The influence on sample preparation on spectral induced polarization of unconsolidated sediments

Spectral induced polarization (SIP) measurements in the laboratory are in many cases intended to provide representative and comparable results of complex electrical conductivity. This is not invariably the case when using unconsolidated sediments, as the sample preparation influences several SIP-relevant properties of the samples, including the pore geometry. The pore space is supposed to control the polarization effect and therefore a change in the pore space will change the measured parameters. We analysed the influence of the sample preparation on SIP measurements by testing various filling methods, each defined by a sequence of particular steps, with regard to the reproducibility of the spectra. The measurements were performed on three different sample materials. Variations of the spectra due to different filling methods were obtained, indicating the importance of considering the sample preparation. Methods that improved the reproducibility compared to loose packing of the samples were found, but the most suitable packing procedure depended on the material properties. The mean relaxation times and normalized chargeabilities were obtained from a Debye decomposition. Although unidentified processes caused scattering of the measured parameters, a relation of the IP-parameters to the porosities was present. A decrease of the porosity reduced both the relaxation times and the normalized chargeabilities.