsand radionuclides were taken for analysis.The eluates were then contacted with thesand aliquots for 7 days, followed byradionuclide analysis of the aqueous sample.The second portion of the ES9 groundwatersample was treated identically to the schemein Figure 2.3, except that the ion exchangesteps were omitted. Commercial tracers(134Cs, 57Co, 94Nb, 239I24OPu and 241Am) wereadded to the solutions .(Table 2.2). Thesolution volumes were 200 mL, and werecontacted with 50 g of sand. Theequilibration time was also 7 days. Thissample pre-treatment is henceforth referredto as "conventional batch" Kdmeasurements.

The third portion of the ES9 sample wasfiltered using a reverse osmosis unit(Osmonics Model 19E, with a 112-HRmembrane, 200 nominal molecular weightcutoff - NMWCO), to remove colloidal-sizedmaterial, high molecular weight dissolvedorganics, and many dissolved ionic species.All other features, similar to the

"conventional batch" method, were used.This sample is referred to henceforth as"reverse osmosis" Kd measurements.

2.5.4 Speciation Checks for SolubleRadionuclides

The fundamental assumption behind thiswork was that the chemical species of theradionuclides of interest remain stable forthe duration of the experiment.Radionuclide speciation, operationallydefined as the relative proportion ofnegatively charged versus positively chargedspecies, waschecked at various points in the sampleprocessing to determine if the speciation hadchanged with time or sample manipulation.The top part of the flowchart in Figure 2.4constitutes the sample manipulation for aspeciation check. The key steps duringsample processing, where samples weretaken for speciation checks, are indicated inFigure 2.3

33 33 NUREGICR-6627

Figure 2.3 Simplified flow chart of sample manipulation and analysis

Nnn-r~ linnAn

Sample •Filtration (0.45 prm)

Major ionsgamma and alpha analysis

Anion exchangecolumn procedure:

o.0 similar to left.

Cn

CD M<0

ic analyses Anionic components Radiogenic analyses.............. . .="................. ........ .. ...RadioloaLcal analysis:

total alpha and gammaemitters (before spiking)

r ion _ _ _ _peciation checkilysis

Equilibration withgeological medium

(7 day contact period)

aR plcae 1 qu libain for 7 daysdium without geological medium

(blank)

• • •Speciation check

SoeciationMajor ion checanalysis Radionuclide

analysis

6

"aoana

Equilibration for 7 dawithout geological me

(blank)

Majdr ionanalysis

NUJREG/CR-6627 334

Table 2.2 Levels of radionuclides present in contaminated and uncontaminated groundwater (measured),and levels of artificial tracers added to each solution (calculated)

ES16 contaminated groundwater ES9 uncontaminatedRadionuclide (June 1996 sampling) groundwater

Initial amount tracer Detection limit tracer detectionreported (pCi/L) added (pCi/L) (pCi/L)* added limit

(pCiIL) (pCi/L)**

Co-60 3080 -1.9 11-65

Ru-106 460 11-54 140-810

Cs-137 1050 4.0-8.0 11-95

Pu-238 38 -0.05 0.3

Pu-239 184 -0.05 1.02E4 0.3

Am-241 5.4 5-50 (y)t 9530 35-110 (y)t-0.050.3 (a)

Cm-244 0.54 --0.05 . • 0.3

Artificial tracers

Co-57 4.1 1.42E4 -2.7 5.3E4 -8

Nb-94 8,210 2.7-5.4 5.68E4 -11

Cs-134 2.9 8370 1.4-14 5.43E4 1.4-54

*Based on a I L sample and a 3 h count for the gamma emitters, and a 16 h count for the alphas.**Based on a 0.2 L sample and a 3 h count for the gamma emitters, and a 6 h count for the alphas.

tBased on gamma spectrometry.tBased on alpha spectrometry

35 NUREG/CR-6627

Figure 2.4 Flowchart showing the procedure for blanks (with uncontaminatedgroundwater) and for speciation checks (note: for speciation checks, major ionanalysis is also omitted)

Sample

0.

C1 ........

x . ...

hDýFiltration (0.45 ptm)

b~.

Major ionsgammra and alpha analysis

.Q3

<)

Anionic components Cationic components

Major ionsRadionudideanalysis

" Major ions" Radionuclide

analysis

Speciationcheck:

this part isomitted.

00

SSSSSSSSS

SSSSSSSSSSSSSS

Equilibration with geological redium(7 day contact period)

W Radionuclide analysis No(aqueous solution)

NUREGICR-6627 336

The initial speciation check for the ES16groundwater used the original sample afterfiltration, whereas the "anionic" and "cationic"species were collected immediately at theoutlet of their respective columns. Anadditional speciation check was done after thetracers were added, before equilibration withthe sand. After equilibration with sand, thesupernatant solution, used for totalradioisotope content (Ceq in Eq. (2)), was alsoused as the "total" for the speciation check.Two of the replicates of equilibrated solutionwere saved for the "anionic" and "cationic"species determination.

Finally, two additional groundwater sampleswere taken from ES-9 and ES-16approximately 5 months (November 1996)after the original sampling, to check forspeciation changes in the solution alone (i.e.,without being in contact with sand). Thesamples were treated identically as before(Sections 2.5.2 and 2.5.3), and spiked with thesame quantities of radionuclides as before(Table 2.2). The equilibrations were done overa period of 7 days, which was the same as thecontacting period for the initial Kddeterminations.

2.5.5 Analyses of Soluble GroundwaterConstituents

The cations were analyzed by ICP-AES, theanions by ion chromatography (HPLC, IonPacAS4A column, and AI-450 processing software,all from Dionex), and DIC + DOC by totalcarbon analyzer (Dohrmann DC-80). The pHwas measured as soon as possible after eachstep or immediately before equilibration withgeological material.

Each solution was prepared for radiochemicalanalysis by pouring the sample into a tray,double-lined with a polyethyelene sheet, andevaporated to dryness under an infra-redlamp. The dried sample, in the polyethyelenesheet, was folded and pressed with a hydraulicpress to make a 4 cm diameter disc in a plasticpetri dish. This sample was counted on ahyperpure Ge detector for 3 or 5 h. Aftercounting, each disc was recovered for analysisof the alpha emitters. Approximately 0.1 Bqof yield tracer (236Pu or 24 2Pu) was added to

each disc, which was then ashed overnight at4500C in a pyrex beaker. After this firstashing, -0.5 g of NH 4NO3 was added to theashes in each beaker, to minimize problemswith undesirable precipitates in the samplere-dissolution in the subsequent steps. Afterthis addition, the residue was ashed again at4500C for at least 6 hours. A proceduresimilar to the method of Rao and Cooper[1995] was used to prepare NdF3 co-precipitated actinide sources. The samplesfrom ES-16 were all processed to separate theAm + Cm fraction from the Pu + Np fractions,whereas the sources from the samples fromES-9 (uncontaminated groundwater) wereprepared for total actinide analysis. All thesamples were counted on 9 different alphaspectrometers, individually calibrated for 7hours (ES-9 samples) or 16 hours (ES-16samples).

2.5.6 Addition of Spikes for KdMeasurements

The ES16 groundwater contained variouslevels of 6000, 1°6Ru, 13 7Cs, 2 39 /24 0Pu and 24 1AM

(Table 2.2). Tracers that were not present inthe groundwater (57Co, 94Nb, 134CS), wereadded after ion exchange processing tocompare their KI values to those of thecontaminants already present in the plume.The same tracers, plus 241Am and 2 39/ 24 0 Pu,

were added to the ES-9 uncontaminatedgroundwater to compare the results of the Iafter ion exchange processing to that of atypical "classical" batch Kd.

The tracers were obtained either commercially(57Co, from Amersham Canada) or fromGeneral Chemistry Branch at CRL (94Nb,134

CS, 2 4 1AM, 2 4 2 Pu, 2 39/ 240Pu). The Pu used in

the sorption experiment was adjusted to its+IV oxidation state with NaNO2. All tracerscame in strong acidic solutions (HC1, HFand/or HNO3), and dilutions had to be made,in some cases, prior to addition to the finalsolutions. Small volumes of spikes were used,which translated into no significant drop inpH and small additions of Na+, Cl-, F- andNOs" ions (Tables 2.3 and 2.4).

37 NUREG/CR-6627

0~

00

Table 2.3 Concentrations of major ions in ES-16 contaminated groundwater at various stages of sample processing

ES-16 contaminated groundwater compositionConcentration (all concentrations in mg/L)

Ion increase after Original filtered After cation exchange resin step After anion exchange resin steptracer addition groundwater Column before after column before after

Eluent equilibration equilibration eluent equilibration equilibrationCa'* 1.72 ND 0.60 0.58 1.72 1.73 3.60Fe 3+ 1.56 0.58 0.35 0.07 ND ND NDK* 1.40 0.50 0.30 1.20 1.70 1.70 2.50

Mg2÷ 0.72 ND ND 0.16 0.74 0.72 1.00Na÷ 6.80 13.90 13.60 13.60 6.70 6.80 9.00Si 5.80 5.70 5.70 6.20 5.60 5.70 6.70F- 0.01 0.28 0.31 0.29 0.35 0.05 0.06 0.10Cl- 0.1 1.25 1.32 1.32 3.12 24.95 24.15 26.01

N0 3 0.4 0.26 0.48 0.10 0.35 ND ND n.15SQ42- 8.46 8.72 8.77 14.80 ND ND 4.48

DIC (as C) 5.09 5.07 3.90 1.98 0.09 0.07 0.27DOC (as C) 4.17 3.49 4.03 5.03 9.80 9.67 10.47

pH 7.31t 6.28 6.80 6.88 7.34 4.13 4.19 5.974.14t I I_ II

tTo be compared to eluent from the cation exchange resin step.TTo be compared to eluent from the anion exchange resin step.

Table 2.4 Concentrations of major ions in ES-9 uncontaminated groundwater at various stages of sample processing

ES-9 uncontaminated groundwater composition(all concentrations in mg/L)

Conc. Ion exchange processing Kd determination Kd determination, after reverseincrease (for blanks purposes) "conventional batch" method osmosis processing

Ion after tracer "nanofiltered batch" methodaddition Original after cation after anion Original before after Original before after

filtered exchange exchange filtered contact contact nanofiltere contact contactd

Ca'+ 4.90 0.09 4.80 4.80 4.80 3.30 0.75 0.99 1.49Fe 3÷ 2.00 1.77 0.01 1.89 1.74 0.07 0.02 0.03 0.01K÷ 1.90 0.35 1.40 1.50 1.30 1.90 0.60 1.30 1.50

Mg2÷ 1.88 ND 1.86 1.90 1.88 1.06 0.46 0.52 0.44Na÷ 0.1 2.12 12.50 2.04 2.04 2.00 4.50 1.18 1.38 5.50Si 7.40 6.40 7.20 7.70 7.70 8.00 4.50 4.30 5.40F- 0.1 0.08 0.09 0.07 0.09 0.07 0.16 0.06 0.08 0.11CI- 2.1 1.43 1.24 26.31 1.13 1.23 2.68 0.73 0.86 4.66

N0 3 3.3 0.08 0.10 ND 0.06 ND 0.13 0.12 0.08 3.41S042 10.39 8.97 ND 10.42 10.39 14.89 2.95 2.89 8.28

DIC (as C) 3.28 3.09 0.27 2.77 4.33 1.30 0.74 0.47 0.16DOC (as C) 5.32 5.51 17.60 4.84 3.11 3.38 1.74 1.40 1.57

pH 6.85t 6.88 7.52 4.22 6.79 6.89 7.02 6.10 6.45 6.496.69t I_ _

tTo be compared to "batch Kd determination, before contact".$To be compared to "nanofiltered batch, before contact".

I%

3 Results and Discussion

3.1 Concentrations andChemical Speciation inGroundwater

The results of the large-volume watersampling conducted down-gradient fromthe Chemical Pit between 1983 and 1994for partitioning radionuclides betweenparticulate, cationic, anionic, and non-ionicspecies are shown in Figures 3.1 to 3.4.Total concentrations of 60Co, 106Ru, 125Sb,and 239,240pu decreased significantlybetween 1983 and 1993-94. However, therelative distribution of these radionuclidesbetween the various anionic, cationic, andnon-ionic chemical forms remainedrelatively constant over this eleven yeartime frame. This indicated that theformation mechanism(s) of the moreabundant mobile anionic species of theseradionuclides has apparently not changedappreciably over the past ten or moreyears. Even though major changes in thequantity and composition of liquiddischarges to the Chemical Pit occurredover this time period, they apparently didnot have a major impact on the chemicalspeciation of radionuclides migrating fromthis site.

