Conceptual Models and Approaches to Understanding Long ...€¦ · Understanding Long Term Performance of Cementitous Waste Forms Presented by David S. Kosson, Ph.D., Professor Vanderbilt

Conceptual Models and Approaches to Understanding Long Term Performance of

Cementitous Waste FormsPresented by

David S. Kosson, Ph.D., ProfessorVanderbilt University

Consortium for Risk Evaluation with Stakeholder Participation (CRESP)

To

Advisory Committee on Nuclear Waste

Nuclear Regulatory Commission

July 18, 2006

Department of Civil and Environmental Engineering


NRC/ACNW - July 18, 2006 (DRAFT)

Consortium for Risk Evaluation with Stakeholder Participation

Collaborations

Vanderbilt University

D. Kosson, A. Garrabrants, S. Mahadevan, F. Sanchez, J. Clarke, S. Lopez

Netherlands Energy Research Centre (ECN)

H. van der Sloot, R.Comans, J.C.L. Meeussen, A. van Zomeren, P. Seignette

DHI (Denmark)

O. Hjelmar

Savannah River National Lab

C. Langton, G. Flach

Pacific Northwest National Lab

J. Serne, T. Brouns





Generic Vault Disposal System

ReinforcingSteel

Waste Form

Clean Grout(high strength)

Muli-layer Cap and Infiltration Control

Drainage Layeror Capillary Break

PerchedWater

Seepage

Infiltration





Motivation

Need for realistic (as practical) estimates of long-term constituent release for near-surface disposal of cementitious and other non-vitrified waste forms.

ApplicabilityPerformance Assessments and 3116 Determinations

HLW tank closure using groutDisposal of saltstone & similar wastes at SRNL, INL, ORPPrimary and secondary waste streams from steam reformingSecondary waste streams from vitrification

Waste Treatment Acceptance CriteriaOperational ControlsManagement of future wastes from reprocessing (GNEP)

Primary Constituents of ConcernLong lived & Mobile: Tc-99, Np-237, Se-79, I-129, C-14, U, Mobile: Cs-137, Sr-90, Nitrate, tritium





Broader Questions

What basis should be used to

Define the appropriate type of waste form for specific wastes?

Estimate long-term waste form and disposal system performance?

Establish treatment (operational) criteria?

Define monitoring requirements that are pre-emptive to system failure?





Constituent Release by Leaching

Primary Factors System Integrity

Engineered and Institutional Systems

Waste Form PerformancePhysical IntegrityWater ContactMoisture StatusOxidation Rates and ExtentConstituent Chemistry and Mass Transport





Processes and Impacts

Conceptual ModelMicro-cracks develop, increasing solid-liquid surface area

Bridging of micro-cracks create macro-cracks

Through-cracks develop over time, leading to convective flow

Ultimate end state may be permeable matrix –equilibrium release

Physical Integrity & Water Contact Monolithic Matrix

Flow-aroundLow interfacial areaDiffusive release

Stressed MatrixFlow-around/throughHigher interfacial areaDiffusion-convection

Spalled MatrixHigh permeabilityVery high interfacial areaEquilibrium-based release

ImpactNeed to account for the sequence of physical states and rate of changes Influences chemical reactions and constituent releaseBoth “intact” & “degraded” cases are not realistic





Processes and ImpactsMoisture Transport

Full Saturation

Capillary SaturationContinuous LiquidDiscontinuous Gas

Transition ZoneContinuous Liquid Continuous Gas

Insular SaturationDiscontinuous LiquidContinuous Gas

Completely Dry

Hamb

RH=100%

Hamb

RH=100%

Hamb

RH=100%

Hamb

RH=100%

Hamb

RH





Processes and ImpactsOxidation

Rates and Extent

1.4E+012.6E-048.9E-03Conc of O2 [mole/L]

1.1E+040.0000190.21DO2 [cm2/s]

Ratio (A/W)WaterAir

(1) Wilke and Chang, 1955(2) www.swbic.org/education/ env-engr/gastransfer/gastransf.html

oxidation front

O2

occludedpore

Conceptual ModelWaste form pores – two phase system of gas and liquid; depends on moisture content (saturation)O2 transport via gaseous diffusion may be important depending on saturation.Oxidation may lead to change in leaching behavior

