An Experimental Approach for Delayed Stress Corrosion

Failures of structural and mechanical components have long been attributed to environmentally assisted cracking (EAC). The umbrella of EAC encompasses several phenomena, including stress corrosion cracking (SCC), corrosion fatigue (CF), hydrogen embrittlement (HE) and liquid metal embrittlement (LME). The latter, LME, has resulted in the failure of components in petrochemical and aeronautical industries, among others. The effects are detrimental, with crack tip velocities on the order of centimeters per second and failures occurring rapidly. Previous research has provided numerous underlying microstructural failure mechanisms aimed at identifying the true failure mode. Conflicting experimental data has extended the debate over the true mechanism promoting renewed interest in novel experimental regimes. Utilizing fracture mechanics specimens, the solid-liquid Al-Hg couple was analyzed to extend or reject current theories. Through the implementation of an original environmental chamber capable of testing notched and pre-cracked components in corrosive environments, C(T) specimens were subjected to experiments submersed in liquid mercury. Upon the application of an initially applied stress intensity factor (under load-control), incubation periods preceding failure were observed. Crack initiation and propagation were observed to occur along the starter notch, as well as other regions on the specimen. Results provided evidence that additional factors, such as a critical load or critical microstructural orientation, were factors in crack initiation and propagation. In the quest to observe the influence of these additional factors, a variation of the experimental setup was implemented and initial tests have begun. INTRODUCTION Environmentally assisted cracking (EAC) is a widely known phenomenon that has been extensively researched throughout the 1900’s [1,2]. An encompassing term, EAC includes hydrogen embrittlement (HE), corrosion fatigue (CF), stress corrosion cracking (SCC) and liquid metal embrittlement (LME). Each subset of EAC has been observed to cause failures in structural members of mechanical designs, sometimes with catastrophic results [1,2]. While some modes of EAC are not as common as others, e.g. LME as opposed to SCC or HE, the result is still the same; failure more rapidly than anticipated. Stress corrosion cracking became a mainstream issue during the World War II era, in which several Liberty ships experienced through hull failures. It was determined that a combination of the new weld techniques coupled with the cold saltwater environment and microstructure of the hull material, cracks were able to originate and propagate, resulting in the highly publicized rupture of the hulls. Notable consideration was given to the crack velocities, as failures were described as instantaneous. During this time, fracture mechanics was developed to help explain such failures [3]. Fracture mechanics is largely based upon the development of mathematical models of Inglis and Griffith [4]. Inglis implemented the theory of elasticity to solve the problem of an elliptical hole, while Griffith used energy methods to quantify brittle fractures. Taking this theory, Irwin was able to successfully apply it to ductile materials, as well as define the stress intensity factor, SIF or K [4]. In the presence of a crack, K defines the stress field surrounding the defect. Under plane-strain conditions, K is considered a material constant. Upon reaching a critical value known as the plane-strain fracture toughness, KIc, cracks can be expected to grow and propagate. The speed at which they propagate is largely affected by the surrounding environment, making it an extremely useful tool in the study of EAC. Studies aimed at quantifying the effects of SCC have implemented blunt notch specimens, as well as fracture mechanics pre-cracked specimens. Upon placing smooth tensile specimens in 3.5% NaCl, Gordon et al. observed crack initiation resistance was lower than that of specimens in air [5]. Similarly, by introducing specimens to a humid environment, Spiedel demonstrated the effect that moisture had on crack velocities [6]. By introducing water, crack tip velocities were observed to increase as the level or moisture in the environment increased, Fig. 1. From a completely dry environment to a 1 Copyright © 2010 by ASME Proceedings of the ASME 2010 Pressure Vessels & Piping Division / K-PVP Conference PVP2010 July 18-22, 2010, Bellevue, Washington, USA