Currently, the use of either zinc galvanized or bare cold rolled steel
sheet is increasingly in demand in the painted state in order to incr
ease the corrosion resistance. The problem is that corrosion can howev
er develop along the diffusion paths that already exist in the paint r
esulting in reduced service life. In every case, corrosion products fo
rm at the metal/paint interface well before the visible appearance bec
omes apparent to the naked eye. With this perspective, it would be ext
remely useful to detect, as early as possible, the corrosion initiatio
n in order to enhance control. The technique of electrochemical impeda
nce is an appropriate method by which corrosion phenomena can be demon
strated under paint; however this technique does not provide localized
information. To overcome this problem, the new methods of infrared th
ermography and surface potential measurements are being developed. A v
ariation of the latter technique will be discussed. While all classica
l methods of dielectric characterization have been applied to the orga
nic paints under test, these were supplemented by surface potential me
asurements. This consisted of following the evolution of the external
charge density. This technique has generated a great deal of interest.
This is due to the simple observation that changes in charge applied
to a material over time, depend upon the physico-chemical characterist
ics of the material, and in particular its composition and structure.
In consequence, a metal/polymer system would not have the same dielect
ric behaviour if corrosion products were present at the interface. Thu
s, ii is possible to conduct surface potential mapping studies which d
istinguish between corroded and non corroded areas. It has been demons
trated furthermore that the presence of corrosion products manifests i
tself as either an area of charge stockage or as an area of charge dep
letion, depending upon the precise system. From the polarisation/depol
arisation equation of an ideal material, a simple model has been devel
oped in order to address these differences in behaviour. If the materi
al under test is considered as a composite (paint/pigment or paint/cor
rosion products), the models for mixed dielectric kinetics can be appl
ied to the resultant permittivity and resistivity to determine the beh
aviour of the integral. Depending on the metal/paint substrate, the be
haviour of the material is determined by either the resistivity (reduc
tion in dielectric properties, conductivity...), or by the permittivit
y (potential to stock charge) which translates into either an increase
or decrease in the material relaxation. This simple model thus allows
an interpretation of the different behaviour observed as a function o
f the type of paint. In the case of corroded systems, the paint can be
considered as the matrix of a composite and the corrosion products as
a second phase with variable volume fraction. In systems characterize
d by high dielectric properties corrosion products in macroscopic term
s give rise to a reduction in the material time constant. The opposite
is observed for materials with low dielectric properties. This model
explains why the presence of corrosion products translates into a zone
of charge retention or indeed charge depletion. The developed method
allows the nondestructive precise localization of corroded areas on pa
inted steel sheet as well as a differentiation between paints. With th
e use of the model system, the differences in behaviour of the differe
nt paints can be described in terms of their permittivity and resistiv
ity.