ABM Project meeting 2012 The upcoming TR report: Content and mayor findings Some thoughts about the continuation Daniel Svensson

Content – The experiment -The included clay materials and their origin -The experimental configuration -The test site -Excavation procedure and sampling

Content - Hydromechanical tests -Water content and density -Swelling pressure and hydraulic conductivity -Unconfined compression tests

Content – Chemical and mineralogical analysis - Reference materials -Cation exchange capacity (CEC) -Anion analysis -Reductive dissolution extraction (CBD) -Chemical data and evolved gas analysis -Powder X-ray diffraction (XRD) bulk and clay fractions

Content – Chemical and mineralogical analysis – Package 1 -Focus on MX-80, Asha, DepCAN and Calcigel -Cation exchange capacity (CEC) -Anion analysis -Exchangable cations (EC) -Chemical composition of bulk samples -Powder X-ray diffraction (XRD) - bulk samples

Content – Microbiology - Reference materials -Quantification of microbes by culturing -Microscopy -Identification of microbes DNA extraction Cloning and sequencing - Special emphasis was put on aerobic heterotrophs and anaerobic sulphate and iron reducers and autotrophic acetogens

Content – Microbiology – Parcel 1 Asha 505, Deponit CA-N and MX-80, were analysed with focus on: -Cultivable heterotrophic aerobic bacteria -Heterotrophic sulphate-reducing bacteria -Autotrophic sulphate-reducing bacteria -Iron-reducing bacteria

Mayor findings – Hydromechanical I The hydro-mechanical standard tests (water content, free swelling, liquid limit and grain density) showed the large diversity of the materials. The liquid limit varies between 68 % (Friedland) up to 545 % (MX-80). There was also a clear difference between the different materials regarding the grain density. Eight of the tested materials had a grain density between 2678 and 2753 kg/m 3 while three of the materials (Asha, Rokle and Friedland) had a grain density between 2828 and 2940 kg/m 3

Mayor findings – Hydromechanical II At excavation the degree of saturation was very close to 100% in all samples. Hydraulic conductivity and swelling pressures were determined on MX-80, Asha 505 and Deponit CAN. No change was identified.

Mayor findings – Hydromechanical III A significant decrease in swelling pressure was seen on Asha 505 and Deponit CAN. The largest deviation was noticed on samples from the innermost part. One possible explanation is the cation exchange process in the clays.

Mayor findings – Hydromechanical IV Mechanical properties were determined by unconfined compression tests on MX-80, Asha and Deponit CAN. Reduced strain was observed on re-saturated specimens of all three materials from the field experiment. Regarding maximum deviator stress and compared to reference material no large deviations was seen on specimens of MX-80 and Deponit CAN while a decreased maximum deviator stress was seen on Asha specimens and especially on specimens taken from the innermost position. Unconfined compression tests were also run on air-dried, ground, re-compacted and re-saturated material of MX-80, Asha and Deponit CAN from the field experiment. Those tests showed strain of the same size or larger than observed on the reference specimens while the maximum deviator stress was of the same size.

Mayor findings – Chemistry and mineralogy - reference clays I Most clays had a rather high CEC values which is typical for bentonites. Friedland and Callovo-Oxfordian was much lower in CEC which is compatible with their different mineralogy. The water extractable amount of sulphate and chloride was not negligible. Ikosorb, Friedland and Asha 505 had wt% chloride. MX-80, IbecoSeal, Deponit CA-N and Friedland had wt% sulphate, the most likely source is gypsum which is a water soluble hydrated calcium sulphate. Fluoride and bromide are very low in all cases. CBD-extracts showed elevated amounts of Mn in Rokle indicate possible presence of undetected Mn- phase. Rather high Si in Calcigel and Rokle (compared to Kunigel that is low but high in quartz) possibly indicates the presence of amorphous silica.

Mayor findings – Chemistry and mineralogy - reference clays II The chemical data and evolved gas analysis confirmed the variety in the clays regarding e.g. iron content. Kunigel with only 1.9 wt% Fe2O3 can be compared to Asha 505 and Rokle with 12.2 and 13.7 wt% respectively. Organic carbon was low in most cases, but higher in Callovo-Oxfordian (0.7) and Friedland (0.5) wt%. The total sulfur content was very low (<0.2 wt %) in Rokle, Ikosorb, Febex, Calcigel, Asha 505, low (< 0.5 wt %) in MX80, Ibecoseal and Kunigel, and higher in Friedland (0.53) and Callovo-oxfordian (0.68). X-ray diffraction was used to show the variability in the crystalline phases of the clays. Most clays were found to be dominated by dioctahedral smectites.