Anionic chemical speciation was thepredominant charge-form observed for60C0, 106Ru, 125Sb, and 239,240pu during their

transport in the groundwater down-gradient from the Chemical Pit. An earlierstudy (Champ et al., 1985) of the ChemicalPit contaminant plume also showed, that inaddition to the above group ofradionuclides, anionic species of 55Fe, 63Ni,95Zr-Nb, and 137Cs were the dominantchemical forms being transported in thegroundwater. This was not too surprizingfor the 95Zr-Nb, l°6Ru, 125Sb, and 239,240Pu,

where thermodynamic calculations showanionic species to be predominant inoxygenated groundwater systems.

However, the predominantly anionicspecies of 55Fe, 60C0, 63Ni, and 137Cs

observed in this groundwater wereunusual, since thermodynamic calculationsgenerally indicate that cationic speciesshould predominate.

There is strong evidence to indicate thatthe negatively charged radionuclide speciesare primarily organo-radionuclidecomplexes formed with either: 1) organicmacromolecules of the natural humic/fulvicmaterials originating from thedecomposition of plant and animalresidues, and/or 2) anthropogenic organicchelating agents, such as ethylenediaminetetraacetic acid (EDTA), because a numberof which are known to have been disposedof within the Chemical Pit (Killey, et al,1984; Robertson, 1986). Other potentialforms of these migrating radionuclide-bearing anions may include inorganiccolloidal materials and/or inorganiccomplexes composed of common aquo-anionligands (sulfato-, chloro-, hydroxo-, etc., orcombinations thereof).

It was also of interest to determine if therelative distribution of radionuclidesbetween soluble anionic, cationic, and non-ionic forms changed as the groundwatermoved down-gradient from the oxidizingenvironment of the pit area to the morereducing environment of the East Swamparea. Figure 3.5 shows that the relativeabundances of cationic, anionic, and non-ionic forms of 60Co changed very little asthe radiocobalt moved from the pit area tothe swamp. Apparently, the changes in theoxidizing conditions of the groundwaterhad little effect on the charge-formspeciation of 60C0. Similar behavior wasnoted for 106Ru and 239,24 0Pu, but 125Sb

showed significant changes between non-ionic and anionic forms during thismigration.

NUREG/CR-6627 440

Figure 3.1 Changes in 60Co concentration and chemical form with time in well ES-16which is in the CRL Chemical Pit groundwater plume. Although the totalconcentration of 60 Co decreased significantly with time, the relativeabundance of the cationic, anionic, and non-ionic species remained similar

over the 11-year study period.

100000

cationic Coanion Co

* nonionic Co

U

,.

z0

I-

zU1

z0

0(0

I-UJ

100007

1000

100

1011983 1991 1993 1994

SAMPLING DATES

41 41 NUREGICR-6627

Figure 3.2 Changes in 106Ru concentration and chemical form with time in well ES-16which is located in the CRL Chemical Pit groundwater plume

100000

W PARTICULATECATIONIC

* ANIONIC[ NONIONIC

z

0

zLUI0z00

10000

1000

100-

10

.1 ! -I--

1983 1991 1993

SAMPLING DATES

NUREG/CR-6627 42

Figure 3.3 Change in 125Sb concentration and chemical form with time in well ES-16which is located in the CRL Chemical Pit groundwater plume

10000

C)

z0

LI-z0

Uf)C1

Wo

1000

100

N PARTICULATECATIONIC

* ANIONIC[ NONIONIC

.1

.01

.001 -I--

1983 1991 1993

SAMPLING DATES

43 43 NUREG/CR-6627

Figure 3.4 Change in 239,240Pu concentration and chemical form with time in groundwaterat well ES-16 which is located in the CRL Chemical Pit groundwater plume

100

CATIONICANIONIC

0.

z

z0

zrCý

10

.1

.01

1983 1991 1993 1994

SAMPLING DATES

NUREG/CR-6627 444

Figure 3.5 Change in 60Co concentration and chemical form in the CRL Chemical Pitgroundwater plume as a function of distance from the pit. These results showthat the relative abundance of cationic, anionic, and non-ionic forms of 6OCochanged very little during transport from the oxidizing groundwaterenvironment at wells CP4 and ES39 to the relatively reducing environment atwell ES16 which is located in East Swamp.

10000*:

CATIONICANIONIC

* NONIONIC

U%.w

Z0

Z

mgc

LI-Uw

00

0oI,0o.

1000 -

100-

10

I i

CP-4(10 meters)

ES-39(22 meters)

ES-16(40 meters)

SAMPLING WELLS

45 45 NUREG/CR-6627

3.2 Radionuclide Association withColloids in Groundwater

3.2.1 Evolution of Colloid SpeciationAlong the Flowline

The results of particle size analysis carried outwith Nuclepore membranes are summarized inTable 3.2 for samples collected from wells CP4,ES39 and ES16. Except for the higherconcentration of suspended particles (> 450 nm)in CP4, colloid and suspended particleconcentrations were low in the groundwater.Highest suspended particle concentrations werefound in the most recently installed piezometer(CP4), and were lowest in the most developed(i.e. oldest and most heavily pumped)piezometer (ES16). When CP4 was sampledagain in 1994 the concentrations of bothparticles and colloids were lower. Since theparticle concentrations in these waters were toolow to scatter sufficient light, the UPA requiredsamples of particle concentrates to producemeaningful particle size distributions. Sizeanalysis with the UPA indicated that CP4contained particles from 20 to 2500 am, with ahigh concentration of particles around 50 nm.Colloidal particles in ES39 were dominated bycolloids ranging in size from 6 to 15 nm. TheUPA could only detect particles smaller than400 nm in ES16.

By analyzing the filtrate and retentate samplesproduced during groundwaterfiltration it was possible to calculate thegroundwater concentration of a given elementassociated with dissolved species, colloids from1 to 10 nm, colloids from 10 to450 am, and suspended particles larger than450 nm. The concentration of dissolvedelements is given by

colloids (1-10 nm) = (R2-F2)/C2 (2)

where R2 represents element concentrations inretentate R2, and C2 is the colloid concentrationfactor during filtration with the 10,000 NMWLmembrane. The concentration factor is given bythe sample volume beforefiltration divided by the volume of retentate.The concentration of elements associated withcolloids between 10 and 450 nm can becalculated from equation 3 or equation 4. Thereported value for these concentrations is theaverage of equations 3 and 4.

colloids (10-450 nm) = (F3-Fl)/C1 (3)

colloids (10-450 am) = (R1-F1)/C1 - (4)particles (>450 aun)

In the above equations F3 represents elementconcentrations filtered through the 450 nmfilter, F1 is the filtrate from the 100,000 NMWLmembrane, R1 is the retentate from the 100,000NMWL filtration, and C1 is the colloidconcentration factor for the 100,000 NMWLfiltration.

The concentration of elements associated withsuspended particles is given by the average ofequations 5 and 6.

particles (>450 nm) = (R3-F3)/C3/C1 (5)

particles (>450 nm) = (R1-F3)/C1 (6)

Since the fraction of dissolved salts that wasconcentrated along with the colloids wasbetween 0 and 4 percent, it is necessary toaccount for salt retention when calculatingcolloid compositions from particle concentrates.First, the percent of a given element'sgroundwater concentration, which was enrichedin the particle concentrate was calculated.Then the measured percent salt retention,determined by conductivity, was subtractedfrom this value. The amount of the givenelement associated with colloids was calculatedfrom the "corrected" percent.

dissolved concentration = F2 (1)

where F2 represents element concentrations inthe F2 filtrate (see Fig. 2.1). The groundwaterconcentrations ofelements associated with colloids between 1 and10 m is given by

NUREG/CR-6627 446

Table 3.1 Gravimetric particle size analysis of micro-particulates in groundwatercollected down-gradient from the Chemical Pit

CP 4 1993 SIZE (nm)

10000500010004001005015

total particles (>450 rnm)total colloids (10 - 450 nm)

CP4 1994 SIZE (nm)

ES 39

10000500010004001005015

total particles (> 450 nm)total colloids (10 - 450 nm)

SIZE (nm)

10000500010004001005015

total particles (> 450 in-)total colloids (10 - 450 rim)

SIZE (n3m)

10000500010004001005015

total particles (> 450 nrm)total colloids (10 - 450 tim)

mg/L

0.0970.7100.6910.0260.0480.0720.0841.5230.2

mg/L

0.1030.0590.1060.012

0.0030.0590.0000.2800.062

mg/L

0.0640.0580.3160.0700.0730.0970.1340.5090.304

mg/L

0.0530.0500.0460.0640.0510.0580.0640.2140.173

ES 16

The results of radiochemical analysis carriedout at CRL on filtrate and retentate samples

were used to determine radionuclide activitiesfor dissolved and colloidal species. The results

47 NUREG/CR-6627

of the determinations for the gamma-emittersare given in Table 3.2. It is evident from thedata that dissolved species (< 10,000 NMWLor 1 nm) dominated in the cases of all themeasured gamma-emitters in these samples.The results for dissolved 60Co and 125Sbshowed a downward trend away from the pit.However, dissolved 106Ru shows a trend inconcentration in the opposite direction. Thisprobably reflected the lack of recent inputs ofthis mobile and short-lived radionuclide to thepit (i.e. 106Ru was being flushed out of theaquifer).

The results of size analyses of the actinidesare given in Table 3.3. The concentrations ofactinides in the plume were very low andconsequently the errors in the analyses werequite high, particularly for the minorfractions. This accounted for some values inthe table being negative. Since the well atCP4 was freshly installed in 1993 and showeda high particulate content, it was decided tore-sample it again in March of 1994.Unfortunately, due to a process upset in thewaste treatment plant late in 1993 and

freezing of transfer lines early in 1994, theChemical Pit received some unusual inputsbetween the two sampling periods. As hadbeen anticipated, the 1994 results for CP4showed a lower particulate component for theactinides than the 1993 results. Theconcentrations of the dissolved components forall of the actinides, except 2 44Cm were higherin CP4 in 1994. Although this may be due tothe addition of wastes to the pit prior to thesecond sampling, it might also be due to aseasonal trend to higher concentrations in thewinter, when the ground was frozen and therewas no water infiltration from the surface.This trend had been observed in previouswork in the same plume. The concentrationsof dissolved 238U, 237Np and 244 Cm showed adownward trend away from the pit, but239124 0Pu and 238Pu+24iAm (which is probablymostly 238Pu) showed an upward trend.

In summary, visual inspection of thegroundwater indicated that the suspendedload was very small. This was confirmed bythe particle size analyses (Table 3.2) wheretotal particles were only a small fraction of

Table 3.2 Size Fractionation of Gamma-Emitters

Radionuclide Concentration (pCi/kg)

Sample Element Dissolved Colloid

(1-10 nm)

Colloid

(10-450 nm)

3.480.0

140

SuspendedParticles

(>450 nm)

1.409.268.02

CP4 Sb:125Ru-106Co-60

ES 39 Sb-125Ru-106Co-60

ES 16 Sb-125Ru-106Co-60

10708610

13,200

35090209260

21014300

7510

0.0508286

16.1713122

14.5721178

5.24621237

8.91302102

0.0092.90.00

1.409.268.02

NUREG/CR-6627 48

Table 3.3. Size fractionation of actinides found in groundwater collected down-gradientfrom the Chemical Pit

Radionuclide Concentration pCi/L (numbers in parentheses are % of total species)

Sample Element Dissolved

CP41994

U-238

Np-237

Pu-239/240

Pu-238 +Am-241

Cm-244

U-238

Np-237

Pu-239/240

Pu-238 +Am-241

Cm-244

0.474 (82)0.772 (79)

3.37 (61)

4.87 (76)

0.481 (81)

0.549 (62)

0.482 (47)

2.27 (33)

1.67 (31)

0.483 (59)

Colloid

1-10 rim

0.072 (12)0.123 (13)1.35 (24)1.22 (19)

0.076 (13)

0.213 (24)0.180 (18)2.57 (37)2.20 (41)0.250 (31)

Colloid Particulate

10-450 um

0.032 (6)0.079 (8)0.845 (15)0.356 (5)0.037 (6)

0.100(11)0.136 (13)1.34 (19)0.266 (5)

0.085 (10)

-0.019-0.046-.039-0.160-0.028

0.024 (3)0.222 (22)0.766 (11)1.23 (23)-0.017

CP41993

ES391993

ES161993

U-238

Np-237

Pu-239/240

Pu-238 +Am-241

Cm-244

U-238

Np-237

Pu-239/240

Pu-238 +Am-241

Cm-244

0.438 (68)

0.494 (80)

1.99 (64)

3.98 (82)

0.926 (71)

0.262 (70)0.315 (71)

3.30 (83)

7.75 (93)

0.290 (84)

0.110 (17)

0.073 (12)0.411 (13)

0.424 (9)

0.216 (17)

0.072 (19)

0.055 (12)

0.422 (11)

0.450 (5)

0.033 (9)

-0.064

-0.020

0.301 (10)0.216 (4)

0.120 (9)

0.001(0.3)

0.024 (5)

0.158 (4)

0.161 (2)

0.023 (7)

0.095 (15)

0.048 (8)0.402 (13)

0.255 (5)

0.044 (3)

0.037 (10)

0.047 (11)

0.110 (3)

-0.099

-0.007

49 NUREG/CR-6627

the groundwater sampled. For the gamma-emitting radionuclides (Table 3.2), thedissolved fraction accounted for 86-99% of thetotal radioactivity of these isotopes present inthe groundwater. For the actinideradionuclides (Table 3.3), the dissolvedfraction became increasingly large in thegroundwater as a function of distance fromthe Chemical Pit, and accounted for 70 to 93%of the total actinide elements present in thegroundwater. Thus, the dissolved fraction ofall radionuclides was the predominant formbeing transported in the groundwater, andcolloidal transport amounted to only 7-30 %depending upon the radionuclide. Similarresults were observed in 1993 for the gamma-emitting radionuclides, and these data werereported by Schilk, et al., 1995.

