Presence of sulphate reducing bacteria near a Boom ...

EXTERNAL REPORT SCK•CEN-ER-79 09/SAe/P-12

Presence of sulphate reducing bacteria near a Boom Clay-steel interface

Sven Aerts, Pierre De Cannière, Hugo Moors

Report prepared by SCK•CEN in the framework of ONDRAF/NIRAS programme on geological disposal, under contract CCHO-2004-2470/00/00, DS 251-B62 February, 2009

SCK•CEN Boeretang 200 BE-2400 Mol Belgium

RDD

EXTERNAL REPORT OF THE BELGIAN NUCLEAR RESEARCH CENTRE SCK•CEN-ER-79 09/SAe/P-12

Presence of sulphate reducing bacteria near a Boom Clay-steel interface

Sven Aerts, Pierre De Cannière, Hugo Moors

Report prepared by SCK•CEN in the framework of ONDRAF/NIRAS programme on geological disposal, under contract CCHO-2004-2470/00/00, DS 251-B62 February, 2009 Status: Unclassified ISSN 1782-2335

SCK•CEN Boeretang 200 BE-2400 Mol Belgium

RDD

© SCK•CEN Studiecentrum voor Kernenergie Centre d’étude de l’énergie Nucléaire Boeretang 200 BE-2400 Mol Belgium Phone +32 14 33 21 11 Fax +32 14 31 50 21 http://www.sckcen.be Contact: Knowledge Centre [email protected]

RESTRICTED All property rights and copyright are reserved. Any communication or reproduction of this document, and any communication or use of its content without explicit authorization is prohibited. Any infringement to this rule is illegal and entitles to claim damages from the infringer, without prejudice to any other right in case of granting a patent or registration in the field of intellectual property. SCK•CEN, Studiecentrum voor Kernenergie/Centre d'Etude de l'Energie Nucléaire Stichting van Openbaar Nut – Fondation d'Utilité Publique - Foundation of Public Utility Registered Office: Avenue Herrmann Debroux 40 – BE-1160 BRUSSEL Operational Office: Boeretang 200 – BE-2400 MOL

Table of contents List of abbreviations ...................................................................................................................6 Abstract.......................................................................................................................................7 1 Background and objective of the study ...................................................................................9 2 Materials and Methods ............................................................................................................9 2.1 Sampling of clay...............................................................................................................9 2.1.1 Direct sampling with swab ........................................................................................9 2.1.2 Direct sampling with knife ......................................................................................10 2.1.3 Clay core sampling using cutting edges ..................................................................10 2.2 Sample selection for bacterial and chemical analyses....................................................11 2.3 Chemical analyses of clay samples ................................................................................11 2.4 Bacterial analyses of clay samples .................................................................................12 3 Results ...................................................................................................................................12 3.1 Chemical analyses ..........................................................................................................12 3.2 Bacterial analyses ...........................................................................................................13 3.2.1 Direct clay sampling................................................................................................13 3.2.2 Clay core analyses ...................................................................................................14 4 Discussion..............................................................................................................................15 5 Main outcomes of the experiments........................................................................................15 6 Prediction of the impact of SRB activity on the supercontainer ...........................................15 7 Conclusions and recommendations .......................................................................................18 References ................................................................................................................................19 Appendix 1: Cultivation media ................................................................................................20 .1 Preparation of synthetic media .........................................................................................20 .2 Composition of the media ................................................................................................20 .3 Reference composition of RBCW (EG/BS piezometer) ..................................................21 Appendix 2: Sulphide measurement.........................................................................................22 Appendix 3: Detection of SRB: apsA.......................................................................................23

5

List of abbreviations AFt apsA BFS EBS EDZ EGBS OPC PCR RBCW SRB

calcium tri-sulphoaluminate hydrate α subunit of adenosine-5'-phosphosulfate reductase Blast furnace slag Engineered barrier system Excavation disturbed zone Extension Gallery Bottom Shaft piezometer. Piezometer located just beneath the first access shaft of the HADES URL. Ordinary Portland Cement Polymerase chain reaction Real Boom Clay water Sulphate reducing bacteria

6

Abstract Within the frame of high-level waste disposal Boom Clay is currently studied as the reference host rock in Belgium. Sulphate reducing bacteria (SRB), which are present in Boom Clay, can transform sulphate into sulphide. Sulphide can cause corrosion of for instance the carbon steel overpack surrounding the waste canister which could have a severe influence on the efficiency of the repository. Steel tubes which have been in contact with Boom Clay for 20 years were examined for traces of corrosion [Kursten, 2009]. Simultaneously, in this study the clay-steel interface was investigated for the presence of different sulphur species and SRB. Culture experiments showed that the SRB concentrations in Boom Clay were highest at the clay-steel interface. The analysis of sulphur species only revealed the presence of sulphate in the clay cores with varying concentrations depending on the distance from the clay-steel interface. Although the steel tubes showed signs of corrosion only very limited amounts of sulphur were detected [Kursten, 2009] Finally, the possible influence of SRB activity through the influence of sulphide on the supercontainer design is summarized based on discussions conducted during the workshop on Sulfur-Assisted Corrosion in Nuclear Waste Disposal Systems (SACNUC, October 2008). Indeed, the supercontainer's integrity is particularly important during the thermal phase of the repository to prevent a premature release of radionuclides while the temperature of the formation is still elevated.

