Clay barriers in radioactive waste disposal

This article was downloaded by: [Northeastern University]On: 11 November 2014, At: 17:44Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

Revue Française de Génie CivilPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tece18

Clay barriers in radioactive waste disposalAntonio Gens a & Sebastià Olivella aa Universitat Politècnica de Catalunya, Departament d'Enginyeria del Terreny , JordiGirona 1–3, Edifici D-2, 08034, Barcelona, Spain E-mail:Published online: 05 Oct 2011.

To cite this article: Antonio Gens & Sebastià Olivella (2001) Clay barriers in radioactive waste disposal, Revue Françaisede Génie Civil, 5:6, 845-856

To link to this article: http://dx.doi.org/10.1080/12795119.2001.9692329

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shallnot be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and otherliabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/loi/tece18

http://dx.doi.org/10.1080/12795119.2001.9692329

http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

RFGC – 5/2001. Environmental Geomechanics, pages 845 to 856

Clay barriers in radioactive waste disposal

Antonio Gens — Sebastià Olivella

Universitat Politècnica de CatalunyaDepartament d’Enginyeria del TerrenyJordi Girona 1-3, Edifici D-208034 Barcelona, Spain

[email protected]

ABSTRACT. Deep geological is one of the preferred options for the disposal of high levelradioactive waste. In most designs, the canisters placed in drifts or boreholes are surroundedby an engineered barrier usually made of compacted swelling clay. The barrier undergoessevere heating from the canisters and hydration from the host rock. In this situation a numberof interacting thermal, hydraulic and mechanical phenomena arise. The behaviour of anengineered barrier in a particular disposal scheme has been analysed by coupled THManalyses. The results of the computations allow a more detailed examination and a betterunderstanding of the complex transient phenomena occurring in the near field.

KEYWORDS: radioactive waste, geological disposal, clay barriers, thermohydromechanicalphenomena, coupled numerical analysis

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014

846 RFGC – 5/2001. Environmental Geomechanics

1. Introduction

One the favoured options for the disposal of high-level heat emitting radioactivewaste is deep geological disposal. The aim of geological disposal is to remove theradioactive waste from human environment and to ensure that any radionucliderelease rates remain below prescribed limits [CHA 87]. Figure 1 shows a typicalscheme for an underground mined repository. It involves the sinking of deep shaftsdown to a depth of several hundred meters. The depth will, of course, be controlledby local geological conditions. The shafts provide access to a network of horizontaldrifts that constitute the main repository area. Part of those drifts will be accesstunnels and part will be devoted to nuclear waste disposal. At present, hardcrystalline rocks, argillaceous rocks and saline rocks are the type of host formationsmost likely to be selected, although other rock types are also investigated.

As an example of conceptual design, Figure 2 shows the longitudinal section of adisposal tunnel in granite. It can be observed that the canisters containing high-levelwaste are emplaced in the centre of an horizontal drift. A concrete plug separates thedisposal area from the access tunnel. The space between canisters and the host rockis filled by a suitable material to constitute an engineered barrier. The material mostusually considered is compacted swelling clay, normally some kind of bentonite onits own or mixed with other materials like sand. However, cement-based materials(special concretes) and crushed salt (for repositories in salt rock) are also beingexamined.

Figure 1. Typical scheme of a deep geological repository for nuclear waste

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014

Clay barriers in radioactive waste disposal 847

To study the problems associated with a deep geological repository, it isconvenient to differentiate the near field from the far field. The near field can bedefined as the zone altered by the presence of the radioactive waste. It includes theengineered barriers and a portion of the host rock adjacent to the waste location. Thefar field extends from the end of the near field (not a precise location) to the groundsurface. In this Chapter, coupled THM analyses carried out to study the behaviour ofthe near field for a particular repository conceptual design are presented. From theexamination of the results, a number of observations concerning the THM behaviourof the clay barrier can be made. Before describing the analyses, however, it isconvenient to discuss, in a qualitative manner, the phenomena and behaviour likelyto occur during the lifetime the repository. In the work to be presented, mainattention will focus on the early transient phase, the stage in which the actions on thebarrier are stronger.