3.2.2 Evolution of Chemical SpeciationAlong the Flowline

The soluble radionuclide speciation in portionsof the <10,000 NMWL (1 nm) filtrates fromthe three wells was measured by anionexchange chromatography using the techniquedescribed in Section 2.3.2 of this report. Thethree samples were loaded andchromatographed with a 25 cm analyticalcolumn at a flow rate of 6 mL per minute. The6OCo.chromatograms for these samples areshown in Figure 3.6. Seven peaks wereidentified across all of the chromatograms andthese are numbered in the figure. The speciesthat eluted in KC1 were likely being displacedby competition from C1- ions. In previous work(Cooper et al., 1995) size exclusionchromatography showed that some of thesespecies have higher molecular weights thansimple cations, indicating that they werebeing displaced from the columns ascomplexes. On the other hand, species thateluted in 0.1 M HCl had lower molecularweights. These species, which are stronglyretained by the AG MP-1 resin, were probablybeing destabilized at the lower pH. Once thecomplexes broke down, the cations liberatedfrom them were eluted from the columns.

There was clearly an evolution in chemicalspeciation of 60Co as the groundwater flowedfrom the pit (CP4) towards the wetland (ES39and ES16). The integrated counts (CPM/L) inthe peak regions are given in Table 3.4. The

dominant peak was number 4 and it reached amaximum at ES39. Peak 3 was at itsminimum at ES39. It is possible that these 2peaks were poorly resolved in thechromatogram for the sample from ES39 andthis accounted for the observed differences inthe counts at ES39. Peaks 1,2, 5, 6 and 7declined as the groundwater flowed towardsthe wetland. The dissolved 60Co species thateluted in KC1 dominated over the ones thateluted in HC1.

The chromatograms obtained for 106Ru areshown in Figure 3.7. Since the countingstatistics for 106Ru were poorer, thechromatograms showed more scatter thanthose of 60Co. Peaks 2 and 4 were fairlyconstant down the flow system. Peak 1declined with distance from the pit, whilepeaks 3 and 5 increased. The heights of thepeaks in KC1 and HC1 were similar, althoughmore 1°6Ru eluted in the KC1.

Chromatograms obtained for 238U appear inFigure 3.8. Most of the 238U eluted in HCl.The peak height showed a minimum at ES39.Most of the 23 7Np species shown in Figure 3.9also eluted in HCL. However, the maximumpeak height appeared at ES39. There wasalso evidence for peaks in the KC1 gradient atCP4 and ES16 (Figure 3.10), but littleevidence of the same peak at ES39. Theresults for 237Np were corrected for 234U byusing the counts from the 238U peak andassuming that the two U isotopes were inequilibrium.

Figure 3.11 shows the chromatogramsobtained for 2391240 Pu. Here again most of thespecies eluted in HCl, but peaks in KCl werebecoming more intense. These are shown inmore detail in Figure 3.12. Although CP4 andES16 showed only a single peak in thegradient, there were two distinct peaks atES39.

Chromatograms for 241Am are shown in Figure3.13. Since these results were obtained fromtotal actinide sources, the 24 1Am counts had tobe corrected for 2 38Pu. The 238Pu/ 239/24 0Pu ratiowas established from fractions which showedseparation of Pu from Am and Cm. This ratiowas then used to correct the results from therest of the fractions. Having to apply the

NUREG/CR-6627 550

correction increased the uncertainty in theresults for fractions that eluted in HC1, so itwas difficult to interpret this region in thechromatograms. Americium eluted earlierthan plutonium in the KC1 gradient and twopeaks appeared to be present. The first peakdeclined with increasing distance from the pit,while the second one declined towards ES39and increased again at ES16.

Figure 3.14 shows the chromatogramsobtained for 244Cm. Most of the species elutedin the KC1 gradient and there were two peaks,which were similar to 241Am. The first peakreached a maximum at ES39, while the secondpeak was greatest at ES16. The elution of theES16 sample in HU1 showed the presence of adelayed peak, which was not present in theother two chromatograms.Figures 3.8 through 3.13 show an interestingtrend in speciation going up the actinideseries. The lower actinides were stronglyretained by the resin and were mostly

removed by the HCl elution, probably throughdestabilization of the complexes. The higheractinides were less strongly retained by theresin and were eluted earlier in the KC1gradient. Fewer of their complexes remainedon the columns long enough to be eluted inHCL.

In summary, the anion chromatographicseparations of the soluble (< 1 nm) fractionfrom the ultra-filtration of groundwater fromwells CP4, ES39, and ES16 demonstrated thepresence of multiple chemical species for eachradionuclide. For 60Co and 106Ru, at least 5-7specific chemical species were observed. Formost of the actinide radionuclides, the numberof individual species was fewer than observedfor the 60Co and 106Ru, with about 2-4 speciesgenerally being observed. This points out howcomplicated the chemical speciation ofradionuclides being transported ingroundwater can be, and clearly shows thatnumerous natural organo-radionuclidecomplexes simultaneously exist ingroundwater at this site.

Table 3.4 Sunmary of counts in the peaks of chromatograms from ultra-filtered samples

Peak #

1

2

3

4

5

6

7

CP4-F2

100

216

502

1203

232

122

283

ES39-F2

60

191

266

1419

247

74

125

ES-16-F2

62

68

450

1072

131

75

115

ES16

30

78

402

1022

108

140

406

51 NUREG/CR-6627

Figure 3.6 Co-60 Chromatograms for samples from the three wells

CP4F2 Co-60 CPM/L

0

600

400

200

00 20 40 60 80

Fraction number100 120 140 160

I ES39F2 Co-60 CPM/L I8oo

600

400

ILQ 200

0

-2000 20 40 60 80 100 120 140 160

Fraction number

ES16F2 Co-60 CPM/L I. B

0

600

400

200

0

-2000 20 40 60 80

Fraction number100 120 140 166

Vertical lines show eluent changes

NUREG/CR-6627 52

Figure 3.7 Ru-106 chromatograms for samples from the three wells

I ES39F2 Ru-106 CPM/L I40 -

30

202

o 10 -

0

.100 20 40 60 80

Fraction number100 120 140 160

ES16F2 Ru-106 CPM/L I40 -(

30 -

20 2

0 10 -

-100 20 40 60 80 100 120 140 160

Fracdion number

Vertical lines show eluent changes

53 53 NURLEG/CR-6627

Figure 3.8 U-238 chromatograms for the samples from the three wells

...-CP4F2, U-238mBqAo of loaded groundwater

5

4

3

E 2

I

00 20 40 60 80 100 120 140

Fraction No.

ES39F2, U-2388mBq/L of loaded groundwater

2

E0.5-

00 20 40 60 80 100 120 140

Fraction No.

ES16F2, U-238mBq/L of loaded groundwater

4

3

Mcr 2E

00 20 40 60 80 100 120 140

Fraction No.

NUREGICR-6627 554,

Figure 3.9 Np-237 chromatograms for samples from the three wells

CP4F2, Np-237 ImBq/L of loaded groundwater

3

2

E 00 20

0 040 60 80 100 120 140

Fraction No.

ES39F2, Np-237mBqAL of loaded groundwater

8

6

4-

E 2

0

-20 20 40 60 80 100 120 140

Fraction No.

ES16F2, Np-237mBq/L of loaded groundwater

4

3

.EI

0A-1 6 80 1 1, 140

0 20 40 60 80 100 120 140

Fraction No.

55 55 NUREG!CR-6627

Figure 3.10 Np-237 chromatograms for samples from the three wells

CP4F2, Np-237 ImBq/L of loaded groundwater

0.2

0.15

E 0.1

E3 0.05 ,~LA

0 N

-0.05 F,,0 20 40 60 80 100 120 140

Fraction No.

ES39F2, Np-237imBq/L of loaded groundwater

0.06

0.04 -

,gr .02

0 V vv U

-0.020 20 40 60 80 100 120 140

Fraction No.

ES16F2, Np-237 imBq/L of loaded groundwater

0.4

0.3

9 02

E 0.1

-0.10 20 40 60 80 100 120 140

Fraction No.

NTJREGICR-6627 556

Figure 3.11 Pu-239+240 chromatograms for samples from the three wells

CP4F2, Pu-239+240 ImBq/L of loaded groundwater

20

10

E5

A0 __20 20 40 60 80 100 120 140

Fraction No.

ES39F2, Pu-239+240 ImBqAL of loaded groundwater

10

8 -

E 4

2

00 20 40 60 80 100 120 140

Fraction No.

ESI 6F2, Pu-239+240Im~q/L of loaded groundwater I

40

30

g20E

10

00 20 40 60 80 100 120

Fraction No.140

57 57 NUREG/CR-6627

Figure 3.12 Pu-239+240 chromatograms for samples from the three wells

CP4F2, Pu-239+240ImBq/L of loaded groundwater

4

3

E

I

0 2"00 20 40 60 80 100 120 140

Fraction No.

ES39F2, Pu-239+240mBqiL. of loaded groundwater

0.5

0.4

,0.3

E 0.2

0.1

00 20 40 60 80 100 120 140

Fraction No.

ES16F2, Pu-239+240mBq/L of loaded groundwater

10

8

E 4

2

00 20 40 60 so 100 120 140

Fracton No.

NLTREG/CR-6627 558

Figure 3.13 Am-241 chromatograms for the samples from the three wells

CP4F2, Am-241mBq/L. of loaded groundwater15.

10

! .5

-5

-100 20 40 60 80 100 120 140

Fracton No.

ES39F2, Am-241mBq/L of loaded groundwater

10

8

6cr 4.

E 22

0

-20 20 40 60 80 100 120 140

Fracton No.

ES16F2, Am-241mBq/L of loaded groundwater

4

-20 20 40 60 80 100 120 140

Fraction No.

59 59 NUREG/CR-6627

Figure 3.14 Cm-244 chromatogrc ms for samples from the three wells

CP4F2, Cm-244 ImBq/L of loaded groundwater

6

00 20 40 60 80 100 120 140

Fraction No.

ES39F2, Cm-244mBq/L of loaded groundwater

20

15 -

ot10 -15

0 20 40 60 80 100 120 140Fraction No.

ES16F2, Cm-244mBq/L of loaded groundwater

4

3

g2

1

00 20 40 60 80 100 120 140

Fraction No.

NLTREG/CR-.6627 660

3.3 Speciation of Radionuclidesand Organics in Groundwater

As described in Section 2.4, a series of anionexchange chromatography experiments wereconducted to separate a number of solublephysicochemical species of 60Co and 137Cs fromgroundwater sampled at well ES16. In thistechnique, the anionic radionuclide specieswere first sorbed from groundwater onto anionexchange resin and then eluted off the resinwith increasingly stronger solutions to removethe different physicochemical species.Organic chemical species, as determined byUV absorption, were also separated andmeasured by this technique.

