Miniature Canister (MiniCan) Corrosion Experiment · Miniature Canister (MiniCan) Corrosion Experiment Progress Report 2 for 2008-2009 . ... SKB agree in general to this conclusion

Miniature Canister (MiniCan) Corrosion Experiment

Progress Report 2 for 2008-2009

Foreword to progress reporting of the Mini-Can Project

The Swedish Radiation Safety Authority (SSM) has carried out a quality audit on two projects related

to copper corrosion research, in a series of quality reviews of SKB. The audit was conducted by

Galson Sciences Ltd and the results are published by SSM in the report SSM 2010:17, Quality

Assurance Review of SKB´s Copper Corrosion Experiments.

Two status reports prepared by the contractor to inform the client (SKB) of the status of the project

were found to contain data not presented in a public report. The Galson Science auditors stated that in

their opinion, according to good scientific practice these data should also have been presented

publicly, even though they accepted that the interpretation of the measured data had not been

concluded and the results could be misleading due to the possibility of unexplained experimental

artefacts. SKB agree in general to this conclusion and have decided to make the reports available.

The two status reports were prepared after the technical report SKB TR 09-20. These status reports

were not prepared with a view to publishing them as SKB Technical reports and they were therefore

not reviewed in the customary way for such reports.

The reports have now been reviewed and the review comments will be taken into account in

forthcoming future reports. To make the presentation of the results as transparent as possible, the

reports are published in their original form, together with a recent set of review comments.

The quality audit revealed that both the projects were conducted in accordance with the SKB quality

system. The status reports of the Mini-Can project were reviewed according to the quality system of

the company managing the project.

Stockholm 2010-10-19

Christina Lilja Peter Wikberg

SERCO COMMERCIAL

Miniature Canister (MiniCan) Corrosion Experiment Progress Report 2 for 2008-2009

Prepared for SKB

Prepared by Serco

Your Reference 1013

Our Reference SERCO/TAS/E.003110.02/Issue 01

Classification SERCO COMMERCIAL

February 2010

Design I Advise I Integrate I Deliver SERCO COMMERCIAL

SERCO/TAS/E.003110.01/Issue 01 SERCO COMMERCIAL

Title Miniature Canister (MiniCan) Corrosion Experiment

Progress Report 2 for 2008-9

Prepared for

SKB

Your Reference

1013

Our Reference

SERCO/TAS/E.003110.02/Draft issue 01

Confidentiality, copyright and reproduction

SERCO COMMERCIAL This report is submitted by Serco Technical Services (hereafter referred to as Serco) in connection with a contract to supply goods and/or services and is submitted only on the basis of strict confidentiality. The contents must not be disclosed to third parties other than in accordance with the terms of the contract. To minimise our impact on the environment, Serco prints on 100% recycled paper

Contact Details Materials and Component Research Laboratories Serco Technical Services F7 Culham Science Centre Abingdon Oxfordshire United Kingdom OX14 3DB Telephone 01635 280385 Facsimile 01635 280389 www.serco.com/technicalassurance

Name Signature Date

Author(s) N.R. Smart A.P. Rance B. Reddy

19/2/2010

Reviewed by R.J. Winsley

22/2/2010

Approved by N.R. Smart

26/2/2010

SERCO/TAS/E.003110.01/Issue 01 SERCO COMMERCIAL Page 2


Executive Summary To ensure the safe encapsulation of spent nuclear fuel rods for geological disposal, SKB of Sweden are considering using the Copper-Iron Canister, which consists of an outer copper canister and a cast iron insert. Over the years a programme of laboratory work has been carried out to investigate a range of corrosion issues associated with the canister, including the possibility of expansion of the outer copper canister as a result of the anaerobic corrosion of the cast iron insert. Previous experimental work using stacks of test specimens has not shown any evidence of corrosion-induced expansion. However, as a further step in developing an understanding of the likely performance of the canister in a repository environment, Serco Technical Services (formerly Technical and Assurance Services) has set up a series of experiments in SKB’s Äspö Hard Rock Laboratory (HRL) using inactive model canisters, in which leaks were deliberately introduced into the outer copper canister while surrounded by bentonite, with the aim of obtaining information about the internal corrosion evolution of the internal environment. The experiments use five small-scale model canisters that simulate the main features of the SKB canister design (hence the project name, ‘MiniCan’). The main aim of the work is to examine how corrosion of the cast iron insert will evolve if a leak is present in the outer copper canister. This report describes the progress on the five experiments running at the Äspö Hard Rock Laboratory and the data obtained from the start of the experiments in late 2006 up to Winter 2009. The full details of the design and installation of the experiments are given in a previous report and this report concentrates on summarising and interpreting the data obtained to date. This report follows an earlier progress report describing work up to June 2009. The current report presents the results of the water analyses obtained in 2007 and 2008, including gas composition and microbial activity. These data show an increase in the dissolved iron concentration inside the support cage, together with a decrease in the pH. Both these observations may be due to microbial activity. The electrochemical measurements provide an in situ Eh value which is comparable with published data. The Eh value inside the boreholes decreased with time as oxygen was consumed by microbial activity and reaction with minerals in the surrounding rock. Corrosion rates have been obtained using a range of electrochemical methods. The copper corrosion rate was initially measured as being <3.5 µm yr-1, but in recent months the corrosion rates of iron and copper appear to have accelerated to unexpectedly high values. The electrochemically measured corrosion rate of iron was considerably higher (>1 mm yr-1 in some cases) than expected on the basis of laboratory experiments in the absence of microbial activity. These corrosion rates will be corroborated using the weight loss coupons when the experiments are dismantled for examination in due course. It has been agreed that Experiment 3 should be removed for analysis to confirm corrosion measurements made so far using electrochemical methods. An outline plan for carrying out this operation is included in the report. The information gained from the analysis will aid interpretation of the electrochemical data obtained to date; for example with a knowledge of the film structure on the corrosion coupons it will be possible to construct equivalent circuit models to support analysis of the AC impedance data. The information obtained would also allow better definition of supporting laboratory experiments and aid with development of a more detailed mechanistic understanding of the corrosion processes occurring in the MiniCan experiments.





Contents

1 Introduction 7

2 Experimental 7

2.1 Test layout and test conditions 7

3 Monitoring performance of model canisters 8

3.1 Water analysis 9

3.2 Pressure 9

3.3 Strain 9

3.4 Corrosion coupons 10

3.5 Mounting system for sensors and corrosion test pieces 11

3.6 Electrical connections 11

3.7 Monitoring equipment 11

4 Results 11

4.1 Water analysis 11

4.2 Electrochemical potential measurements 13

4.3 Electrochemical corrosion rate measurements 15

4.4 Copper wire resistance measurements 15

4.5 Pressure and strain gauge data 15

4.6 Evolution of LPR and AC impedance data 15

4.7 Evolution of noise data 16

5 Discussion 16

5.1 Water analysis and microbial activity 16

5.2 Electrochemical measurements 17

5.3 Corrosion rate measurements 17



6 Removal of Experiment 3 for analysis 18

6.1 Removal and transport 18

6.2 Analysis 19

7 Conclusions 21

8 Acknowledgments 21

9 References 21

Appendix 1 Microbial and Gas Analysis 131



1 Introduction To ensure the safe encapsulation of spent nuclear fuel rods for geological disposal, SKB of Sweden are considering using the Copper-Iron Canister, which consists of an outer copper canister and a cast iron insert. A programme of work has been carried out over a number of years to investigate a range of corrosion issues associated with the canister, including the possibility of expansion of the outer copper canister as a result of the anaerobic corrosion of the cast iron insert. Experimental work using stacks of copper and iron test specimens has not shown any evidence of corrosion-induced expansion [1]. However, as a further step in developing an understanding of the likely performance of the canister in a repository environment, Serco Technical Services has set up a series of experiments in SKB’s Äspö Hard Rock Laboratory (HRL) using inactive model canisters, in which leaks were deliberately introduced into the outer copper canister while surrounded by bentonite, with the aim of obtaining information about the internal corrosion evolution of the internal environment [2].

The experiments use small-scale model canisters that simulate the main features of the SKB canister design (hence the project name, ‘MiniCan’). The main aim of the work is to examine how corrosion of the cast iron insert will develop if a leak is introduced into the outer copper canister. The study addresses issues such as:

• Does water penetrate into the annulus through a small defect?

• How does corrosion product spread around the annulus from the leak point?

• Does the formation of corrosion product in a constricted annulus cause any expansive damage to the copper canister?

• What is the effect of water penetration on the insert lid seal?

• Is there any detectable corrosion at the copper welds?

• Are there any deleterious galvanic interactions between copper and cast iron?

• Does corrosion lead to failure of the lid on the iron insert?

The experiments were set up between September 2006 and February 2007 and the detailed design, experimental setup and initial results up to May 2008 are reported in reference [3]. The five experiments are automatically monitored and the scheduled experimental data were obtained and analysed. This progress report includes the data obtained up to Winter 2009 and followprevious progress report that presented data up to June 2009 [

s a 4]. Further progress reports will be

produced in due course and consideration is being given to removing one or more of the ‘MiniCan’ experiments for analysis.

2 Experimental The five experiments setup in the boreholes at Äspö HRL are being logged automatically by computer-controlled data logging equipment. The detailed experimental layout and test conditions are given in reference [3] and summarised below.

2.1 Test layout and test conditions

Figure 1 shows the layout of the experiments in the boreholes in the Äspö HRL, Figure 2 shows the layout of the canisters and the other test pieces inside the support cages and Figure 3 shows the



insertion of the copper model canisters into the support cages. A summary of the test environments inside the five experiments is given in Table 1. In three out of the five experiments a layer of bentonite was mounted inside an annulus between two concentric cylinders around the model canister. The purpose of the bentonite is to condition the chemistry of the incoming groundwater so that it is representative of the repository situation. The cage prevents direct contact between the bentonite and the surface of the canister in order to avoid blockage of the defect introduced into the copper container. The bentonite holder was filled with compacted bentonite pellets mixed with bentonite powder at a density designed to give a high permeability and low swelling pressure, with the aim of allowing bentonite-conditioned groundwater to reach the model canister rapidly.

In the fourth experiment the model canister was surrounded by compacted bentonite, to simulate as exactly as possible the real exposure conditions in the repository, although in this situation it is expected that the extent of corrosion will be limited by the supply of water through the compacted, low permeability bentonite [5]. In this experiment, where the model canister was in direct contact with the compacted bentonite, there was no inner cylinder and hence no annulus of bentonite, but rings of saturated compacted bentonite were machined to slide over the model canister, inside the outer cylinder. Clay Technology (Lund, Sweden) machined the rings to size. The use of fully saturated bentonite should minimise the time required to achieve full saturation and hence to achieve the full swelling pressure of the bentonite. The thicknesses of the outer cylinder of the support cage and the outer filter cylinder were selected to withstand the swelling pressure exerted by the compacted bentonite (7 MPa). Small holes or slots were machined in the compact bentonite rings to accommodate electrodes and corrosion coupons in the compacted bentonite, and strain gauges on the surface of the model canister.

The fifth experiment was set up without any bentonite around the model canister, to examine whether a biofilm develops on the surface of the canister and to examine its effect on corrosion behaviour. For this experiment, where the canister is directly exposed to groundwater, the perforated steel cage was used to hold the model canister, but no bentonite was placed inside it. Consequently the groundwater enters directly into the vicinity of the model canister.

3 Monitoring performance of model canisters A number of parameters have been measured and monitored over the period since the experiments were set up in late 2006 to Winter 2009, as follows:

• Electrochemical potential of gold and platinum, and the redox potential, Eh;

• Corrosion potential of cast iron, copper and the miniature canister itself;

• Response of strain gauges on two experiments;

• In situ corrosion rate measurements for copper and cast iron;

• Water composition (including gas analysis and microbial measurements).

The reference electrodes used were:

• Two small silver-silver chloride reference disc electrodes inside each support cage (World precision instruments EP series);

• Long-life reference electrodes for cathodic protection (CP) purposes (Silvion silver-silver chloride reference electrodes), which were mounted outside the support cage but immersed in the borehole water.



The redox potential of the environment was measured by means of a gold wire and a platinum flag located near to the model canister inside the support cage and an Eh sensor located outside the supporting cage (Cumberland Electrochemical Ltd Pt/Ir mixed metal oxide, MOx, probe, which has been successfully used embedded in cementitious grout to monitor fluctuations in redox conditions).

The sensors were supported from a nylon support rack inside the void above the model canister (see Figure 2). All reference electrodes were calibrated before installation.

3.1 Water analysis

After installation, water samples were taken for analysis at periodic intervals. Äspö staff carried out the required water analysis. Water samples were extracted through stainless steel tubes that passed out through the borehole flange. For each borehole it was possible to extract samples from both within the support cage itself and from the borehole surrounding the model canister experiment. There was a continuous bleed of gas from each borehole flange to prevent the development of a gas-filled space within the borehole. This gas is assumed to be nitrogen. The concentration of dissolved gases was determined by Microbial Analytics Sweden AB. The methods used are summarised in Appendix 1. The concentration of the following gases was analysed: H2, He, O2, N2, Ar, CH4, CO2, CO, C2H6, C2H4, C2H2, C3H8, C3H6. The microbial analysis was also carried out by Microbial Analytics (see Appendix 1 for details) and was used to measure the following:

1. The total number of microorganisms;

2. The number of aerobic cultivable bacteria;

3. Biomass measured as adenosine-tri-phosphate;

4. The most probable number of the following types of bacteria were determined: sulphate reducing bacteria (SRBs) and autotrophic acetogens. These bacteria were chosen because they are likely to have the strongest effect on the corrosion behaviour of metals.

A programme of water analysis was carried out regularly in the period February to August 2007 and then no more water samples were taken until October 2008, in order to allow the water chemistry to stabilise without any perturbations. The results from the 2007 analyses are reported in reference [3] and the new data from the October 2008 analyses are included in the current report. No additional water analyses have been carried out since October 2008.

3.2 Pressure

The water pressure in the boreholes was initially measured by means of an analogue pressure gauge that was attached to the flanges on each borehole. Later an electrical pressure gauge was attached to the outlet pipe on the flange and the output was recorded on the datalogging equipment.

3.3 Strain

Strain gauges were applied to measure changes in the strain on the outer surface of two of the copper canisters (Tests 1 and 4). The aim of these measurements was to provide an indication of whether expansion caused by internal corrosion has caused any dimensional changes. Standard strain gauge monitoring technology using bi-axial strain gauges was applied (e.g. Techni-Measure Ltd) using cyanoacrylate adhesives and protected with waterproof coatings.



3.4 Corrosion coupons

A number of types of corrosion coupons were mounted within the cage used for the model canister experiments, as follows:

• Plain corrosion coupons of copper and cast iron. The corrosion rate of the materials will be determined from weight loss measurements at the end of experiments. One weight loss sample for each material was installed in each experiment. These coupons were suspended from the nylon mounting rack using polypropylene thread.

• Coupons of copper and cast iron that are electrically connected to the exterior. These were designed to allow the corrosion potential of the electrodes to be measured. It was also possible to carry out electrochemical measurements of the corrosion rate of these coupons using; linear polarisation resistance, LPR, AC impedance, ACI, and electrochemical noise, ECN. Platinised titanium gauze was used as the counter electrode in a conventional 3-electrode electrochemical cell. Electrical connections were made to the coupons using copper wire for the copper coupons and carbon steel wire for the cast iron coupons.

• Copper electrical resistance wire probes. These were set up to measure the corrosion rate of copper using the technique proposed previously by VTT (e.g. [6]). Each coil consisted of a total length of 112.5 cm of 1 mm diameter copper wire (99.9% purity, Advent Cu513918). The length of wire was divided into three regions each of 37.5 cm length by applying heat shrinkable, adhesive-lined polymer tubing to the end sections, leaving the middle section exposed to the test environment. The complete wire was formed into a spiral with a diameter of ~1 cm. The two screened lengths were used as reference resistances and the change in the resistance of the exposed length was used to calculate the corrosion rate. This information was processed by the ACM Field Machine electrochemical unit.

• Stress corrosion test specimens. Four Wedge Opening-loaded (WOL) specimens were mounted in the boreholes. They were machined from a scrap copper lid provided by SKB and they were pre-cracked to give a range of stress intensity factors. In addition four U-bend specimens were manufactured from the same lid material, with dimensions of 2.5 × 80 × 20 mm. Two U-bend specimens and two WOL specimens were mounted in each of experiment boreholes 3 and 4, by loosely suspending them from the stainless steel push rod using plastic connectors. They were thus exposed directly to the groundwater. They will be examined for any indications of SCC at the end of the experiments.

• Copper-cast iron-copper sandwich specimens to investigate jacking effects. These specimens were based on the multi-crevice assembly specimen used in previous galvanic corrosion experiments [7], in which cast iron castellated nuts were tightened against sheets of copper. The aim of these specimens was to investigate a number of possible corrosion mechanisms, including crevice corrosion, galvanic corrosion and expansive corrosion. They consisted of a sheet of copper which was clamped against a block of cast iron using a ring of nylon bolts. To investigate the effect of separation distance between mating surfaces a series of steps was machined into the surface of the cast iron to give separation distances between the mating surfaces ranging from direct contact around the edges of the specimens, increasing in steps of 10 µm to 30 µm. The materials used for thsandwich specimens were prepared from copper sheet and the same type of cast iron used for the insert in the model canisters. The specimens will be removed and examined at the end of the expo

e

sure period.