2. THM phenomena in the near field

The near field is an area of complex phenomena and interactions. The swellingclay making up the barrier is compacted so, initially, it is in an unsaturated state.After placing the canisters, the main actions that affect the bentonite barrier (at leastin the short term) are the heating arising from the canisters and the hydration fromthe surrounding rock. At the inner boundary, the barrier receives a very strong heatflux from the canister. The dominant heat transfer mechanism is conduction thatoccurs through the three phases of the material. A temperature gradient willtherefore develop in the near field and heat dissipation will be basically controlledby the thermal conductivity of the barrier and host rock. Maximum temperaturesenvisaged in repository design can be quite high. Some designs limit the maximumtemperature to 100oC but other concepts allow temperatures as high as 200oC.

Figure 2. Longitudinal section of a disposal drift for a deep geologicalrepository scheme

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


In the inner zone of the barrier, the heat supplied by the heater results in atemperature increase and in strong water evaporation that induces drying of thebentonite. Degree of saturation and water pressure will reduce significantly in thisregion. Vapour arising from bentonite drying will diffuse outwards until finding acooler region where vapour will condense, causing a local increase in watersaturation. Vapour diffusion is a significant mechanism of water transfer mechanismand, to a much lesser extent, of heat transport.

Due to low water pressures existing initially in the unsaturated material thatconstitutes the backfill, hydration will take place with water moving from the hostrock to the barrier. The distribution of water potential is also affected by thephenomena of bentonite drying and vapour transport described above. Hydrationwill eventually lead to saturation of the barrier, but saturation times can often bevery long due to the low permeability of the bentonite and/or host rock.

In addition to the thermo-hydraulic behaviour, there are important mechanicalphenomena also occurring. Drying of the bentonite will cause shrinking of thematerial whereas hydration will produce swelling that may be quite strong inbentonite barriers. Because the barrier is largely confined between canister and rock,the main result of hydration is the development of swelling pressures, in a processquite akin to a swelling pressure test. The magnitude of the stresses developed iscritically dependent on the emplacement density of the bentonite and may reachvalues of several MPa.

The complexity of THM behaviour increases further when the interaction of thebentonite barrier with the host rock is taken into account [GEN 98a], but this aspectof the problem is not discussed here. The discussion above strongly indicates that allthose phenomena are strongly coupled, interacting with each other in a complexmanner. As described in the next Section, coupled analyses are required for a properstudy of the problem.

3. THM analysis of the near field in a nuclear waste disposal scheme

The formulation presented in [GEN 01] has been applied to the detailed analysisof a scheme for disposal of vitrified radioactive waste envisaged for the Est site inFrance, where the host rock is mudstone. The computer code CODE_BRIGHT[OLI 96] has been used to perform the analyses.

The concept for storage in this site is based on the construction of parallel driftswith a diameter of 6 m at 70 m intervals. These drifts will be excavated at a depth of490 m. Every 18 m two horizontal boreholes will be drilled on opposite sides of the

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


tunnel with a length of 19 m (Figure3). Eight canisters containing vitrified wastewill be placed in each borehole. These canisters are located inside a steel liner 1 cmthick. The space between the liner and the borehole wall will be backfilled with anengineered barrier 0.80 m thick made up of compacted bentonite. The diameter ofthe canister is 0.59 m and its effective length is 1.71 m. Therefore, eight canistershave a total length of 13.68 m, the rest of the borehole will be completely backfilledwith compacted bentonite.

The aim of the coupled THM analyses is the simulation, both in the short and inthe long term, of the behaviour of the engineered barrier and immediatelysurrounding rock when subjected to the simultaneous effects of heating from theradioactive waste and of hydration from the host rock. Because of the limited extentof the area of interest and the significant length of the drifts, a one-dimensionalgeometry with radial symmetry has been adopted for the analysis. This requires theuse of special boundary conditions as described in [GEN 98b].

The following operational phases are considered:– Isothermal phase without canisters (0-5 years). The boreholes have been

drilled and the liner and bentonite have been installed. The bentonite in thebarrier starts to hydrate under isothermal conditions.