7

8

1 Background and objective of the study In Belgium, geological disposal is considered the primary option for the final disposal of high-level waste. In this frame Boom Clay is studied as the reference host rock. Different lab and in-situ experiments have revealed the presence of viable bacteria in Boom Clay [Aerts, 2008; Aerts, 2009]. The possible effect of these bacteria on the supercontainer design is investigated with a special focus on the possible influence of SRB. Previously, the presence of SRB in the Boom Clay and the correlation with the oxidation of Boom Clay was investigated [Aerts et al., 2009]. Subsequently, the influence of conditions imposed by the supercontainer design near the overpack on the activity/presence of SRB was studied [Aerts, 2009]. Finally, the influence of SRB activity on the corrosion of steel was examined. As a first step, the corrosion of steel which was exposed to Boom Clay for over 20 years was investigated. The steel itself was analysed to evaluate the impact of 20 years of exposure to Boom Clay [Kursten, 2009]. Simultaneously, the clay near the steel surface was also analysed with respect to the presence of different sulphur species and SRB.

2 Materials and Methods 2.1 Sampling of clay After drilling through the steel and the removal of the steel sample the exposed Boom Clay could be sampled. Boom Clay near the clay-steel interface was sampled using two different techniques: a sterile swab or a sterile knife. Clay was sampled at three different positions relative to the gallery lining: 0.53, 0.83 and 1.90 m. Besides clay at the clay-steel interface, clay cores (± 20 cm) were collected at 0.53 and 1.90 m using cutting edges (Figure 1).

Figure 1: Schematic drawing of the different sampling positions.

The samplings were conducted on June 16 and 17 2008. The selected tube was situated at ring 5, oriented westward. It should be stressed that this part of the gallery was excavated manually while the Boom Clay was frozen (1982-1983) and that the cast iron lining was completely air and water tight. 2.1.1 Direct sampling with swab After removal of the steel disc, the Boom Clay at the interface was exposed to air and easily accessible for sampling. Clay on the interface was sampled using sterile cotton swabs. The swabs were moistened by immersing them shortly in a sterile isotonic solution (0.9 % NaCl) before carefully rubbing them over the clay surface.

9

2.1.2 Direct sampling with knife Alternatively, the clay samples were taken using a knife and a spatula which were flame sterilised on site and the clay samples were collected in different pre-sterilised hungate tubes (Figure 2). The weight of the different tubes was determined prior to the sampling to allow easy determination of the amount of Boom Clay sampled.

a.

b.

c.

Figure 2: (a.) View of the Boom Clay immediately after the removal of the steel sample, (b.) sampling of the clay at the interface using a sterile spatula and a hungate tube to collect the sample and (c.) flushing of the collected clay sample using filter sterilized nitrogen.

This method was used to sample clay at all three positions. After sampling, the hungate tubes were flushed using N2 which was filter sterilised (Figure 2). No more than 4 hours after sampling the hungate tubes were placed inside an anaerobic glovebox (Ar-H2). 2.1.3 Clay core sampling using cutting edges Besides clay sampling at the clay-steel interface, clay was also sampled up to a depth of ~ 20 cm at two different positions using cutting edges. The cutting edges were sterilised prior to the sampling. They were packaged in autoclavable polypropylene bags, which were subsequently closed. The cutting edges in the bags were autoclaved (20 min at 121°C) and afterwards the bags were put in an oven at 90°C overnight to dry and to allow water that penetrated the bag (the bags are not gastight to water vapor) to re-evaporate again. The sterile bag containing a cutting edge was opened just prior to the sampling. The cutting edges were pushed in and retrieved from the clay using a hydraulic press (Figure 3). Immediately after sampling the cutting edges were packaged in a sterile polypropylene bag and a aluminum-coated polyethylene bag. The bags were flushed with filter sterilized nitrogen to remove oxygen from the pack (Figure 3). No more than 4 hours after the sampling the bags were placed inside an anaerobic glovebox with an Ar-H2 (95-5) atmosphere. Clay was sampled at two positions (0.53 and 1.90 m distance relative to the gallery lining) using the cutting edges.

10

a.

b.

c.

Figure 3: (a.) Cutting edge being pushed into the clay and (b.) extracted from the clay. (c.) Flushing of aluminium coated bag containing cutting edge using filter sterilized nitrogen.

