Skip to main content

Table 2 Case-specific analogues for EBS containment and radionuclide retardation in the far-field reviewed for KBS-3 V repository planned to be constructed at Olkiluoto, Finland: current status (mainly based on Posiva 2012a, Appendix C) and recommendations for future work

From: Complementary considerations in the safety case for the deep repository at Olkiluoto, Finland: support from natural analogues

NAs for:

FEP

Current status—case specific

Future outlook

Fuel

Criticality

Few relevant studies have been carried out

It is unlikely that an appropriate NA other than Oklo exists, so no further studies necessary unless criticality could weaken the safety case

Radiolysis

Data available from several ore bodies implying that radiolytic change will be minimal. But not enough information on fluxes of oxidants at natural analogue sites is known to allow a direct comparison with the likely situation in and around a SF canister

No further studies necessary unless a more appropriate site can be shown to exist. Best characterised site, Cigar Lake, has now been heavily disturbed by mining

SF corrosion

Numerous NA studies available. Dissolution rates cannot be quantified readily from natural analogue data, but abundance of naturally occurring uraninite may be taken as a general indication of its stability in the geological environment. Extrapolation of this apparent longevity to spent nuclear fuel should be done cautiously, due to differing boundary conditions

No further studies necessary unless a more appropriate site can be shown to exist (see also McKinley et al. 2015, this issue)

Canister

Cu-corrosion

Numerous studies available, including studies of localised corrosion of copper, such as Kronan (quantitative; Hallberg et al. 1987; King 1995), Littlehampton (qualitative; Milodowski et al. 2002) and rare pitting studies

Additional studies would be useful to support the understanding of localised corrosion in safety assessment. Laboratory copper pitting studies suggest that that the assumed pitting factors in safety assessment are conservative, although arguably additional NA studies in this area would strengthen these arguments

Buffer

Fe-corrosion

Wide range of studies providing quantitative and qualitative data which support the conservative assumptions in safety assessment exists. No pitting data are available. Best examples of use are in the Kristallin-1 and H12 safety assessments

Studies would be very useful to support safety assessment treatment of localised corrosion (see also Posiva 2012a)

Thermal alteration/cementation

Generally, the vast majority of the natural analogue data available is of little value as the environments studied have little relevance to repository conditions

Return to more relevant sites (e.g. Busachi) and apply a range of more relevant analytical techniques to better understand the site history and temperature profiles. Only then should the site be re-sampled and standard clay analyses carried out. For detailed discussion, see Reijonen and Alexander (2015, this issue)

Freezing/thawing

Only general considerations are available

Some deposits quarried are prone to seasonal freeze/thaw cycles and are located in areas of current or palaeo permafrost. Additional information is possible to seek if required (see also Reijonen and Alexander 2015, this issue)

Canister sinking

N/a

The approach of Keto (1999), focussing on deformation of natural bentonites, should be repeated plus the process should be included into current URL experiments (e.g. FEBEX). See also comments in Reijonen and Alexander (2015, this issue). Recently, new sites have been identified in Cyprus

Chemical alteration–saline

Various sites have been studied but no full characterisation of ‘reacted’ samples have been published

Find an appropriate site and carry out a more rigorous analysis of all relevant parameters. For detailed discussion, see Reijonen and Alexander (2015, this issue)

Interactions with copper

Qualitative data is available from Kronan cannon

Data could be more rigorously integrated with existing/new laboratory data

Interaction with iron

No relevant data available

Existing natural analogue are not relevant enough, although general observations from natural systems and the near-field processes URL experiment suggest uptake of iron could reduce swelling pressures. These data need to be integrated with a more appropriate natural analogue study. Possibility exists in the Philippines bentonite study; see Fujii et al. (2010)

Interaction with cement

Qualitative data available from Khushaym Matruk

Needs study of samples from outwith the thermally altered zone. Important to do this as it is the only OPC analogy available (see also comments in Jackson et al. 2014)