3.5 Mounting system for sensors and corrosion test pieces

The corrosion coupons and environmental sensors were supported in a nylon rack which was placed inside the stainless steel support cage above the model canisters before the support cage was sealed. It rested on the top of the model canisters by means of support legs. Electrical connections were taken out through Conax compression fittings in the support cage lid. For Test 4, where the model canister was embedded in compacted bentonite, the corrosion coupons and sensors were placed in slots that were machined into the compacted bentonite before it was loaded into the supporting cage.

3.6 Electrical connections

A summary of the sensors in each of the experiments and the conduits required for the connecting cables and details of output from each experiment is reported in reference [3]. Connecting wires from metal samples inside the canister cage (i.e. corrosion coupons and the model canister itself) were passed through compression glands (316L stainless steel Conax fittings) and connected to cables that were taken out to the external environment through stainless steel tubes which were attached to the supporting cages using Swagelok-type pressure fittings. All connections were made using soldered joints which were then sheathed in heat shrink. It should be noted that the electrochemical measurements rely on the integrity of the sheathing system throughout the experiments and the success of the sheathing will only be confirmed when the experiments are dismantled and the sheathing can be examined. The tubes were passed through the stainless steel flange at the entrance to the boreholes using bulkhead compression fittings.

3.7 Monitoring equipment

The electrochemical monitoring is being performed using an ACM Ltd Field Machine and an Agilent datalogger. The ACM is being used to carry out the electrochemical measurements of corrosion rate (10 channels) and measurements of electrical resistance of the copper wire electrodes (2 channels), and the Agilent datalogger is used to monitor the potentials of the various electrodes, together with monitoring the strain gauges. The datalogging equipment is located in a control room near the boreholes and data is then transmitted via the Internet to Serco’s Culham laboratory for analysis.

4 Results 4.1 Water analysis

4.1.1 Inorganic analysis

The results from the water analyses (SO42-, HCO3

-, S2- and Fe2+) for the five experiments for samples taken in October 2008 are given in Table 2, together with the May 2007 results. In October 2008, data from outside the support cages was only obtained in Experiment 2, because it was found previously that there is not very much variation between the different boreholes [3]. Table 2 is a compilation of analyses for water samples (SO4

2-, HCO3-, S2-, Fe2+, pH only) taken from

both within the support cage and external to the support cage in the borehole. The previous data obtained up to October 2007 [3] (Figure 4-Figure 11) showed a slight variation in the internal and external compositions. The sulphide concentrations were in the range 0.02 to 0.03 mg/L. The values measured in 2008 were lower than the values measured in 2007, in some experiments by a factor of two. The overall sulphide concentrations are generally low (at least three orders of magnitude less than the sulphate concentration). The greatest variation between the internal and external ion concentrations was found in the dissolved iron analyses with 0.15 mg/L of Fe 2+



external to the support cage compared to values up to 19.2 mg/L inside the support cage, in experiments with low density bentonite (Experiments 1 to 3) or no bentonite at all (Experiment 5). Similar variation was noted previously in 2007. The pH values of the water samples taken from inside and outside the support cages are also shown in Table 2. These data show a slight decrease in the pH of water inside the support cages compared to the water in the boreholes. Table 3 gives the measured concentrations of a range of metallic elements and anions, including those present in trace quantities, together with the total organic content (TOC) and conductivity of the water. It is noticeable that chloride, sodium, calcium, bromide, silicon, total sulphur, manganese, lithium, strontium, barium, cobalt, chromium, nickel and conductivity were all significantly higher in Experiment 5 than in other experiments or in the borehole of Experiment 2. The concentrations of nickel and chromium were at least two orders of magnitude greater than in borehole water, suggesting that in Experiment 5 (no bentonite) corrosion of the stainless steel support cage was occurring, as well corrosion of the cast iron insert. Low concentrations of nickel and chromium were observed in Experiments 1, 2 and 3, indicating that most of the increased concentration of iron in these experiments was due to corrosion of the cast iron insert rather than the stainless steel support cage (unless any nickel and chromium released had precipitated as an insoluble corrosion product). The higher concentrations of chloride, sodium, calcium, bromide, silicon, total sulphur, manganese, lithium, strontium, barium, in Experiment 5 indicate that there are local variations in water chemistry, which are presumably due to different water flow patterns through the rock fractures.

The results of the dissolved gas analyses are shown in Table 4. The dissolved oxygen concentration in the groundwater was essentially zero [3]. The dominant dissolved gas is nitrogen, with significant quantities of helium, argon, carbon dioxide and methane also present. The concentration of hydrogen was particularly high in Experiment 1 compared to data obtained in 2007; however the high concentration of hydrogen (215 µL L-1) previously observed in Experiment 1 was much lower (7.3 µL L-1) in 2008. Also the levels of both carbon dioxide and the alkalinity were lower in 2008.

4.1.2 Microbial analysis

The results of the microbial analyses for sulphate reducing bacteria and autotrophic acetogens are summarised in Table 5 and Table 6. The details of the experimental measurement and the analyses, which were provided by Microbial Analytics AB, are given in the Appendix 1. The most recent sample was taken in October 2008. These data show that the autotrophic acetogens were the dominant species, but that sulphate reducing bacteria were also present in significant numbers. No other types of microbes were analysed in detail, because these two varieties are believed to be the most significant in relation to corrosion. The results of the following measurements are included in Table 5 and Table 6:

• Most probable numbers (MPN) and metabolites produced by sulphate-reducing bacteria (SRB) and autotrophic acetogens (AA);

• Biomass measured as adenosine-tri-phosphate (ATP) content;

• Total number of cells;

• Number of cultivable heterotrophic aerobic bacteria (CHAB) and

• Molecular biology analyses which investigate the RNA and DNA pools in the samples.

The total number of cell levels remained unchanged in 2008 but the ATP level increased by almost 10-fold inside the model canister, where the water was in contact with cast iron. Further details and discussion of the results are given in Appendix 1.



4.1.3 Pressure readings

The results from the pressure gauges in the boreholes are shown in Figure 12. Unfortunately there was a loss of data due to a fault with the datalogging system during the Winter of 2007/8. The data show that there is a general tendency for the pressure to fall with time, although the rate of decrease has fallen and the pressures have stabilised in recent months. The low pressure is mainly due to water loss through the rock face surrounding the boreholes.

4.2 Electrochemical potential measurements

This section presents the electrochemical potential measurements for each of the model canister experiments. Some of the figures also include the results of corrosion rate measurements (see Section 4.3). The exposure period for all experiments at the time of writing was ~25,000 hours (i.e. ~3 years).

4.2.1 Experiment 1: Low density bentonite

In this experiment the defect in the outer copper shell (located near the top cap weld), is pointing vertically upwards. Figure 13 shows the results from the potential measurements for gold, platinum, Eh probe and the miniature canister. These data show the Eh values using gold and platinum, which are inside the support cage, and the Eh measured using the mixed-metal oxide electrode outside the support cage. Initially the internal silver-silver chloride reference electrodes were used for the Au and Pt electrodes, but these failed and it was necessary to switch to the Silvion reference electrode, which was mounted in the borehole outside the support cage (see change at 4,000 hours). The Au, Pt and Minican potential data up to 4,000 hours should be discounted. It is not clear why the Ag-AgCl disc electrodes failed in these experiments, since they had behaved satisfactorily during the laboratory autoclave tests [3]. To date, the Silvion reference electrodes have performed satisfactorily.

The mixed metal oxide Eh values are reliable and show a decrease in the Eh with time. This is expected, since oxygen is consumed by microbial activity and mineral reactions. The Eh value is lower inside the cage than outside, indicating more reducing conditions, possibly due to the ongoing corrosion reactions. Recently there have been some fluctuations in the Eh values; the reasons for these are not clear at present.

The electrochemical potential for the model canister represents a mixed potential for the copper canister itself and the inner cast iron insert, which would probably have been partially wetted by water passing through the 1 mm defect in the outer copper canister.

Figure 14 shows the potentials for the iron and copper coupons inside Experiment 1, up to Winter 2009. Again there was a problem with the reference electrodes up to 27 June 2007, but after that date the measurements were changed to be with respect to the Silvion reference electrode, at which point reasonable values of about -450 mV vs NHE were obtained, for both iron and copper. The iron and copper potentials are consistent with anoxic conditions existing in the boreholes.


In this experiment the defect in the outer copper shell is at the top end of the canister (i.e. near the lid) but pointing downwards. Figure 15 shows the results from the potential measurements for gold, platinum, Eh probe and the model canister.

With Experiment 2, as for Experiment 1, there was a problem with the small Ag-AgCl reference electrodes mounted inside the support cages, but when the measurements were carried out against the external Silvion reference electrode reasonable values were obtained. The Eh value is again slightly more reducing inside the support cage than outside.



The electrochemical potential values for iron and copper are shown in Figure 16. As for Experiment 1 it was necessary to measure the values against the Silvion reference electrode because the internal silver-silver chloride electrodes failed. This correction was made after June 2007. The potentials have been stable since the reference electrode system was changed; they are consistent with anoxic reducing conditions.


In this experiment there are two defects present in the outer copper shell; the defect near the top lid is pointing upwards, while the defect near the bottom cap is pointing downwards. Figure 17 shows the results from the potential measurements for gold, platinum, Eh probe and the model canister. In Experiment 3 the internal Ag-AgCl electrodes worked properly until approximately 8,000 hours had elapsed, when it was necessary to change to the Silvion reference electrode outside the support cage. The early fluctuations in Eh inside the support cage appear to be genuine. The Eh value external to the support cage stabilised at ~-300 mV after 3500 hours.

The electrochemical potential values for iron and copper in Experiment 3 are shown in Figure 18. These data show a decrease in the corrosion potential of the copper initially, stabilising at highly negative values. The potentials were subsequently measured against Silvion after failure of the internal silver-silver chloride reference electrodes after approximately 7,500 hours. The potentials have been stable for the last 18 months but have attained more positive values than initially (~-430 mV SHE).

4.2.4 Experiment 4: Compacted bentonite

In Experiment 4 the defect in the outer copper shell is near the top lid but pointing downwards. The whole model canister is surrounded by compacted bentonite, with the test electrodes placed in slots cut into the compacted bentonite. Figure 19 shows the results from the potential measurements for gold, platinum, Eh probe and the model canister. The potentials for the platinum exhibited a considerable amount of noise. It was necessary to use the external Silvion reference electrodes to measure the potentials after about 3,500 hours. It is noticeable that the Eh values are not as negative as in the fully saturated bentonite tests.

The electrochemical potential values for copper in Experiment 4 are shown in Figure 20. Contact has been lost with the iron electrode in this experiment. This may be caused by expansion of the compact bentonite leading to crushing of the electrical connections.

4.2.5 Experiment 5: No bentonite

In this experiment there are two defects in the outer copper shell, both of which are near the top lid. One defect is pointing vertically upwards, the other is pointing downwards. In this experiment there is no bentonite around the model canister (i.e. it is in direct contact with the Äspö groundwater). Figure 21 shows the results from the potential measurements for gold, platinum, Eh probe and the model canister. In this experiment there was a problem with both the internal silver-silver chloride reference electrodes and the external Eh probe and the Silvion electrode and so it was necessary to set up an external reference electrode and Eh probe, mounted on the borehole flange. The iron and copper potentials in this experiment are shown in Figure 22. Again there were problems with the small internal Ag-AgCl reference electrodes up to June 2007, after which it was necessary to switch to using the stainless steel flange as a pseudo reference electrode and then to an external reference electrode. The Eh probe data and the recent iron and copper potentials are unusually high and there is some doubt about the reliability of this reference electrode system.



4.3 Electrochemical corrosion rate measurements

Corrosion rate measurements were carried out using the following electrochemical techniques: linear polarisation resistance, LPR, AC impedance, ACI, and electrochemical noise, ECN. The LPR measurements were carried out over a potential range of ±10 mV with respect to the corrosion potential at a scan rate of 10 mV min-1 (starting with the electrode polarised to -10 mV). The corrosion rate is obtained by measuring the slopes of the plots and applying the Stern-Geary approximation [8], with a Stern-Geary constant of 26 mV. The AC impedance measurements were carried out using a perturbation of ±10 mV, starting at an initial frequency of 10 kHz start and a final frequency of 10 mHz finish, with 100 readings taken per test. The electrochemical noise measurements were carried out at a sampling rate of ten readings per second for a period of one hour. Examples of LPR and AC impedance measurements for copper and iron electrodes in Experiment 1 are shown in Figure 23 and Figure 24. Good agreement was obtained between the corrosion rates derived from the two techniques. The measured corrosion rate data for copper and cast iron are presented in Figure 25 to Figure 28.

The corrosion rate values obtained for copper in Experiments 1, 2, 3 and 5 are summarised in Figure 25. The values from electrochemical measurements on cast iron Experiments 1, 2, 3 and 5 are shown in Figure 27. The corrosion rate data for Experiment 4 was plotted separately for copper and cast iron in Figure 26 and Figure 28 respectively because they yielded considerably higher corrosion rate values. The data show that in most cases there has been a general increase in corrosion rate over time. Recently, very high corrosion rates have been measured for both copper and iron, particularly in Experiments 2 (low density bentonite) and 4 (compacted bentonite). In Experiment 4, the corrosion rate values obtained towards the end of 2008 were extremely high (40 mm/yr) but the values decreased in 2009. There is some uncertainty about the very high values obtained after August 2007. It is possible that there has been some disruption of the electrodes and the connections due to the swelling pressure exerted by the bentonite.

4.4 Copper wire resistance measurements

The results from the electrical resistance measurements on coils of copper wire are shown in Figure 29 (Experiment 2) and Figure 30 (Experiment 5). The resolution of these measurements is expected to increase with increasing measurement period. It appears that the long-term corrosion rate for copper is close to zero in these measurements, but the resolution is rather poor.

4.5 Pressure and strain gauge data

The strain gauge results for Experiments 1 and 4 are shown in Figure 31. Gauges 1a and 1b showed an initial increase in tensile strain and we are not clear about the explanation for this behaviour. It seems to be too early to be due to expansion caused by a corrosion process. It is possible that the variation is caused by small local temperature fluctuations.

4.6 Evolution of LPR and AC impedance data

Figure 32 to Figure 41 show how the LPR plots have evolved with time in each experiment and the evolution of the AC impedance data are shown in Nyquist format in Figure 42 to Figure 53. Figure 54 to Figure 80 present the AC impedance data in Bode format. An increasing slope in the LPR plots corresponds to an increasing corrosion rate. Similarly a decrease in the diameter of the AC impedance Nyquist plots corresponds to an increase in corrosion rate. It is possible to see some fine structure in the Nyquist plots at the high frequency end of the scans (e.g. Figure 42), which may represent the development of features in the surface films, or it could be due to an electrical artefact.



4.7 Evolution of noise data

Electrochemical noise is a technique used to look for indications of localised corrosion; an example of the data acquired to date is shown in Figure 81. These data were obtained by coupling two copper electrodes together and measuring the current and potential noise between them. If localised corrosion were occurring transients would be expected in the results. A series of plots generated from the noise data acquired in each experiment for cast iron and copper electrodes over the period of exposure is shown in Figure 82 to Figure 101. To date, there are no indications of localised corrosion occurring in the MiniCan corrosion coupons (e.g. sharp transients lasting 10s of seconds), but analysis of the noise data is continuing to determine whether corrosion rate data can be extracted.

5 Discussion 5.1 Water analysis and microbial activity

There are a number of interesting features regarding the results of the water analysis from the model canister experiments. In general, the groundwater composition outside the support cages does not vary significantly between boreholes. Firstly, there is a marked increase in the iron concentration inside the model canister support cage compared to the surrounding water in the borehole. This is accompanied by an increase in the concentration of nickel and chromium. This increase in dissolved metal concentration could be due to corrosion of the stainless steel support vessel, the cast iron coupons, and/or the cast iron insert itself, which will be in contact with groundwater as a result of the defect in the outer copper canister. The concentration of chromium and nickel is much lower than the concentration of iron, suggesting that most of the iron results from corrosion of the cast iron, rather than stainless steel, unless there is some kind of selective dissolution of the stainless steel occurring.

Table 2 shows that there has been a decrease in the pH inside the support cage compared to outside the support cage. This could be a result of the microbial activity inside the support cage, as shown in Table 6, or it could be due to hydrolysis reactions involving dissolved metal ions. The gas analysis data (Table 4) show the very low oxygen content of the groundwater and the presence of helium and methane from geological sources.

The enhanced corrosion of the iron and/or stainless steel, as shown by the high concentrations of dissolved metals, would not be expected in anoxic near-neutral water at low temperature, even with the high concentrations of chloride present, based on the results of laboratory experiments in the absence of microbial activity. This result suggests that the corrosion is enhanced by microbial activity. This may also account for the high corrosion rates for iron and copper measured electrochemically. It should be noted that the concentration of dissolved copper was very low.

The microbial analysis indicates that autotrophic acetogens and sulphate reducing bacteria, SRBs, are active in the vicinity of the model canisters. The key metabolic reactions for these species are:

Autotrophic acetogens: 4H2 + 2CO2 ⇒ CH3COO- + H+ + 2H2O

Sulphate reducing bacteria: CH3COOH + SO42- ⇒ H2S and Fe 2+ + H2S ⇒ FeS + 2H+

The acetogens produce acetate, which can cause stress corrosion cracking (SCC) of copper and sulphur from SRB activity is detrimental to copper. In principle, hydrogen produced by the anaerobic corrosion of iron could drive the activity of the acetogens and hence there is a possibility of autocatalytic behaviour (see Appendix 1 for further discussion). The processes summarised above are consistent with the measured fall in pH and the increase in corrosion rate and release of iron observed during these experiments.



