Technical Report NTB 85-05

During drilling and testing at Böttstein, samples were collected for geochemical and isotopic analysis. The sampling and analytical procedures and chemical and isotopic results are described in detail in other Nagra Technische Berichte. This report reviews the procedures and analyses and provides a geochemical interpretation of the results.

Two sedimentary horizons were sampled: the upper Muschelkalk and the Buntsandstein/weathered crystalline zone. Four sets of samples for geochemical analyses and five for isotope analyses were taken from the crystalline section above 1,000 m. They represented the three zones of highest hydraulic conductivity in that interval. The zones of highest conductivity below 1,000 m were also sampled but yielded so little water that only mixtures of formation water and borehole fluid could be collected.

Samples were collected by pumping or from artesian outflow at the surface, and by using pressure vessels at depth. The crystalline section was drilled with deionised water and (or) formation water which contained uranine and MTFMBA tracers. The residual drilling fluid present in samples could thus be monitored precisely.

The sedimentary section was drilled with mud which contained no tracer. Environmental tritium was used as a tracer for samples from both sedimentary and crystalline zones.

The amount of drilling fluid present in the Muschelkalk and upper crystalline samples was negligible for geochemical interpretation. The samples from the Buntsandstein/weathered crystalline zone also contained very little drilling fluid, but, because the interval had been drilled with brine, the sample was not suitable for geochemical interpretation.

The zone of highest hydraulic conductivity in the lower crystalline section was sampled several times. No sample contained less than 40 percent drilling fluid, but the tracer in the drilling fluid made it possible to calculate back to the composition of formation water and so make some geochemical conclusions.

The analyses were examined for charge balance, concordance between measured and calculated total dissolved solids content, and for consistency between duplicate samples or analyses. All showed excellent internal consistency and agreement among duplicates. The results were also examined for their geochemical consistency using the geochemical computer program PHREEQE.

Calculations using measured pH values resulted in dissolved CO₂ concentrations below those measured, and indicated that all samples are oversaturated with respect to calcite. From this it seems all the measured pH values are too high, probably as a result of CO₂ gas loss during sampling or analysis.

The pH of Muschelkalk formation water was calculated using its analyzed dissolved CO₂ content. At this pH, Muschelkalk water is at equilibrium with both calcite and dolomite.

There is isotopic and petrographic evidence that calcite is precipitating from water now present in some fractures in the upper crystalline section. pH values for the samples from this section were calculated assuming the formation water is saturated with respect to calcite.

All samples were affected by the iron packers, tubing and newly-installed casing in the borehole. Iron and hydrogen were added to the samples by corrosion of this iron. Dissolved iron values representative of the formation waters were estimated from analyses from older wells sampled during the Regional Program. Oxidation potentials calculated from the H₂/H⁺ couple are more reducing than those from any other indicator and are not representative of the formation waters.

Small concentrations of dissolved oxygen were reported in almost all samples. The oxygen was most probably introduced during sampling and was not present in the formation water.

Oxidation potential values were derived from measured platinum electrode potentials, calculations using analyzed concentrations of members of redox couples, and calculations based on mineral-water and mineral­-mineral reactions. The waters appear to be responding to a range of oxidation potentials. The magnetite/hematite mineral pair and the redox pairs comprising the dissolved species NH₄⁺/N₂ and CH₄/CO₂ give similar values and define the most reducing conditions to which the waters respond.

The potential at which the uranium dissolved in the waters would be in equilibrium with uraninite is more positive and represents the most oxidizing conditions to which the waters respond. The potentials measured with the platinum electrode were not consistent with any other single indicator of oxidation potential. The measured potentials of some samples corresponded to values calculated from the H₂O/O₂ couple using Sato's (1960) relationship between dissolved oxygen and oxidation potential. Potentials measured in other samples were similar to values calculated from the dissolved uranium concentrations.

Aluminium and silica concentrations of waters from the upper crystalline were calculated assuming saturation with respect to kaolinite and chalcedony. The calculated aluminium concentrations agreed with those measured within the analytical precision reported. Two dissolved silica analyses agree with the calculated values but two are significantly higher. It is possible that the high-silica samples were not filtered before silica analysis.

Calculations suggest that all samples are at equilibrium with fluorite. However, no fluorite was observed in the core.

Water in the Muschelkalk is saturated with respect to calcite, dolomite and gypsum. Its chemistry is interpreted as evolving through a process of dedolomitization driven by gypsum dissolution. The stable carbon isotopic composition of the water agrees with the values calculated from a model of this dedolomitization process and thus supports it. With the model, the ¹⁴C content can be used to estimate a water residence time of 17,000 ± 6,000 years.

Samples for geochemical and isotopic analyses were taken from zones in the upper crystalline centered on depths of 399, 621, 619 and 792 m. The 399 and 621 m samples were taken during drilling, while the 619 and 792 and a sample from 649 m for isotopes alone were collected during the testing phase, after the completion of drilling.

The chemistry of water in the upper crystalline section is virtually identical in all samples. It is a sodiumbicarbonate water with pH of about 8 and a total dissolved solids content of 1,000 to 1,110 mg/l. It appears to be responding to oxidation potentials in the range of -0.1 to -0.3 volts.

Samples from the Buntsandstein/weathered crystalline and those taken during drilling have decreasing ¹⁴C contents with depth. The ¹⁴C from the Buntsandstein/weathered crystalline at about 316 m was 12 to 14 percent modern carbon (pmc), at 399 m it was 8 pmc and at 621 m it was

Two later samples were taken from the same water-bearing zone as the 621 m sample. The 619 m sample, taken about 200 days later had about 8 pmc and the sample from 649 m taken about 260 days after the 621 m sample had about 14 pmc.

This behaviour suggests that the water-bearing zones of the upper crystalline are well connected. While under natural conditions, deeper waters at Böttstein have lower ¹⁴C contents and so longer residence times, the flow in the system induced by the borehole was enough to bring water from the Buntsandstein/weathered crystalline to below 620 m in no more than 250 days.

It is clear from these results that future sampling in boreholes of this type must be done as soon after drilling as possible.

The deep crystalline section is represented by a series of samples from an interval centered on 1,326 m. These samples are of mixtures of formation water with traced, de-ionized drilling fluid. The composition of formation water was estimated by extrapolating the compositions of the mixtures to a tracer concentration of zero.

Water from the deep crystalline is a sodium-chloride water with a total dissolved solids content of over 13,000 mg/l. The origin and evolution of this water cannot be evaluated without further study, but it is likely that it has a very high residence time in the crystalline rock and may have its origin in, or be influenced by, water in the Permo-Carboniferous trough.