In earlier work the elution gradient wasgenerated with a pump and mixing chamber,which made it difficult to generate complexgradients, so a simple linear 0 to 0.5 M KC1gradient was used. The availability of thegradient programmer for the present workallowed more flexibility in generatinggradients. The decision was made to split theKC1 gradient into two segments because thereseemed to be few distinct species eluting inthe latter part of the gradient in previouswork. By splitting the gradient, the slopecould be kept lower in the first half (0 to 0.25M) where the peaks eluted. The slope of thesecond half (0.25 to 0.5 M) was increased tosave time. The direct transition to 0.1 M HC1used in initial runs in this series resulted in ashock which mobilized a lot of species andproduced a peak in the UV chromatogram thatwent off scale. This shock was lessened byusing a gradient for the transition from KC1 toHCL.

A sample of ES16 groundwater was collectedon April 22, 1996 and chromatographed onMay 7. The chromatogram obtained for 60Co isshown in Figure 3.15. The vertical lines showthe changes in eluent. These are from left toright: pH 4.5 HC1, 0 to 0.25 M KC1 gradient,0.25 to 0.5 M KC1 gradient, 0.5 M KC1, 0 to 0.1M HCl gradient and 0.1 M HCl. There werefive distinct peaks in the first KC1 gradientand one in the HCl gradient. A similarchromatogram obtained with a samplecollected from the same well in 1993 is shown

in Figure 3.16 for comparison. In this case theloading and elution were carried out at 6mL/min. The vertical lines again showchanges in eluent. In this case these are fromleft to right: pH 4.5 HC1, 0 to 0.5 M KC1, 0.5 MKC1 and 0.1 M HC1. Although the KC1gradient was changed in 1996, the 0 to 0.25 Mgradient was comparable to the first half ofthe 0 to 0.5 M HC1 gradient in 1993, since theslopes in KCI concentration with eluentvolume were the same. The 1993chromatogram was dominated by two distinctpeaks in the first half of the KC1 gradient.These were much higher than the peaks in the1996 chromatogram, which reflected thedecline in 60Co concentration since the use ofthe pit was discontinued in 1996. Theretention times of the two peaks in the 1993chromatogram corresponded to peaks 3 and 4in the 1996 chromatogram. The 1993chromatogram also showed two peaks in 0.1 MHC1 compared to only one in the 1996chromatogram. The retention times could notbe compared in the HC1 portions of the twochromatograms, since the elution was changedin 1996. The two chromatograms showed thatthere were distinct changes in the speciationover the period of almost three years betweenthe sampling dates. This was reflected in thenumbers and intensities of the peaks in thetwo chromatograms.

The UV chromatogram from the April 22,1996, sample is shown in Figure 3.17. Sevenpeaks in absorbance can be identified. Peaks1 to 4 occur in the first KC1 gradient wherefive peaks appear in the 60Co chromatogram.Peak 1 is very sharp, but peaks 3 and 4 arepoorly resolved. Peak 5 is a very broad peakwhich starts in the KC1 gradients and declinesin the 0.5 M KC1. The sixth peak, whichelutes in the 0 to 0.1 M HC1 gradient is verysharp. Peak 7 is a broad peak, which starts inthe HC1 gradient and continues into the 0.1 MHCl. Since the peak had not declined to thebaseline by the end of the HC1 elution, thevolume of 0.1 M HCl used for elution wasincreased in subsequent runs.

The speciation of the organics was examinedin more detail by measuring the UVabsorbance spectra of selected fractions. Thespectra are shown in the figures in Appendix

61 61 NUREGICR-6627

3. The noise between 350 and 400 nm is anartifact of the UV detector. The spectraobtained from fractions eluted with KC1generally don't show remarkable features.The peaks in absorbance numbered 1 to 5seem to be due to a broad-band increase inabsorbance across the UV spectrum. Thespectrum from fraction 65 shows a broadeningof the initial peak. This indicates an increasein absorbance between about 210 and 250 nm,but there is not a distinct peak. Theabsorbance appears to increase between 250and 300 nm, but again there is no distinctpeak. The spectra of fractions eluted with HC1show more features, with an increase inabsorbance around 240 nm. A broad peakaround 340 nm is also evident in some of thesefractions. Spectra obtained from surfacewater from Mainstream are also shown at theend of Appendix 3. This stream, which is inthe same basin as ES16 has highly coloured,organic rich water. The sample had to bediluted with water to bring the absorbance onscale. Addition of KC1 to the sample resultedin the appearance of the peak at 200 nm,which appeared in the fractions as well. Thespectrum obtained with Mainstream waterafter addition of KC1 exhibited a slow declinein absorbance with increasing wavelength.This is unlike most of the fractions whichshowed a fairly constant absorbance withincreasing wavelength, particularly above 400nm.

On May 27 a sample of groundwater wascollected from the well ES9, which is outsidethe plume. This sample was filtered, loadedand chromatographed in the same manner asthe ES16 sample, except for the increasedvolume of 0.1 M HCl used for elution. The UVchromatogram is shown in Figure 3.18. Itshows some features that are similar to thechromatogram in Figure 3.17, although theintensities of the peaks are different. Thereare broad peaks in the second KCl gradientand 0.5 M KC1, as well as in the 0.1 M HCI.The two peaks in the first KC1 gradient haveretention times similar to peaks 1 and 4 fromthe ES16 sample. The HCI gradient also gavea sharp peak like the one in Figure 3.17.

UV absorbance spectra of the fractions fromthe ES9 run were measured and the spectraare shown in Appendix 4. With this samplethe peaks in the UV chromatogram thateluted in KCI were due to a broad increase inUV absorbance across the spectrum. This issimilar to what was observed with the ES16sample. In this case the broadening of thepeak at 200 nm in fraction 65 is not soapparent. The fractions that eluted in HClshow evidence of broad peaks around 240 and340 nm, which is similar to the same fractionsfrom the ES16 sample.

NUREG/CR-6627 662

Co-60 CPM/kg loaded.ES16, Apr 22, run # 3, May 07

60 "

40

a"

tm

CA

i

- " 40

CL

CO*2 0

6

20

0 20 40 60 80 100 120 140 160Fraction No.tO _ _ _ _ __ _ _ _ _ _ _

[vertlCal lnes snow eluent cnanges I

-1 0

M

0

"1

0

0

0'

I-A

ES16, August 17,1993 - Co-60 CPM/Ll600

400

Q.0

200

00 20 40 60 80 100 120 140

Fraction number160

lVertical lines shown eluent changesI

UV Absorbance at 254 nmES16, Apr. 22, run # 3, 1996 May 7

2.5 A'

CA

0

2

0

o

< 0.5

w0 20 40 60 80 100 120 140 160,•= =Fraction NO.,bi lVerlical lines show eluent ch-an-ge .

00

00

0~

0'1

ON 0C\

CO

UV Absorbance at 254 nm1ES39, May 27, run # 1, May 28

2.5

2

0

1.5

0.5

0

-0.50 20 40 60 80 100 120 140

Fraction No.160

IVertical lines show changes in eluentx

On May 13, 1996 another sample of ES16groundwater was pumped and filtered.Five portions of this sample were loadedonto anion exchange resin andchromatographed between May 14 andMay 23. The UV chromatograms, shown inFigures 3.19 and 3.20, showed goodconsistency in shape and peak retentiontimes from run to run. The chromatogramswere similar to the one in Figure 3.17.There were four peaks in the first KC1gradient, one broad peak in the second KC1gradient and 0.5 M KC1, and two peaks inHCl. The first narrow peak in HC1 seemedto decline between the first and secondruns, while the second, broad peakincreased in the later runs.

The fractions from the first two runs wereanalysed for 60Co using the MINAXIScounter and the chromatograms are showin Figure 3.21. They were similar to thechromatogram of the earlier ES16 sampleshown in Figure 3.6. Six peaks could beidentified in the chromatograms. Fivewere in the first KC1 gradient, while thesixth was in the HC1 gradient. Selectedfractions from the May 15 run were alsoanalysed on the high resolutionspectrometer. These spectra gave resultsfor 137Cs and 6oCo, which are shown in thechromatograms in Figure 3.22. In earlierwork it was also possible to constructchromatograms for '0 6Ru, but in 1996 theconcentration has declined to the pointwhere the peaks from '0 6Ru could nolonger be identified in the spectra from thefractions.

The chromatogram for 137Cs had a peak

which eluted in the first KC1 gradient.This confirmed the presence of uniqueanionic 137Cs species, which had beendetected when anionic/cationic speciationhad been measured with cation and anionresins in samples collected early in 1996.The chromatograms for the two 60Co peaksconfirmed the one obtained earlier withresults from the MINAXIScounter.

In summary, the anion chromatographyseparations of 6 0Co have provided resultsvery similar to those observed for the ultra-

filtered groundwater samples (see Section3.2.2), and confirm the presence of multiplesoluble anionic species of organo-radionuclide complexes existing ingroundwater. These results also confirmthe similar observations reported earlierfor this groundwater plume (Cooper andMattie, 1993; Cooper, Haas, and Mattie,1995).

3.4 Adsorption of Anionic andCationic Radionuclide Speciesonto Soil

Previous large-volume groundwatersampling and radionuclide analyses haveshown that anionic chemical species of alarge suite of radionuclides are transportedmost rapidly in the subsurface saturatedenvironment. In order to more accuratelypredict the "first arrival" transport rates(and conversely, the retardation) of themobile anionic radionuclide species ingroundwater, a series of Kd measurementswere conducted using as tracers: 1) naturalanionic radionuclide species isolated fromwell ES16 groundwater, and 2) forcomparison purposes, commerciallyavailable cationic species of radioisotopes(see Section 2.5). The results of thesesorption experiments are described in thissection.

3.4.1 Groundwater Characterization

Groundwater sampled from well ES16 wascharacterized for radionuclide chemicalspeciation and elemental composition. Theambient concentrations of radionuclidecontaminants in ES16 contaminatedgroundwater (see Table 3.5) weresubstantially lower than those observed inprevious samplings [Champ et al., 1984;Cooper and Mattie, 1993]. Theradionucide chemical speciation in theoriginal contaminated groundwater fromES16 was generally consistent with theprevious sampling episodes as well, exceptperhaps for 137Cs, which showed adominant anionic character (Figure 3.23,top). It was not known whether this was

67 67 NUR.EG/CR-6627

representative of the groundwater, as thetotal was -1050 pCi/L and theconcentration of the anionic species was250 pCi/L. The concentrations of Co-57,Nb-94 and Cs-134 in this groundwatersample were very low. The levels of thesethree nuclides were fairly close to thedetection limits, and these values were

negligible compared to the amount ofspikes used, thus these small amounts donot affect the speciation results. Theseinitial values were subtracted whensorption values were calculated. Theconcentration of all radionuclides werebelow detection limits in ES9(uncontaminated groundwater).

NUREG/CR-6627 668

<. 1.5

0.0

0.5

V- 00)Pa4) -0.5

UV Absorbance at 254 nmES16, May 13, run# 1, May 14

0 20 40 60 80Fraction No.

100 120 140 160

I Vertical ines show eluent changes

UV Absorbance at 254 nmES16, May13, run #2, May15

0.

.0U)0

.@Verieca lInes show elen ahan-ges 4 0 rC I 6

UV Absorbance at 254 nmESI6, May 13, run # 3, May 16

1.55.

04)

@2.00@2

.0

I

¶1 ~

0.5

0

-0.5

IB i

0 20 40 60 80Fraction No.

100 120 140 160

IVerfical lInes show eluent Et ngesJI m

Figure 3.19 UV absorbance chromatograms for successive runs of May 13, 1996 samplesfrom well ES16

69 69 NUREG/CR-6627

UV Absorbance at 254 nmES16, May 13, run # 4, May 221

1.55.

?A 0.o.5go

~-0.5I

UV Absorbance at 254 nm JE$16, May 13, run # 4, May 22 J

:1 itm • ql I

0 20 40 60 80Fraction No.

100 120 140 160

I Verlical ines show eluent changes I

UV Absorbance at 254 nm IES16, May 13, run # 5, May23

5.SUS.00

1.5

0.5

0

-0.5

i-I

I0 20 40 60 80

Fraction No.100 120 140 160

I Vertical Ines show eluent changes

Figure 3.20 UV absorbance chromatograms for successive runs (continued)

NUTREG/CR-6627 770

Co-60 CPM/kg loadedES16, May 13, run #2, May 15

55

45

S35

o 25

C. 15

5

-5

0 20 40 60 80Fraction No.

100 120 140 160

I Vertical Enes show eluent changes Im

Figure 3.21 Cobalt-60 chromatograms for successive runs

71 71 NUTREGICR-6627

.C{ -137 counts per hour200-

o• 150

IO 0 •.S100

0-50

c0 00AA

-50 -Vv0 20 40 60 80 100 120 140 160

Fraction No.

