Corrosion of carbon steel casing foreseen in the high level radioactive waste facility monitored by acoustic and electrochemical techniques
Abstract
The French national radioactive waste management agency (Andra) is designing an industrial centre
for geological disposal (CIGEO1) of Intermediate Level Long-Lived radioactive waste (ILL-LW) and High
Level Waste (HLW). The facility is based on a multi-barrier waste confinement concept which must be
reversible and monitored for at least 100 years. First, vitrified waste are introduced in a stainless steel
container and set into an overpack made of low alloyed steel. Then, the overpacks are introduced into
micro-tunnels drilled in the claystone and cased with a micro-alloyed steel pipe (API 5L X65 grade). The
pipe is in direct contact with a cementitious bentonite grout material. This material is expected to
neutralize the potential acidity resulting from the excavation of the micro-tunnel and the progressive
resaturation of claystone.
The present work takes part of the SCCoDRa project (French acronym for “Monitoring of metallic
components corrosion for radioactive waste disposal). This four-year industrial project is partly
devoted to the improvement of corrosion monitoring methods applied to corrosion processes
foreseen in the High Level Waste facility.
Many parameters of the environment will evolve during the operating phase: aerated conditions
followed by a progressive consumption of oxygen by various processes, temperature, pH, water
resaturation of the host rock…, are expected. So during this period, different corrosion modes may
occur like uniform corrosion, crevice corrosion and smaller corrosion attacks. The present work focuses
on the capacity of Acoustic Emission (AE) and Electrochemical Noise (EN) techniques to detect and to
distinguish these two corrosion modes.
Acoustic Emission is known to be a sensitive technique to corrosion elementary processes such as
hydrogen evolution, thick corrosion products breakdown, plastic deformation, cracking… [1] Yet, the
technique presents some limitations: dissolution processes are not energetic enough to be detected
by piezoelectric sensors, and only evolution of corrosion processes can be detected; no information
can be obtained about the geometry of the damage by AE. Moreover, AE is sensitive towards
elementary mechanisms that can be common to several corrosion modes (e.g. gas evolution), thus it
cannot allow identifying the corrosion modes that occurs. Finally, it is strongly dependant on the
acquisition chain: sensor, amplification, acquisition parameters. On the other hand, EN features
observation (both temporal and frequential) can allow identifying corrosion mode, its kinetics and
transitions from one to another [2].
In this context, the aim of the present work is to evaluate potentialities of both techniques, individually
and coupled, to detect, identify and quantify corrosion damage of carbon steel in the context of
nuclear waste storage. Experimental devices and procedures are developed to control corrosion
processes, in terms of geometry and rate, while being representative of actual damage conditions. The
variation of parameters related to the medium (temperature, pH, nature of the species) allows this
control. The variations in the parameters of the resulting EA and EN signals are analyzed and correlated
with observations of damage evolution.
References
[1] S.Yuyama, “Fundamental Aspects of Acoustic Emission Applications to the Problems Caused by
Corrosion”, presented at the Corrosion Monitoring in Industrial Plants Using Nondestructive Testing
and Electrochemicals methods, Phiadelphia, 1986, vol. ASTM STP 908, pp. 43-74.
[2] A.M Homborg, T. Tinga, E.P.M van Westing, X. Zhang, G.M Ferrari, J.H.W de Wit, J.M.C. Mol, “A
Critical Appraisal of the Interpretation of Electrochemical Noise for Corrosion Studies”, Corrosion The
Journal of Science and Engineering, 2014, vol. 70, No. 10, pp. 971-987.