5.2 Electrochemical measurements

The electrochemical potentials of the gold, platinum and the miniature canisters are indicative of very low oxygen concentrations. The electrochemical potential values obtained for the gold in the Model Canister experiments are comparable to the electrochemical potential values obtained during laboratory investigations of galvanic corrosion of iron-copper couples in anoxic conditions [7,9], and the anaerobic corrosion of iron [10,11]. They are also comparable to Eh values reported in the literature for the Eh value of deep Swedish groundwaters [12,13]. The data show that for Experiments 1 to 3, with low density bentonite, the Eh outside the cage is slightly higher than inside the cage, whereas for Experiment 4, with compacted bentonite the situation is reversed.

The corrosion potential of the copper and the model canisters can be compared with other measurements and modelling activities [12]; for example the predicted long-term corrosion potential of a copper canister in aerated compacted bentonite reported previously [12]. The modelling calculations showed that the predicted short-term corrosion potential for a copper canister is of the order of -300 mV vs SCE (i.e. ~-50 mV vs NHE). This is more positive by approximately 100 mV than the value measured in Experiment 4 for the model canister embedded in compact bentonite under anoxic conditions (Figure 19). The rest potential of the canister became more negative towards the end of 2009. The potential of the copper electrode in the same experiment was similar to that of the copper canister (Figure 20).

The potentials of the iron electrodes are below the hydrogen evolution potential at pH 7 and are consistent with the occurrence of anaerobic corrosion.

5.3 Corrosion rate measurements

The corrosion rates for iron are quite high and appear to be increasing with time. They are considerably higher than would be expected from laboratory experiments on the anaerobic corrosion on iron [14,15]. They have to be treated with some caution as electrochemical measurements tend to overestimate the true values, since they only provide an instantaneous measure of corrosion rate, which is based on a number of assumptions about the nature of the electrochemical interface. The overall corrosion rates will be checked against weight loss measurements at the end of the experiments (i.e. a long-term integrated measurement of corrosion rate). There is also a possibility that crevice corrosion might be occurring if the environmental conditions are correct, although it is normally believed that crevice corrosion is not possible in anoxic conditions. The high rates reported here may support the observation of increased iron concentrations in the water analyses. It is possible that the high rates may be a result of microbial activity inside the support cage (see Section 4.1.2 and Appendix 1).

The electrochemical measurements of the initial corrosion rate indicate corrosion rate values for copper up to values of ~3 µm yr-1 (Figure 25). This is based on a Stern-Geary constant of 26 mV, whereas other authors (e.g. [16]) have used lower values. Thus the figures reported here are conservative values (i.e. they may exaggerate the corrosion rate). These are consistent with the electrical resistance measurements which indicate corrosion rate values close to zero, but with a large scatter. The resolution of these measurements should improve with elapsed time. The derived corrosion rate values can be compared with values given in the literature for experiments on copper in compacted bentonite; for example values of up to 2.5 µm yr-1 have been reported by Rosborg et al. [16] for measurements during the LOT experiment and 4.7 µm yr-1 by Saario et al. [17] for weight loss measurements on copper in compacted bentonite in laboratory tests.

However more recent measurements (e.g. Figure 25) suggest that higher corrosion rates of copper may be occurring. The electrochemically measured corrosion rate for copper in Experiment 2 is much higher than that measured using the copper coil electrical resistance method (see Figure 25 and Figure 29), which suggests that there may be some experimental difficulties associated with the electrochemical corrosion rate measurements. Nevertheless this cannot be confirmed until one



of the experiments is dismantled. The possible effect of enhanced microbial activity on the corrosion rates should also be seriously considered.

6 Removal of Experiment 3 for analysis In view of the high corrosion rates being measured in some of the experiments it has been suggested that consideration should be given to removing one of the ‘MiniCan’ experiments to investigate the extent of any corrosion, both on the coupons and on the model canisters itself. This will also give an opportunity to characterise any surface biological activity (e.g. the presence of biofilms, type of bacteria colonising the surfaces, etc.) and assessing the condition of the bentonite in the experiments. The experiment could then be replaced with a new MiniCan specimen, using a refurbished support cage. Some consideration would need to be given to the timing of replacement; for example it may be beneficial to wait until the analysis of the removed canister has been completed, so that the information gained can be used to identify improvements in the experimental design for subsequent tests. It is recommended that automated monitoring of the remaining four experiments should continue, using the current monitoring protocol.

At a planning meeting held in June 2009 it was agreed that Experiment 3 should be removed for investigation [18]. Following initial discussions, an outline plan for removal and analysis of the MiniCan Experiment 3 is presented below. The overriding aim of the proposed procedure is to use a method that will allow removal of the experiment while avoiding exposure to the atmosphere, in order to minimise changes in the composition of any air-sensitive corrosion products or other materials. Having removed the experiment from the borehole the support cage and the miniature model canister will be transported to the U.K. for analysis at Serco’s laboratories. The main stages of the proposed procedure are described below.

6.1 Removal and transport

The first stage of the procedure would be to build a protective shroud around the entrance to the borehole, which would be purged with inert gas to minimise exposure to air. This would be designed and set up by Äspö staff. Preliminary discussions have already been held on this subject. The shroud could also include a gas-purged glovebag arrangement to surround the experiment on removal to minimise contact with air.

In order to further minimise contact with air, the borehole would be purged with nitrogen while the assembly was being removed from the borehole. It would probably be necessary to move some of the equipment currently situated near the boreholes to provide sufficient working room (e.g. the emergency evacuation vehicle store). It would be necessary to ensure that sufficient lifting equipment was in place to lift the experiment into a transfer flask.

A separate transfer chamber would be fabricated by Serco; it would be slightly larger in diameter than the support cage / model canister assembly. After removal of the support cage assembly from the borehole it would be loaded into the transfer flask and purged with an inert gas, but it would not be dried. An oxygen getter material would be placed in the transfer chamber (e.g. iron filings or steel wire) to minimise the residual oxygen concentration. The transfer flask would then be sealed with a bolted flanged lid and be transported to the U.K., using a specialist transport company. It would be necessary to ensure that there are no customs issues or restrictions that would impede or prevent transport of the experiment to the U.K.

Once in the U.K. at Serco’s laboratories the transfer flask would be placed in a specially constructed inert gas-purged glovebox, which would include a lifting device to enable the model canister to be withdrawn from the transfer flask. A cutting system, such as a rotary saw or band saw, would be mounted inside the glovebox, so that the model canister assembly could be sectioned in a controlled manner, without the use of lubricating fluids. Preliminary investigations and discussions held with suppliers indicate that such cutting systems and gloveboxes would be



available on a commercial basis. Once procured, this equipment would be available for dismantling other experiments in the future.

6.2 Analysis

It is suggested that a further campaign of water analysis should be carried out on all five experiments before removal of Experiment 3. This would cover all the species already analysed to date, as described in the current report, including ion analysis, gas analysis and microbial characterisation. The surface of the support cage would be swabbed for microbial analysis immediately after removal from the borehole and further samples for microbial analysis would be taken once the experiment was opened up in the U.K. Microbial Analytics would be involved in the microbial analysis.

After transfer of the experiments to the U.K., a number of analyses should be carried out on the components of the experiments, in order to extract as much information as possible from them. All dismantling and preparation of the samples would take place under an inert gas atmosphere (nitrogen or argon), in order to prevent alteration of the samples before analysis. Samples may be transferred from the purpose-built dismantling glovebox to an existing glovebox at Culham using small transfer flasks or glove bags, for further sample preparation. It is suggested that the analytical steps on the samples would include the following, working from the outside of the support cage to the inside:

Stainless steel support cage

• Overall visual examination, photographic record.

• Cut samples for sectioning and optical microscopic examination, to determine the extent of any pitting or possible microbial corrosion attack.

Bentonite in annulus of support cage

The support cage around the model canister would be carefully dismantled and photographed (i.e. the end caps lids will be removed). If cutting of the cylindrical filters is required it would probably be necessary to remove the support cage from the glovebox. Where possible, the homogeneity of the bentonite in the annulus would be examined (e.g. to determine whether any ‘channels’ have formed in the bentonite during water sampling). Samples of the bentonite would be removed for analysis. This analysis would initially concentrate on visual observation to assess whether any alteration appeared to have taken place in the bentonite. If required, a range of analytical techniques could be applied to the bentonite, such as those used in the recent NF-PRO project [19]. It is suggested that a decision on the exact analyses applied to the bentonite is postponed until the condition of the bentonite can be assessed visually. It may also be of interest to characterise microbial activity within the bentonite.

Corrosion coupons (weight loss)

The main aim of the analyses would be to measure weight loss, in order to derive a value of corrosion rate, for comparison with the electrochemically measured values. There is one weight loss specimen for copper and one for iron. These will be used for characterising the surface using non-destructive techniques prior to making the weight loss measurement. After the weight loss measurement has been completed the sample would be sectioned and examined microscopically for signs of localised corrosion. The techniques used for characterising the coupons (both iron and copper) would be as follows:

• Visual examination and photographs

• Raman spectroscopy in sealed inert gas-filled glass tubes to identify the composition of the corrosion product



• SEM/EDX examination of corrosion product

• If the surface films are found to be too thin to measure using SEM/EDX, surface analysis techniques such as X-ray photoelectron spectroscopy could be used to characterise the surface specific composition of the films and to measure film thickness by depth profiling.

• Weight loss measurement.

• Sectioning and optical microscopy to determine the depth of general or localised corrosion; SEM if localised corrosion is apparent.

Corrosion coupons (electrode samples)

The protocol for analysis of the coupons would be similar to that for the weight loss coupons but it would not be possible to measure the weight changes on the electrodes because wires are already attached to the coupons. Analysis of the composition of the films on the surface of the coupons will aid interpretation of the electrochemical data. For example, if it can be shown that the corrosion product films on the surface of the coupons are composed of metal sulphides, it may be possible to construct equivalent circuit models based on the assumed structure of the films.

Copper-cast iron-copper sandwich specimens

The dimensions of the specimens will be checked using vernier callipers and then carefully dismantled in an inert atmosphere to examine the extent of corrosion and the thickness of any corrosion products. Samples of corrosion product will be analysed using Raman spectroscopy and both the copper and cast iron specimens will be sectioned and examined using optical microscopy to investigate the extent of any localised corrosion.

Copper canister and cast iron insert

One of the main objectives of the programme is to investigate the evolution of corrosion around the annulus between the cast iron insert and the outer copper canister. After a thorough visual examination of the intact canister, including accurate measurement of the outer dimensions to check for any expansion or contraction, the outer copper canister will be carefully cut using a mechanical saw mounted in the custom-built glovebox. The canister will be cut into slices and the copper will then be peeled away from the cast iron to reveal the distribution of corrosion product around the annulus in relation to the position of the defect in the copper container. The distribution of corrosion product in the annulus will be photographed and documented. The questions raised at the outset of the project will be addressed (see Section 1), namely:

• Does water penetrate into the annulus through a small defect?

• How does corrosion product spread around the annulus from the leak point?

• Does the formation of corrosion product in a constricted annulus cause any expansive damage to the copper canister?

• What is the effect of water penetration on the insert lid seal?

• Is there any detectable corrosion at the copper welds?

• Are there any deleterious galvanic interactions between copper and cast iron?

• Does corrosion lead to failure of the lid on the iron insert?

Samples of both the copper and cast iron will be prepared for analysis to determine the composition of any surface films using SEM/EDX and Raman spectroscopy.




Sections from the electron beam welds in the copper canister will be prepared so that they can be investigated to determine whether there are any indications of localised corrosion (e.g. pitting, cracking, intergranular corrosion etc.).

The lid region of the cast iron inset will be carefully dismantled and examined for any signs of crevice corrosion around the gasket material.

Stress corrosion test specimens

The copper SCC WOL specimens present in the borehole of Experiment 3 will be examined to measure the length of the cracks, to determine whether exposure to the borehole water had led to any increase in crack length. Similarly, the U-bend specimens will be examined for indications of SCC initiation.

7 Conclusions The main conclusions from the project to date are as follows:

1. Water analysis has shown that there are compositional differences between the water inside the support cages compared to the external borehole water. These can be explained on the basis of enhanced corrosion of the iron and microbial activity inside the support cages.

2. Microbial analysis has demonstrated that SRBs are active in the experimental boreholes.

3. The electrochemically measured corrosion rates have accelerated in recent months for both copper and iron. These data need to be confirmed by weight loss measurements, when the experiments are dismantled. If they are found to be correct, they could be due to microbially influenced corrosion.

4. It is necessary to remove Experiment 3 for analysis to confirm the measurements made remotely to date using electrochemical methods.

8 Acknowledgments The authors gratefully acknowledge assistance provided by the following during the reporting period:

• SKB: Christina Lilja, Lars Werme, Richard Bäck, Mats Lundqvist, Teresita Morales, SKB chemistry laboratory staff at Äspö.

• Microbial Analytics: Karsten Pedersen and Sara Eriksson

The authors also gratefully acknowledge financial support provided by SKB for conducting this project.

9 References

1 N.R. Smart, P.A.H. Fennell and A.P. Rance, Expansion Due to Anaerobic Corrosion of Iron, SA/EIG/15031/C001, 2005.


2 N.R. Smart, R. Bäck, P.A.H. Fennell, G. Knowles, M. Lundqvist, A.P. Rance, B. Reddy, D. Spencer and L.O. Werme, In Situ Corrosion Testing Of Miniature Copper-Cast Iron Radioactive Waste Canisters, Corrosion 2007, Extended abstract, Research in Progress Symposium, Nashville, March 2007.

3 N.R. Smart and A.P. Rance, Miniature Canister Corrosion Experiment – Results of Operations to May 2008, SKB Technical Report TR-09-20, 2009.

4 N.R. Smart, A.P. Rance and B. Reddy, Miniature Canister (MiniCan) Corrosion Experiment Progress Report 1 for 2008-9, SERCO/TAS/E.003110.01/Issue 01, January 2010.

5 A.E. Bond, A.R. Hoch, G.D. Jones, A.J. Tomczyk, R.M. Wiggin and W.J. Worraker, Assessment of a Spent Fuel Disposal Canister. Assessment Studies for a Copper Canister with Cast Steel Inner Component, SKB Report TR-97-19, 1997.

6 T. Saario, Effect of the Degree of Compaction of Bentonite on the General Corrosion Rate of Copper, presented at EuroCorr 2004, Nice, 2004.

7 N.R. Smart, A.P. Rance and P.A.H. Fennell, Galvanic Corrosion of Copper-Cast iron Couples, SA/EIG/13974/C001, 2004.

8 ASTM standard G59-97e1, Standard Test Method for Conducting Potentiodynamic Polarization Resistance Measurements, 2003.

9 N.R. Smart, P.A.H. Fennell, A.P. Rance, L.O. Werme, Galvanic Corrosion of Copper-Cast Iron Couples in Relation to the Swedish Radioactive Waste Canister Concept, presented at Eurocorr 2004, Nice, September, 2004; Published in a separate volume: pg. 52 in Prediction of Long Term Corrosion Behaviour in Nuclear Waste Systems, Proceedings of 2nd International Workshop, Nice, September 2004, Eurocorr 2004, published by Andra in Science and Technology Series.

10 R. Peat, S. Brabon, P.A.H. Fennell, A.P. Rance and N.R. Smart, Investigation of Eh, pH and Corrosion Potential of Steel in Anoxic Groundwater, AEAT/R/PS-0028, issue 1, 2000 and SKB report TR-01-01, 2001.

11 N.R. Smart, P.A.H. Fennell, R. Peat, K. Spahiu and L. Werme, Electrochemical Measurements During The Anaerobic Corrosion Of Steel, Materials Research Society Symposium Proceedings Volume 663, Scientific Basis for Nuclear Waste Management XXIV, K.P. Hart and G.R. Lumpkin (eds.), p. 487-495, 2001.

12 F. King, L. Ahonen, C. Taxen, U. Vuorinen and L. Werme, Copper Corrosion Under Expected Conditions in a Deep Geological Repository, SKB report TR-01-23, 2001.

13 I. Puigdomenech, J-P Ambrosi, L Eisenlohr, J-E Lartigue, S.A. Banwart, K. Bateman, A.E. Milodowski, J.M. West, L. Griffault, E. Gustafsson, K. Hama, H. Yoshida, S. Kotelnikova, K. Pedersen, V. Michaud, L. Trotignon, J. Rivas Perez and E-L. Tullborg, O2 Depletion in Granitic Media. The REX project, SKB report TR-01-05, 2001.

14 N.R. Smart, D.J. Blackwood and L. Werme, Anaerobic Corrosion of Carbon Steel and Cast Iron in Artificial Groundwaters: Part 1–Electrochemical Aspects, Corrosion 58(7), 547, 2002.

15 N.R. Smart, D.J. Blackwood and L. Werme, Anaerobic Corrosion of Carbon Steel and Cast Iron in Artificial Groundwaters: Part 2-Gas Generation, Corrosion 58(8), 627, 2002.


http://www.skb.se/ppw_en/document.asp?ppwAutnRef=PUB-18352&id=3515&prevUrl=





16 B. Rosborg, D. Eden. O. Karnland, J. Pan and L. Werme, Real-time Monitoring of Copper Corrosion at the Äspö Laboratory, presented at Eurocorr 2004, Nice, September, 2004; Published in a separate volume: pg. 10 in Prediction of Long Term Corrosion Behaviour in Nuclear Waste Systems, Proceedings of 2nd International Workshop, Nice, September 2004, Eurocorr 2004, published by Andra in Science and Technology Series.

