Laboratory experiments

In order to test the theoretical considerations presented above, laboratory experiments were conducted to determine the resistivity-temperature curves for saturated and dry material from the field site at Schilthorn. For these experiments, a set of miniature electrodes and cables was developed for the use on the laboratory scale (Figure 6.4). This miniature DC resistivity tomography system was originally developed to monitor the migration of contaminants in soils in scaled centrifuge experiments (Depountis etal. 1999). The system includes a standard DC resistivity meter and a set of 24 miniature electrodes with 3-cm spacing, which were connected to a switchbox allowing rapid measurements with different electrode combinations. The general setup of the experiments and a more detailed discussion of the results can be found in Hauck (2001). The material sample was contained in an 80 cm x 60 cm x 50 cm plastic box with a water outlet at the bottom and a punctuated tube across the bottom floor for water injection. Temperature was measured using UTL-1 mini-loggers (see e.g. Hoelzle etal. 1999). The box containing the sample was placed in a cold room, and measurements were recorded from the outside.

Figure 6.5 shows the resistivity-temperature curves for the saturated and dry Schilthorn material. Thereby, the terminus ''dry'' is used in a qualitative way, as the material was not completely dried to ensure electrical coupling with the electrodes. Two-dimensional tomographic interpretation of the measured resistivity values showed a heterogeneous distribution of the initial water content prior to freezing (not shown here). In Figure 6.5, mean resistivity values along the miniature electrode array were plotted versus the temperature. For temperatures above the freezing point, the small increase of resistivity with decreasing temperatures (Equation (6.3)) is clearly seen.

Figure 6.4 Measurement setup of the laboratory experiments with the miniature DC resistivity system

For temperatures below the freezing point, the resistivity increases exponentially with decreasing temperature. From these curves, the factor b in Equation (6.4) can be estimated for both cases. The exponential increase is larger in the saturated experiment than for the unsatu-rated material (b = 0.735°C-1 vs b = 0.273°C-1). This is in good agreement with previous studies, where the largest resistivity increase was found for samples with comparatively high water content (Olhoeft 1978, Seguin 1978).

Calculating the unfrozen water content S using Equation (6.5) and plotting resistivity p, unfrozen water content S and temperature T against time for the saturated Schilthorn experiment, the processes described qualitatively above are visualised (Figure 6.6). Thereby, the direction of the p-axis is reversed to facilitate the interpretation. As the temperature is approaching the freezing point, the resistivity is increasing slightly, due to a diminished mobility of the ions in the pore water. At the freezing point, the temperature curve flattens, as ice begins to form in the pore spaces. At the same time, the unfrozen water content starts to decrease. Because of the migration of the ions from the freezing phase to the still unfrozen parts of the sample, the freezing point is lowered and the temperature is still decreasing slightly. The resistivity is increasing only slowly, because the

Figure 6.5 Resistivity-temperature relationship determined in the laboratory for two different samples: (a) Schilthorn material saturated with water and (b) Schilthorn material in its initial ''dry'' state

Figure 6.6 Resistivity p (broken line), temperature T (dashed) and unfrozen water content S (solid) as a function of time for the laboratory experiment with the saturated Schilthorn material. Note that the p-axis is reversed

Time (hrs)

Figure 6.6 Resistivity p (broken line), temperature T (dashed) and unfrozen water content S (solid) as a function of time for the laboratory experiment with the saturated Schilthorn material. Note that the p-axis is reversed migration of the ions decreases the resistivity of the unfrozen water, which nearly cancels out the effect of the decrease of S. After more than half of the water is frozen, the temperature curve decreases faster (after ca 12 hours) accompanied by a fast increase in resistivity, as less and less unfrozen water is available for electric conduction. Finally, the temperature curve flattens again accompanied by a sharp bend in the curve for the unfrozen water content. As the resistivity is very sensitive to further reductions of the unfrozen pore water, it is still increasing at this point.

0 0

Post a comment