2.2 Sample selection for bacterial and chemical analyses The two cores were removed from the cutting edges outside the anaerobic box using a hydraulic press. The aluminum-coated bag containing the clay core in a nitrogen atmosphere was opened using a sterile knife and the clay core was collected in another sterile plastic bag after removal from the cutting edge. Immediately after the collection, the clay core was transferred to an anaerobic box (Ar-H2 atmosphere (95-5)) where different experiments were started. In order to exclude a possible influence of the oxidation of the Boom Clay which could occur during the removal of the clay from the cutting edges only the inner part of the clay was used in the further analyses. Two types of analyses were started: chemical and bacterial. The chemical analyses was performed on 4 clay segments: 0, 2, 10 and 18 cm from the clay-steel interface. The selected samples had a thickness of ± 1 cm. For the bacterial analyses three clay segments were selected: 2, 10 and 18 cm from the clay-steel interface (bacterial analysis of the clay at the clay-steel interface was performed using the samples obtained from the direct sampling). Precautions were taken to perform all manipulations of the Boom Clay samples under aseptic conditions. Sterile tweezers, knifes and recipients were used and the work surface was sterilised. 2.3 Chemical analyses of clay samples The sulphur species (SO42-, S2O32-, HS-) present in the clay cores were analyzed as a function of distance to the steel surface. In an anaerobic box (Ar-H2 atmosphere (95-5)) about 2.3 g of clay was placed in a centrifuge tube (PE Nalgene tube, 50 ml). The clay was not dried to prevent loss of sulphide. 20 ml of de-oxygenated milli-Q water1 was added to the 'wet' clay and the slurries were shaken in a anaerobic box for 4 weeks. Subsequently, a part of the clay slurry was transferred to a hungate tube which was analyzed immediately for the presence of sulphide using a Spectroquant Sulphide Test (1.14779.0001, Merck) For a more detailed description of the sulphide measurement: see Appendix 2. The rest of the slurries was centrifuged for 2 hours at 21000 g and the supernatant was analyzed using ion chromatography (IC25 (Dionex) equipped with an AS4A-SC column) for the presence of SO42- and S2O32-. The presence of SO32- can be detected but the exact concentration can not be determined because of its limited stability (SO32- is very rapidly converted to SO42-). 1

Milli-Q Plus PF system using Millipore QPAK®1 column.

11

2.4 Bacterial analyses of clay samples The presence of bacteria in the different clay samples was investigated. The number of SRB present in the different clay samples was estimated by decimal dilution series using two different media: Postgate and Real Boom Clay Water (RBCW) medium (Appendix 1 for more detailed information on media preparation). The clay samples collected at the clay-steel interface were inoculated 6 days after sampling. From each sampling site 2 samples were incubated using Postgate medium and 2 using RBCW medium. A volume of 4.5 ml of the medium was added to the different clay samples and after homogenization a fivefold decimal dilution series were prepared for each clay sample. Subsequently, the samples were incubated in closed gastight hungate tubes stored outside the anaerobic box at room temperature. The SRB load of the different segments from the two clay cores were investigated using the same approach. The analyses were performed three weeks after the sampling. Meanwhile the clay cores were stored in an anaerobic box (Ar-H2 atmosphere) while packaged in the aluminum-coated bags containing a N2 atmosphere. At the chosen positions (2, 10 and 18 cm from the surface) the outer surface of the clay core was removed and about 0.25 g of clay (the exact amount was weighed) was transferred into a hungate tube containing 4.5 ml of medium using a sterile knife and a spatula. After homogenization of the clay-medium mixture a fivefold decimal dilution series was prepared. The cultures were incubated outside the anaerobic box at room temperature in closed hungate tubes. The incubation experiments contained a blank control (inoculated with sterile milli-Q water) and a positive control (MORPHEUS water sample known to contain SRB [Aerts et al., 2007]).

3 Results The collected Boom Clay samples were analysed chemically and microbiologically. First, the concentrations of different sulphur species present in the clay samples were measured. Secondly, the bacterial load of the samples was determined via culture tests. 3.1 Chemical analyses The results of the chemical analysis of the sulphur species present in the different clay samples are given in Table 1. Sulphite was not detected in any of the samples. Table 1: Chemical analysis of the sulphur species present in the Boom Clay samples. Values give the concentrations in the original 2.3 g of Boom Clay. Clay core position

Distance to steel-clay interface

SO42concentration [mM]

S2O32concentration [mM]

S2concentration [mM]

0.53 m 0.53 m 0.53 m 0.53 m

0.00 m 0.02 m 0.10 m 0.18 m

4.90 0.12 < 0.10 < 0.10

< 0.36 < 0.36 < 0.36 < 0.36

-

1.90 m 1.90 m 1.90 m 1.90 m

0.00 m 0.02 m 0.10 m 0.18 m

1.43 11.40 5.30 4.85

< 0.36 < 0.36 < 0.36 < 0.36

-

-: indicates that no sulphide was detected in the samples.