17 T. Saario, I. Betova, J. Heinonen, P. Kinnunen and C. Lilja, Effect of the Degree of Compaction of Bentonite on the General Corrosion Rate of Copper, presented at Eurocorr 2004, Nice, September, 2004; Published in a separate volume: pg. 45 in Prediction of Long Term Corrosion Behaviour in Nuclear Waste Systems, Proceedings of 2nd International Workshop, Nice, September 2004, Eurocorr 2004, published by Andra in Science and Technology Series.

18 N.R. Smart, Minutes of Model Canister Planning Meeting, 2 June 2009, SKB offices, Stockholm.

19 N.R. Smart, F. Bate, L. Carlson, M.R. Cave, K. Green, T.G. Heath, A.R. Hoch, F. M. I. Hunter, O. Karnland, S.J. Kemp, A.E. Milodowski, S. Olsson, A. M. Pritchard, A. P. Rance, R.A Shaw, H. Taylor, B Vickers, L. O. Werme and C.L. Williams,, Interactions Between Iron Corrosion Products And Bentonite, Serco/TAS/MCRL/19801/C001 Issue 2, 2008.



Experiment number (Borehole number) Environment and defect details

1 (KA3386A02) Low density, high permeability bentonite in annulus around model canister. Single defect

near top weld, pointing vertically upwards.

2 (KA3386A03) Low density, high permeability bentonite in annulus around model canister. Single defect near top weld, pointing vertically downwards.

3 (KA3386A04) Low density, high permeability bentonite in annulus around model canister. Two defects, one near top weld, pointing vertically upwards,

near bottom weld, pointing vertically downwards.

4 (KA3386A05) Compacted bentonite. Single defect near top weld, pointing vertically downwards.

5 (KA3386A06) No bentonite, direct contact of groundwater on model canister. Two defects, both near top

weld, one pointing vertically upwards, the other vertically downwards.

Table 1. Summary of test conditions for model canister experiments.



SO42- (mgL-1)

HCO3- (mgL-1)

S2- (mgL-1)

Fe2+ (mgL-1)

pH

Experiment

05-2007

10-2008

05-2007

10-2008

05-2007

10-2008

05-2007

10-2008

05-2007

10-2008

1 486 417 28 19 0.036 0.021 11.1 19.2 7.4 7.3

2 506 413 27 20 0.057 0.020 2.33 4.15 7.6 7.4

2G 354 481 28 23 0.062 0.023 0.16 0.15 7.6 7.6

3 439 410 51 38 0.037 0.022 0.82 15.7 7.6 7.2

5 605 567 14 8 0.051 0.030 6.30 11.1 7.6 6.8

Note: It is not possible to extract any water for analysis from inside the support cage of Experiment 4, because of the presence of compacted bentonite.

Table 2. Chemical composition in the water from inside and the outside (2G) of the Model Canister experiments sampled in 05-2007 and 10-2008.




Experiment 1 Experiment 2 Experiment 2 (borehole) Experiment 3 Experiment 5

Cl (mg/l) 7887 7674 7926 6895 11362

Na (mg/l) 2350 2350 2330 2130 2930

Ca (mg/l) 1910 2470 2460 1930 3840

Total S (mg/l) 129 186 188 157 233

Br (mg/l) 49.8 44.7 48.3 35.7 71.3

Mg (mg/l) 47.9 51.9 51.3 76.0 42.9

Sr (mg/l) 36.7 46.6 46.6 37.8 73.2

K (mg/l) 9.01 11.6 11.4 11.0 13.5

Si (mg/l) 4.01 6.00 6.04 5.91 9.28

F (mg/l) 1.83 1.83 1.75 1.23 1.87

Li (µg/l) 1410 1820 1750 1360 2670

Mn (µg/l) 362 387 370 490 678

Ba (µg/l) 89.0 80.3 78.3 89.8 101

Mo (µg/l) 54.0 60.1 57.4 60.5 53.4

Zn (µg/l) 1.08 4.03 1.19 <0.8 4.33

Ni (µg/l) 0.866 2.87 <0.2 0.971 322

Al (µg/l) 0.745 4.42 3.20 1.37 1.17

Cu (µg/l) 0.251 0.767 <0.2 <0.2 0.283

Cu (µg/l) 0.251 0.767 <0.2 <0.2 0.283

V (µg/l) 0.199 0.275 0.291 0.215 0.0455

Pb (µg/l) 0.179 0.383 0.152 0.174 0.220

Cr (µg/l) 0.169 1.80 0.159 0.191 248

Co (µg/l) 0.0391 0.152 0.0526 0.022 8.12

NO2- (mg/l) 0.0013 0.0006 <0.0002 0.0008 0.0012

NO3- (mg/l) 0.0010 <0.0003 0.0015 0.0021 0.0006

PO43+ (mg/l) <0.0005 <0.0005 <0.0005 <0.0005 <0.0005

TOC (mg/l) 3.2 1.9 1.3 3.6 1.2

Conductivity, mS/m

2230 2178 2197 1975 2960

Table 3. Chemical composition of the water from inside and the outside (Experiment 2 only) the support cage around the model canister experiments, as sampled in October 2008

(measurement uncertainty ~±12%).


Gases Date Experiments

1 2 2 G 3 5

09-07 68 128 76 69 112 Gas/water (mL L-1) 10-08 90 109 103 53 86

09-07 0.22 0.52 0.22 0.10 215 H2 (µL L-1) 10-08 90.1 0.57 0.52 0.55 7.3 09-07 0.58 1.08 3.39 0.72 1.14 CO (µL L-1) 10-08 0.52 0.53 0.70 0.69 6.8 09-07 326 317 210 245 117 CH4 (µL L-1) 10-08 500 593 344 215 216 09-07 913 786 186 1850 843 CO2 (µL L-1) 10-08 338 284 338 568 394 09-07 0.33 0.20 0.09 0.12 0.65 C2H6 (µL L-1) 10-08 bda 018 bda 0.28 1.08 09-07 0.14 0.17 bda 0.02 bda C2H2-4 (µL L-1) 10-08 bda bda bda bda bda

C3H8 (µL L-1) 09-07 bda bda bda bda bda 10-08 bda bda bda 0.05 0.05 C3H6 (µL L-1) 09-07 bda bda bda bda bda 10-08 bda bda bda bda bda Ar (µL L-1) 09-07 838 1,060 649 849 710 10-08 1,110 1,190 740 185 bda He (µL L-1) 09-07 7,980 8,420 8,110 4,920 14,200 10-08 9,990 9,100 9,070 6,600 14,900 N2 (µL L-1) 09-07 57,800 118,000 67,100 61,100 95,600 10-08 77,400 91,900 98,200 44,900 71,200 O2(µL L-1) 09-07 0.00 0.00 0.00 0.00 0.00 10-08 0.00 0.00 0.00 0.00 0.00

Table 4. Dissolved gas in the water from the model canister experiments sampled in September 2007 and October 2008 (a - below detection level).



Experiment ATP

(amol mL-1) a

TNC (mL-1) a

Q-PCR 16S DNA

(mL-1)

CHAB (mL-1) a

MPN SRB (mL-1) c

Q-PCR active SRB

(relative activity)

Q-PCR total SRB

(mL-1)

MPN AA (mL-1) c

Q-PCR AA (relative activity)

Acetate (mg L-1)

1 2,400 ± 450

130,000 ± 16,000

na b 227 ± 42 5,000 (2,000 – 17,000)

10 na b 30,000 (2,000 – 17,000)

631 8.5

2 2,800 ± 240

95,000 ± 39,000

na b 200 ± 57 1,400 (600 – 3,600)

2 na b 50,000 (20,000 – 200,000)

10 8

2 G 3,300 ± 710

31,000 ± 11,000

na b 20 ± 35 3 (1 – 12) 3 na b 3 (1 – 12) 3 6.9

3 15,500 ± 810

150,000 ± 61,000

na b 530 ± 28 230 (90 – 860)

2 na b 7,000 (3,000 – 21,000)

336 6.9

5 3,000 ± 680

13,000 ± 4900

na b 17 ± 12 300 (100 – 1100)

1 na b bd 1 11

a ± standard deviation b not analysed c 95% confidence interval)

Table 5. The microbial composition inside and the outside (2G) the model canister experiments sampled in 05-2007.



Experiment ATP

(amol mL-1) a

TNC (mL-1) a

Q-PCR 16S DNA (mL-1)

CHAB (mL-1) a

MPN SRB (mL-1) c

Q-PCR active SRB

(mL-1)

Q-PCR total SRB

(mL-1) MPN AA (mL-1) c

Q-PCR active

AA (mL-1)

Acetate (mg L-1)

1 17,400 ± 2,500

180,000 ± 4,600

6,700,000 bd b 5,000 (2,000 – 17,000)

5,500 540,000 1.7 (0.7 – 4)

bd b 1.7

2 10,400 ± 690

180,000 ± 18,000

4,500,000 3 ± 6 800 (300 – 2,500)

31 37,000 17 (7 – 48)

bd b 1.6

2G 4,400 ± 570

2,500 ± 880

8,000,000 3 ± 6 2.3 (0.9 – 8.6)

bd b bd b 0.4 (0.1 – 1.7)

bd b 2.9

3 27,400 ± 1820

200,000 ± 22,000

5,000,000 20 ± 17 70 (30 – 210)

350 74,000 90 (30 – 290)

bd b 1.8

5 18,600 ± 1,270

71,000 ± 11,000

14,000,000 bd b 110 (40 – 300)

bd b 11,000 2700 (1200 – 6700)

bd b 2.3

a ± standard deviation b below detection limit c 95% confidence interval)

Table 6. The microbial composition inside and outside (2G) the model canister experiments sampled in October 2008.



Figure 1. Layout of model canister experiments in borehole.



Figure 2. Layout of model canister and sensors inside support cage.



Figure 3. Left: Strain gauges mounted on model canister; Centre: Positioning of bentonite pellets inside annulus of support cage; Right: insertion of model canister assembly into support cage.



Experiment 1: low density bentonite

0.1

1

10

100

1000

10000

18.1.07 9.3.07 28.4.07 17.6.07 6.8.07 25.9.07

Sampling date

Con

cent

ratio

n (m

g/l)

NaKCaMg HCO3ClSO4BrFSiMnLiSr

Figure 4. Summary of water analyses: Experiment 1. For any given sampling date the first concentration shown when approaching along the line from the left hand side of the diagram refers to the analysis in the borehole, but outside the support cage, and the point on the right hand side refers to the concentration on the inside of the support cage. The results of the iron analyses are shown in Figure 10.



Experiment 2: Low density bentonite

0

1

10

100

1000

10000

18.1.07 9.3.07 28.4.07 17.6.07 6.8.07 25.9.07

Sampling date

Con

cent

ratio

n (m

g/l)





Experiment 3: Low density bentonite

0.1

1

10

100

1000

10000

18.1.07 9.3.07 28.4.07 17.6.07 6.8.07 25.9.07

Sampling date

Con

cent

ratio

n (m

g/l)





Experiment 4: Compacted bentonite

0

1

10

100

1000

10000

24.12.06 12.2.07 3.4.07 23.5.07 12.7.07 31.8.07 20.10.07

Sampling date

Con

cent

ratio

n (m

g/l)


Figure 7. Summary of water analyses: Experiment 4. It was not possible to remove any water from inside the support cage for analysis, so al data refer to a water sample taken from the borehole outside the support cage. The results of the iron analyses are shown in Figure 10.



Experiment 5: no bentonite

0.1

1

10

100

1000

10000

100000

18.1.07 9.3.07 28.4.07 17.6.07 6.8.07 25.9.07

Sampling date

Con

cent

ratio

n (m

g/l)





Sulphide concentration

0

0.05

0.1

0.15

0.2

0.25

18/01/2007 09/03/2007 28/04/2007 17/06/2007 06/08/2007 25/09/2007

Sampling date

Con

cent

ratio

n (m

g/l)

Expt 1 - in cageExpt 2 - in cageExpt 3 - in cageExpt 5 - in cageExpt 1 - in boreholeExpt 2 - in boreholeExpt 3 - in boreholeExpt 4 - in boreholeExpt 5 - in borehole

Figure 9. Summary of sulphide concentration in water samples.



Iron content of water

0.1

1

10

100

18.1.07 9.3.07 28.4.07 17.6.07 6.8.07 25.9.07

Sampling date

Con

cent

ratio

n (m

g/l)

Expt 1 - in cage

Expt 2 - in cage

Expt 3 - in cage

Expt 5 - in cage

Expt 1 - in borehole





Figure 10. Summary of iron analysis in water samples.



6.4

6.6

6.8

7

7.2

7.4

7.6

7.8

18.1.07 9.3.07 28.4.07 17.6.07 6.8.07 25.9.07 14.11.07Sampling date

pH

Expt 1 - in cage

Expt 2 - in cage

Expt 3 - in cage

Expt 5 - in cage






Figure 11. pH of water samples taken from model canister experiments.



Figure 12. Pressure reading for bore hole 1 to 5 (pressure inside support cage, no measurements possible for Experiment 4 because of presence of compacted bentonite).



Figure 13. Results of Eh and canister potential measurements from Experiment 1 (low density bentonite).



Figure 14. Results of corrosion potential and corrosion rate measurements for cast iron and copper electrodes in Experiment 1 (low density bentonite).



Figure 15. Results of Eh and can potential measurements from Experiment 2 (low density bentonite).






Figure 17. Results of Eh and can potential measurements from Experiment 3 (low density bentonite).






Figure 19. Results of Eh and can potential measurements from Experiment 4 (compact bentonite).



Figure 20. Results of corrosion potential and corrosion rate measurements for copper electrodes in Experiment 4 (compacted bentonite).



Figure 21. Results of Eh and can potential measurements from Experiment 5 (no bentonite).



Figure 22. Results of corrosion potential and corrosion rate measurements for cast iron and copper electrodes in Experiment 5 (no bentonite).



(a) (b)

Figure 23. Typical examples of linear polarisation resistance (LPR) measurements on Experiment 1: (a) cast iron, (b) copper.



(a) (b)

Figure 24. Typical examples of AC impedance measurements (ACI) on Experiment 1: (a) cast iron, (b) copper. Top figure: blue line = theta plot; red line = impedance plot.



Figure 25. Summary of copper corrosion rates obtained by AC impedance and LPR measurements.



Figure 26. Summary of copper corrosion rates obtained by AC impedance and LPR measurements (Experiment 4).



Figure 27. Summary of cast iron rates obtained by AC impedance and LPR measurements.



Figure 28. Summary of cast iron rates obtained by AC impedance and LPR measurements (Experiment 4)



Figure 29. Results of corrosion measurement from copper wire electrical resistance measurement (Experiment 2: low density bentonite).



Figure 30. Results of corrosion measurement from copper wire electrical resistance measurement (Experiment 5: no bentonite).




Figure 31. Strain gauge results from Experiments 1 and 4.


Experiment 1(Cast Iron) LPR plots

-0.04

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

0.04

0.05

-100 -50 0 50 100 150 200

Potential (mV)

Cur

rent

(mA

/cm

2 )07-Mar-0725-May-07

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

-700 -690 -680 -670 -660 -650 -640 -630 -620 -610 -600

Potential (mV)

Cur

rent

(mA

/cm

2 )

11-Jul-0711-Oct-07

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

-700 -690 -680 -670 -660 -650 -640 -630

Potential (mV)

Cur

rent

(mA

/cm

2 )

23-May-0820-Jun-0831-Jul-0812-Aug-0811-Oct-0812-Nov-0810-Feb-0923-Apr-0901-Sep-0927-Oct-09

Figure 32. LPR plots of cast iron in Experiment 1 (low density bentonite).



Experiment 1(Cu) LPR plots

-0.003

-0.002

-0.001

0

0.001

0.002

0.003

-60 -10 40 90 140

Potential (mV)

Cur

rent

(mA

/cm

2 )07-Mar-0723-May-07

-0.004

-0.003

-0.002

-0.001

0

0.001

0.002

0.003

0.004

-700 -690 -680 -670 -660 -650 -640 -630 -620 -610 -600

Potential (mV)

Cur

rent

(mA

/cm

2 )

29-Jun-0701-Nov-0711-Nov-07

-0.02

-0.015

-0.01

-0.005

0

0.005

0.01

0.015

0.02

-690 -680 -670 -660 -650 -640 -630 -620 -610 -600

Potential (mV)

Cur

rent

(mA

/cm

2 )

23-May-0820-Jun-0831-Jul-0822-Aug-0811-Oct-0814-Nov-0810-Feb-0923-Apr-0901-Sep-0927-Oct-09

Figure 33. LPR plots of copper in Experiment 1 (low density bentonite).