Co-60, 1173 keV counts per hour600500T

C 400 -

iD 300.3200

o 1000 .

-100 . . .0 20 40 60 80 100 120 140 160

Fraction No.

Vertical lines show changes in eluent

Figure 3.22 Cs-137 and Co-60 chromatograms for well ES16 1996 May 5 sample

NUREG/CR-6627 72

Table 3.5 Levels of radionuclides present in contaminated and uncontaminated groundwaters (measured),and levels of artificial tracers added to each solution (calculated).

ES16 contaminated groundwater(June 1996 sampling)

ES9 uncontaminatedgroundwaterRadionuclide

Initial amount tracerreported (pCi/L) added (pCi/L)

detection limit(pCi/L)*

traceradded(pCi/L)

detectionlimit

(pCi/L)**

Co-60 3080

Ru-106 460

Cs-137 1050

Pu-238 38

Pu-239 184

Am-241 5.4

Cm-244 0.54

-1.9 ;•y , 11-65

11-54 140-810

4.0-8.0 11-95

-0.05 0.3

-0.05 1.02E4 0.3

5-50 (yf 9,530 35-110 (y)t-0.05 (a)* 0.3 (ac)

-0.05 * 0.3

Artificial tracers

Co-57 4.1 1.42E4 -2.7 5.3E4 -8

Nb-94 8,210 2.7-5.4 5.68E4 -11

Cs-134 2.9 8370 1.4-14 5.43E4 1.4-54

*Based on a 1 L sample and a 3 h count for the gamma emitters, and a 16 h count for the alphas.**Based on a 0.2 L sample and a 3 h count for the gamma emitters, and a 6 h count for the alphas.

tBased on gamma spectrometry.t•ased on alpha spectrometry

73 73 NUREG/CR-6627

Speciation checks: after cation exchange processing ("anionic" species)Original nuclide speciation in ES-16

: 1000

.. 1000S 10

0.1

0.0ola 100

O00

.2o~

10

C

= 0.1

z

0.001

Co-60 Ru-106 Cs-1 37 Cm-244 Pu-238Nuclide

ES-16 anionic component (spiked) before contact

Co-60 Ru-106 Cs-137 Cm-244 Pu-238Nuclide

ES-16 anionic component (spiked) after contact

-1000

100

• 10C

0

. 0.1

z 0.01Nb-94 Cs-134 Am-241 Pu-2391240

Co-60 Ru-106 Cs-I137 Cm-244 Pu-238Nuclide

Total Anionic R71 Cationic1...J comnponentst • components

Figure 3.23 Speciation of nuclides and tracers in contaminated groundwater (from wellES16) before and after pretreatment through a cation exchange resin

NUR.EG/CR-6627 774

Tables 3.6 and 3.7 show the groundwatermajor ion chemistry of the ES-16 and ES-9samples, respectively, at various stages ofsample manipulation. The total concentrationof ions, before contacting with sand, was-1200 ,Eq/L for both ES-16 and ES-9samples, whereas the nanofiltered sampletotal ionic content decreased to -230 gEq/L.The ion concentrations increased by -5 to 20%after the groundwater had equilibrated withthe sand, except for Cl- ion, which increasedby -200-300%. The total ion concentrationsincreased from -230 to 740 gEq/L in thenanofiltered sample after contact with sand.Additions of radiotracers, dissolved in acids,increased the anion levels by -9 pEq/L (ES16samples) and -120 AEq/L (ES9 samples). ThepH did not change appreciably after additionof tracers (see Tables 3.6 and 3.7; the pHvalues should be compared to "after ionexchange processing"), and the changes indissolved ions probably did not affect sorption,assuming that ion exchange was the mainmechanism controlling sorption. In fact, thesand has a cation exchange capacity (CEC) of-1.2 mEq/100 g [Jackson and Inch, 1983],which exceeds the total cations present in

each flask by a factor of about five 2, or morein the case of the nanofiltered solution. Theadditions of ions from the spikes would notexceed the sand exchange capacity.

3.4.2 Kd Measurements Using Cationicand Anionic Radionuclide Species

Table 3.8 shows the sorption results obtainedfor the various types of groundwater sampletreatment. The 1 a refer to the standarddeviation among replicate measurements,except for 2 3 9 12 4 0pu and 241Ajn(a) in ES9uncontaminated groundwater, where twoseparate measurements are listed. The valuesobtained with 57Co, 94Nb and 134Cs for all thesamples, plus 239/24 0Pu and 24lAm for the ES9samples, in principle, should be equivalent for

2 For example, in the sorption experiments, 200 g of

sand were contacted with 800 mL of solution. Thesand has a total capacity of 2.4 mEq (1.2mEq/100g), and the solution has 0.8 L x (1200izEq/L)/2 for the cations, for a total content of-0.48 mEq, or an excess exchange capacity of -5.

all types of sample pre-treatments, becausethe radiotracers were all added at the last stepprior to equilibration with the sand. Thevalues obtained for the contaminants in ES16(60C0, 10 6Ru, 137Cs, 2 3 8Pu, 2 39 /24 0Pu, 24 1Ain and2 4 4Cm) represent the KI's for the separatespecies ("cationic" and "anionic") present inthe contaminated plume.

This experiment demonstrated quiteconvincingly that the Kd values of the nucidespresent in the contaminated aquifer (Co, Cs,Pu, Am), were much smaller than the KIvalues determined with commercial tracers ofthe same elements, on the same geologicalmaterial. The difference in Kd valuesdetermined by the two sets of tracers wasabout 2-3 orders of magnitude for Cs, up to 2orders of magnitude for Co and Am, and atleast a factor of -5 to -30 for Pu. Niobium-94was not detected in the ES16 groundwaterafter anionic pre-treatment, so the commercialtracer results cannot be compared to a naturalaquifer value. The high sorption value for Nbwas not surprising [Ticknor et al., 1996;Caron, unpubl.]. Two determinations eachusing Pu and Am commercial tracers arelisted in Table 3.8 for ES9, as only two valuesof the triplicates were available. This largeexperimental variability would lead torelatively high 1 sigma uncertainties for theaverage Kd values.

The levels of nucides in the cationic fractionwere detectable for only two nuclides, Co andCs. The Kd for cationic Co was 5.5, versus 0.8for the anionic form of the element, whereasthe "less than" value reported for Cs wouldsuggest similar Kd values. These values arestill well below the Kd's obtained with thecommercial tracers for these radionuclides.The Kd values obtained with the commercialtracers added to the solutions, were in thesame range as other values obtained for CRLsand in a separate study (Table 3.9), althoughour values are significantly lower than thosereported by Ticknor et al. [1996] for Co. Thevalues are comparable to the-recommendedvalues for a low-level radioactive waste(LLRW) near-surface disposal concept at CRL(Caron, 1995). Although the range obtainedwith the four different measurements with thetracers is large (-1 order of magnitudedifference), these values are clearly higher

75 75 NUREG/CR-6627

than the Kd's measured for the same elements(Co, Cs, Pu and Am) present in the plume. Thecase is not so clear for Co, for which in situ Kdvalues ranged from 0.5 to 130 mL/g in thesandy material of the same plume [Killey etal., 1984]; the values of the current study(contaminated groundwater versus addedtracer) are at both ends of this range.

The variations of Kd values obtained with thespikes in the four measurements do notappear to be related to the sample pre-treatment. One would expect thatnanofiltration with the RO unit (200 Daltons

NMWCO) would remove colloidal materialand some high molecular weight organics,both of which contribute to keepingradionuclides in solution or suspension, andthis would effectively give a lower sorptionvalue than the "true"one. However, this is notthe case, and it suggests that high molecularweight organics and/or colloidal matter did notplay an important role in sorption, at least forthe duration of this experiment.

NUREG/CR-6627 776

Table 3.6 Concentrations of major ions in ES16 contaminated groundwater at various stages of sample processing.

ES16 contaminated groundwater compositionConcentration (all concentrations in mg/L)

Ion Increase after Original filtered After cation exchange resin step After anion exchange resin stepTracer addition groundwater Column before after column before after

Eluent equilibration equilibration eluent equilibration equilibrationCa 2+ 1.72 ND 0.60 0.58 1.72 1.73 3.60Fe 3+ 1.56 0.58 0.35 0.07 ND ND NDK+ 1.40 0.50 0.30 1.20 1.70 1.70 2.50

Mg2+ 0.72 ND ND 0.16 0.74 0.72 1.00Na+ 6.80 13.90 13.60 13.60 6.70 6.80 9.00Si 5.80 5.70 5.70 6.20 5.60 5.70 6.70F- 0.01 0.28 0.31 0.29 0.35 0.05 0.06 0.10Cl- 0.1 1.25 1.32 1.32 3.12 24.95 24.15 26.01

NO- 0.4 0.26 0.48 0.10 0.35 ND ND 0.15S0 42- 8.46 8.72 8.77 14.80 ND ND 4.48

DIC (as C) 5.09 5.07 3.90 1.98 0.09 0.07 0.27DOC (as C) 4.17 3.49 4.03 5.03 9.80 9.67 10.47

PH 7.31t 6.28 6.80 6.88 7.34 4.13 4.19 5.974.14_1 1

iTo be compared to eluent from the cation exchange resin step.ITo be compared to eluent from the anion exchange resin step.

C)

0~~0~'t~J

Table 3.7 Concentrations of major ions in ES9 uncontaminated groundwater at various stages of sample processing

-.4

-4

ES9 uncontaminated groundwater composition(all concentrations in me[L)

Conc. Ion exchange processing KI determination Ki determination, after reverseIncrease after (for blanks purposes) "conventional batch" method osmosis processing

Ion tracer "nanofiltered batch" methodAddition Original after cation after anion Original before after Original before after

filtered exchange exchange filtered contact contact nanofiltered contact contactCa 2÷ 4.90 0.09 4.80 4.80 4.80 3.30 0.75 0.99 1.49Fe3 ÷ 2.00 1.77 0.01 1.89 1.74 0.07 0.02 0.03 0.01

K+ 1.90 0.35 1.40 1.50 1.30 1.90 0.60 1.30 1.50Mg2

+ 1.88 ND 1.86 1.90 1.88 1.06 0.46 0.52 0.44Na÷ 0.1 2.12 12.50 2.04 2.04 2.00 4.50 1.18 1.38 5.50Si 7.40 6.40 7.20 7.70 7.70 8.00 4.50 4.30 5.40F- 0.1 0.08 0.09 0.07 0.09 0.07 0.16 0.06 0.08 0.11Cl- 2.1 1.43 1.24 26.31 1.13 1.23 2.68 0.73 0.86 4.66

N03" 3.3 0.08 0.10 ND 0.06 ND 0.13 0.12 0.08 3.41S042- 10.39 8.97 ND 10.42 10.39 14.89 2.95 2.89 8.28

DIC (as C) 3.28 3.09 0.27 2.77 4.33 1.30 0.74 0.47 0.16DOC (as C) 5.32 5.51 17.60 4.84 3.11 3.38 1.74 1.40 1.57

pH 6.85t 6.88 7.52 4.22 6.79 6.89 7.02 6.10 6.45 6.496.69t .....

iTo be compared to "batch Kd determination, before contact".fTo be compared to "nanofiltered batch, before contact"

Table 3.8 Kd values obtained for the ambient radionuclides and the tracers for thedifferent sample manipulations

Sample source and treatment

Contaminated GW Contaminated GW Uncontaminated Uncontaminatedfrom ES16, after from ES16, after GW from ES9, GW from ES9,

Nuclide cation exchange anion exchange spiked wI tracers, spiked w/tracers,(anionic (cationic Batch sorption Batch sorption

components) components) (unaltered GW) (nanofiltered GW)Kd ±1lo Kd ±lra K ±1 o K ±1 o

60Co 0.8 0.1 5.5 0.2

'OGRu 0.7 1.0

137Cs 5.3 6.2 <8.6 -

23 8PU 25 2239/240 Pu 1 25 20 844239/ 240PU 212

241AMi (y)* 0. 956

241AMi (a).* 1 1.824~(*..................................189... ..

244CM 37 38

Artificial tracers57Co 352 10 32.8 5.4 >31.4 - 122 5

94Nb 103 5 1250 180 439 226 2440 270

134Cs >635 - >850 - 3820 1760 580 42

NOTE: Blank spaces mean that K& was not determined because the detection limits have beenreached, or the radioisotope was not present. A "-2 means "not applicable". The shaded areasindicate that these results are not available at the time the report was written.*: Am (Y) means that Kd was determined using the gamma results, Am (a) is based on alphameasurements.