Increased Tc-99 release; other constituents

ImpactGas phase transport not considered

Flux of O2 (gas) ~105 > liquid phase flux

Impact to Tc-99 oxidation minimized





Processes and ImpactsCarbonation

CO2

carbonation front

Conceptual ModelCO3-2 + Ca+2 → CaCO3 (s)

Gas phase diffusion of CO2Liquid phase diffusion of HCO3-

Pore water pH decreasedAlters solubility of constituents (increase or decrease depending on species).

CarbonationExpansive precipitate – internal stress (cracking)Pore blocking – increases diffusional resistance (decreases oxidation, release rates).Extent and pore effects depend on waste form alkalinity and saturation

0.0001

0.001

0.01

0.1

1

10

100

2 4 6 8 10 12 14Leachate pH

As

[mg/

L]

NoncarbonatedCarbonated

ImpactPotential for speciation changes (e.g., As)Pore structure changesMay have either positive (e.g., pore capping) or detrimental (i.e., increased solubility) impacts





VaultWall

WasteForm

(high SO4)


Conceptual ModelTransport described by moving dissolution fronts

Precipitation/reaction processes near external boundaries may significantly impact release (+ or -)

Dissolution/diffusion of Ca(OH)2 and CSH control pore water pH

pH gradients alter trace species release

SO4 leaching from waste into vault attacking concrete physical structure.

Source of SO4 may be waste or external environment

ImpactMass transport estimates do not reflect the dynamic chemistry and mineralogy of the waste form.Release rates and extents mechanistically different from simplified assumptions, limiting predictability.

Leaching of Major Constituents

leac

han

t

Ca moving front

effHD

CCa=CCa,0SCa=Sp,0

SSO4=SSO4,0ε=ε0

Ca(OH)2dissolution

pH

CCa=CCaSCa=0ε>ε0

effCaD

SO4 moving front

effSO4

D

SO4 moving front

effSO4

D

Sulfate species precipitate in cracks and large pores in vault concrete.






Conceptual ModelRelease based on coupled chemistry and mass transport.Release dependent on:

Moisture conditionspH gradientsRedox chemistryBoundary layer formation

ImpactPerformance assessments may grossly over- or under-predict release

Leaching of Trace Constituents

leac

han

t

Ca moving front

effHD

CCa=CCa,0SCa=Sp,0SMe=SMe,0

ε=ε0

Ca(OH)2dissolutionpH

Me moving front

effHDeffMe

D

CCa=CCaSCa=0

CMe=f{pH}

effCa

D

AMD - Selenium

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

0.01 0.1 1 10 100 1000Mean Interval [days]

Mea

n Fl

ux [m

g/m

2s]

AMD-AAMD-BMean

Diffusion Model predicts

flux 102greater than measured

after ~1 year





Integrated Long-Term Degradation

Chemical degradation and physical stress effects are coupled and integrated.

Physical stressCyclic loadingFlexural bendingDrying shrinkageSeismic eventsSettlement

Chemical degradationOxidationLeachingExpansive reactions

CarbonationSulfate attackRebar corrosion

Microcracks• Increase porosity

• Increase interaction pore water/surface

Through-cracks• Preferential flow path• Diffusive and

convective release• Loss of strength

Spalling• Loss of cohesiveness

• Two body problem• Eventual release from

“granular” material





Current Studies on Secondary Waste from ORP

Reducing GroutGround Steel Slag 43 wt%Class F Fly Ash 42 OPC 7 DI Water 7

Synthetic Hanford GroundwaterCaSO4 1.20 mmol/L NaHCO3 1.04Mg(HCO3)2 0.62 CaCl2 0.34 KHCO3 0.19 Ca(HCO3)2 0.18

MotivationTc-99, I-129 in secondary wastes from vitrification

ObjectiveLeaching assessment of reducing grout for secondary waste treatment. Comparison with “ANS16.1-type” testing in synthetic ground water.

