In this series, an epoxy varnish for the corrosion protection of carbon steel was analysed in the dry state by broadband dielectric spectroscopy (part 1) to describe the molecular mobility of the epoxy network. Electrochemical impedance spectroscopy (EIS) measurements were then performed in a NaCl solution (part 2), with the intent to detect the signature of the molecular mobility in the wet state. The dielectric manifestation of the glass transition (a mode), previously characterized in part 1 of this series for the dry varnish, was evidenced from EIS measurements through the use of the dielectric permittivity formalism. This a mode showed the characteristic Vogel Fulcher Tamman dependence, shifted towards higher frequencies when compared to that of the dry varnish, consistently with the plasticization of the epoxy network due to the water uptake. Moreover, it was shown that taking the a mode into account when fitting the EIS impedance data with equivalent circuits led to a much better fit in the capacitive resistive transition region. It was shown that the dc charge transport processes are not only triggered but also governed by the molecular mobility of the epoxy network. In the case of thick coatings (several hundreds of micrometers), this dependence demonstrated the absence of through pores, as it is often discussed in the literature. Indeed, under the assumption of current flowing through free electrolyte via open pores, the temperature dependence of the electrolyte's dc conductivity (Arrhenius law) should be found instead of the temperature dependence of molecular mobility.

In the present work, a commercial high solids epoxy-based paint used for the corrosion protection of steel was characterized by electrochemical impedance spectroscopy (EIS). The aim was to analyze the influence of water absorption on the physical structure of the paint, through the impedance data analysis. First, the water uptake was followed by gravimetric measurements on free-standing films as a function of the exposure time to a 0.5 M NaCl solution. Gravimetric measurements indicated a mass loss, linked to a release of a plasticizer from the formulation. This compound was dosed in the NaCl solution. The water volume fraction was also determined from the high-frequency part of the impedance diagrams, employing a linear rule of mixtures. The water fraction values obtained from the two methods were in acceptable agreement. The glass transition temperature (T g ) was determined by differential scanning calorimetry for the dry coating as well as for the coating after different immersion times in the NaCl solution. The use of the dielectric permittivity formalism allowed the mean relaxation times to be obtained from the impedance data analysis. The real part of the electrical conductivity was also considered. The variation of these parameters values as a function of the exposure time to the NaCl solution, was correlated with the T g variation. They accounted for plasticization at the beginning of immersion (water uptake) and antiplasticization for longer exposure times (plasticizer release). For the studied system, it was possible to extract parameters representative of the molecular mobility from the impedance spectra analysis and to link them to the modification of the coating's physical structure.