1, 2: Kd determined by two different calculations using the same Co in Eq. (2) and two Ceq from twoduplicate solutions that gave inconsistent results.

79 79 NUREG/CR-6627

Table 3.9 Comparison of values obtained in this work with ES16 groundwater (pre-treated with cation exchange resin to isolate the anionic species) and with artificialtracers vs. other works.

Element Kd (mL/g)

This work, This work, Compiledcontaminated "batch" with Ticknor et al. [1996]* databases for

plume (anionic commercial sand**species) tracers

Am 1.8 159-956 1900(AB)

Cm 37 - 4000(B)

Co 0.8 -32-352 638; 982 (n = 2) 600,B)

Cs 5.3 580-3820 2490; 2450 (n = 2) 280(,B)Nb - 103-2440 892; 740 (n =2) 750(A)

Pu 25 120; 844 306 - 563 (n = 4) 550(AB)Ru 0.7 1 55(B)

*Ticknor et al. [1996]: values obtained experimentally with CRL sand.**Compiled databases: A, [Caron, 1995]; B, [Sheppard and Thibault, 1990]

3.4.3 Speciation Checks During KdMeasurements

Figures 3.24 to 3.28, show the nuclidespeciation for the different sorptionexperiments before and after equilibrationwith CRL sand. Figures 3.27 and 3.28 showthe speciation in ES16 and ES9 groundwater,respectively, for the solutions stored underambient room conditions (in the dark) for a 7-day period, which was the same duration asthe equilibration with sand. The initialspeciation of the radionucides is also shownin ES16 contaminated groundwater.In all situations, the artificial tracers 57Co and134CS were almost exclusively cationic,

whereas 94Nb, 239124 0Pu and 24lAin were mostlyanionic. Among the species initially presentin the contaminated plume, 60Co, 238Pu,239/240Pu, 2 4 1

Am and 244Cm were mostly anionic,only negative species of 106Ru have beendetected, and 1

37Cs had a strong anioniccomponent, which is unusual. Thesespeciation results are consistent with previousstudies done at CRL [Killey et al.,

1984; Cooper and McHugh, 1983; Cooper andMattie, 1993; Champ and Robertson, 1986;Champ et al., 1985].

After treatment with the cation exchangeresin, and upon equilibration of contaminatedES16 groundwater with CRL sand (Fig. 3.23),6 0 C0, 106Ru, 24 1Am and 2 4 4Cm retained their

initial relative speciation, whereas the 137Csrelative speciation had shifted to displayapproximately the same amount betweentotal, anionic and cationic species, just beforeequilibration. The Pu speciation results wereunclear; their relative speciation seemedconsistent, but the Pu levels droppedsignificantly after the ion exchange step.

After treatment with the anion exchangeresin, the 106Ru and the anionic 137Cs levelsdropped below detection limits. The negative137Cs species were not detected after

equilibration with sand.

In Figure 3.24, a significant proportion of 60Co

decreased, especially the anionic species,which dropped from 2620 to 324 pCi/L, aftertreatment with anion exchange resin. Itremained at -324 pCi/L after equilibration. It

NUREG/CR-6627 880

is unclear whether the small KI observed forCo (5.5 mL/g) could be attributable to cationicor neutral species present in the "cationicfraction". One would expect that cationicspecies, if present, would exhibit a muchhigher Kd value.

The speciation results were as expected for the"conventional batch" and "conventional batch-nanofiltered" experiments for 57Co, 94Nb, 134Cs,

and 239/240Pu (Figs 3.25 and 3.26). The resultsfor 24 1Am were more ambiguous, showing thepredominance of anionic species beforeequilibration (Fig. 3.25) and the predominanceof cationic species after equilibration.

The aqueous speciation in the solutions, storedunder room conditions, did not changeappreciably for the 7-day equilibration period(Figs. 3.27 and 3.28). In a few cases, a cationicor anionic component appeared for the Csisotopes or 94Nb, but this was insignificantcompared to the total (note the log scales inthe figures). The amounts of Am and Pu inES-9 seemed to have increased afterequilibration (Fig. 3.28); the reason for thisanomaly is presently unknown.

In all experiments, the spiked tracers 57Co,94Nb and 134Cs were not detected in the initialsolutions (except in the initial ES-16 solution),or their values were probably not significant,as they were close to the detection limit.These tracers were not detected after ionexchange processing. After addition of thesespikes, both before and after equilibration,

almost all of the 57Co and 134Cs were cationic,

whereas 94Nb was anionic. In all cases afterequilibration, the speciation of 60Co wasalways different from that of 57Co, whichsuggests that the 60Co aqueous species werenon-labile for the duration of the experiment,as no isotopic exchange seemed to have takenplace. The data on 137Cs was too limited andinconsistent to make a judgement.

Ruthenium was probably present in thegroundwaters as RuO4- [Champ et al., 1984].The case is unclear for Nb. EH-pH diagramssuggest the dominance of the pentavalentoxidation ion, Nb(OH)50, which would becomethe negative species Nb(OH)6- at pH 7.4. Baesand Maesmer [1976] have mentioned thataqueous Nb species would be a pH-drivenequilibrium between the polynuclear speciesNb6017(OH)26-, Nb60ls(OH)7-, and Nb60l98-.

The polynuclear oxyl ion Hx(Nb6O19)8x(aqa hasalso been listed (Cotton and Wilkinson, 1980;Baston et al., 1992; and references therein). Itis suspected that this complex polymeric ion isthe starting point of the formation of colloidalmaterial. In a sand column experimentpreviously done at CRL (Caron, unpublished),small losses of Nb were observed, and the bulkof the Nbcontacted with the column was essentiallyimmobile. The calculated Kd was higher than7800 mL/g.

81 81 NUREG/CR-6627

Speciation checks: after anion exchange processing ("cationic" species)Original nuclide speciation in ES-16

C

0

V.

(U

Nuclide

Nuclide speciation in ES-16 before equilibrationA

1000

C1000

0.1

0.01z

Nuclide

AjNuclide speciation in ES-16 after equilibration

1000 1t1

. 1000

10

0u 10

0.1

z 001zCo-57 Co-60 Nb-94 Ru-106 Cs-134 Cs-137

Nuclide

Tota-l Anionic rTE Cationic10 .Dcomponents L-components

Figure 3.24 Speciation of nuclides and tracers in contaminated groundwater (from well ES-16) before andafter pretreatment through an anion exchange resin column. The S7Co, 94Nb, and 1MCs were thecommercial tracers

NUREG/CR-6627 882

Conventional "batch" Kd determination:unaltered groundwater

Nuclide speciation in ES-9, after addition of tracers (before equilibration)

:rooo

o 100

z 0.1

1000

S10-CD

z

Co-57 Nb-94 Cs-134 Am-241-g Pu-239/240Co-60 Ru-106 Cs-137 Am-241-a

Nuclide

Nuclide speciation in ES-9, after equilibration

0.1

Co-60 Ru-106 Cs-137 Arn-241-aNuclide

1T0 a compoen t s I I opoet

Figure 3.25 Nuclide speciation (tracers only) in the samples for the conventional "batch"Kd determinations before and after equilibration with CRL sand. Note: "Am-241-g" refers to the results obtained with gamma spectrometry, and "Am-241-a" were obtained with alpha spectrometry.

83 NUREG/CR-6627

Conventional "batch" Kd determination:nanofiltered grounawater

Nuclide speciation in ES-9 (all spikes)Before equilibration

001

001

Co-57 Co-60 Nb-94 Cs-134 Cs-137 Am-241Nuclide

Nuclide speciation in ES-9 (all spikes)After equilibration

2 1000

Cr 100

N 10

.= IS 0.10

0.01 ........Co-57 Co-60 Nb-94 Cs-134 Cs-137 Am-241Nuclide

-.. r--pAnionic •I CationicntTotal I I components I components

Figure 3.26 Nuclide speciation (tracers only) in the samples for conventional "batch" Kddeterminations before and after equilibration with CRL sand (after nanofiltration of theuncontaminated groundwater to remove organics and colloidal particles). Note: "Am-241results were obtained by gamma spectrometry.

NUREG/CR-6627 84

Nuclide speciation stability during the experiment:contaminated groundwater from ES-1 6 with added tracers

ES-16 anionic solution stability (day 1)

2r

C0

C0

0

z

C0

C0

1000

100

10

I

0.1

0.01

1000

100

10

1

0.1

0.01

Co-57 Nb-94 Cs-134 Am-241 Pu-239/240Co-60 Ru-I 06 Cs-1 37 Cm-244 Pu-238

Nuclide

ES-16 anionic solution stability (day 7)

Co-57 Nb-94Co-60

Cs-1 34 Am-241 Pu-239/240Ru- 106 Cs- 137 Cm-244 Pu-238

Nuclide

*Total Dnini F-Jaýn

Figure 3.27 Nuclide speciation stability in aqueous solutions during the experiment(ES16 with tracers).

85 85 NUREGICR-6627

Nuclide speciation stability during the experiment:uncontaminated groundwater from ES-9

Speciation stability in ES-9 for tracers (dayl)

1000

0 100

Cu0 10

0

U

0.1

- 10000"

.o 100

Cu0 10-C

0

z

Co-57 Nb-94 Cs-I 34Co-60 Ru-I 06

Nuclid(

Am-241-g Pu-239/240Cs-1 37 Am-241 -a

0.1Co-57 Nb-94 Cs-1 34 Am-241-g Pu-239/240

Cs-137 Am-241-aCo-60 Ru-106Nuclide

Total -lAnion°iCents Cationic1 .o Lcomponents

Figure 3.28 Nuclide speciation stability on aqueous solutions during the experiment(ES16 uncontaminated groundwater, with tracers).

NUREG/CR-6627 886

4 Summary and Conclusions

The results of this work show changes incolloid and chemical speciation ofradionuclides as they migrate down thegroundwater flow path from the Chemical Pitto the nearby wetland. Dissolved speciesdominate throughout the system. This islikely due to complexation of the radionuclidesin this system, since there is much evidencefor natural complexation (Killey et al., 1984,Cooper and McHugh, 1983). While someradionuclides, such as 6°Co and 244Cm showeddeclining concentrations in the dissolvedphase as the plume moves away from the pit,others, such as 106Ru and Pu showed increasedconcentrations in the wetland.

The anion exchange chromatograms producedfrom the soluble (<10,000 NMWL) fractionsfrom each well show changes in speciationdown the flowline. Cobalt-60 showed the mostcomplex speciation with at least 7 speciespresent. This may be partly due to the factthat the elution scheme had been optimizedfor 6OCo (Cooper et al., 1993). The peaksshowed variations in intensity in the samplesfrom the three wells, indicating that chemicalforms were evolving in the aquifer.

There was an interesting trend in speciationof the actinide elements. The lower elementsin the actinide series were strongly retainedby the anion exchange resin and were mostlyremoved by elution with HC1, probablythrough destabilization of the complexes. Thehigher actinides, which were less stronglyretained by the resin, were eluted earlier bythe KCl. Fewer of their complexes remainedon the columns long enough to be eluted inHCI. This trend probably indicated that thelower actinides were in higher oxidation states(i.e. V and VI) and consequently were lesshighly positively charged (i.e. MO2+ andMO2-) than Am and Cm, which would have atriple positive charge. When complexed withnegatively charged ligands, the species of thelower actinides would have a higher netnegative charge than similar species of thehigher actinides, Species with a higher netnegative charge would be more strongly

retained by the anion exchange resin thanthose with a lower net negative charge.

Anion chromatography measurements madein 1993 and 1996 showed that theconcentration of dissolved 60Co in the plumehad declined substantially since use of the pitshad been discontinued in 1996. The speciationof anionic forms of 60Co has also changed.Some species declined with time and otherspecies appeared in the plume groundwater.

The UV chromatograms showed that organicspecies were also concentrated andchromatographically separated on the anionexchange columns. Peaks occurred in similarregions as peaks in the 60Co chromatogram.The peaks that eluted in KOl were due tospecies which exhibited a broad absorbanceacross the UV spectrum. Species that elutedin HCO showed broad peaks at 240 nm and 340nm. The organics in uncontaminatedgroundwater gave generally similarchromatograms and UV spectra. This showedthat most of the organic species were ofnatural origin.

The anion exchange chromatograms from theMay 15 sample showed that thechromatographic separations were quiteconsistent and that the technique could beused to measure small changes in speciationwith time. The technique also showed thatanionic forms of 137Cs were present in thegroundwater plume.