mg/kg Added AsAg 243 AgNO3

As(V) 1000 Na2HAsO4▪7H2OBa 500 Ba(NO3)2

Cd 1000 Cd(NO3)2▪4H2OCu 1000 Cu(NO3)2▪2.5H2OCs 1000 CsClI 1214 NaI

Pb 1000 Pb(NO3)2Re 971 KReO4Sb 952 Sb2O3Se 751 KSeO4Zn 1000 Zn(NO3)2

Contaminants in Reducing Grout





Equilibrium – Trace Species

1,000

10,000

100,000

1,000,000

2 4 6 8 10 12 14

Leachant pH

Rhe

nium

[ug/

L]

AMD-AAMD-BAMG-AAMG-B

0.01

0.1

1

10

100

1,000

2 4 6 8 10 12 14

Leachant pH

Ura

nium

[ug/

L]

ML

MDL


0.01

0.1

1

10

100

1,000

1,0000

2 4 6 8 10 12 14

Leachant pH

Iodi

de [u

g/L]

Iodate


Solid-Liquid Partitioning

ComparisonAMD – DI WaterAMG – Synthetic Hanford Ground Water

1

10

100

1,000

10,000

100,000

2 4 6 8 10 12 14

Leachant pH

Stro

ntiu

m [u

g/L]

ML


waste form11.5





Mass Transport TestsAMD - Rhenium

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

1.E+00


Mea

n Fl

ux [m

g/m

2s] AMD-A

AMD-BMean

AMD - Selenium

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01


Mea

n Fl

ux [m

g/m

2s]

AMD-AAMD-BMean

MT001 TestTank Leaching in DI WaterConstituent FluxConstituent Release

ComparisonAMD – DI WaterFlux @ constant diffusivity (green dash)AMD - Calcium

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

1.E+00


Mea

n Fl

ux [m

g/m

2s]

AMD-AAMD-BMean M

ean

Flux

[mg/

m2

s]AMD - Strontium

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02


AMD-AAMD-BMean





Synthetic Groundwater

Precipitate on sample AMG-Bafter 6th leaching interval

MT001 Test

ComparisonAMD – DI WaterAMG – Synthetic Hanford Groundwater

Hanford GroundwaterCaSO4 1.20 mmol/L NaHCO3 1.04Mg(HCO3)2 0.62 CaCl2 0.34 KHCO3 0.19 Ca(HCO3)2 0.18

Rhenium

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

1.E+00


Mea

n Fl

ux [m

g/m

2s] PNL1-AMD

PNL1-AMG

6th leachateMea

n Fl

ux [m

g/m

2s]

Cesium

Mean Interval [days]

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

1.E+00

0.01 0.1 1 10 100 1000

PNL1-AMDPNL1-AMG

6th leachate

Mea

n Fl

ux [m

g/m

2s]

Selenium

1.E-09

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01


6th leachate

PNL1-AMDPNL1-AMG





Process- and Mechanism-BasedExperimentation & Modeling

Long-TermPerformance

Estimates

SensitivityAnalysis

UncertaintyAnalysis

Conceptual Model(chemistry & physics)

Mathematical Model & Computer Simulation

Model Verification(comparison to other models & limit cases)

ModelValidation

ObservationalExperiments

Parametric Experiments(individual processes to obtain parameter values & constitutive

relations)

Integrative Experiments(multiple processes & field tests)

Field Scenarios

Independent data





Measure intrinsic leaching characteristics of material (aqueous-solid partitioning (pH and LS); release kinetics)

Batch extractions & tank leaching (monoliths)Constituent fraction readily leachedControlling mechanism for release (mineral dissolution and solubility, solid phase adsorption, aqueous phase complexation)

Release kinetics for mass transfer parameters

Evaluate release in the context of field scenarioExternal influencing factors such as carbonation, oxidation HydrologyMineralogical changes

Use geochemical speciation and mass transfer models to estimate release for alternative scenarios

Model complexity to match information needsMany scenarios can be evaluated from single data set

Overarching Framework(Kosson, van der Sloot, Sanchez & Garrabrants, 2002, Environ. Engr. Sci.,

19, 159-203)