12

3.2 Bacterial analyses 3.2.1 Direct clay sampling Boom Clay samples were collected at the clay-steel interface at three different locations (Figure 1). The amount of clay sampled at each position is shown in Table 2. Table 2 also shows the range of the size of the SRB population present in the different Boom Clay samples determined through culture experiments. Table 2: Amount of Boom Clay [mg] sampled at the clay-steel interface at the different drilling positions (distance measured from the beginning of the steel pipe). The medium used to cultivate SRB present in the clay samples is given in column 4. Sampling position

Tube number

Mass of Boom Clay [mg]

Cultivation medium

Number of SRB in sample

Number of SRB in 1 g of Boom Clay [CFU/g]

0.53 m 0.53 m 0.53 m 0.53 m

1 3 2 4

52.5 226.9 268.4 263.7

Postgate Postgate RBCW RBCW

10 - 102 103 - 104 104 - 105 > 106

2 102-2 103 4 103 - 4 104 4 104 - 4 105 > 4 106

0.83 m 0.83 m 0.83 m 0.83 m

5 7 6 8

103.3 165.6 260.8 214.8


102 - 103 103 - 104 103 - 104 102 - 103

103 - 104 6 103- 6 104 4 103- 4 104 5 102- 5 103

1.90 m 1.90 m 1.90 m 1.90 m

11 13 12 14

282.1 408.8 38.1 551.6


103 - 104 103 - 104 10 - 102 102 - 103

4 103- 4 104 2.5 103 - 2.5 104 2.5 102 - 2.5 103 2 102 - 2 103

Direct sampling at the steel-clay interface was also performed using sterile cotton swabs at 0.83 m and 1.90 m and these clay samples were cultivated using both media. After two months incubating all cultures showed significant formation of FeS as shown in Figure 4.

a.

b.

c.

Figure 4: Boom Clay cultures from clay sampled at the steel-clay interface. (a.) Swabs sampled at 0.83 (9 (Postgate medium) & 10 (RBCW medium)) and 1.90 m (15 (Postgate medium) & 16 (RBCW medium)). Clay sampled at 0.53 m incubated in (b.) RBCW medium and (c.) Postgate medium.

Besides the formation of a black deposit (FeS) SRB presence was also proven through the isolation of DNA from the different cultures. The presence of SRB was confirmed via the

13

detection of the apsA gene (PCR reaction using APS7-F and APS8-R [Friedrich, 2002], see Appendix 3 for reaction conditions). 3.2.2 Clay core analyses Besides the clay sampled at the steel-clay interface, SRB presence was also investigated at three different positions (2, 10 and 18 cm from the steel-clay interface) in the two clay cores which were collected using cutting edges. These clay cores were collected at 0.53 m and 1.90 m from the gallery lining. Table 3 shows the amount of bacteria present in the different clay samples. Table 3: The amount of bacteria present in 0.25 g of Boom Clay sampled at different positions with respect to the steel-clay interface. Clay core position*

Distance to steelclay interface

Cultivation medium

Number of SRB in sample (0.25 g)

Number of SRB in 1 g of Boom Clay [CFU/g]

0.53 m 0.53 m 0.53 m 0.53 m 0.53 m 0.53 m

0.02 m 0.10 m 0.18 m 0.02 m 0.10 m 0.18 m

Postgate Postgate Postgate RBCW RBCW RBCW

1 - 10 0 0 1 - 10 1 - 10 10 - 100

4 - 40 0 0 4 - 40 4 - 40 40 - 400

1.90 m 1.90 m 1.90 m 1.90 m 1.90 m 1.90 m

0.02 m 0.10 m 0.18 m 0.02 m 0.10 m 0.18 m

Postgate Postgate Postgate RBCW RBCW RBCW

1 - 10 1 - 10 1 - 10 1 - 10 1 - 10 1 - 10

4 - 40 4 - 40 4 - 40 4 - 40 4 - 40 4 - 40

*

: Distance to the of the gallery lining.

Figure 5 summarizes the results from Table 1 and 2 and shows the trend for the amount of bacteria present in Boom Clay at different positions from the steel-clay interface. 10000000 0,53 m (Postgate) 0,53 m (RBCW) 1,90 m (Postgate) 1,90 m (RBCW)

Number of SRB [CFU / g clay]

1000000 100000 10000 1000 100 10 1 0m

0,02 m

0,10 m

0,18 m

Distance to steel-clay interface [m]

Figure 5: The amount of SRB present in Boom Clay at different positions from a steel-clay interface.

14

4 Discussion Sulphide, thiosulphate and sulphite were not detected in the Boom Clay cores but sulphate was measured. The measurements show a different trend: the clay core sampled near the gallery contains less sulphate than the clay core sampled further away from the gallery except for the sulphate concentration at the clay-steel interface. However, normally, it would be expected that the clay core near the gallery at 0.53 m would contain the higher sulphate concentrations due to oxidation during the excavation of the gallery and an additional oxidation during the installation of the steel tube. The clay core situated outside the excavation disturbed zone (at 1.90 m), should contain lower sulphate levels which are only caused by oxidation during the installation of the steel tube. Based on the limited data from the microbial experiments there appears to be no difference between the RBCW medium and the Postgate medium concerning the growth of SRB. Investigation of the two Boom Clay cores sampled using cutting edges reveals that the SRB concentration at the clay-steel interface is significantly higher than the SRB concentrations inside the Boom Clay at a few centimeters from the interface. The difference could be explained by the fact that the SRB at the interface had the necessary space to grow while further away from the interface growth of SRB was significantly less to non-existing due to space restrictions. However, the presence of the metallic surface could also account for the increased bacterial concentration.