A novel approach for determining the sorptioncoefficient Kd of nuclides with geologicalmaterial was used in this study. Theapproach consisted of passing separatealiquots of contaminated groundwaterthrough a cation exchange resin (to removecationic species of radionuclides) and an anionexchange resin (to remove anionic species ofnuclides). Each isolated solution was thencontacted with uncontaminated sand toseparately determine Kd values of the anionicspecies and cationic species of nuclides. Thevalues were compared with Kds measured,after spike additions of different isotopes of

87 87 NUREG/CR-6627

the same set of elements, to determine theimportance that aqueous speciation has onsorption. These values were also compared toa conventional "batch" Kd determination withonly tracers added.

The results showed that Kd's for spikeadditions of Co, Pu and Am were at least oneorder of magnitude higher than the values forthe same nuclides present in the contaminantplume. In the case of Cs, the difference wastwo orders of magnitude or more. Thisdemonstrated that using tracer solutions todetermine K& values in the laboratory couldoverestimate sorption compared to the fieldconditions.

The nuclide speciation, defined as the relativeabundance of dissolved negative species

-versus positive species, was stable during theexperiment. The 5 7Co speciation, afterequilibration, was similar to the originalspeciation, but different from that of 6°Co atany point, which suggested that the speciesresponsible of complexing 6oCo were stableand non-exchangeable.

Performance assessment modelers would beable to use the "field" Kd values developed inthis study for modeling radionuclide transportat similar LLW disposal sites. More accuratepredictions of the "first-arrival" times of themobile complexed organo-radionuclide speciescould then be made.

NUREG/CR-6627 888

5 References

ASTM. 1984. Distribution Ratios by the Short-Term Batch Method, American Society forTesting and Materials, Philadelphia, PA,Method D 4319-83.

Baes, C.F., and R.E. Maesmer. 1976. TheHydrolysis of Cations. New York, NY: Wiley-Interscience.

Baston, G.M.M., JA. Berry, A.K. Littleboy,and N.J. Pilkington. 1992. "Sorption ofActivation Products on London Clay andDungeness Aquifer Gravel". Radiochim. Acta58/59: 225-233.

Boggs, S., D. Livermore, and M.G. Seitz. 1985.Humic Substances in Natural Waters andTheir Complexation with Trace Metals andRadionuclides: A Review. ANL-84-78,Argonne, Illinois.

Bondietti, E.A. 1982. "Mobile Species of Pu,Am, Cm, Np, and Tc in the Environment." In:'Environmental Migration of Long-LivedRadionuclides, IAEA-SM-257/42,International Atomic Energy Agency, Vienna,Austria.

Buddemeier, R. W. and J. R. Hunt. 1988."Transport of Colloidal Contaminants inGroundwater; Radionuclide Migration at theNevada Test Site". Applied Geochem. 3: 535-548.

Cabaniss, S.E., and M.S. Shuman. 1988."Fluorescence Quenching Measurements ofCopper- Fulvic Acid Binding." Anal. Chem.60:2418-2412.

Caron, F., CA. Delmastro, and M.K. Haas.1997. Determination of Sorption Coefficients(Kd) for 6OCo and 94Nb in Chalk River Sand,Sand/Dochart Clay and Sand /ClinoptiloliteMixtures. Unpublished data, file number0501-0620 Niobium, Atomic Energy ofCanada, Ltd., Chalk River Laboratories,Chalk river, Ontario, Canada.

Caron, F. 1995. Partition Coefficients forBackfill and Geosphere in the PerformanceAssessment Calculations of IRUS, AECL

Report, TR-660, Atomic Energy of Canada,Ltd., Chalk River Laboratories, Chalk River,Ontario, Canada.

Champ, D.R., and D.E. Robertson. 1986."Chemical Speciation of Radionuclides inContaminant Plumes at the Chalk RiverNuclear Laboratories. In: Speciation ofFission and Activation Products in theEnvironment. Bulman, RA., and J.R. Cooper(Eds.). Elsevier Appl. Sci., London, UK.

Champ, D.R., J.L. Young, D.E. Robertson, andK.H. Abel. 1985. "Chemical Speciation ofLong-Lived Radionuclides in a ShallowGroundwater Flow System." Water Poll. Res.J. Canada 19, 2.

Choppin, G.R. 1988. "Humics andRadionuclide Migration." Radiochim. Acta44/45:23-28.

Cooper, E.L., and J.F. Mattie. 1993. Report onSpeciation Studies of Radionuclides in theChemical Pit Plume, AECL Report, RC-978,Atomic Energy of Canada, Ltd., Chalk RiverLaboratories, Chalk River, Ontario, Canada.

Cooper, E.L. and J.O. McHugh. 1983."Migration of Radiocontaminants in aForested Wetland on the Canadian Shield:Nuclide Speciation and Arboreal Uptake".Science Tot. Environ., 28, 215-230.

Cooper, E.L., P. Vilks, M.K. Haas and J.F.Mattie. 1993. Study of RadionuclideSpeciation in Groundwater Using AnionExchange Resins. AECL Report, RC-1150,Atomic Energy of Canada, Ltd., Chalk RiverLaboratories, Chalk River, Ontario, Canada.

Cooper, E.L., M.K. Haas and J.F. Mattie.1995. "Studies of the Speciation of Plutoniumand Other Actinides in Natural GroundwaterUsing Anion Exchange Resin". Appl. Radiat.Isot., 46, 1159-1173.

89 89 NUREG/CR-6627

Cooper, E.L. 1999. Personal Communication,Chalk River Laboratories, Chalk River,Ontario, Canada.

Cotton, F. A., and G. Wilkinson. 1980.Advanced Inorganic Chemistry (4th Edition).New York, NY: J. Wiley and Sons.

Ewald, M., C. Belin, and P. Berger. 1983."Corrected Fluorescence Spectra of FulvicAcids Isolated from Soil and Water." Environ.Sci. Technol. 17:501-504.Florence, T.M. 1982. "The Speciation of TraceElements in Waters." Talanta 29:345-364.

Jackson, R.E., and K.J. Inch. 1983."Partitioning of Strontium-90 Among Aqueousand Mineral Species in a ContaminatedAquifer". Environ. Sci. Technol. 17: 231-237.

Jensen, B.S., and H. Jensen. 1988. "ComplexFormation of Radionuclides with OrganicLigands Commonly Found in Groundwater."Radiochim. Acta 44/45:45-49.

Kersting, A. B., D. W. Efurd, D. L. Finnegan,D. J. Rokop, D, K, Smith, and J. L. Thompson.1999. "Migration of Plutonium in GroundWater at the Nevada Test Site", Nature 397,Jan. 7: 56-59.

Killey, R.W.D., J.0. McHugh, D.R. Champ,E.L. Cooper and J.L. Young. 1984."Subsurface Cobalt-60 Migration from a Low-Level Waste Disposal Site". Environ. Sci.Technol., 18(3), 148-157.

Killey, R.W.D., and J.H. Munch. 1984.Subsurface Contaminant Transport from theLiquid Disposal Area, CRNL: (1)Hydrogeology and Tritium Contaminationnear the Chemical Pit. AECL-7691, AtomicEnergy of Canada, Ltd., Chalk RiverLaboratories, Chalk river, Ontario, Canada.

Killey, R.W.D., and J.F. Mattie. 1993.Carbon-14 in the Vicinity of WasteManagement Area C: Results of a ScopingStudy. AECL-10795, Atomic Energy ofCanada, Ltd., Chalk River Laboratories,Chalk River, Ontario, Canada.

Killey, R.W.D., and J.H. Munch. 1993.Performance of the Impermeable Cover at AreaC - Status to 1992. TR-506, Atomic Energy ofCanada, Ltd., Chalk River Laboratories,Chalk River, Ontario, Canada.

Killey, R.W.D., R.R. Rao, and S. Eyvindson.1993. Radiocarbon Speciation andDistribution in an Aquifer Plume andGroundwater Discharge Area, Chalk River,Ontario. RC-1135, Atomic Energy of Canada,Ltd., Chalk river Laboratories, Chalk River,Ontario, Canada.

Levesque, M. 1972. "Fluorescence and GelFiltration of Humic Compounds." Soil Sci.113(5):346-353.

Lieser, K.H., B. Gleitsmann, S. Peschke, andT.H Steinkopff. 1986. "Colloid Formation andSorption of Radionuclides in NaturalSystems." Radiochim. Acta 40:39-47.

Lieser, K.H., A. Ament, R. Hill, R.N. Singh, U.Stingl, and B. Thybusch. 1991. "Colloids inGroundwater and Their Influence onMigration of Trace Elements andRadionuclides." Radiochim. Acta 49:83-100.

McCarthy, J.F. 1988. Role of ColloidalParticles in the Subsurface Transport ofContaminants. DOE/ER-0384, U.S.Department of Energy, Washington, D.C.

McCarthy, J.F., and J.M. Zachara. 1989."Subsurface Transport of Contaminants."Environ. Sci. Technol. 23:496-502.

Olsen C.R., P.D. Lowry, S.Y Lee, I.L. Larsen,and N.H. Cutshall. 1986. "Geochemical andEnvironmental Processes AffectingRadionuclide Migration from a Formerly UsedSeepage Trench." Geochim. Cosmochim. Acta44/45:165-170.

Patterson, R.J., and T. Spoel. 1981."Laboratory Measurements of the StrontiumDistribution Coefficient for Sediment from aShallow Sand Aquifer." Water Resources Res.17(3):513-520.

NUREG/CR-6627 990

Penrose, W.R., W.L. Polzer, E.H. Essington,D.M. Nelson, and KA. Orlandini. 1990."Mobility of Plutonium and AmericiumThrough a Shallow Aquifer in a Semi-aridRegion". Environ. Sci. Technol. 24: 228-234.

Petit, J.C. 1990. "Migration of Radionuclidesin the Geosphere: What Can We Learn fromNatural Analogues?", Radiochim. Acta51:181-188.

Pickens, J.F., R.E. Jackson, and K.J. Inch.1981. "Measurement of DistributionCoefficients using a Radial InjectionDual-Tracer Test." Water Resources. Res.17(3):529-544.

Rai, D., and R.J. Serne. 1978. Solid Phasesand Solution Species of Different Elements inGeologic Environments. PNL-265 1, PacificNorthwest Laboratory, Richland, Washington.Ramsay, J.D.F. 1988. "The Role of Colloids inthe Release of Radionuclides from NuclearWaste." Radiochim. Acta 44/45:165-170.

Rao, R.R., and E.L. Cooper. 1995. "Separationof Low Levels of Actinides by SelectiveOxidation/Reduction and Co-Precipitationwith Neodymium Fluoride". J. Radioanal.Nucl. Chem. (Art.) 197: 133-148.

Riley, R.G., and J.M. Zachara. 1992. ChemicalContaminants on DOE Lands and Selection ofContaminant Mixtures for Subsurface ScienceResearch. DOE/ER-0547T, U.S. Departmentof Energy, Washington, D.C.

Robertson, D. E. 1986. "Evaluation of a LargeVolume Water Sampling Technique forDetermining the Chemical Speciation ofRadionuclides in Groundwater". In:Speciation of Fission and Activation Productsin the Environment. Bulman, R.A., and J.R.Cooper (Eds.). Elsevier Appl. Sci., London,UK.

Robertson, D.E., M.P. Bergeron, D. Holford,K.H. Abel, C.W. Thomas, D.A. Myers, D.R.Champ, R.W.D. Killey, D.L. Moltyaner, J.L.Young, and T. Ohnuki. 1989. Demonstrationof Performance Modeling of a Low-Level WasteShallow-Land Burial Site. NUREG/CR-4879vol. 2, U.S. Nuclear Regulatory Commission,Washington, D.C.

Robertson, D.E., and R.W. Perkins. 1975."Radioisotope Ratios in Characterizing theMovement of Different Physical and ChemicalSpecies through Natural Soils." In: IsotopeRatios as Pollutant Source and BehaviorIndicators, International Atomic EnergyAgency, Vienna, Austria.

Robertson, D.E., M.P. Bergeron, D. Holford,K.H. Abel, C.W. Thomas, DA. Myers, D.R.Champ, R.W.D. Killey, D.L. Moltyaner, J.L.Young, and T. Ohnuki. 1989. Demonstration ofPerformance Modeling of a Low-Level WasteShallow-Land Burial Site. NUREG/CR-4879vol. 2, U.S. Nuclear Regulatory Commission,Washington, D.C.