Integrated Use of Testing and Simulation





LeachXSSoftware-based system for evaluating leaching

Incorporates multiple processes and system configurationsData management/interpretationGeochemical analysis via ORCHESTRA (Meeussen, 2003)Database of material leaching information

User Input, Test Results

and ParametersDatabase Access

Calculation Engines

Scenarios(e.g., fill

characteristics,geometry, infiltration,

hydrology)

Materials(Leaching data,

Composition, Physicalcharacteristics)

Regulatory(Regulatory

thresholds andcriteria from different

jurisdictions)

Thermo-dynamic

Databases

ScenarioDatabase

MaterialsLeachingDatabase

RegulatoryDatabase

LeachXS(Materials and

ScenariosEvaluation)

Orchestra(GeochemicalSpeciation and

Reactive TransportSimulator)

Reports(Figures, Tables,

Scenario and MaterialDescriptions)

ExcelSpreadsheets(Data, Figures)

Other Models(Source Term and

Parameters forFate, Transport,and Risk Models)

Output Reporting and

Graphing

User Input, Test Results

and ParametersDatabase Access

Calculation Engines



hydrology)





jurisdictions)

Thermo-dynamic

Databases

ScenarioDatabase


RegulatoryDatabase












hydrology)



hydrology)







jurisdictions)



jurisdictions)

Thermo-dynamic

Databases

Thermo-dynamic

Databases

Thermo-dynamic

Databases

ScenarioDatabaseScenarioDatabase



RegulatoryDatabase

RegulatoryDatabase

RegulatoryDatabase























Output Reporting and

Graphing LEACHXS

CHROMIUM SPECIATION IN MORTAR AND WATER

Cr as function of pH

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

Con

cent

ratio

n (m

ol/l)

Cement mortar Model prediction CaCrO4[aq].diss Cr+3.diss Cr[OH]+2.diss Cr[OH]2+.diss

Partitioning liquid and solid phase, Cr

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

Con

cent

ratio

n (m

ol/l)

Free DOC-boundPOM-bound FeOxideClay Ba[SCr]O4[96%SO4]Cr[OH]3[A] Cr-Ettringite

Cr fractionation in solution

0%

20%

40%

60%

80%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

Frac

tion

of to

tal

conc

entra

tion

(%)

DOC-bound CaCrO4[aq].diss Cr+3.diss Cr[OH]+2.diss Cr[OH]2+.diss Cr[OH]3.diss

Cr fractionation in the solid phase

0%

20%

40%

60%

80%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

Frac

tion

of to

tal

conc

entra

tion

(%)

POM-bound FeOxideClay Ba[SCr]O4[96%SO4]Cr[OH]3[A] Cr-Ettringite LEACHXS

Leachant Simulation – Boundary Effects

DI Water Hanford GW

DI Water Hanford GW DI Water Hanford GW

DI Water Hanford GW

CO2 equilibrium

CementMaterial

Acidic Soil

Leachant

DI Water Hanford GW LEACHXS

MODELLING OF 3 LAYER SYSTEM WITH FULL CHEMICAL SPECIATION AND TRANSPORT

MSW Bottom Ash

Cement

Soil

LEACHXS





Suggested Path Forward

Process of continuous improvement, such that assessments incorporate “state of the art” understanding to extent practical

Important for current assessments and future nuclear waste management (legacy and future wastes) Need to define short-term and long-term needs

Experimental studies coupled with model development and validation

Formation/effect of boundary layers (e.g., CaCO3, oxidized layer)Moisture transport and statusOxidation ratesFull geochemical model (equilibrium & mass transfer) for key systemsPhysical changes considering key geochemistry and mass transfer





Conclusions

Significant processes are not included in current DOE performance evaluations that can have major impacts constituent release.

It is important to have a more robust system understanding and model for near-term and longer-term DOE waste management decisions.

CRESP and SRNL, along with others, are currently working together to provide the needed evaluation system components.

Documents

Conceptual Models and Approaches to Understanding Long ...€¦ · Understanding Long Term Performance of Cementitous Waste Forms Presented by David S. Kosson, Ph.D., Professor Vanderbilt