– Ventilation phase (5-35 years). The canisters have been emplaced but theaccess drifts have not been backfilled yet. The drift walls are assumed toremain at a constant temperature due to ventilation. This enhances heatdissipation.

– Backfilled phase (35-1000 years). The drifts have been backfilled and theyare no longer ventilated.

GalleryT=28oC

Boreholes withwaste canisters,liner andbentonite

50 m

9 m

35 m

19 m 3 m

oT=28 C

q=0q=0

Figure 3. Scheme of the EST site disposal concept

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


3.1. Material properties

The thermal, hydraulic and mechanical properties of the host rock and of thebentonite have been determined from an extensive characterization programme thathas included both field and laboratory tests. The most characteristic rock parametersare:

– Thermal conductivity, O = 1.6 W/mK (Specific heat, c = 700 J/kgK)– Intrinsic permeability, k = 1.14 x 10-21 m2

– Young’s modulus, E = 3800 MPa

The properties of the bentonite depend significantly on the compaction density.The mechanical behaviour has been modelled using a thermo-elasto-plastic model[GEN 95]. Since the liner and the host rock provide fairly stiff boundaries to theengineered barrier, parameters have been selected to provide a good reproduction ofthe swelling pressure of the bentonite at various densities.

The main hydraulic constitutive parameters concern the definition of intrinsicpermeability, relative permeability and retention curve. Intrinsic permeability is afunction of dry density or porosity. The following expression has been used tocalculate intrinsic permeability as a function of porosity:

)1(

)1( 3

2

2

3

o

ookk

I

I

I

I �

�

[1]

where ko =6u10-21 m2 and Io =0.40. It should be mentioned that, because ofhydromechanical coupling, porosity variations due to volumetric deformationscomputed during the analysis cause variations of intrinsic permeability inaccordance with [1].

The relative permeability law:3lrl Sk

[2]

has been derived from backanalysis of infiltration tests. The relationship betweensuction and degree of saturation (retention curve) is described using Van Genuchtenequation with parameters P=18 MPa and O = 0.38.

Gas flow must be considered in this problem since gas pressure variations are farfrom negligible in many cases. A high gas mobility is considered for bentonite,justified by the type of micro structure set up during compaction in this type ofmaterials [OLI 00].

Thermal parameters are used to define heat capacity and thermal conductivity.Heat capacity of the bentonite is computed as an additive function of the specificheats of each component (cs, cl, ca, cv), solid, liquid, air and vapour.

> @ olvalavlvlllss LSTcScScSch IUIUIUIUIU )1()1()1()1( �� [3]

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


where cs,= 1100J/KgK, cl,= 4180J/KgK, ca,= 1000J/KgK and cv,= 1900J/KgK. Lo =2.5x106 J/Kg is the latent heat of vaporization and accounts for the thermal effectsof phase change.

Thermal conductivity is considered a function of degree of saturation accordingto:

W/mK1.28ë W/mK 0.50ë ëëë satdrySsat

)S(1dry

ll

� [4]

Finally, the diffusion coefficient controlling vapour pressure in Fick’s law iscalculated as:

)15.273(109.5

where)1(3.26

g

oolg

P

TDDSD

�u �

�

UIW[5]

Pg is the gas pressure expressed in Pa and W is the tortuosity. Effective vapourdiffusion values have been determined in constant temperature gradient testsperformed in the laboratory. A tortuosity value approximately equal to one has beenbackcalculated from those tests.

3.2. Initial and boundary conditions

Thermal conditions. An initial temperature of 28oC is considered throughout thedomain. A variable heat boundary condition is prescribed at the inner boundary toreflect the varying heat generating capacity of the canisters. Figure 4 shows thevariation of heat power per canister with time. The plot starts at a time of 30 yearsbecause this is the period that the waste will be left to cool before being placed in anunderground storage.

0

100

200

300

400

500

600

700

800

10 100 1000

Time (years)

Pow

er (

W/c

anis

ter)

Time powerfunctions

Exponential decayfunctions

Figure 4. Variation of heating power of vitrified waste canisters with time

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


Hydraulic conditions. The proposed depth for waste disposal in the EST site is490 m. A pore water pressure value of 4.5 MPa has been assumed as initialcondition in the host rock. The initial water pressure in the bentonite backfill is -23MPa, the value that leads to the prescribed initial degree of saturation of 0.71.

