the coupled environment models for localized corrosions; crevice corrosion and stress corrosion cracking
Open Access
- Author:
- Lee, Sang-kwon
- Graduate Program:
- Materials Science and Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 20, 2013
- Committee Members:
- Digby D Macdonald, Dissertation Advisor/Co-Advisor
Digby D Macdonald, Committee Chair/Co-Chair
Kwadwo Osseo Asare, Committee Member
James Hansell Adair, Committee Member
Mirna Urquidi Macdonald, Committee Member - Keywords:
- the coupled environment fracture model
localized corrosion
crevice corrosion
stress corrosion cracking
coupling current - Abstract:
- This dissertation details my investigation of crevice corrosion and stress corrosion cracking based upon the differential aeration hypothesis, currently considered the physical basis of virtually all localized corrosion phenomena. The differential aeration hypothesis attributes localized corrosion to a spatial separation of local anodes and local cathodes. The former occurs in the region of a given system (e.g. within a crevice or crack) that has the least access to a cathodic depolarizer (e.g. O2). The latter occurs in the region (e.g. on an external surface) with the greatest access to a cathodic depolarizer. This hypothesis, combined with the natural law of charge conservation, yields the measurements of the electron coupling current that flows from local anodes to local cathodes, which contains valuable information concerning the processes that occur within local anodes. A simple crevice corrosion monitor was developed to monitor crevice corrosion in 1018 mild steel, Type 304 stainless steel, and Type 410 stainless steel in deionized water and in NaCl solutions with and without the addition of a chemical corrosion inhibitor. The monitor, which measures the electron coupling current that flows from the crevice to the external surface, followed the evolution of crevice activity in a manner that can be understood in terms of the cathodic process that occurs on the external surface and the partial anodic process that develops within the crevice, due to the accumulation of H+ and Cl-. The crevice initiation time is typically very short, but varies depending upon chloride concentration and, possibly, inhibitor concentration. After initiation, the coupling current increases over time, passing through a maximum, then decreasing and eventually changing sign, from positive to negative, which indicates crevice inversion. This inversion is attributable to the gradual build-up of H+ within the crevice to the extent that proton reduction within the crevice becomes the principal cathodic reaction in the system, while the anodic reaction moves to the external surface. In addition, amines are effective corrosion inhibitors of crevice corrosion of mild steel and stainless steels in NaCl solutions by forming a protective inhibitor film, thereby inhibiting the cathodic reaction occurring on the external surface, provided that they are present in sufficiently high concentrations. The shape evolution of sensitized Type 304 stainless steel surface cracks in boiling water reactor primary coolant circuit piping at 288 °C was explored as a function of environmental variables—such as electrochemical potential, solution conductivity, flow velocity, and multiplier of standard exchange current density for O2 reduction—using the coupled environment fracture model (CEFM). For more accurate prediction of crack growth rate, Shoji’s approach for calculating crack tip strain rate, together with a precise treatment of the stress intensity factor for semi-elliptical surface cracks, has been integrated into the CEFM. This revised CEFM accurately predicted the dependence of crack growth rates on the stress intensity factor and offers an alternative explanation for the development of semi-elliptical cracks to those suggested by fracture mechanics alone. Moreover, the CEFM predicted that the minor axis of the ellipse should be oriented perpendicular to the surface, in agreement with observation. The development of the observed semi-elliptical cracks with the minor axis perpendicular to the surface is therefore attributed to the dependence of the crack growth rate on the electrochemical crack length. The CEFM, which has been used extensively to predict intergranular stress corrosion cracking in sensitized austenitic stainless steel components in the heat transport circuits of nuclear power reactors, has been modified and calibrated to predict crack growth rates in Al-Mg alloys in marine environments, and Alloy 22 in saturated NaCl solutions. Calibration involves optimization of the CEFM on measured crack growth rate data of the materials of interest and, after extracting values for essential parameters, calculation of crack growth rate data as a function of independent variables of interest. The customized CEFM provided quantitative predictions of the effects of O2 concentration, electrochemical potential, stress intensity factor, flow velocity, multiplier of the standard exchange current density for O2 reduction, and conductivity on crack growth rates in lightly sensitized AA5083-H321 in 3.5 wt.% NaCl solutions. The importance of external environment properties—such as conductivity, oxidant/reductant concentration, and the kinetics of cathodic reactions on the surfaces external to the crack—has been confirmed. Crack growth is attributable to a sequence of micro-fracture events at the crack front. Micro-fracture frequency is determined by the mechanical conditions that exist at the crack tip and are governed by the stress intensity. Micro-fracture dimension is determined by hydrogen-induced fracture. The crack growth rate is the product of these two quantities. Another area of exploration is the crack shape evolution of surface stress corrosion cracks in lightly sensitized AA5083-H321 in 3.5 wt.% NaCl solutions. Evolution of semi-elliptical surface cracks is attributable to the dependence of the crack growth rate on the electrochemical crack length, rather than on variation in the local stress intensity along the crack front. Accordingly, the evolution of the semi-elliptical shape depends upon environmental variables—such as electrochemical potential, solution conductivity, flow velocity, and multiplier of standard exchange current density for O2 reaction. The CEFM for lightly sensitized AA5083-H321 also predicts that the local stress intensity along the crack front, assuming constant load conditions, has an impact on the crack shape evolution after a certain time, due to the large differences in stress intensities at the edge and the center of surface cracks. Nevertheless, the CEFM predicts that all the aspect ratios at failure examined in this study fall within the range of 0.7 to 0.9, which indicates a crack that is semi-elliptical in shape, with the long axis parallel to the surface. Theoretical aspects of the stress corrosion cracking of Alloy 22 in contact with a saturated NaCl solution was explored in terms of the CEFM, which was calibrated based upon available experimental data. Crack growth rate was predicted as a function of stress intensity, electrochemical potential, temperature, and electrochemical crack length. The CEFM for Alloy 22 predicts that the crack growth rate increases with temperature, passing through a maximum at about 80 °C, and that its electrochemical potential increases with temperature. Based upon the dependence of the crack growth rate on the electrochemical crack length, the evolution of a semi-elliptical surface crack in a planar surface under constant loading conditions over the course of one million years was demonstrated. One million years is the service horizon for some high level nuclear waste repositories of the Yucca Mountain type. The CEFM suggests that stress corrosion cracking is possible in Yucca Mountain-type repositories, provided that the crack can nucleate. It predicts that a crack could penetrate the 2-cm thick outer shell of the waste package in a time of 1057 years at the selected temperature (110 °C), suggesting that through-wall cracking in the outer shell of the waste package is dominated by the crack initiation. The CEFM’s success in explaining the stress corrosion cracking of aluminum alloys and Alloy 22, supports the notion that internal and external environments are strongly coupled. Moreover, the CEFM, which was originally developed to describe intergranular stress corrosion cracking in sensitized stainless steels is equally applicable for describing the stress corrosion cracking in lightly sensitized aluminum alloys and Alloy 22.