-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

-800 -780 -760 -740 -720 -700 -680 -660 -640

Potential (mV)

Cur

rent

(mA

/cm

2 )

25-May-0711-Jul-0711-Nov-07

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

-670 -665 -660 -655 -650 -645 -640 -635 -630 -625 -620

Potential (mV)

Cur

rent

(mA

/cm

2 )

20-Feb-0831-Mar-0823-May-0820-Jun-0831-Jul-0822-Aug-0811-Oct-0812-Nov-0810-Feb-0923-Apr-0901-Sep-0926-Oct-09





-0.15

-0.1

-0.05

0

0.05

0.1

0.15

-700 -690 -680 -670 -660 -650 -640 -630

Potential (mV)

Cur

rent

(mA

/cm

2 )

11-Jul-0701-Nov-0710-Nov-07

-0.05

-0.04

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

0.04

0.05

-660 -655 -650 -645 -640 -635 -630 -625 -620 -615 -610

Potential (mV)

Cur

rent

(mA

/cm

2 )

20-Feb-0830-May-0820-Jun-0831-Jul-0822-Aug-0811-Oct-0812-Nov-0810-Feb-0923-Apr-0901-Sep-0927-Oct-09





-0.008

-0.006

-0.004

-0.002

0

0.002

0.004

0.006

0.008

-800 -790 -780 -770 -760 -750 -740

Potential (mV)

Cur

rent

(mA

/cm

2 )25-May-0729-Jun-0711-Jul-07

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

-50 -40 -30 -20 -10 0 10 20 30 40 50

Potential (mV)

Cur

rent

(mA

/cm

2 )

01-Nov-0711-Nov-07

-0.08

-0.06

-0.04

-0.02

0

0.02

0.04

0.06

0.08

-655 -650 -645 -640 -635 -630 -625 -620 -615 -610

Potential (mV)

Cur

rent

(mA

/cm

2 )

28-Mar-0820-Jun-0831-Jul-0822-Aug-0811-Oct-0812-Nov-0810-Feb-0901-Sep-0924-Apr-0926-Oct-09




Experiment 3 (Cu) LPR plots

-2.00E-03

-1.50E-03

-1.00E-03

-5.00E-04

0.00E+00

5.00E-04

1.00E-03

1.50E-03

2.00E-03

2.50E-03

-780 -770 -760 -750 -740 -730 -720

Potential (mV)

Cur

rent

(mA

/cm

2 )

25-May-0729-Jun-0711-Jul-07

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

0.04

-645 -640 -635 -630 -625 -620 -615 -610 -605 -600

Potential (mV)

Cur

rent

(mA

/cm

2 )

28-Mar-0830-May-0831-Jul-0811-Oct-0812-Nov-0810-Feb-0923-Apr-0901-Sep-0926-Oct-09





-8

-6

-4

-2

0

2

4

6

8

-600 -500 -400 -300 -200 -100 0

Potential (mV)

Cur

rent

(mA

/cm

2 )

29-Jun-0701-Nov-0710-Nov-07

-1.5

-1

-0.5

0

0.5

1

1.5

-530 -520 -510 -500 -490 -480 -470 -460 -450

Potential (mV)

Cur

rent

(mA

/cm

2 )

23-May-0820-Jun-0831-Jul-0822-Aug-0811-Oct-0810-Feb-0923-Apr-0901-Sept-0926-Oct-09

Figure 38. LPR plots of cast iron in Experiment 4 (compacted bentonite).




-6.00E-04

-5.00E-04

-4.00E-04

-3.00E-04

-2.00E-04

-1.00E-04

0.00E+00

1.00E-04

2.00E-04

3.00E-04

-420 -320 -220 -120 -20 80 180 280 380

Potential (mV)

Cur

rent

(mA

/cm

2 )

25-May-0731-Jul-0707-Mar-07

-1

-0.5

0

0.5

1

1.5

-520 -510 -500 -490 -480 -470 -460 -450

Potential (mV)

Cur

rent

(mA

/cm

2 )

10-Nov-0723-May-0831-Jul-0822-Aug-0803-Oct-0814-Nov-0810-Feb-0923-Apr-0901-Sep-0926-Oct-09

Figure 39. LPR plots of copper in Experiment 4 (compacted bentonite).




-0.06

-0.04

-0.02

0

0.02

0.04

0.06

-400 -300 -200 -100 0 100 200 300

Potential (mV)

Cur

rent

(mA

/cm

2 )

07-Mar-0724-May-0711-Jul-0728-Feb-08

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

-800 -750 -700 -650 -600 -550 -500 -450

Potential (mV)

Cur

rent

(mA

/cm

2 )

28-May-0820-Jun-0831-Jul-0811-Oct-0810-Feb-09

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

-310 -305 -300 -295 -290 -285 -280 -275 -270 -265 -260

Potential (mV)

Cur

rent

(mA

/cm

2 )

23-Apr-0901-Sep-0926-Oct-09

Figure 40. LPR plots of cast iron in Experiment 5 (no bentonite).



-1.E-03

-8.E-04

-6.E-04

-4.E-04

-2.E-04

0.E+00

2.E-04

4.E-04

6.E-04

8.E-04

1.E-03

-300 -250 -200 -150 -100 -50

Potential (mV)

Cur

rent

(mA

/cm

2 )11-Jul-0710-Nov-0719-Feb-0831-Mar-08

-1.0E-03

-8.0E-04

-6.0E-04

-4.0E-04

-2.0E-04

0.0E+00

2.0E-04

4.0E-04

6.0E-04

8.0E-04

1.0E-03

-650 -600 -550 -500 -450 -400 -350

Potential (mV)

Cur

rent

(mA

/cm

2 )

28-May-0831-Jul-0822-Aug-0828-Aug-0803-Oct-0810-Feb-09

-1.0E-03

-8.0E-04

-6.0E-04

-4.0E-04

-2.0E-04

0.0E+00

2.0E-04

4.0E-04

6.0E-04

8.0E-04

1.0E-03

-250 -230 -210 -190 -170 -150 -130 -110 -90 -70 -50

Potential (mV)

Cur

rent

(mA

/cm

2 )

23-Apr-0901-Sep-0926-Oct-09

Figure 41. LPR plots of Copper in Experiment 5 (no bentonite).



Experiment 1 (Cast Iron) ACI plots

0

50

100

150

200

250

0 50 100 150 200 250 300 350 400 450 500

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

17-May-0707-Mar-07

0

1

2

3

4

5

6

7

8

9

10

0 5 10 15 20 25 30Z' (ohm cm2)

Z'' (

ohm

cm

2 )

13-Jul-07

0

10

20

30

40

50

60

0 50 100 150 200 250

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

14-Nov-0725-Jan-0821-Feb-0828-Feb-0828-Mar-0820-Jun-0830-Jul-0822-Aug-0812-Nov-0820-Feb-0924-Apr-0901-Sep-0926-Oct-09

Figure 42. ACI plots of cast iron in Experiment 1 (low density bentonite).



Experiment 1 (Cu) ACI plots

0

20000

40000

60000

80000

100000

120000

0 10000 20000 30000 40000 50000 60000 70000 80000 90000

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

08-Mar-0711-Mar-0717-May-0708 Jun 0726-Jun-0713-Jul-0715-Nov-07

0

2000

4000

6000

8000

10000

12000

0 2000 4000 6000 8000 10000 12000 14000Z' (ohm cm2)

Z'' (

ohm

cm

2 )

27-Jan-0822-Feb-0802-Apr-0826-May-08

0

100

200

300

400

500

600

700

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Z' (ohms cm2)

Z'' (

ohm

s cm

2 )

21-Jun-0830-Jul-0823-Aug-0812-Nov-0816-Nov-0821-Feb-09

Figure 43. ACI plots of copper in experiment 1 (low density bentonite).




0

20000

40000

60000

80000

100000

120000

0 10000 20000 30000 40000 50000 60000 70000 80000 90000

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

08-Mar-0711-Mar-0717-May-0708 Jun 0726-Jun-0713-Jul-0715-Nov-07

0

2000

4000

6000

8000

10000

12000

0 2000 4000 6000 8000 10000 12000 14000Z' (ohm cm2)

Z'' (

ohm

cm

2 )

27-Jan-0822-Feb-0802-Apr-0826-May-08

0

100

200

300

400

500

600

700

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Z' (ohms cm2)

Z'' (

ohm

s cm

2 )

21-Jun-0830-Jul-0823-Aug-0812-Nov-0816-Nov-0821-Feb-0923-Apr-0901-Sep-0926-Oct-09

Figure 44. ACI plots of copper in Experiment 1 (low density bentonite).




0

200

400

600

800

1000

1200

1400

1600

0 200 400 600 800 1000 1200

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

17-May-07

08-Jun-07

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300Z' (ohm cm2)

Z'' (

ohm

cm

2 )

13-Jul-07

13-Nov-07

0

20

40

60

80

100

120

140

0 50 100 150 200 250 300

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

25-Jan-0821-Feb-0828-Feb-0828-Mar-0820-Jun-0830-Jul-0822-Aug-0812-Nov-0820-Feb-0923-Apr-0902-Sep-0926-Oct-09





0

200

400

600

800

1000

1200

0 500 1000 1500 2000 2500 3000 3500

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

17-May-07

08-Jun-07

13-Jul-07

13-Nov-07

0

20

40

60

80

100

120

140

160

0 100 200 300 400 500 600

Z' (ohms cm2)

Z'' (

ohm

s cm

2 )






0

2000

4000

6000

8000

10000

12000

14000

0 5000 10000 15000 20000 25000

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

17-May-07

08-Jun-07

0

50

100

150

200

250

300

350

0 100 200 300 400 500 600Z' (ohm cm2)

Z'' (

ohm

cm

2 )

13-Jul-07

13-Nov-07

0

20

40

60

80

100

120

0 50 100 150 200 250 300 350

Z' (ohm cm2)

Z'' (

ohm

cm

2 )






0

50000

100000

150000

200000

250000

300000

350000

400000

0 50000 100000 150000 200000 250000 300000 350000

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

08-Mar-07

17-May-07

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

0 10000 20000 30000 40000 50000 60000 70000 80000Z' (ohm cm2)

Z'' (

ohm

cm

2 )

26-Jan-08

21-Feb-08

28-Feb-08

0

20

40

60

80

100

120

0 50 100 150 200 250 300 350

Z' (ohm cm2)

Z'' (

ohm

cm

2 )






0

1000

2000

3000

4000

5000

6000

0 2000 4000 6000 8000 10000 12000

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

17-May-07

08-Jun-07

-2

0

2

4

6

8

10

12

0 2 4 6 8 10 12Z' (ohm cm2)

Z'' (

ohm

cm

2 )

13-Jul-07

13-Nov-07

-2

3

8

13

18

23

28

33

0 10 20 30 40 50 60

Z' (ohm cm2)

Z'' (

ohm

cm

2 )


Figure 49. ACI plots of cast iron in Experiment 4 (compacted bentonite).




0

50000

100000

150000

200000

250000

0 20000 40000 60000 80000 100000 120000 140000 160000

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

17-May-07

08 Jun 07

13-Jul-07

13-Nov-07

08-Mar-07

-2

0

2

4

6

8

10

12

0 2 4 6 8 10 12 14 16 18 20Z' (ohm cm2)

Z'' (

ohm

cm

2 )

25-Jan-0821-Feb-08

01-Apr-0824-May-08

-2

0

2

4

6

8

10

12

14

0 2 4 6 8 10 12 14 16 18 2

Z' (ohms cm2)

Z'' (

ohm

s cm

2 )

0


Figure 50. ACI plots of copper in Experiment 4 (compacted bentonite).




0

500

1000

1500

2000

2500

0 500 1000 1500 2000 2500

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

8-Mar-0716-May-0701-Jun-07

0

100

200

300

400

500

600

0 100 200 300 400 500 600 700 800 900

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

12-Jul-0702-Oct-0712-Nov-07

-20000

0

20000

40000

60000

80000

100000

120000

140000

160000

0 20000 40000 60000 80000 100000 120000 140000

Z' (ohm cm2)

Z'' (

ohm

cm

2 )

25-Jan-0821-Feb-0828-Feb-0828-Mar-0820-Jun-0830-Jul-0822-Aug-0812-Nov-0820-Feb-0902-Sept-0926-Apr-0927-Oct-09

Figure 51. ACI plots of cast iron in Experiment 5 (no bentonite).




0

500

1000

1500

2000

2500

0 500 1000 1500 2000 2500

Z' (ohm cm2)

Z'' (

ohm

cm

2 )08-Mar-0717-May-0708 Jun 0713-Jul-0715-Nov-07

-10

10

30

50

70

90

110

130

150

170

0 50 100 150 200 250 300 350Z' (ohm cm2)

Z'' (

ohm

cm

2 )

27-Jan-0822-Feb-0802-Apr-0826-May-08

Figure 52. ACI plots of copper in Experiment 5 (no bentonite)




0

20

40

60

80

100

120

0 20 40 60 80 100 120 140 160

Z' (ohms cm2)

Z'' (

ohm

s cm

2 )

21-Jun-0830-Jul-0823-Aug-0812-Nov-0816-Nov-0821-Feb-09

0

20000

40000

60000

80000

100000

120000

140000

160000

0 10000 20000 30000 40000 50000 60000 70000 80000 90000 100000

Z' (ohms cm2)

Z'' (

ohm

s cm

2 )

26-Apr-09

02-Sep-09

26-Oct-09

Figure 53. ACI plots of copper in Experiment 5 (no bentonite)


Experiment 1 (Cast Iron) Bode Plot

0

50

100

150

200

250

300

350

400

450

500

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

Thet

a (o )

23-May-07 (Impedence)23-May-07 (Theta)


0

5

10

15

20

25

30

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )

13-Jul-07 (Impedence)13-Jul-07 (Theta)


0

20

40

60

80

100

120

140

160

180

200

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )

11-Nov-07 (Impedence)11-Nov-07 (Theta)


0

50

100

150

200

250

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )

27-Jan-08 (Impedence)27-Jan-08 (Theta)

Figure 54. Bode plots of cast iron in Experiment 1 (low density bentonite).




0

50

100

150

200

250

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )

29-Feb-08 (Impedence)29-Feb-08 (Theta)


0

20

40

60

80

100

120

140

160

180

200

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )



0

20

40

60

80

100

120

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )

22-Aug-08 (Impedence)22-Aug-08 (Theta)


0

20

40

60

80

100

120

140

160

180

200

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )






0

20

40

60

80

100

120

140

160

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )



0

20

40

60

80

100

120

140

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

Thet

a (o )

23-Apr-09 (Impedence)23-Apr-09 (Theta)


0

10

20

30

40

50

60

70

0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-4

-2

0

2

4

6

8

10

12

14

16

18

Thet

a (o )

01-Sept-09 (Impedence)01-Sept-09 (Theta)


0

10

20

30

40

50

60

70

0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

Thet

a (o )

26-Oct-09 (Impedence)26-Oct-09 (Theta)




Experiment 1 (Cu) Bode Plot

0

20000

40000

60000

80000

100000

120000

140000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

80

90

Thet

a (o )



0

10000

20000

30000

40000

50000

60000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

2000

4000

6000

8000

10000

12000

14000

16000

18000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

Thet

a (o )


Figure 57. Bode plots of copper in Experiment 1 (low density bentonite).




0

100

200

300

400

500

600

700

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

Thet

a (o )



0

2000

4000

6000

8000

10000

12000

14000

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

Thet

a (o )



0

500

1000

1500

2000

2500

3000

3500

4000

4500

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

Thet

a (o )



0

100

200

300

400

500

600

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

30

35

40

45

Thet

a (o )






0

200

400

600

800

1000

1200

1400

1600

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

200

400

600

800

1000

1200

1400

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

50

100

150

200

250

300

350

400

450

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-10

0

10

20

30

40

Thet

a (o )



0

50

100

150

200

250

300

350

400

450

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-10

0

10

20

30

40

Thet

a (o )

26 Oct-09 (Impedence)26-Oct-09 (Theta)





0

100

200

300

400

500

600

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

Thet

a (o )



0

20

40

60

80

100

120

140

160

180

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )



0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

45

Thet

a (o )


Figure 60. Bode plots of Cast Iron in Experiment 2 (low density bentonite).




0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

45

Thet

a (o )



0

50

100

150

200

250

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

10

20

30

40

50

60

70

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

Thet

a (o )



0

10

20

30

40

50

60

70

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

Thet

a (o )






0

10

20

30

40

50

60

70

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

Thet

a (o )



0

20

40

60

80

100

120

140

160

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

45

Thet

a (o )



0

10

20

30

40

50

60

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-4

-2

0

2

4

6

8

10

12

14

16

Thet

a (o )



0

10

20

30

40

50

60

70

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-4

-2

0

2

4

6

8

10

12

14

16

Thet

a (o )






0

500

1000

1500

2000

2500

3000

3500

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

Thet

a (o )



0

20

40

60

80

100

120

140

160

180

200

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

-5

0

5

10

15

20

25

30

35

40

45

Thet

a (o )



0

100

200

300

400

500

600

700

800

900

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

Thet

a (o )



0

100

200

300

400

500

600

700

800

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

Thet

a (o )






0

100

200

300

400

500

600

700

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

45

Thet

a (o )



0

50

100

150

200

250

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

50

100

150

200

250

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

100

200

300

400

500

600

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )






0

50

100

150

200

250

300

350

400

450

500

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )



0

20

40

60

80

100

120

140

160

180

200

0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-15

-10

-5

0

5

10

15

20

25

30

Thet

a (o )



0

100

200

300

400

500

600

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

-5

0

5

10

15

20

25

30

Thet

a (o )



0

20

40

60

80

100

120

140

160

180

200

0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

-5

0

5

10

15

20

25

30

35

Thet

a (o )






0

20

40

60

80

100

120

140

160

180

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-10

0

10

20

30

40

50

Thet

a (o )



0

5000

10000

15000

20000

25000

30000

35000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

Thet

a (o )



0

100

200

300

400

500

600

700

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

Thet

a (o )



0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )






0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )

28-Mar-08 (Impedence)28-Mar-08 (Theta)


0

50

100

150

200

250

300

350

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

Thet

a (o )



0

20

40

60

80

100

120

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )



0

20

40

60

80

100

120

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )






0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20



0

50

100

150

200

250

300

350

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

Thet

a (o )



0

20

40

60

80

100

120

140

160

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-15

-10

-5

0

5

10

15

20

25

Thet

a (o )



0

20

40

60

80

100

120

140

160

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-15

-10

-5

0

5

10

15

20

25

Thet

a(o )






0

50000

100000

150000

200000

250000

300000

350000

400000

450000

500000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

80

90

Thet

a (o )



0

500

1000

1500

2000

2500

3000

3500

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

10000

20000

30000

40000

50000

60000

70000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )






0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

70

Thet

a (o )

20-Jun-08 (Impedence)20-Jun-08 (Theta)


0

500

1000

1500

2000

2500

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

Thet

a (o )



0

500

1000

1500

2000

2500

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

Thet

a (o )



0

500

1000

1500

2000

2500

3000

3500

4000

4500

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

60

Thet

a (o )






0

500

1000

1500

2000

2500

3000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

0

10

20

30

40

50

Thet

a (o )



0

50

100

150

200

250

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-10

0

10

20

30

40

Thet

a (o )



0

50

100

150

200

250

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-10

0

10

20

30

40

50

Thet

a (o )



0

200

400

600

800

1000

1200

1400

1600

1800

2000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

45

Thet

a (o )






0

50000

100000

150000

200000

250000

300000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

90

Thet

a (o )

17-May-07 (Impedence)"23-May-07 (Theta)


0

20000

40000

60000

80000

100000

120000

140000

160000

180000

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

90

Thet

a (o )



0

2

4

6

8

10

12

14

16

18

20

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

2

4

6

8

10

12

14

16

18

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )


Figure 72. Bode plots of copper in Experiment 4 (compacted bentonite).