Mechanical conditions. A confining stress of 12.5 MPa has been consideredwhich corresponds approximately to the lithostatic stress. This confining pressurehas been used as boundary condition at r=9.5 m. The excavation of the borehole issimulated before starting the isothermal phase.

3.3. Results of the analysis

The analysis presented here corresponds to the Base Case of a comprehensivecomputational programme involving the performance of a large set of sensitivityanalyses [GEN 00]. The variation of temperatures with time at some selected pointsof the buffer and the rock are shown in Figure 5. After 5 years of isothermal phasethe temperatures rise very quickly when the canisters are placed in the borehole. Themaximum temperature (190.4oC) is achieved at the contact with the liner after 8.6years (3.6 years after the placement of the canisters). The rest of the domain reachestheir peaks at increasingly longer times as the distance to the canisters increases.When the gallery is backfilled there is a change in the thermal regime, but theinfluence is small. After 1000 years, temperatures are close to the initial onesalthough some differences still remain. It should be noted, however, that the use of a1-D geometry implies some uncertainty on the accuracy of the cooling periodduration.

0

25

50

75

100

125

150

175

200

225

250

0.1 1.0 10.0 100.0 1000.0

Time (years)

Tem

pera

ture

(ºC

)

Contact with liner Barrier near rock Rock (r=2.9 m)Rock (r=6.7 m)

Figure 5. Variation of rock and bentonite temperatures with time

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


The process of hydration can be followed in Figure 6 that shows the evolution ofthe total mass of water present in the bentonite. Hydration starts at a modest rateduring the isothermal phase (A-B). After heating starts at 5 years, water reaches theboundary between rock and bentonite from two opposite directions:

– Water arrives from the bentonite by vapour transfer and condensation. Thiswater is extracted from the inner part of the barrier that undergoes drying.

– Water arrives from the rock by flow. This flow is enhanced by the increasein water pressures set up in the rock due to the temperature rise.

Initially there is, in fact, a slight water transfer out of the barrier into the rock (B-C), most of it in form of vapour. After some time, the pressures in the rockpredominate as the vapour transfer becomes less significant because there is lesswater available in the inner part of the barrier and the temperature rise slows down.From that moment on, buffer hydration resumes and proceeds uninterruptedly up tofull saturation (C-D). The end of the ventilation phase with the closure of the accessgallery is barely noticeable. Because of the large thickness of the barrier and thevery small value of permeability, hydration times are very long, 99% saturation isonly reached throughout the domain at 330 years. The increase of water mass aftersaturation (D-E) is very small and exclusively due to cooling effects.

The evolution of the degree of saturation for three representative points of thebarrier is presented in Figure 7. The effects of vapour transfer from the inner to theouter regions of the barrier are evident. There is a major drying of the bentoniteclose to the liner and the vapour moves to increase the degree of saturation of thecentral and outer parts of the barrier. As a consequence, the barrier near the rock

1000

1100

1200

1300

1400

1500

1600

0.1 1.0 10.0 100.0 1000.0

Time (years)

Mas

s of

wat

er (

kg)

Water in the bentonite

B

A

C

DE

Figure 6. Variation of mass of water in the engineered barrier with time

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


hydrates relatively quickly although it does not quite reach saturation. The evolutionof the degree of saturation of the central part of the barrier exhibits a small peak thatcorresponds to a transient phenomenon of condensation and subsequent evaporation.

When the temperature decreases, the vapour transfer effects also reduce and thebarrier near the liner increases its degree of saturation. After about 150 years thedegree of saturation is practically uniform across the barrier, the temperature isalready quite low and vapour effects are no longer significant. The final stage ofhydration progresses uniformly. This is a clear indication that the permeability of therock and not that of the bentonite controls the progress of hydration.