Robertson, D. E., A. J. Schilk, K. H. Abel, E. A.Lepel, C. W. Thomas, S. L. Pratt, E. L. Cooper,P. Hartwig, R. W. D. Killey. 1995. "ChemicalSpeciation of Radionuclides Migrating inGroundwaters". J. Radioanal. Nuc. Chem.194:237-252.

Schilk, A. J., D. E. Robertson, K. H. Abel, E. L.Cooper, R. W. D. Killey, P. G. Hartwig, C. W.Thomas, S. L. Pratt, P. Vilks, J. F. Mattie, M.K. Haas, E. A. Lepel,, R. Matzner, and K.King. 1996. The Role of Organic Complexantsand Microparticulates in the FacilitatedTransport of Radionuclides. NUREG/CR-6429,U. S. Nuclear Regulatory Commission,Washington, D. C. 20555-0001.

Sheppard, M.I., and D.H. Thibault. 1990."Default Soil Solid/Liquid PartitionCoefficients, Kds, for Four Major Soil Types: aCompendium". Health Phys. 59: 471-482.

Summers, R.S., P.K. Cornel, and P.V. Roberts.1987. "Molecular Size Distribution andSpectroscopic Characterization of HumicSubstances." Sci. Total Environ. 62:27-37.

Ticknor, K.V., L. Boisvenue, and T.T.Vandergraaf. 1996. The Sorption ofRadioelements on Sand from the Chalk RiverSite, AECL Report, TR-755, Atomic Energy ofCanada, Ltd., Chalk river Laboratories, ChalkRiver, Ontario, Canada.

91 91 NtTREG/CR-6627

Warwick, P., A. Hall, P. Shaw, J.J.W. Higgo,G.M. Williams, B. Smith, D. Haigh, and D.Noy. 1991. "The Influence of Organics in FieldMigration Experiments (Part 2)." Radiochim.Acta 52/53:465-471.

Welch, S.J. and R.W.D. Killey. 1997. SoilCoring and Bitumen Block Sampling inInfiltration Lysimeter 5. AECL Report, TR-689 (in prep), Atomic Energy of Canada, Ltd.,Chalk River Laboratories, Chalk River,Ontario, Canada.

Williams, G.M., J.J.W. Higgo, MA. Sen, W.E.Falck, D.J. Noy, G.P. Wealthall, and P.Warwick. 1991. "The Influence of Organics inField Migration Experiments (Part 1)."Radiochim. Acta 52/53:457-463.

Wobber, F.J. 1991. Subsurface ScienceProgram: Program Overview and ResearchAbstract.- DOE/ER-0432, U.S. Department ofEnergy Washington, D.C.

Lirn, S. 0., K. J. King, M. Benz, W. G.Evenden, D. E. Robertson. 1999.Environmental Dynamics of Carbon-14 Near aLow-Level Radioactive Waste Burial Ground.NUREG/CR-6587, U. S. Nuclear RegulatoryCommission, Washington, D. C.

NUREG/CR-6627 92

Appendices

Appendix 1

Instructions for Preparation of Chromatographic ColumnsVersion 2, 96-02

Materials:3/8" O.D. polyethylene tubing3/8" tubing to 1/4"-28 th chromatography end fittingsSlurry reservoirporous polyethylene frits- 3/8" diameter (cut from larger discs with leather punches)AG MP-1 resin 200-400 meshdouble distilled water(purge with He for 5 minutes)Omnifit 2-way valvelengths of Teflon tubing with 1/4"-28 th fittings

Equipment:Alcott model 760 HPLC pump400 to 600 mL beakers.disposable plastic pipettes

Method:

1) Cut a length of polyethylene tubing: 7 mm for guard columns, 25 mm for separation columns. Thelonger columns can be straightened if they are first heated in an oven(don't melt them).

2) Fit an end fitting on one end of the column. Replace the frit if need be.

3) Fit an end fitting to the other end of the column. Replace the frit if need be.

4) Remove the top fitting, but leave the nut on the tubing. Attach the slurry reservoir to the top of thecolumn.

5) Attach a 2-way Omnifit valve to the bottom of the column. Fit a length of Teflon tubing into the valveand insert the other end into a beaker. Clamp the assembly vertically to a stand with the valve at thebottom.

6) Weigh out approximately 10 g of AG MP-1 resin into a plastic bottle and slurry it with about 50 mLof distilled water. If fines are floating on the top of the water, then decant them off. Using a plasticpipette, transfer portions of resin slurry to the column (note: experience has been that the slurry willhave to be slightly pressurized initially with a syringe to wet the frits and establish gravity flow.Alternatively, wet the frits by forcing water through them prior to assembling onto the columns.).Allow the water to drain between additions, but do not allow the liquid level to fall below the top of thebed. Continue adding resin until the resin bed is slightly above the top of the polyethylene tube.

7) Fill the slurry reservoir with distilled water and allow to drain almost to the resin bed under gravityflow. If the resin bed is still above the top of the column then shut off the Omnifit valve. If the resinbed has compacted to below the top of the column, then add more slurry and repeat the packing withdistilled water under gravity flow.

93 93 NUREG/CR-6627

8) Remove the slurry reservoir and replace it with the end fitting. During assembly of these fittingsdistilled water will have to be added to displace air. Connect the adapter to the pump with Teflontubing(purge the air out of the tubing before connecting).

10) Open the Omnifit valve and pump several hundred mL of distilled water through the column at aflow rate of 10 mL/minute.

11) Stop the pump and close the Omnifit valve. Remove the end fitting from the top of the column. Ifthe bed has packed down below the top of the column, then refit the slurry reservoir and carefully addmore resin slurry to top it up. Drain away excess water through the Omnifit valve. Continue addingslurry until the bed is at or above the top of the tubing.

12) Fill the upper end fitting with water and fit it onto the column. This is the inlet end of the column.If the column is to be stored, then fill the inlet with water and fit a plug part way into the end fitting. Ifthe valve is removed from the outlet, then it will have to be similarly plugged.

Appendix 2

Procedure for Chromatographic Separation, 96-05

Instructions for Chromatographic Separation, 96-05

1) Set up the apparatus as shown in the figure, but leave out the Nal flow cell and use 0.5 M KCI for thegradient. Use the gradient programmer instead of the system shown in the figure. Set up the UVdetector between the columns and the fraction collector.

2) De-gas 4 L of ES16 water by purging with helium for 5 minutes. There is a tank of helium in JimYoung's lab by the hall tree.

3) Connect the tube to the groundwater reservoir to position 1 of the Scanivalve.

4) Load columns overnight at 3 mLmin.

5) Rinse the columns with 20 mL (6.5 min.) of distilled water.

6) Begin collecting fractions and start recording the signal from the LUV detector. Rinse the columnswith 30 mL (10 min.) of pH 4.5 HCL.

7) Elute the columns with a KCI gradient: 85 mrin. to 50% 0.5 M KCI.

8) Elute the columns with a KCI gradient: 40 min. to 100% 0.5 M KC1.

9) Elute the columns for 60 min. with 100% 0.5 M KCI.

9a) Elute the columns with a 0-100% 0.1 M HCl gradient for 20 minutes.

10) Elute the columns for 50 min. with 100% 0.1 M HCI.

11) Rinse the columns with double distilled water for 90 min. and shut down the system. The rinse canbe collected in a bottle and discarded.

12) Remove the samples from the racks, cap and number them, and store in the coldroom.

NUREG/CR-6627 994

13) Count the samples on the Packard auto gamma counter and return to storage in the coldroom. Tryto minimize the amount of time that the samples are out of the coldroom.

Elution Scheme

Eluentdistilled waterpH 4.5 HC1gradient to 0.25 M KC1gradient to 0.5 M KC10.5 M KC1gradient to 0.1 M HC10.1 M HC1distilled water

Volume(mL)203025512018060270150

Time(min)6.510854060209050

Position1234567

Scanivalve Ports

Connectiongroundwater reservoirdistilled waterpH 4.5 HCIgradient programmer0.5 M KC10.1 M HCIdistilled water

ins9605a.doc

95 95 NUREG/CR-6627

NRC FORM 335 U.S. NUCLEAR REGULATORY COMMISSION 1. REPORT NUMBER(2189) (Assigned by NRC, Add Vol, Supp., Rev.,N20C 1202 BIBLIOGRAPHIC DATA SHEET and Addendum Numbers, if any.)

(See instructions on the reverse)

2. TITLE AND SUBTITLE NUREG/CR-6627

The Role of Organic Complexants and Colloids in the Transport of Radionuclidesby Groundwater - Final Report 3. DATE REPORT PUBLISHED

MONTH YEAR

January 20004. FIN OR GRANT NUMBER

L19355. AUTHOR(S) 6. TYPE OF REPORT

D. E. Robertson, C. W. Thomas, S. L. Pratt, K. H. Abel, E. A. Lepel, A. J. Schi Ik (a) TechnicalE. L. Cooper, F. Caron, K. J. King, R. W. D. Killey, P. G. Hartwig, J. F. Matti e, M. K. Haas, (b)E. Romaniszyn, M. Benz, W. G. Evenden (b) 7. PERIOD COVERED (nclusive Dates)

S. 0. Link (c)P. Vilks (d)

8. PERFORMING ORGANIZATION - NAME AND ADDRESS (I/NRC, provide Division, Ofhce or Region, U.S. Nuclear Regulatory Commission, and mailing address: if contractor,provide name and mailing address.)

(a) Pacific Northwest National Laboratory, Box 999, Richland, WA 99352, (b) Chalk River Laboratories, Chalk River, Ontario KOJ1JO,

Canada, (c) Washington State University at Tri-Cities, Richland, WA 99352, (d) Whiteshell Laboratories, Whiteshell, Manitoba ROE1LO,

Canada

9. SPONSORING ORGANIZATION-NAME AND ADDRESS (If NRC. type 'Same as above~ if contractor, pro wide NRC Division, Office or Region. U.S Nuc/ear Regulatory Commission.9. SPONSORING ORGANIZATION - NAME AND ADDRESS (If NRC, "ap 'Same as above', dfcontractor, provide NRC Division, Office or Region. U.S. Nuclear Regulatory Commission.

and mailing address.)

Division of Risk Analysis and ApplicationsOffice of Nuclear Regulatory ResearchU.S. Nuclear Regulatory CommissionWashington, DC 20555-0001

10. SUPPLEMENTARY NOTES

F. O'Dnnnell. NIRC Project MRn~oerC Proiect Manarlermill

11. ABSTRACT (200 words or less)

This final report describes the results of field and laboratory work conducted during the last two years of the project.These studies will provide performance assessment modelers and regulators with additional information to better assessthe rates and mechanisms of radionuclide transport from LLW disposal facilities that become infiltrated with water.The field studies were conducted at two low-level radioactive waste (LLW) management areas within the boundaries of

the Chalk River Laboratories, Chalk River, Ontario, Canada, under a cooperative arrangement with the Atomic Energyof Canada, Ltd. At these locations, several well-defined, slightly radioactive groundwater contaminant plumes havedeveloped over the past 25 to 40 years, providing excellent field sites for conducting studies of radionuclide transport by

groundwater. These studies have primarily addressed: 1) characterization of the sub-surface geochemical environment

near the disposal facilities, 2) identification and quantification of the migrating radionuclides and their chemicalspeciation in groundwater, 3) the sorption behavior (Kd measurements) of the mobile radionuclide species (cationic and

anionic) onto site soils, 4) identification of colloidal radionuclides, and 5) the environmental dynamics of ' 4C in thevicinity of a solid LLW disposal facility (published earlier as NUREG/CR-6587).

12. KEY WORDS/DESCRIPTORS (List words or phrases that will assist researchers in locating the report.) 13. AVAILABILITY STATEMENT

radionuclide transport unlimitedgroundwater 14. SECURITY CLASSIFICATION

colloids (This Page)

organic complexation unclassifiedradionuclide speciation (This Report)

low-level radioactive waste unclassified

15. NUMBER OF PAGES

16. PRICE

NRC FORM 335 (2-89)

,jIW~ -~

Printed q ,on recycledA

paper

Federal Recycling Program

NUREG/CR-6627 THE ROLE OF ORGANIC COMPLEXANTS AND COLLOLDS IN THE JANUARY 2000TRANSPORT OF RADIONUCLIDES BY GROUNDWATER

UNITED STATESNUCLEAR REGULATORY COMMISSION

WASHINGTON, D.C. 20555-0001

SPECIAL STANDARD MAILPOSTAGE AND FEES PAID

USNRCPERMIT NO. G-67