5 Main outcomes of the experiments The steel samples were investigated for corrosion and although the steel samples showed severe corrosion only very limited traces of sulphur were detected on the steel surface [Kursten, 2008]. Therefore, the observed corrosion of the steel pipe can not be directly related to the activity of SRB. However, the highest SRB concentrations were observed at the clay-steel interface, while in the Boom Clay a few cm away from the interface much smaller populations were detected. Although sulphide was not detected in the Boom Clay at the interface, the increased amount of SRB at the clay-steel interface and the presence of sulphate at this position implies that SRB induced corrosion can not be ruled out. However, the elevated concentrations of SRB at the interface with the corroded surface combined with a lack sulphur compounds on the metallic surface could indicate some other mechanism of microbially induced corrosion.

6 Prediction of the impact of SRB activity on the supercontainer Research concerning the influence of the conditions imposed by the supercontainer design, i.e. high pH and elevated temperature, on SRB was previously performed [Aerts, 2008]. The experiments showed that viable SRB can be present inside the concrete buffer but that these SRB will very likely never be active inside the buffer due to the high pH of the cementitious environment (pH > 12.5). However, SRB activity at the concrete-clay interface could not be excluded since the pH (± 10), the temperature (ranging between 40 and 70°C) and the available space create an environment suitable for the growth of SRB isolated from Boom Clay water. SRB could influence the supercontainer performance through the production of sulphide. Culture experiments with SRB isolated from Boom Clay water have shown sulphide concentrations in the range of 4.7 to 9.4 mM [Aerts, 2008]. One can not exclude that these sulphide concentrations could be produced by SRB developing in residual voids existing

15

around the galleries or in the annular space between the supercontainer and the gallery lining in case of poor backfill scenarios. Due to the concentration gradient, sulphides could then diffuse inside the cementitious buffer towards the carbon-steel overpack. To what extent the metallic barrier and the concrete buffer will be affected by the sulphide remains largely unknown and certainly deserves thorough study. Different aspects of the sulphur issue were discussed during the SACNUC workshop (21-23 October 2008, Brussels) and an overview of these issues is presented here. Source term: Is it possible to predict the sulphide concentration at the overpack surface? Is it correct to assume that sulphur transport is diffusion-controlled? In the absence of buffer materials, the metallic overpack is directly exposed to air during the operational phase and to water in the long term, there is no diffusion barrier and the sulphide concentration expected in the water could be very high because of the possible development of SRB. The role of compacted materials (bentonite or cementitious materials) installed around containers is to limit the different processes favourable to corrosion: • Transport of aggressive sulphur species (HS–, S2O32–) is controlled by diffusion. • Microbial activity is limited by several factors, such as: low water activity, space restriction and diffusion-limited transport of nutrients/toxins. At dry densities above 2 000 kg/m3 no significant bacterial activity can be observed [Pedersen et al., 2000]. Also, at pH higher than 12.5, no microbial activity is expected. In the absence of microbes, there is no longer production of free sulphides close to the overpack surface. • In the presence of Fe2+ in the compact buffer, no high concentrations of HS– are expected because of the precipitation of poorly soluble iron sulphides such as FeS or FeS2. To predict the transport and the reactivity of sulphur species at the surface of canisters, it is also necessary to better understand the water chemistry of reduced sulphur species. A lot is known but not everything, a.o. the nature of adsorbed surface species. How can short-term experimental results be extrapolated to the long time-frame of geological disposal? Sufficient knowledge of elementary mechanisms is necessary to develop models for long-term assessment for supporting semi-empirical models mostly used in canister lifetime prediction. Mechanistic understanding for Cu is much more advanced than for carbon steel. Sensitivity analysis with different concentrations of sulphides are needed as for chlorides, i.e. do threshold values exist below which no effects are observed? Then, extrapolation to real conditions should be feasible. For the Belgian concept, which foresees a pH of at least 12.5 for an extended period of time, can bacterial activity be excluded? Which bacteria can be active at this pH? Under which conditions does microbial activity not play a role? How can we prove this experimentally? No significant microbial activity is expected at high pH (> 12.5). It is much more difficult for bacteria to cope with high pH than with low pH. To survive, alkalophiles must maintain a relatively low alkaline level around pH 8 inside their cells by constantly pumping hydrogen ions (H+) across their cell membranes into their cytoplasm [Horikoshi, 1999]. Energetically and from the viewpoint of mass transfer (very limited number of available H+ at high pH), the proton pump in cell membranes functions much less efficiently at high pH than at low pH. Moreover, membrane protein degradation and hydrolysis reactions are more severe under alkaline conditions than in acidic media [Krulwich, 2006].