0

2

4

6

8

10

12

14

16

18

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

2

4

6

8

10

12

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-3

-2

-1

0

1

2

3

4

5

6

7

8

Thet

a (o )

21-Jun-08 (Impedence)21-Jun-08 (Theta)


0

2

4

6

8

10

12

14

16

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-3

-2

-1

0

1

2

3

4

5

Thet

a (o )



0

2

4

6

8

10

12

14

16

18

20

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )






0

2

4

6

8

10

12

14

16

18

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

-5

0

5

10

15

20

25

30

35

Thet

a (o )



0

2

4

6

8

10

12

14

16

18

20

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )



0

1

2

3

4

5

6

7

8

9

10

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-8

-6

-4

-2

0

2

4

6

8

Thet

a (o )



0

1

2

3

4

5

6

7

8

9

10

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-8

-6

-4

-2

0

2

4

6

Thet

a (o )






0

500

1000

1500

2000

2500

3000

3500

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

30

35

40

45

50

Thet

a (o )



0

100

200

300

400

500

600

700

800

900

1000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

30

35

Thet

a (o )



0

50

100

150

200

250

300

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

Thet

a (o )



-5

0

5

10

15

20

25

30

35

0.0001 0.001 0.01 0.1 1 10 100 1000 10000

Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

50

100

150

200

250

300

350

Thet

a (o )


Figure 75. Bode plots of cast iron in Experiment 5 (no bentonite).




0

50

100

150

200

250

300

350

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

Thet

a (o )

30-Mar-08 (Impedence)30-Mar-08 (Theta)


0

50

100

150

200

250

300

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

Thet

a (o )



0

10

20

30

40

50

60

70

80

90

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )



0

20

40

60

80

100

120

140

160

180

200

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

30

35

40

Thet

a (o )






0

20

40

60

80

100

120

140

160

180

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

30

35

40

45

Thet

a (o )



0

20

40

60

80

100

120

140

160

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

5

10

15

20

25

30

35

40

45

Thet

a (o )



0

10

20

30

40

50

60

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-5

0

5

10

15

20

25

Thet

a (o )



0

5

10

15

20

25

30

35

40

45

50

0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-10

-5

0

5

10

15

20

25

Thet

a (o )






0

20000

40000

60000

80000

100000

120000

140000

160000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

90

Thet

a (o )



0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

90

Thet

a (o )



0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

90

Thet

a (o )



0

20000

40000

60000

80000

100000

120000

140000

160000

180000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

Thet

a (o )


Figure 78. Bode plots of copper in Experiment 5 (no bentonite).




0

20000

40000

60000

80000

100000

120000

140000

160000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

20000

40000

60000

80000

100000

120000

140000

160000

180000

200000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

90

Thet

a (o )



0

20000

40000

60000

80000

100000

120000

140000

160000

180000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

5000

10000

15000

20000

25000

0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

Thet

a (o )







0

20000

40000

60000

80000

100000

120000

140000

160000

180000

200000

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-10

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

20000

40000

60000

80000

100000

120000

140000

160000

180000

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

0

10

20

30

40

50

60

70

80

Thet

a (o )



0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000Frequency (Hz)

Impe

denc

e (o

hm.c

m2 )

-20

-10

0

10

20

30

40

50

60

70

80

Thet

a (o )




Figure 81. Results from electrochemical noise measurements for a pair of copper electrodes in Experiment 2 (low density bentonite).



Figure 82. Results from electrochemical noise measurements for cast iron in Experiment 1 (low density bentonite).

Figure 83. Results from electrochemical noise measurements for cast iron in Experiment 1 (low density bentonite).



Figure 84. Results from electrochemical noise measurements for copper electrode in Experiment 1 (low density bentonite).



Figure 85. Results from electrochemical noise measurements for copper electrode in Experiment 1 (low density bentonite)



Figure 86. Results from electrochemical noise measurements for cast iron electrode in Experiment 2 (low density bentonite).



Figure 87. Results from electrochemical noise measurements for cast iron electrode in Experiment 2 (low density bentonite)





















Figure 94. Results from electrochemical noise measurements for cast iron electrode in Experiment 4 (compacted bentonite).



Figure 95. Results from electrochemical noise measurements for cast iron electrode in Experiment 4 (compacted bentonite).



Figure 96. Results from electrochemical noise measurements for copper electrode in Experiment 4 (compacted bentonite).



Figure 97. Results from electrochemical noise measurements for copper electrode in Experiment 4 (compacted bentonite)



Figure 98. Results from electrochemical noise measurements for cast iron electrode in Experiment 5 (no bentonite).



Figure 99. Results from electrochemical noise measurements for cast iron electrode in Experiment 5 (no bentonite).



Figure 100. Results from electrochemical noise measurements for copper electrode in Experiment 5 (no bentonite).




Figure 101. Results from electrochemical noise measurements for copper electrode in Experiment 5 (no bentonite).


Appendices

Contents

Appendix 1 Microbial and Gas Analysis




Appendix 1 Microbial and Gas Analysis

Contents

1. Introduction 2. Materials and methods 3. Results 4. Discussion 5. Conclusions 6. Future research 7. References

This appendix contains the results of water analysis, gas analysis and microbial analysis and was prepared by Microbial Analytics, Sweden AB

Investigations of MINICAN Sampling and analyses of gases and microorganisms in

the water from MINICAN in 2007 - 2008

Compiled by Sara Eriksson, Microbial Analytics Sweden AB

March 2009


1. Introduction 1.1 The MINICAN experiment

The MINICAN project is located at the depth of 450m in the Äspö Hard Rock Laboratory (HRL) research tunnel. MINICAN consists of five different experiment canisters (Table A-1), denoted experiment A02 – A06. Four of the MiniCan test copper canisters are surrounded by bentonite in an outer cage, of which the bentonite in experiment A05 is fully compacted according to the KBS-3 approach (dry density 1600 kg m-3) and experiments A03, A04 and A06 are compacted with bentonite to a lower density than will be used (dry density 1300 kg m-3). Experiment A02 lacks bentonite. In all the MiniCan copper canisters, holes with a diameter of 1 mm have been introduced to allow Äspö groundwater to come in contact with the interior cast iron inserts.

Table A-1. The MiniCan experiments installed at the Äspö Hard Rock Laboratory

Experiment name

Sampling point Bentonite dry density Introduced leak into the copper canister

A02 (Expt 1) KA3386 A02 Highly permeable, low density bentonite (1300 kg m-3) in cage around test canister

Hole (1 mm in diameter) located at the top of the test canister


Hole (1 mm in diameter) located at the bottom of the test canister


Holes (1 mm in diameter) located both at the bottom and top of the test canister

A05 (Expt 4) KA3386 A05 Fully compacted bentonite (1600 kg m-3) from blocks around the test canister

Hole (1 mm in diameter) located at the top of the test canister

A06 (Expt 5) KA3386 A06 No bentonite, groundwater in direct contact with the test canister

Two holes (1 mm in diameter) located at the top of the test canister

1.2 Microbial corrosion and hydrogen gas metabolism

The experimental part of the MiniCan project started in 2007 which aims at, during a five year period, to examine how the corrosion of the cast iron insert develops inside perforated copper canisters. In the real waste repository, corrosion of the cast iron will, in a worst-case scenario, expose the spent nuclear fuel to groundwater and release radionuclides into the surroundings. Another potential risk with corrosion is that hydrogen gas can be produced because when iron comes into contact with anoxic water, cathodic hydrogen is formed at the iron surface (King and Miller, 1971). Hydrogen gas is unwanted in the KBS-3 storage for two reasons. The first is that an increase in gas volume will build up the pressure inside the system. The second is that development of hydrogen gas is closely linked to activity of sulphate-reducing bacteria (SRB). Hydrogen gas may be used as an excellent energy source for many of the microbes in the deep granitic subsurface (Pedersen, 1999), in particular the SRB.



SRB utilise both inorganic and organic energy sources, of which the major subsurface sources are hydrogen gas and acetate, respectively. Cathodic hydrogen is believed to be directly scavenged by SRB if they possess the enzyme hydrogenase (Cord-Ruwisch and Widdel, 1986). In addition, there are indications that SRB (and methanogens) can use metallic iron directly in their metabolism as their electron donor (Dinh et al., 2004).

The direct corrosion net formula is given as Equation 1:

4Fe + SO42- + 4H2O FeS + 3Fe2+ + OH-

(Dinh et al., 2004)

SRB living on organic compounds such as acetate can produce sulphide, which can indirectly corrode the iron chemically and form FeS.

The indirect reaction net formula is given as Equation 2:

2[CH2O(organic carbon)] + 1⅓Fe + 1⅓SO42- + ⅔H+ 2HCO3

- + 1⅓FeS + 1⅓H2O

(Dinh et al., 2004)

Acetate can be produced by subsurface bacteria called autotrophic acetogens (AA), with hydrogen gas and carbon dioxide as building blocks.

Consequently, one of the main questions to be answered in this project was whether microbial activity could be connected to high (or low) levels of hydrogen gas in the MINICAN experiments and what other factors could affect as well as be the consequence of this activity. This question has been approached by comparing the gas composition to the microbial presence and activity in the groundwater close to the canisters. In May, August and September 2007 and in October 2008, analyses of the microbial presence and activity, chemistry and dissolved gas in the water from the space between the copper canisters and the bentonite cage in the MiniCan experiments A02 – A04, A06 and in the groundwater in experiment A03 were performed. The analyses aimed to investigate if and how different types of microbes thrived inside the MiniCan experiments, and how they could potentially be linked to corrosion processes. Sampling and analysis of gases (hydrogen, carbon monoxide, argon, carbon dioxide, methane, ethane, ethane, ethylene, propane, propene, propyn, helium, oxygen and nitrogen) and microorganisms (i.e. quantitative most probable number (MPN) of sulphate-reducing and acetogenic microbes, culturable heterotrophic aerobic bacteria (CHAB) analysis and adenosinetriphosphate (ATP) measurements) were performed. In addition molecular methods determining the DNA and RNA pools for detection of species-specific microbial presence as well as activity were developed and successfully used.

2. Material & Methods 2.1 Sampling occasions and analyses performed

Samplings of water for gas-, microbe- and chemical analyses from the MINICAN experiments were performed both in 2007 and 2008. In Table A-2 and Table A-3 the sampling dates and analyses performed are shown.



Table A-2. Analyses performed in May, August and September 2007 and in October 2008 in the groundwater and water by the canister in the MINICAN experiments.

Experiment name

Sampling point

Sampling date dissolved gas

Sampling date microbes

Sampling date water chemistry

A02-C KA3386 A02-canister

2007-09-28, 2008-10-15 2007-08-21, 2008-10-15 2007-05-22, 2008-10-15


2007-09-28, 2008-10-15 2007-08-21, 2008-10-15 2007-05-22, 2008-10-15

A03-G KA3386 A03-groundwater

2007-09-28, 2008-10-15 2007-08-21, 2008-10-15 2007-05-22, 2008-10-15


2007-09-28, 2008-10-15 2007-08-21, 2008-10-15 2007-05-22, 2008-10-15


a a a


2007-09-28, 2008-10-15 2007-08-21, 2008-10-15 2007-05-22, 2008-10-15

a Water could not be extracted from experiment A05-C.

Table A-3. Parameters analysed in the groundwater and water by the canister in the MINICAN experiments.

Gas analyses Microbe analyses Chemical analyses Sampling

vessel Analyses Sampling

vessel Analyses Sampling

vessel Analyses

PVB sampler H2, CO2, CO, CH4, C2-3H2-8, O2, He, Ar, N2

10 - 100-mL anaerobic tube or flask

TNC, ATP, CHAB, MPN SRB and AA, 16S rDNA cloning and sequencing or 16S rDNA Q-PCR enumeration, Q-PCR SRB and AA, acetate

According to SKB standard

SO42-, HCO3

-

Fe2+, S2-

pH

2.2 Sampling procedures

A sterile tube and junctions with a mounted valve, stopcock and a needle was attached to the connection of each experimental sampling point listed in Table A-2. An anaerobic sampling vessel was attached to the sterile tube and the needle was penetrated through the septa of the vessel. During sampling, the stopcock and the outflow from the SKB MINICAN connections were opened and the water filled the sampling vessels. An additional needle was penetrated through the septum of the sampling vessel shortly after the sampling begun, to eliminate dangerous pressure build-up.

In 2007, the first 50 mL of water sampled from the MINICAN experiments flushed the sample equipment and were discarded. Thereafter, the following volumes of water were sampled for analyses according to Table A-3: Q-PCR and 16S rDNA cloning and sequencing (350 mL), ATP and TNC sampling (30 mL), MPN SRB (15 mL), CHAB (10 mL), acetate (10 mL), MPN AA (10 mL) and dissolved gas (200 mL). The chemistry sampling was performed by SKB in 2007-05-22.



In 2008, the first 30 mL of water sampled from the MINICAN experiments flushed the sample equipment and were discarded. Thereafter, the following volumes of water were sampled for analyses according to Table A-3: acetate (10 mL), MPN SRB (10 mL), MPN AA (10 mL), ATP (30mL), TNC (30 mL), CHAB (30 mL), Q-PCR (600 mL), dissolved gas (250 mL) and chemistry (2500 mL).

2.3 Analyses and statistical evaluation

Gas analysis was performed as previously described by Pedersen et al. (2008b) and TNC, MPN SRB, MPN AA, CHAB and ATP analyses were performed according to Hallbeck & Pedersen (2008). Acetate was determined using the ATP Biomass Kit HS for determining total ATP in living cells (no. 266-311; BioThema, Handen, Sweden). 16S rDNA cloning and sequencing, 16S rDNA enumeration, Q-PCR SRB and Q-PCR AA was performed as described in Pedersen et al. (2008a).

Statistical analyses and graphics were performed using STATISTICA software, version 8.0 (Statsoft, Tulsa, OK, USA).

3. Results 3.1 Gas composition in MINICAN

The gas compositions inside and outside the MINICAN experiments are shown in Table A-4 and Table A-5.

Total gas volume

The total gas volume dissolved in the water inside the four MINICAN experiments where water was extractable showed a gas volume increase from 2007 to 2008 (with 32%) in one of the experiments, A02-C. Experiments A03-C, A04-C and A06-C instead showed a decrease in gas volume per water volume (with 15 – 25%).

Hydrogen gas

Generally in 2007, the water both outside and inside the experiments contained fairly low concentrations of hydrogen gas (0.10 – 0.52 µL L-1), with the exception of the water in experiment A06-C, which contained 215 µL L-1, approximately 130-2200 times more than in the other samples at the time. Interestingly, in 2008 the hydrogen content in experiment A02-C had increased to 90 µL L-1. This was approximately 160-170 times more than in experiments A03-C and A04-C (which still contained 0.57 and 0.55 µL L-1), and 12 times more than in experiment A06-C in which the hydrogen content had decreased to 7 µL L-1.

Methane

In 2007, the methane concentrations in the experiments ranged from 104 to 326 µL L-1. The methane contents were slightly higher in MINICAN experiments A02-C, A03-C and A04-C than in A03-G (1.2 – 1.6 times), and on the opposite lower inside MINICAN experiment A06-C compared to A03-G (0.6 times). In 2008, the methane content had increased in experiment A02-C and A03-C (500 – 593 µL L-1) compared to the year before and in experiment A04-C and A06-C (215 and 216 µL L-1). The methane content in A03-G was 344 µL L-1.

Carbon dioxide

Carbon dioxide content in the MINICAN experiments in 2007 was approximately four to ten times higher (786 – 1850 µL L-1) inside experiments A02-C, A03-C, A04-C and A06-C than in




A03-G (186 µL L-1). In 2008, the carbon dioxide contents were generally lower than in 2007. Experiments A02-C, A03-C and A06-C contained 284 – 394 µL L-1. Experiment A04-C contained 568 µL L-1, slightly more than in the other experiments but less than the previous year. A03-G contained more carbon dioxide than previous year, 338 µL L-1.