Mayor findings – Chemistry and mineralogy - reference clays III Close up of the 060 region

Mayor findings – Chemistry and mineralogy - Parcel 1 - I The distribution of chloride indicated that the wetting with groundwater resulted in laterally and vertically smoothed and almost constant concentrations. The sulfate distribution displayed no consistent pattern, but some blocks had a sulfate maximum at the heater, along with depletion in the peripheral parts, indicating a lateral transfer of sulfate. The test conditions also resulted in significant changes in the exchangeable cation pool. Whereas lateral gradients within the blocks were insignificant, a vertical gradient in the relative cation distribution had developed, dividing the package into an upper, Ca-dominated, and a lower, Na-dominated part.

Mayor findings – Chemistry and mineralogy - Parcel 1 - II The distribution of acid soluble carbon indicated carbonate dissolution close to the heater in the carbonate-bearing bentonites. The iron content had increased in several of the blocks All four blocks examined displayed a deficit in magnesium in the peripheral parts and a more or less distinct gradient towards to the heater. The XRD-data for random powders provided no evidence of any significant change of the smectite structure in any of the four blocks examined. Supplementary analyses of clay fractions are, however, necessary for more detailed evaluations..

Mayor findings – Chemistry and mineralogy - Parcel 1 - III A tendency of increasing CEC towards the heater can be seen in particular in calcigel (5) and Deponit CAN (27). Dissolution of carbonates in the samples proximal to the heater was noted and could be one reason. On the other hand, the incorporation of corrosion products of iron in the bentonite would be expected to lower the CEC of samples closest to the Fe- tube due to the dilution effect. Alternatively it is also possible that corrosion products may add to the CEC or reduction of smectite iron.

Mayor findings – Microbiology – Reference materials - I A drastic difference in microbial abundance between the materials was observed; Kunigel V1 was close to sterile while Friedland and Ibeco Seal M-90 contained high numbers of all microbes examined. It was found that 200 – /g aerobic microbes were present in ten of the eleven dry materials, while Kunigel V1 lacked detectable aerobic microorganisms. Iron reducers were detected in all examined materials, ranging from 10 (in Kunigel V1) to 8000/g (in Ibeco Seal M-90). Autotrophic acetogens was detected in the range of 20 – 60/g in the Febex, Friedland and Ibeco Seal materials, and below 10/g in the other materials examined.

Mayor findings – Microbiology – Reference materials - II Sulphate reducers were detected in the range of 10 – 90/g in Asha 505, Calcigel, Deponit CA-N, Febex, Friedland and Ibeco Seal M-90, and was below 10 g-1 in the other materials. The dominant part of sulphate reducers and acetogens in the materials thrived in the thermophilic temperature range and grew faster at 50°C than at 20°C. A correlation between produced acetate and sulphide was observed, suggesting that acetogenesis boosted the sulphate-reducing populations in the clay mineral systems. The studies showed that dry clay is a very potent media for long term survival of both aerobic, anaerobic and thermophilic microbes and as it seems from phylogenetic studies, not spore-formers only.

Mayor findings – Microbiology – Parcel 1 It was found that mesophilic aerobic bacteria were present in the range of 10 2 – 10 3 g -1 wet weight (gww -1 ) in Asha 505, Deponit CA-N and MX-80. All other bacteria were below detection in the materials. Thus, the results showed that bacteria with potential corrosive or buffer degradation properties were present in the raw materials, however failed to survive after the high swelling pressure and temperatures of the experiment.

Some ideas about the continuation: -Further evaluation of collected XRD bulk data -Develop custom designed Siroquant diffraction profiles from observed data on pure smectite fractions to provide good input to quantitative phase analysis Further evaluation of the clay fractions by: -Evaluating the ethylene glycol samples by the Reynolds & Moore method using Newmod -Special pretreatment of samples with kaolinite and/or chlorite to reduce uncertainties -Evaluation of pre-saturation of bulk clay with divalent cations for quantitative XRD and/or use of internal standard -Search for new methods to further characterize the iron-bentonite zone and to locate the iron-phase (s)

Availability The 1st proof has been constructed. A list of corrections is beeing assembled. Probably printed and ready after the summer (?)

Acknowledgement: And thank you all for your time! We want to thank Dietrich Koch For all information about the clays and their origin. And also for making a study visit to the clay mines in Landshut possible. Stephan Kaufhold and Reiner Dohrmann For reviewing the report and hence increasing its quality and readability.