16

Culture experiments conducted with SRB isolated from Boom Clay water revealed no SRB activity above pH 11. Prolonged exposure (four months) of these SRB to high pH environments (pH 11, 12 and 13) showed that viable SRB remained present [Aerts, 2008]. Experiments should be designed to test alkalophile SRB strains, based on a careful overview of the existing literature. Another argument also deals with the poor life evolution at high pH. On Earth, extreme acidic environments are relatively frequent (e.g., volcanic lakes, sulphide ores mine tailing, …), while alkaline environments are much rarer. The pH of soda lake (natron-, trona-rich evaporites in desert environments, …) does not exceed a value of 12 typical for natural sodium carbonate (Na2CO3) salts [Grant, 2006]. In general, life only evolves to adapt to stressing extreme conditions. As high-alkaline environments are rare in nature, life has not been too often exposed to such conditions and as a consequence evolution to adapt to high pH has not much progressed [K. Pedersen, personal communication]. Degradation of cements and concretes is commonly observed under aerobic conditions and could occur in the early phase of repository operations. Sulphur oxidising bacteria oxidise sulphur, sulphides and thiosulphates under aerobic conditions to produce sulphuric acid. These acids can attack the concrete matrix by dissolving calcium silicates hydrate gel and Ca(OH)2. Direct anaerobic corrosion of concretes is not known [West et al., 2002]. Microbes can only develop at the surface of carbonated cement, i.e. the external surface and the surface of large cracks occurring in cement blocks. The presence of biofilms developing inside open fissures in old concrete structures should be investigated. How can the effect of irradiation on sulphide be studied? There is a lack of knowledge to develop radiolysis models integrating a sulphur chemistry module. A gap exists for the sulphur species between (-I) and (III) valences. Irradiation of blast furnace slag concrete pastes in the UK showed the formation of calcium trisulphoaluminate hydrate (AFt). The observation that an AFt type phase is only present in the irradiated blends may imply that the oxidation of reduced sulphur species (S2-) is accelerated by gamma irradiation [Richardson et al., 1990]. Knowledge of fundamental radiochemistry of sulphur is required. What happens when there concurrently occur aerobic and anaerobic zones and can this be avoided? In case of disposal concepts without buffer material installed in close contact with the canisters, anodic area cannot be ruled out. This is the main disadvantage of leaving an open annular space around the canisters for the sake of retrievability. The only way to overcome the problem is to completely backfill the void of the annulus. Another major advantage of the complete backfilling (with dense swelling bentonite or cementitious materials) is to limit microbial growth and to control the transport of sulphide in the system by diffusion. How can we predict what will/can happen during excavation and disposal? What is the impact on corrosion? It is important to know the spatial extent of the oxidized zone in the excavation disturbed zone in a clay formation and to assess the quantity of H2SO4 that could be produced by pyrite oxidation around a gallery. The thickness of the oxidized zone is presently estimated not to exceed about 2 m for the galleries excavated at Mol in the Boom Clay [De Craen et al., 2008]. It is also important to evaluate the impact of the salts accumulated in open and ventilated galleries on the degradation of cementitious buffers (ettringite and thaumasite crystallisation) and on the corrosion of steels.

17

The continuous water flow towards an open and ventilated gallery limits the diffusion of oxygen in the clay formation and flushes back sulphate along with chloride towards the gallery. As a consequence, salts accumulate near the gallery as can be observed by efflorescence of sulphate on the walls of ventilated galleries. High salt amounts accumulated in the engineered barrier system (EBS) might influence the corrosion of the metallic overpack and should therefore also be taken into account. The possibility of ingress of aggressive species within the EBS and its consequences on key safety functions, particularly containment, should be assessed to determine the maximum amount of salt that could accumulate in, or around, an open and ventilated gallery. The maximum time duration acceptable to leave a gallery open could be more limited by the consequences of salt accumulation in the near-field of a gallery than by the extent of the oxidized zone in the formation, i.e., near-field effects might be more limiting than far-field implications. Anaerobic corrosion of iron will also produce hydrogen that can easily be used by bacteria as electron source to reduce CO2 in acetate or to achieve sulphate reduction. Finally, underground man activities can leave various contaminations in, or around, galleries: nitrate as residues of blasting operations, organic matter (cement admixtures, wood beams abandoned behind concrete lining after supporting gallery roof or walls, oil and various technological wastes). Quality assurance should be applied and a strong regulation should be enforced to minimize the introduction of organic matter in a repository. Mass balance of inventories of critical materials before and after each operation in which they are used could be helpful to guarantee that they have been totally removed and to minimize possible surprises. Will mass transport limitations drive the corrosion rate to very low values? Could the presence of S-species lead to the formation of protective sulphide containing films? Corrosion of metal surfaces in intimate contact with compacted bentonite is a diffusion controlled process as very well illustrated in the talk of Fraser King. Coupling point-defect model, or interfacial reactions parameters, at the metal/bentonite interface to reactive transport calculations of corrosive species migrating in compacted bentonite, is certainly an important step to achieve more reliable long-term predictions on the corrosion of metals. The precipitation of an insoluble metallic sulphide film at the surface of metals, or metal oxides, could contribute to the formation of a protective layer passivating the metal surface. However, the barrier properties and the mechanical adhesion of porous sulphide films are far from being sufficiently understood.

7 Conclusions and recommendations The investigation of steel and Boom Clay samples which have been in close contact during 20 years was a first attempt to asses the problem of microbially induced corrosion in Boom Clay. Increased concentrations of SRB were detected at different sampling positions near the claysteel interface but free sulphide was not detected in Boom Clay. Although the steel showed severe signs of corrosion only limited traces of sulphur were detected at the steel surface implying that the corrosion on this steel tube was probably not caused by SRB. However, discussions during the SACNUC 2008 workshop revealed several questions concerning the possible effects of SRB activity and the resulting free sulphide. Therefore, insitu experiments in Boom Clay should be performed to confirm earlier results obtained in lab experiments.