Carbon monoxide

In 2007, the content of carbon monoxide in the MINICAN experiments was two to six times higher in A03-G (3.4 µL L-1) than inside all the MINICAN experiments (0.6 – 1.1 µL L-1). The contents of carbon monoxide were similar inside the MINICAN experiments A02-C, A03-C and A04-C (0.5 – 0.7 µL L-1) in 2008 compared to the previous year. In experiment A06-C, the carbon monoxide content was higher in 2008 compared to 2007 (6.8 µL L-1).

Argon

Argon was detected inside the MINICAN experiments from below detection limit to 1190 µL L-1. No specific trends could be detected from 2007 to 2008.

Helium

Helium was apart from nitrogen the most abundant gas in the MINICAN experiment, ranging from 4920 – 14900 µL L-1. The content of helium in each specific sampling point was fairly constant from 2007 to 2008.

Nitrogen

The most abundant gas in the MINICAN experiment was nitrogen, ranging from 44900 – 118000 µL L-1. The content of nitrogen decreased slightly (with 25%) in most of the samples from 2007 to 2008, except in MINICAN experiment A02-C where the nitrogen content instead increased (with 34%).

Ethane, ethylene, ethane, propane and propene

The gases ethane, ethylene, ethane and propane were detected in trace amounts (in the order of 0.1 – 1.1 µL L-1) in some of the samples while the concentration of propene were below detection limits in all samples.


Table A-4. Dissolved gas in the water from the MINICAN experiments sampled in 2007-09-28. (NB. these data are also shown in Table 3 in the main body of the report).

Experiment name

Gas/water (mL L-1)

H2 (µL L-1)

CO (µL L-1)

CH4 (µL L-1)

CO2 (µL L-1)

C2H6 (µL L-1)

C2H2-4 (µL L-1)

C3H8 (µL L-1)

C3H6 (µL L-1)

Ar (µL L-1)

He (µL L-1)

N2 (µL L-1)

A02-C 68 0.22 0.58 326 913 0.33 0.14 bd a bd a 838 7980 57800

A03-C 128 0.52 1.08 317 786 0.20 0.07 bd a bd a 1060 8420 118000

A03-G 76 0.22 3.39 210 186 0.09 bd a bd a bd a 649 8110 67100

A04-C 69 0.10 0.72 245 1850 0.12 0.02 bd a bd a 842 4920 61100

A06-C 112 215 1.14 117 843 0.65 bd a bd a bd a 710 14200 95600 a below detection limit

Table A-5. Dissolved gas in the water from the MINICAN experiments sampled in 2008-10-15. (NB. these data are also shown in Table 5 in the main body of the report).

Experiment name

Gas/water (mL L-1)

H2 (µL L-1)

CO (µL L-1)

CH4 (µL L-1)

CO2 (µL L-1)

C2H6 (µL L-1)

C2H2-4 (µL L-1)

C3H8 (µL L-1)

C3H6 (µL L-1)

Ar (µL L-1)

He (µL L-1)

N2 (µL L-1)

A02-C 90 90.1 0.52 500 338 bd a bd a bd a bd a 1110 9990 77400

A03-C 109 0.57 0.53 593 284 0.18 bd a bd a bd a 1190 9100 91900

A03-G 103 0.52 0.70 344 338 bd a bd a bd a bd a 740 9070 98200

A04-C 53 0.55 0.65 215 568 0.28 bd a 0.05 bd a 185 6600 44900

A06-C 86 7.30 6.80 216 394 1.08 bd a 0.05 bd a bd a 14900 71200 a below detection limit





3.2 Microbial composition in MINICAN

The microbial compositions inside and outside the MINICAN experiments in 2007 and 2008 are shown in Table A-6 and A-8.

ATP

The amount of ATP inside the MINICAN experiments A02-C, A03-C, A04-C and A06-C in 2007 ranged from 1700 to 3000 amoles mL-1. The water in A03-G contained 3300 amoles ATP mL-1. The ATP content in this water in 2008 was in the same order of magnitude, 4400 amoles mL-1. Interestingly, the sampling in 2008 from inside the MINICAN experiments A02-C, A03-C, A04-C and A06-C consistently showed an increase in ATP content (between 1.7 and 7.3 times). The largest increase in ATP was found in MINICAN experiment A02-C (17400 amole mL-1, 7.3 times the content in 2007).

TNC

The TNC analyses of water from inside the MINICAN experiments in 2007 and 2008 showed approximately the same numbers (between 104 and 105 cells mL-1). There was a slight increase in cell numbers in MINICAN experiment A02-C and A06-C and a slight decrease in MINICAN experiments A03-C and A04-C. However, the TNC in A03-G showed a 10-fold decrease from 2007 to 2008 (from 3 × 104 to 3 × 103 cells mL-1).

Q-PCR 16S rDNA

In 2008, the total number of cells in the MINICAN experiment was also determined by Q-PCR of 16S rDNA. The numbers of 16S rDNA sequences in the waters in the MINICAN experiment were compared to 16S rDNA sequences in standard samples with of a known cell number. Generally, the cell numbers determined by this approach were higher than the numbers determined by counting (TNC). The cell numbers ranged from 4.5 × 106 to 1.4 × 107 and were from 50 to 2500 times larger compared to the TNC counting. The largest cell numbers found by this method were in A03-G and in A06-C.

CLONING AND SEQUENCING

In 2007, microbes in the examined waters were analysed by cloning and sequencing of a fragment of their 16S rDNA sequence. All unique sequences found in the MINICAN experiments were submitted to GenBank and given an accession number that was unique for the specific sequence. Those can be found at http://www.ncbi.nlm.nih.gov/. The accession numbers submitted to GenBank were EU714778-EU714782 and EU714788-EU714797. The sequences were then compared to the corresponding sequences of known microbes in the Gen Bank database. By doing this, it was possible to get an indication of which type of microbes that were dwelling in the MINICAN experiments. The results are shown in Table A-8.

In this evaluation, only microbes with 99% similarity or more to a sequence in the Gen Bank database were taken into consideration, since the 16S DNA sequence on which the assumptions were made can give uncertain results below that percentage. Two distinct types of microbes could be detected. The first group was found in the bentonite experiments and quite unexpectedly contained many human pathogens or close relative to those, i.e. Stenotrophmonas maltophilia, Streptococcus mitis, Veillonella parvula, Heamophilus parainfluenzae, Pseudomonas anguillaseptica and Propionibacterium acnes. The second group was more expected and contained known acetogens, i.e. Acetobacterim wieringue, A. malicum and A. carbinolicum, confirming qualitatively that acetogens were present in the MINICAN experiments.


http://www.ncbi.nlm.nih.gov/


CHAB

The numbers of CHAB decreased from 2007 to 2008. In 2007, the number of CHAB inside MINICAN experiments A02-C, A03-C and A04-C ranged from 200 to 530 mL-1 and in 2008 the same ranged from 0 to 20 mL-1. In A03-G and A06-C very few (3 and 0 mL-1) CHAB were detected in 2008 compared to 20 and 17 mL-1 the year before.

MPN AA

The numbers of AA in the MINICAN experiments determined by MPN in 2007 ranged from 8 to 50000 mL-1. The highest amount of AA was found in MINICAN experiment A02-C, A03-C and A04-C (7000 – 50000 mL-1). In A03-G and in A06-C the numbers of AA were low (below detection limit and 8 mL-1, respectively). However, in 2008 the high numbers of AA inside MINICAN experiments A02-C, A03-C and A04-C were gone. The numbers decreased during the year to 1.7 – 90 mL-1. The numbers in A03-G and in A06-C were 0.4 mL-1 and 2700 mL-1, respectively. The acetate concentrations in 2007 ranged from 6.9 – 11 mg L-1. In 2008, the concentrations ranged from 1.6 – 2.9, reflecting the lower numbers of AA this year.

Q-PCR AA

With slightly different protocols, Q-PCR was performed on the MINICAN experiment waters in both 2007 and 2008. In 2007, relative yet quantitative values of the AA activity were calculated (the water with lowest activity was set to relative activity 1). In 2008, the achieved Q-PCR results were compared to standard curves and absolute values of active AA were calculated. Thus, trends over the two years may be made instead of comparison of absolute numbers. The activity of AA in 2007 was found highest inside MINICAN experiments A02-C, A03-C and A04-C (relative activity 10 – 631). The water in A03-G and in A06-C showed much lower activity in 2007 (relative activity 1 and 3, respectively). In 2008, the situation was drastically different. The levels of AA in all the examined waters had by then decreased until below detection limit, showing the same trend as the MPN method (Section 0).

MPN SRB

The numbers of SRB in the MINICAN experiments determined by MPN in 2007 ranged from 230 to 5000 mL-1. In difference to the CHAB and AA numbers which decreased during the following year, the SRB numbers inside the MINICAN experiments stayed in the same order of magnitude in 2008 (ranging from 70 to 5000 mL-1). The highest amount of SRB was found in MINICAN experiment A02-C (5000 mL-1) both in 2007 and 2008. The amount of SRB in A03-G were low both in 2007 and 2008 (2 and 3 mL-1, respectively).

Q-PCR SRB

As in the case with AA, SRB Q-PCR was performed on the MINICAN experiment waters in both 2007 and 2008. In 2007, relative yet quantitative values of the SRB activity were calculated (the water with lowest activity was set to relative activity 1). In 2008, the achieved Q-PCR results were compared to standard curves and absolute values of all SRB as well as active SRB were calculated. The results from the two years show similarities. The activity was highest inside MINICAN experiments A02-C, A03-C and A04-C. The MINICAN experiment A02-C showed the highest SRB activity both in 2007 and 2008 (Relative activity 10 and number of active SRB 5500 mL-1, respectively). The water in A03-G and in A06-C showed the lowest activity both in 2007 and 2008 (Relative activity 3 and 1 and number of active SRB below detection limit, respectively).

The Q-PCR and the MPN results concerning SRB correlated very well (Figure A-1), showing the robustness of the Q-PCR methods.



In 2008, Q-PCR was also used to determine the total numbers of SRB. Generally, these numbers were higher than the numbers of active SRB. However, the same trend as the activity of SRB was also seen in this analysis. The numbers of SRB were highest inside MINICAN experiments A02-C, A03-C and A04-C (37000 – 540000 mL-1). Once again the highest numbers were found in MINICAN experiment A02-C. The SRB numbers were lower in MINICAN experiment A06-C and A03-G (11000 to below detection limit, respectively).

Figure A-1 Correlation between the results of the Q-PCR and the Most probable number (MPN) of sulphate-reducing bacteria (SRB) in A) 2007 and B) 2008. The correlations were significant in both cases (A) r2= 0.89, n=6, p= 0.0051, B) r2= 0.98, n=6, p= 0.0028




Table A-6. Microbial composition in the water from the MINICAN experiments sampled in 2007-08-21. (NB. these data are also shown in Table 5 in the main body of the report).

Experiment name

ATP (amol mL-1) a

TNC (mL-1) a


CHAB (mL-1) a

MPN SRB (mL-1) c

Q-PCR active SRB (relative activity)

Q-PCR total SRB (mL-1)

MPN AA (mL-1) c

Q-PCR AA (relative activity)

Acetate (mg L-1)

A02-C 2400 ± 450

130000 ± 16000

na b 227 ± 42

5000 (2000 – 17000)

10 na b 30000 (2000 – 17000)

631 8.5

A03-C 2800 ± 240

95000 ± 39000

na b 200 ± 57

1400 (600 – 3600)

2 na b 50000 (20000 – 200000)

10 8

A03-G 3300 ± 710

31000 ± 11000

na b 20 ± 35

3 (1 – 12) 3 na b 3

(1 – 12) 3 6.9

A04-C 15500 ± 810

150000 ± 61000

na b 530 ± 28

230 (90 – 860)

2 na b 7000 (3000 – 21000)

336 6.9

A06-C 3000 ± 680

13000 ± 4900 na b 17 ±

12

300 (100 – 1100)

1 na b bd 1 11

a ± standard deviation b not analysed c 95% confidence interval)

Table A-7. Microbial composition in the water from the MINICAN experiments sampled in 2008-10-15 (NB. these data are also shown in Table 5 in the main body of the report).

Experiment name

ATP (amol mL-1) a

TNC (mL-1) a


CHAB (mL-1) a

MPN SRB (mL-1) c

Q-PCR active SRB (mL-1)

Q-PCR total SRB (mL-1)

MPN AA (mL-1)

c

Q-PCR active AA (mL-1)

Acetate (mg L-1)

A02-C 17400 ± 2500

180000 ± 4600 6700000 bd b

5000 (2000 – 17000)

5500 5400001.7 (0.7 – 4)

bd b 1.7

A03-C 10400 ± 690

180000 ± 18000

4500000 3 ± 6 800 (300 – 2500)

31 37000 17 (7 – 48)

bd b 1.6

A03-G 4400 ± 570

2500 ± 880 8000000 3 ± 6

2.3 (0.9 – 8.6)

bd b bd b 0.4 (0.1 – 1.7)

bd b 2.9

A04-C 27400 ± 1820

200000 ± 22000

5000000 20 ± 17

70 (30 – 210)

350 74000 90 (30 – 290)

bd b 1.8

A06-C 18600 ± 1270

71000 ± 11000

14000000 bd b 110 (40 – 300)

bd b 11000

2700 (1200 – 6700)

bd b 2.3

a ± standard deviation b below detection limit c 95% confidence interval)


Table A-8. Microbial diversity determined by cloning and sequencing in the water from the MINICAN experiments sampled in 2007-08-21

Experiment name

Number of identical clones

Microbial group Accession number Closest relative Identity (%) Length (bp)

A02-C 2 Gammaproteobacteria EU714792 Stenotrophomonas maltophilia AJ293470 99 966

A02-C 3 unknown EU714780 Nitrospirae bacterium clone DQ450806 96 940

A02-C 2 unknown EU714781 Uncultured bacterium clone Thp_B_116 EF444760 95 911

A02-C 1 unknown EU714782 Uncultured bacterium clone TBM.31JAN-47 EF608433 94 993

A03-C 1 Clostridia EU714793 Acetobacterium wieringae X96955 99 946

A03-C 4 Gammaproteobacteria EU714794 Stenotrophomonas maltophilia AJ293474 99 967

A03-C 3 Clostridia EU714795 Veionella parvula AY995769 99 955

A03-C 2 Gammaproteobacteria EU714796 Haemophilus parainfluenzae EU083530 100 949

A03-C 2 Lactobacillales EU714797 Streptococcus mitits AY518677 99 924

A03-C 1 Clostridia EU714798 Acetobacterium malicum X96957 99 923

A03-C 1 Unknown EU714783 Uncultured bacterium AY255076 91 1017

A03-G 13 Gammaproteobacteria EU714788 Stenotrophomonas maltophilia AJ293474.1 100 920




A03-G 3 Gammaproteobacteria EU714789 Pseudomonas gingeri AF320991.1 98 885

A03-G 1 Actinobacteraceae EU714790 Proprioni bacterium acnes AY642051 100 963

A03-G 2 unknown EU714778 Koordiimonas gwangyangensis AY682384 94 886

A03-G 1 Deltaproteobacteria EU714779 Desulfocapsa sulfoexigens AY771932 94 954

A03-G 14 Gammaproteobacteria EU714791 Pseudomonas anguilliseptica AF439803 99 713

A04-C 5 Unknown EU714784 Uncultured bacterium clone TBM 31JAN-47 94 922

A04-C 4 Unknown EU714799 Uncultured bacterium clone Lupin-1130 m-2-pse1

EF200114 99 931

A04-C 3 unknown EU714785 Uncultured bacterium clone ThpB_116 EF 444760 95 922

A04-C 1 Clostridia EU714800 Acetobacterium carbinolicum X96956 99 871

A04-C 1 unknown EU714786 Uncultured bacterium WLEA-02, EF117512 95 929

A04-C 1 unknown EU714787 Uncultured bacterium clone PC06110 EU101173 90 900


3.3 Water chemistry

The water chemistry in the MINICAN experiments in 2007 and 2008 are shown in Table A-9 and Table A-10. The sulphate concentrations inside MINICAN experiments A02-C, A03-C, A04-C and A06-C decreased 7 – 20% during the year, while the concentration increased in A03-G (with 36%). pH was generally lower in 2008 as were the alkalinity in all examined waters compared to 2007 (18 – 43 % lower), consistent with the lower carbon dioxide levels this year (Table A-4 and Table A-5). Sulphide was low both years (0.02 – 0.062 mg L-1). However, ferrous iron levels inside the MINICAN experiments increased from 2007 to 2008 so any sulphide produced would thus precipitate as iron sulphide.

Table A-9. Chemical composition in the water from the MINICAN experiments sampled in 2007-05-22.

Experiment name

SKB sample number

SO42-

(mg L-1) HCO3

- (mg L-1)

S2- (mg L-1)

Fe2+ (mg L-1) pH

A02-C 14287 486 28 0.036 11.1 7.4 A03-C 14288 506 27 0.057 2.33 7.6 A03-G 14283 354 28 0.062 0.16 7.6 A04-C 14289 439 51 0.037 0.82 7.6 A06-C 14290 605 14 0.051 6.30 7.6

Table A-10. Chemical composition in the water from the MINICAN experiments sampled in 2008-10-15.