 

Quantitative data available from Cyprus

Preliminary results strongly suggests minimal reaction of bentonite in low alkali cement leachates (see also Alexander and Milodowski (2014)

 

Qualitative data available from Philippines

Generally supports Cyprus data, but closer study of potential reaction zones is necessary (see also Fujii et al. 2010)

 

Qualitative data available from Searles Lake

Conditions are far from repository-relevant, but results fully consistent with Cyprus and Philippines. Different reaction pathways when compared to Cyprus may be a reflection of the open nature of the system

Chemical erosion

 NA studies not available

Potential exists in the montmorillonite fracture gouges at certain sites (e.g. Forsmark and Olkiluoto). No detailed review has been conducted on the stability of montmorillonite at Olkiluoto. The controls of the mineralogical parageneses and groundwater circulation should be well described in order to assess stability. Paleohydrogeological data could then be utilised to address also dilute conditions at relevant depths. See Reijonen and Alexander (Reijonen and Alexander 2015, this issue) for detailed discussion. See also Puura and Kirsimäe (2011)

Foreign materials

Cements–longevity of OPC

NA data available from Jordan (quantitative) and Scawt Hill (qualitative)

Some very old natural cements, but survival may be dependent on isolation from groundwater (e.g. due to low permeability of the material). This requires further assessment

Cements–longevity of low-alkali cements

Mainly archaeological analogues available, generally restricted to shorter timescales (2–5 ka), e.g. Roman cements (qualitative)

Archaeological cements not usually from environments relevant to a deep geological repository, but worth re-analysing as analogues of modern low alkali cements. New NA site has been recently identified in the Naples region, Italy, and should be investigated

Cement-host rock interaction

NA studies available from Jordan for OPC (quantitative) as well as Cyprus and Philippines for low alkali systems, but both currently qualitative (see also Jackson et al. 2014)

Focussed analysis of the potential impact on the Olkiluoto site would be worthwhile when more data on the site hydrogeology are available (cf. discussion in Alexander and Neall 2007). Low alkali cement leachate/host rock interaction sites available for study if this was felt necessary

Cement carbonation

N/a

Ignoring carbonation in safety assessment may be over-conservative, a short scoping study would be worthwhile to assess if additional natural analogue data are required

Silica Sol longevity

No identified NAs

Potential site has not been selected so far

Radionuclide retardation in the geosphere

Retardation in clays

Numerous NAs available worldwide, several associated with URL projects. Quantitative data on long-term diffusion processes in clay formations and some are on a scale representative of the clay buffer in the engineered barrier system

No further studies are required for most radionuclides. But the case of high pH leachate alteration of clay CEC may impact certain radionuclides (see also Alexander and Milodowski 2014)

Retardation in fractured crystalline rocks

Very few, directly relevant natural analogue studies available. Quantitative data exists for matrix diffusion, but qualitative for other retardation processes

Very difficult to find appropriate sites, perhaps better dealt with in URLs (but note caveats on use of data; see also Alexander et al. 2003)

Matrix diffusion

One of the most thoroughly studied aspects of retardation in fractured rocks. Clear long-term, long-distance quantitative information available

No further generic studies are required, but a small, site specific study would be useful for Olkiluoto

Redox fronts

Numerous NA studies worldwide, but very few, directly relevant natural analogue studies available

A focussed study in an appropriate environment would be very useful, but no further qualitative studies are required for Olkiluoto. Useful for steel canister designs (see also McKinley et al. 2015, this issue)

Colloids

One of the most thoroughly studied aspects of retardation in fractured rocks, but one of the least understood

No more work is necessary until convincing sampling methods are developed (see Alexander et al. 2011)

  1. FEBEX full scale engineered barriers experiment at Grimsel, Switzerland, OPC ordinary portland cement, CEC cation exchange capacity, URL underground rock laboratory