18

References Aerts S., Boven P., Jacops E., De Boever P., Van Geet M. (2007) Sulphate reducing bacteria in Boom Clay, Poster Presentation at "Clays in natural and engineered barriers for radioactive waste confinement", Lille, France, 17-20/09/2007 Aerts S., Jacops E., Dewel A. (2009) Microbial activity around the Connecting Gallery, Revision 1, SCK•CEN report (SCK•CEN-ER-61) Aerts S. (2008) Use of inhibitors to prevent bacterial artefacts in experiments, SCK•CEN report (SCK•CEN-ER-65) Aerts S. (2009) Effect of geochemical conditions on bacterial activity, SCK•CEN report (SCK•CEN-ER-75) De Craen M., Honty M., Van Geet M., Weetjens E., Sillen X., Wang L., Jacques D., Martens E. (2008) Overview of the oxidation around galleries in Boom Clay (Mol, Belgium), NF-PRO report D4.3.24 Friedrich M.W. (2002) Phylogenetic analysis reveals multiple transfers of adenosine-5'phosphosulfate reductase genes among sulfate-reducing microorganisms, Journal of Bacteriology, 184,1, 278-289 Grant W.D. (2006) Alkaline environments and biodiversity, In: Extremophilies, Eds: Gerday C., Glansdorff N., In: Encyclopedia of Life Support Systems (EOLSS), developed under the auspices of the UNESCO, Eolss Publishers, Oxford, UK (http://www.eolss.net) Horikoshi K. (1999) Alkaliphiles: Some applications of their products for biotechnology, Microbiology and Molecular Biology Reviews, 63, 4, 735-750 Krulwich T.A. (2006) Alkaliphilic prokaryotes, In: The Prokaryotes Volume 2, 3rd Edition, Eds. M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, E. Stackebrandt, Springer Science, Singapore Kursten B. (2009) Corrosion studies of structural in situ steel tubes exposed to Boom Clay for 20 years, SCK•CEN report (SCK•CEN-ER-82) Pedersen K., Motamedi M., Karnland O., Sandén T. (2000) Mixing and sulphate-reducing activity of bacteria in swelling, compacted bentonite clay under high-level radioactive waste repository conditions, Journal of Applied Microbiology, 89, 1038-1047 Richardson I.G., Groves G.W., Wilding C.R. (1990) Effect of γ radiation on the microstructure and microchemistry of ggbfs/OPC cement blends, Mat. Res. Soc. Symp. Proc., 176, 31-37 West J.M., McKinley I.G., Stroes-Gascoyne S. (2002) Microbial effects on waste repository materials, In: Interactions of microorganisms with radionuclides, Eds. M.J. Keith-Roach, F.R. Livens, Elsevier, Oxford, UK

19

Appendix 1: Cultivation media .1 Preparation of synthetic media One synthetic medium was used in this report for the enrichment cultures: the Postgate medium. Besides the synthetic medium also a RBCW-based medium was used in order to more closely mimic the in-situ conditions. Since SRB need to be cultivated under anaerobic conditions both media contain ascorbic acid and thioglycolate, reductants, and rezazurin, a redox indicator that will change colour from colourless to pink in the presence of oxygen. The Postgate medium was prepared in an anaerobic box (Ar-H2 atmosphere (95-5)) using deoxygenated milliQ water. The different components were weighed outside the anaerobic box and were subsequently brought inside. Inside the box the components were dissolved in the water and the pH of the mixture was adjusted to the correct pH value (7.8) using NaOH (pH is measured using a glass electrode). The solution is subsequently transferred into hungate tubes (4.5 ml in each tube) while continuously swirling the medium. The RBCW medium was prepared using the same basic approach: components are weighed outside the anaerobic box and the medium mixture is prepared inside the anaerobic box. The Boom Clay water used in the preparation was filter sterilized (GD/X syringe filter, Whatman) prior to the addition of the different salts and after transfer to hungate tubes. These hungate tubes are sterilised in an autoclave at 121°C for 20 minutes. The sterile tubes are stored in a cold room after sterilization. The inoculation of the medium is performed inside an anaerobic box to prevent the contamination of the medium with oxygen. The quality of the Postgate medium was tested prior to the experiments through the incubation of the media with MORPHEUS water samples and Desulfovibrio intestinalis (DSM 11275). Growth of SRB was detected in both cases through the production of sulphide and the isolation of bacterial DNA containing the APS gene1. The quality of the RBCW based medium was evaluated through the incubation of MORPHEUS water samples. Production of sulphide was detected in the medium. .2 Composition of the media The composition of the two media used in this study is shown in Table 4. Table 4: Composition of the Postgate and RBCW medium. Values given in g.