Experiment name

SKB sample number

SO42-

(mg L-1) HCO3

- (mg L-1)

S2- (mg L-1)

Fe2+ (mg L-1) pH

A02-C 14642 417 19 0.021 19.2 7.3

A03-C 14643 413 20 0.020 4.15 7.4

A03-G 14645 481 23 0.023 0.15 7.6

A04-C 14644 410 38 0.022 15.7 7.2

A06-C 14641 567 8 0.030 11.1 6.8



4. Discussion 4.1 Development of successful Q-PCR

The activity and the presence of microbes are generally correlated. In the MINICAN experiments this was observed (Figure A-1). The correlation between sulphate-reducing activity results obtained by the Q-PCR and the numbers of SRB obtained by MPN methods (Table A-6,) in the MINICAN experiments was very good (Figure A-1). This was also a test of the robustness of the Q-PCR methodology; making it a useful tool in future activity studies of SRB in different projects concerning the integrity of the nuclear waste.

4.2 Degradation of carbon present in the bentonite possibly support growth of microbes

In 2007, CHAB were present in higher numbers inside the MINICAN experiments A02-C, A03-C and A04-C that contained bentonite than in A03-G and MINICAN experiment A06-C which lacked bentonite (20-50 times, Table A-6). This shows a possible growth-stimulating effect of the bentonite. The MX-80 bentonite (used in the MINICAN experiments) contains about 0.20-0.25 % organic carbon (Svensson, 2008), which potentially can serve as a carbon and energy source for subsurface bacteria able to utilise organic carbon. When investigating the number of CHAB in the MINICAN experiment water, it was shown that they were positively correlated with the carbon dioxide concentration (Figure A-2). This indicates that they produced carbon dioxide during the degradation of organic carbon. In 2008, the number of CHAB had decreased, concomitant with the levels of carbon dioxide and the alkalinity (Table A-5 , and Table A-10). This indicates that, if the CHAB used organic carbon from in the bentonite, the carbon source was depleted in 2008.

The carbon dioxide levels also seem to correlate with the numbers and activity of AA. In 2007, when the carbon dioxide levels were higher (Table A-4) the numbers and activity of AA were high (up to 50000 mL-1, Table A-6, Section 0). In accordance with the discussion above, the water close to the canister inside the MINICAN experiments A02-C, A03-C and A04-C contained more AA than the groundwater in A03-G and A06-C without the bentonite cage (875-16700 times more, Table A-6). The presence and activity of AA may also partly be controlling the hydrogen gas levels. The energy source for AA is hydrogen gas, so a high abundance of AA should decrease the hydrogen content. Concomitantly, hydrogen could scarcely be detected inside MINICAN experiments A02-C, A03-C and A04-C in 2007 when the numbers and activity of AA were high (Table A-4 and Table A-6) but it could be detected in a much higher concentration (215 µL L-1) close to the canister in A06-C where no AA could be detected (Table A-4 and Table A-6). This suggests the following; hydrogen gas was produced by corrosion of the cast iron, fore sure in A06-C but probably also in experiment A02-C, A03-C and A04-C. The difference between the two systems is that in experiment A02-C, A03-C and A04-C the hydrogen gas was consumed by the AA, which were not present in A06-C.



Figure A-2 Correlation between the number of CHAB (Culturable Heterotrophic Aerobic Bacteria) and the carbon dioxide concentration in the water inside and outside the MINICAN experiments.

The correlation was significant (r2= 0.86, n=12, p= 0.00003).

When the carbon dioxide levels decreased in 2008, the number of AA drastically decreased (Table A-5 and A-7). Concomitant with the decrease of AA, the levels of hydrogen gas begun to increase in MINICAN experiment A02-C (90 µL L-1, Table A-5). There is another possible explanation to the decreasing AA levels inside MINICAN. As seen in Table A-4 and Table A-5, there are indications that the methane levels increase inside the MINICAN experiments. Methanogens, organisms producing methane, are also believed to be able to derive electrons from metallic iron (Dinh et al., 2004) and have previously been shown to out-compete AA in deep groundwater since they also use hydrogen and carbon dioxide in their metabolism (Kotelnikova and Pedersen, 1998).

4.3 Sulphate-reducing bacteria thrives in the MINICAN experiments

Even if the numbers of AA and CHAB decreased in 2008, the TNC levels remained unchanged inside the MINICAN experiments A02-C, A03-C, A04-C and A06-C (Table A-6). In addition the ATP levels increased up to almost 10-fold inside the MINICAN experiments A02-C, A03-C, A04-C and A06-C, but in A03-G. In MINICAN experiment A02-C, the amount of gas increased (Table A-5). This activity can at least partly be explained by SRB activity. Table A-6 and A-7 show that SRB continues to thrive in the MINICAN experiments. In both 2007 and 2008, the MPN numbers of SRB found in the MINICAN experiments A02-C, A03-C, A04-C and A06-C ranged from 70-5000 (Table A-4) and, as shown with SRB Q-PCR, they were active (Table A-6 and A-7). The MPN numbers of SRB peeked inside the MINICAN experiment A02-C (5000 SRB mL-1) but were 30-400 times more abundant also in MINICAN experiments A03-C, A04-C and A06-C, compared to A03-G.

As all examined waters from the MINICAN experiments except the groundwater in A03-G were surrounding an iron cage, obviously the SRB thrived in the vicinity of exposed iron. As mentioned above (Equation 1), SRB can thrive in close connection to iron for two reasons:

• Cathodic hydrogen produced when anaerobic water comes in contact with iron is believed to be directly scavenged by SRB if they posses the enzyme hydrogenase (Cord-Ruwisch and Widdel, 1986).



• SRB can possibly use metallic iron directly as their electron donor (Dinh et al., 2004).

• SRB can also use the acetate produced by AA as a carbon and energy source. The end product is then sulphide which can corrode the iron and form FeS (Equation 2).

In short, all activity of SRB in the vicinity of iron leads to iron deterioration, regardless of the specific SRB metabolism. The question that remains is, however, if SRB will continue to thrive in the MINICAN experiments. To answer this, we must know the specific metabolic pathways utilised by the SRB because if they use acetate and other organic molecules (e.g. from the bentonite) as energy the decreasing AA levels and decreasing CHAB numbers indicate that this compound is to be depleted. On the other hand, if they use cathodic hydrogen or metallic iron these electron donors will not be depleted as long as the cast iron remains. As shown in Table A-6 and A-7, ferrous iron was in abundance in the MINICAN experiments. This suggests that the corrosion processes in MINICAN, at least partly to direct corrosion via SRB hydrogenases or use of metallic iron as an electron donor (Equation 1). More extensive comparison of chemical species such as ferrous iron, alkalinity, pH and SRB activity would be invaluable in the pursuit of answering this question.

5. Conclusions The following conclusions can be drawn from the first two years of microbial examinations in the MINICAN experiments.

• There seemed to be a growth-stimulating effect of bentonite in the MINICAN experiments, possibly bentonite served as a good carbon and energy source for subsurface bacteria able to utilise organic carbon. This phenomena has been observed before in the Prototype Repository (Eriksson, 2008). In 2007, during this putative bentonite degradation, carbon dioxide levels were high. The high carbon dioxide levels correlated with high abundance of AA.

• In 2008, the levels of both carbon dioxide and the alkalinity were lower. Even if precipitation of carbonates could be partly responsible for the lower levels, it is likely explained by decreased degradation of organic carbon. The carbon dioxide levels seemed to control the numbers and activity of AA. In 2008, the numbers of AA drastically decreased when the carbon dioxide levels dropped.

• Even so, the TNC levels remained unchanged in 2008 and the ATP levels increased up to almost 10-fold inside the MINICAN experiments where the water was in contact with the cast iron. The increased activity could at least partly be assigned to SRB based on MPN and Q-PCR results. In the MINICAN experiments, SRB could basically derive electrons to their metabolism in three ways:

1. From cathodic hydrogen produced when anaerobic water comes in contact with iron, if they posses the enzyme hydrogenase.

2. From metallic iron.

3. From organic carbon, probably acetate produced by AA.

• If SRB will continue to thrive in the MINICAN experiments, the electron donor will have to be cathodic hydrogen or metallic iron since these energy sources will not be depleted as long as the cast iron remains. Dropping AA levels indicates that acetate will not be produced on a long term scale. Ferrous iron was in abundance in the MINICAN experiments. This links the corrosion processes in MINICAN, at least partly,



to direct corrosion via SRB hydrogenases or use of metallic iron as an electron donor (Equation 1). More extensive comparison of chemical species such as ferrous iron and sulphate to alkalinity and pH and SRB activity is needed to find the electron donor for the SRB.

• The Q-PCR methodology is an extremely useful tool in future activity studies of both AA and SRB studies because of it instead of several months take days to analyse specific metabolic activities in groundwater. Q-PCR also detects unculturable microbes.

6. Future research • The MINICAN experiments have given us a unique opportunity to examine what

microbial processes that will appear in situ if the copper canisters in a KBS-3 repository fail to hold. However, the assumptions and theories in the report are based on results from two sampling occasions. It is necessary to continue the monitoring programme to be able to further confirm the findings.

• As seen in Table A-4 and Table A-5, there are indications that the methane levels increase inside the MINICAN experiments. Methanogens are also believed to be able to derive electrons from metallic iron (Dinh et al., 2004) and have previously been shown to out-compete AA in deep groundwater since they also use hydrogen and carbon dioxide in their metabolism (Kotelnikova and Pedersen, 1998) Therefore, the presence of methanogens should also be evaluated in future investigations.

• The electron source for the SRB could be revealed by a combination of evaluation of chemical data such as ferrous iron, pH and alkalinity (see Equations 1 and 2) combined with SRB activity studies using Q-PCR. In addition to the Q-PCR methods already applicable, the MINICAN experiments can be a good opportunity to develop e.g. a Q-PCR which detects SRB-hydrogenase activity (Caffrey et al., 2007). In this manner, it is possible to link the consumption of corrosion-produced cathodic hydrogen to SRB even further.

7. References Caffrey, S.M., H.-S. Park, J.K. Vorrdouw, Z. He, J. Zhou and G. Voordouw. 2007. Function of

Periplasmic Hydrogenases in the Sulfate-Reducing Bacterium Desulfovibrio vulgaris Hildenborough. Journal of Bacteriology 189, 6159-6167.

Cord-Ruwisch, R. and F. Widdel. 1986. Corroding iron as a hydrogen source for sulphate reduction in growing cultures of sulphate-reducing bacteria. Applied Microbiology and Biotechnology 25, 169-174.

Dinh, H.T., J. Kuever, M. Mussmana, A.W. Hassel, M. Stratmann and F. Widdel. 2004. Iron corrosion by novel anaerobic microorganisms. Nature 427, 829-832.

Eriksson, S.. 2008. Prototype repository - Analysis of microorganisms, gases, and water chemistry in buffer and backfill, 2004-2007. SKB IPR 08-01

Hallbeck, L. and K. Pedersen. 2008. Characterization of microbial processes in deep aquifers of the Fennoscandian Shield. Applied Geochemistry 23, 1796-1819.

King, R.A. and D.A. Miller. 1971. Corrosion by the Sulphate-reducing Bacteria. Nature 233, 491-492.




Kotelnikova, S. and K. Pedersen. 1998. Distribution and activity of methanogens and homoacetogens in deep granitic aquifers at Äspö Hard Rock Laboratory, Sweden. FEMS Microbiology Ecology 26, 121-134.

Pedersen, K.. 1999. Subterranean microorganisms and radioactive waste disposal in Sweden. Engineering Geology 52, 163-176.

Pedersen, K., J. Arlinger, S. Eriksson, M. Hallbeck, J. Johansson, S. Jägevall and L. Karlsson. 2008a. Microbiology of Olkiluoto Groundwater - Results and Interpretions 2007. Posiva Working Report 2008-34.

Pedersen, K., J. Arlinger, A. Hallbeck, L. Hallbeck, S. Eriksson and J. Johansson. 2008b. Numbers, biomass and cultivable diversity of microbial populations relate to depth and borehole-specific conditions in groundwater from depths of 4 to 450 m in Olkiluoto, Finland. The ISME Journal 2, 760-775.

Svensson, D.. 2008. SKB. Personal communication

1 (5)

Review statement / Response statement

(filled out by author/administrator)

Review statement date

2010-09-25

(reviewer)

Response statement date

2010-

(author/administrator)

Granted

(reviewer)

Approved

(author/admin.)

Review document:

Miniature Canister (MiniCan) Corrosion Experiment Progress report 2 for 2008-2009

Version / other identification of document

Version February 2010

SKBDoc id 1233269, ver 0.1

Review instruction: SKBDoc 1253952 Review plan Version 1.0

Reviewer: Fraser King

Competence of reviewer

(filled out by reviewer)

26 years experience in corrosion of copper as a canister material for the disposal of nuclear waste, involving studies for various

agencies, including: AECL, OPG, NWMO (Canada); SKB (Sweden); Posiva (Finland); Nuclear Decommissioning Authority

(UK); and Nagra (Switzerland).

Review decision (filled out by reviewer)

Summary of important review

comments

Progress Report 2 is an updated version of Progress Report 1. Large portions of the text are identical and virtually all of the

comments on Report 1 apply equally to Report 2. These comments are not repeated here, but they should also be taken into

account for Report 2. Some of the issues found for Report 1 have, coincidentally, also been found by the authors and have been

corrected. Since their number is smaller than the number of items that have not been corrected, those items that have been

corrected are listed here in italic font and a comment has been added to the Corrections/Measure column. Any item from the

review of Report 1 that does not appear here in italics has not been corrected in Report 2 and should be corrected in exactly the

same manner as used for Report 1.

Appendix I is identical to that in Report 1 and, in particular, contains the statement under Future Research "The MINICAN

experiments have given us a unique opportunity to examine what microbial processes that will appear in situ if the copper

canisters in a KBS-3 repository fail to hold" which should be deleted.

2 (5)

Standpoint Granted – Meets the criteria of acceptance, continued handling of errand granted

Granted with reservation – Meets the criteria of acceptance if corrections are made according to comments below, reply

message resent to reviewer

Not granted – Does not meet the criteria of acceptance, additional review needed

Corrections according to comments (filled out by author/administrator)

Motive for standpoint Provided all items from Reports 1 and 2 are appropriately discharged, Report 2 is acceptable for publication.

3 (5)

Review comments/questions (filled out by reviewer) and Response statement (filled out by author/administrator)

No. Section Concerning Comment/Question Corrections/Measure

[ID ] [Page,

headline,

section ]

[Content, e.g. text, figure,

table, etc. that the comment

concerns.

For example, text that is to

be commented may be

quoted.]

[Specification of comment and/or question, including motivation.

If needed, provide advices, instructions and suggestion for improvements]

[Brief description of how the review comment

will be handled ]

1 All

pages,

footer

Report number in footer The Serco report number in the footer refers to Report 1 and should

be updated.

2 4.2.4 The text still states that contact was lost with the cast iron coupons

in Expt 4 but the updated results are given again in Fig 28.

3 4.7,

Figs 82-

101

Noise data It is difficult to derive any information from the noise data

presented, but if the figures are simply provided as representative

data for each of the coupons in the five tests then this is acceptable

for a progress report.

The noise data have not been reviewed in detail.

4 (5)


4 Section

6

Comments on proposed

decommissioning plan

Are we prepared if the inside of the vessel turns out to be a black,

smelly, slimy mess? How are you going to clean the samples for

the detailed surface analysis that is proposed? Are we prepared to

perform detailed characterisation of any biofilm, including the

microbiological and physical characteristics (e.g., spatial

distribution, effective mass transport properties, spatial separation

of anodic and cathodic processes, etc.)? Is it worth doing detailed

analysis if the conditions within the vessels are not relevant to

canisters in highly compacted bentonite? Are we certain that these

conditions are relevant for the case of partially eroded bentonite?

It is proposed that mass-loss measurements on the electrodes will

not be performed because of the uncertain mass of the wire

connection. However, if the corrosion is as extensive as is

reported, then even an approximate mass-loss measurement would

be sufficient to provide a ball-park corrosion rate. It may be worth

determining an approximate mass-loss of the electrodes to compare

with the high corrosion rates currently being measured. That way,

at least we would have an approximate mass-loss rate on the same

specimens as we have electrochemical corrosion rates for, which

would deflect the criticism that we are comparing mass-loss and

electrochemical corrosion rates determined on different specimens.

For the WOL specimens, is the crack extension measured on the

edge of the crack or on the crack surface, following fracturing of

the sample? If the latter, what is the procedure for fracturing the

sample? Timo Saario uses cyclic loading to grow a fatigue crack

because OFP is too ductile to fracture at liquid nitrogen

temperatures.

5 Figs 16,

18

I note that some of the corrosion rates determined by LPR and EIS

are diverging in the most-recent measurements.

5 (5)


6 4.6 Other issues Figs 33-62 Figures 44 and 45 are duplicates.

Y-axis not labelled in Figs 33, 34, 36

Corrected

7 Fig 55 Graph title in the left-hand graph in the centre row. Corrected

8 Fig 58 Panes out of chronological sequence (when read L to R, top to

bottom).

9 Fig 62 Upper LH pane is a repeat of the lower RH pane in Fig 61


bottom).


bottom).

12 Missing data No EIS data are given for cast iron from Expt 4 but corrosion rates

are given in Figure 28.

13 Fig. 74 Panes out of chronological sequence (when read L to R, top to

bottom).


bottom) and caption for lower RH pane is incorrect (as it was also

in Report 1, but I missed it!).


bottom).

Documents

Miniature Canister (MiniCan) Corrosion Experiment · Miniature Canister (MiniCan) Corrosion Experiment Progress Report 2 for 2008-2009 . ... SKB agree in general to this conclusion