In response to hydration and temperature changes, stresses develop inside thebarrier. The evolution of radial stresses is shown in Figure 8. There is an initialincrease associated with the increase of temperature, remaining approximatelyconstant afterwards over a long period of time. However, when the barrier movescloser to saturation stresses start to rise more quickly, with quite a fast increase (inlogarithmic terms) around the time of saturation. This is a consequence of the factthat the constitutive model prescribes increasingly larger strains as saturation isapproached and of the rise in pore water pressures. The effective stresses reached atthe end of the analysis are somewhat below the value of the swelling pressure. Thereason is that in this case the rock is not very stiff and does not provide sufficientconfinement. It can also be noted that radial stresses are quite uniform across thebarrier.

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1.10

0.1 1.0 10.0 100.0 1000.0

Time (years)

Deg

ree

of s

atur

atio

n

Barrier near liner Center barrier Barrier near rock

Figure 7. Variation of degree of saturation in the engineered barrier with time

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


4. Conclusions

The performance of realistic simulations of the near field behaviour inunderground repositories for radioactive waste requires the use of generalformulations and computer codes capable to tackle coupled THM numericalanalyses. An example of application concerning the behaviour of the bentonitebarrier and immediately adjacent rock in a scheme for an underground wasterepository in mudstone has been presented. It demonstrates the applicability of theformulation to engineering cases and provides a good illustration of the power ofthese methods to contribute to an enhanced understanding of the overall behaviourof the system.

In high level radioactive waste disposal, it is impossible to carry out tests that canbe taken to the time scales characteristic of repository life. The only way to bridge thisgap to longer times with some confidence is by means of modelling, using properlyvalidated numerical tools based on sound theoretical formulations. This is an absoluterequirement in the confidence-building process needed in this type of projects.

Acknowledgements

The Authors are grateful to ANDRA for their technical and financial support. Theassistance of the European Commission, ENRESA and the DGICYT is alsoacknowledged.

-10.0

-8.0

-6.0

-4.0

-2.0

0.00.1 1.0 10.0 100.0 1000.0

Time (years)

Rad

ial s

tres

s (M

Pa)

Barrier near liner

Center barrier

Barrier near rock

Figure 8. Variation of radial stresses in the engineered barrier with time

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014


5. References

[CHA 87] CHAPMAN N.A., MC KINLEY I.G., The geological disposal of nuclear waste, 1987,John Wiley & Sons, Chichester.

[GEN 95] GENS A., “Constitutive laws”, in Modern issues in non-saturated soils (Gens,Jouanna, Schrefler, eds.), 1995, Springer-Verlag, Wien, p. 129-158.

[GEN 98a] GENS A., GARCIA MOLINA A.J., OLIVELLA S., ALONSO E.E., HUERTAS F.,“Analysis of a full scale in situ heating test simulating repository conditions”, Int. J. Num.Anal. Meth. Geomech.., 1998, 22, p. 515-548.

[GEN 98b] GENS A., OLIVELLA S., BUXO P., Modelling Of the THM behaviour of anengineered barrier, C RP 0UPC 98-001, 1998, ANDRA, Paris.

[GEN 00] GENS A., OLIVELLA S., “Non isothermal multiphase flow in deformable porousmedia. Coupled formulation and application to nuclear waste disposal”, in Developments intheoretical geomechanics, Smith & Carter (eds.), 2000, Balkema, Rotterdam, p. 619-640.

[GEN 01] GENS A., OLIVELLA S., “THM phenomena in saturated and unsaturated porousmedia. Fundamentals and applications”, Environmental Geomechanics, 2001, HermèsScience Publications, Paris.

[OLI 96] OLIVELLA S., GENS A., CARRERA J., ALONSO E.E., “Numerical formulation for asimulator (CODE_BRIGHT) for the coupled analysis of saline media”, EngineeringComputations, 1996, 13, p. 87-112.

[OLI 00] OLIVELLA S., GENS, A., “Vapour transport in low permeability unsaturated soilswith capillary effects”, Transport in Porous Media, 2000, 40, p. 219-241.

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

17:

44 1

1 N

ovem

ber

2014

Clay barriers in radioactive waste disposal

Documents

Transcript of Clay barriers in radioactive waste disposal