Medium composition [g] Component CaCl2.2H2O FeSO4.7H2O K2HPO4 MgSO4.7H2O Na2SO4 NH4Cl DL-Na-lactate Na-acetate Yeast extract 1

Postgate medium

RBCW medium

0.1 0.5 0.5 2.0 1.0 1.0 2.0 1.0

0.5 0.5 2.0 1.0 0.75 1.0 1.0 1.0

Information on the DNA extraction and the PCR reaction: Appendix 3.

20

Rezazurin Na-thioglycolate Ascorbic acid Milli-Q water RBCW (filter sterile)

0.001 0.1 0.1 1000 -

0.001 0.1 0.1 1000

.3 Reference composition of RBCW (EG/BS piezometer) Mean composition of Boom Clay water sampled from the EG/BS piezometer. Na 17.83 mM (410 ppm) K 0.24 mM (9.3 ppm) Mg 0.12 mM (3.0 ppm) Si 0.11 mM (3.0 ppm) Ca 0.10 mM (3.9 ppm) Fe 0.02 mM (1.1 ppm) HCO3Cl F Br SO42-

17.18 mM (1048 ppm) 0.73 mM (26.0 ppm) 0.17 mM (3.2 ppm) 0.006 mM (0.5 ppm) 0.006 mM (0.6 ppm)

TIC TOC

206 mg C/l 96 mg C/l

21

Appendix 2: Sulphide measurement Since sulphide is extremely sensitive to oxidation several precautions were taken to prevent oxidation of sulphide during the measurement and hence a distortion of the results: ¾ The original sample is diluted in deoxygenated milliQ water. ¾ Syringes used to sample the cultures were stored inside an anaerobic box (Ar-H2) for at least 48 hours with the plunger at the maximum position to fill the syringe with atmosphere from the anaerobic box. Thus, the syringe contains an oxygen-free atmosphere and oxygen traces inside the syringe and needle will diffuse to the surrounding atmosphere. ¾ The culture sample is injected directly into the dilutant, i.e. make sure that the needle is placed in the diluting liquid before injecting the sample, to avoid contact with the atmosphere. ¾ The reagents are added immediately to the diluted sample. The measurement is based on the methylene blue method [] where the formation of methylene blue, which is proportional to the amount of sulphide present in the sample, is measured using a UV-VIS spectrophotometer at 612 nm. Measurements were performed on a Ultrospec 3000 UV/Vis spectrophotometer (Pharmacia Biotech) using plastic cuvets (Greiner Bio-one) with a path length of 1 cm. A calibration curve was determined using Na2S.nH2O (VWR, for analysis, ~35 % Na2S). The Na2S crystals were stored inside an anaerobic box and the calibration solutions were prepared inside the box. The solutions were prepared in hungate tubes to prevent oxidation of sulphide outside the box and poisoning of the catalyst inside the box as much as possible. The calibration curve is shown in Figure 5. 2 1,8 1,6 y = 0,0478x R2 = 0,9989

Absorbance

1,4 1,2 1 0,8 0,6 0,4 0,2 0 0

5

10

15

20

25

30

35

40

Concentration Sulphide [ppm]

Figure 5: Calibration curve for the measurement of sulphide using a Na2S.nH2O dilution series measured at 612 nm.

Due to the limited range of the methylene blue method additional dilution of the samples was sometimes necessary. These dilutions were prepared using an identical approach to the sampling procedure, i.e. dilutions were prepared in deoxygenated water using 'degassed' syringes.

22

Appendix 3: Detection of SRB: apsA The presence of SRB was investigated via isolated DNA through the detection of the apsA gene, a key enzyme in the reduction of sulphate by bacteria. The screening was performed through a PCR reaction employing the APS7-F and APS8-R primer pair [Friedrich, 2002]. DNA isolated from a pure culture of Desulfovibrio intestinalis (DSM 11275) cultivated in Postgate medium was used in all PCR reactions with APS primers as a positive control. The PCR reaction was performed using the GoTaq Green Master Mix (Promega) and the reaction mixtures contained 12.5 µl master mix, 1 µl of both primers (50 ng/µl), 5 µl DNA template and 5.5 µl of nuclease free water. The reaction mixtures were prepared on ice and the reaction was carried out using a GeneAmp PCR System 2700 (Applied Biosystems). The following thermal profile for amplification was used: an initial denaturation step (3 minutes, 94°C) was followed by 35 cycles of denaturation (40 seconds at 94°C), annealing (55 seconds 56°C) and extension (1 minute, 72°C). After a terminal extension (7 minutes, 72°C), the samples were stored at -20°C. Aliquots of the amplicons (2 µl) were analyzed by electrophoresis on 1% agarose gels and visualized after staining with GelRed (Biotium Inc.) using a Molecular Imager Gel Doc XR System (Bio-Rad Laboratories). The DNA ladder (FastRuler DNA Ladder High range, Fermentas), which contains 5 bands: 10000, 4000, 2000, 1000 and 500 bp (Figure 6), was used as internal standard to determine the length of the DNA fragments.

10000 bp 4000 bp 2000 bp 1000 bp 500 bp Figure 6: FastRuler DNA Ladder High range (Fermentas).

23

Date Author:

Sven Aerts

Verified by:

Natalie Leys

QA verification:

Elke Jacops

Approved by:

Mieke De Craen

Approval