Measuring electrical properties of mortar and concrete samples using the spectral induced polarization method: laboratory set-up

Abstract

Abstract Electrical impedance spectroscopy (EIS), also known as Spectral Induced Polarization (SIP) has proven to be a successful non-destructive technique able to characterize the chemical and physical properties of a complex structure (concrete, rocks, and soil) under various environmental conditions. Because this technique aims to measure small amplitude signals, it can easily be influenced by parasitic effects which are not representative of the intrinsic properties of the investigated material. The purpose of this paper is to validate the resolution of our experimental methodology as a preliminary step to ensure an accurate measurement of the bulk complex resistivity response of concrete samples within a wide frequency range (1.43 mHz–20 kHz). A cement mortar sample is first used as a homogeneous (isotropic) material to improve the sample holder design for reducing the errors and controlling the external parameters (such as coupling effect, geometric factor, contact impedance, electrode polarization) during the laboratory measurements. Also, the performance and installation of the measuring components such as the electrodes (current and potential) and the sample holder design are assessed. The specimen dimensions and designs are selected based on a sample standard size that will be used to determine the physico-mechanical properties of concrete (compressive strength, modulus of elasticity, ultrasonic pulse velocity, length variation). The examinations are performed in parallel with the SIP measurement on the same sample. The accuracy of our measurement setup is then validated using aluminum-bronze plates bonded to the sample using a conductive gel to transmit electric current with a density less than 10 - 3 A m−2. The results showed that by controlling the sponge moisture in the Ag/AgCl non-polarizable voltage electrodes, our SIP measurement system is able to measure the phase lag with an error smaller than 1 mrad over a frequency range from 1.43 mHz to 20 kHz.