Induced polarization of volcanic rocks. 3. Imaging clay cap properties in geothermal fields

2019 
Smectite-rich clay caps form permeability seals in geothermal systems. The presence of smec-tite is also responsible for a strong surface (interfacial) electrical conductivity and polarization due to their electrical double layer properties. We developed new complex conductivity models using both differential effective medium (DEM) and volume averaging theories accounting for both conduction and polarization of these high cation exchange capacity (CEC) materials. These models predict that the chargeability is also a non-linear function of the pore water conductivity reaching a constant value at pore water conductivity far above the so-called iso-conductivity point. The iso-conductivity point is characterized by the equality between the conductivity of the rock and the conductivity of the pore water. We apply the DEM conductivity model (which requires only two textural parameters) to smectite-rich volcanic and sedimentary rocks using data sets from the literature. When smectite is present in the volcanic rocks, the CEC of the rock is dominated by the CEC of smectite. The grain conductivity and the normalized chargeability are related to each other by a dimensionless number R = 0.10 (independent of temperature and saturation) and both are controlled by the excess of charge per unit pore volume Q V , which can be determined from the CEC and porosity. Our petro-physical model is also able to predict the permeability of the rock as well from the CEC and the porosity. It is applied to a 3-D data set at Krafla volcano (Iceland). The porosity, the CEC, the percentage of smectite, and the permeability of the clay-cap are imaged by 3-D induced polarization tomography. Electrical conductivity tomography alone does not allow separation of the contribution of the bulk pore space from the interfacial properties related to alteration and therefore should be used with caution.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    76
    References
    14
    Citations
    NaN
    KQI
    []