Developing a Fundamental Understanding of the Relationship Between Environment, Microstructure and Stress on Pre-cursors to Crack Initiation

Environment-Assisted Cracking is a critical environmental degradation mechanism in the oil and gas environment. The specific oil/gas environments result in unique characteristics of the environmental degradation process. This uniqueness is derived from the environment and the nature of films that form on the materials depending on the environments involved: (1) dominated by CO2 or (2) dominated by H2S and the reducing conditions that exist that result in metal dissolution and hydrogen related processes being dominant. FeCO3 (Siderite) is a conductor and FeS (Makinawite) is an insulator. In both cases hydrogen will be present. Additionally, sulphur is a hydrogen recombination poison which will increase the probability of hydrogen entry into the material. The two films are protective if un-breached and at least partially retard the entry of hydrogen. However, if the film is breached the environment gains access to the material.In the case of carbon steel, conditions are often achieved that promote pitting. Metal dissolution and hydrolysis serve to both remove material and generate hydrogen ions, in this respect, the role of metal dissolution and potential hydrogen uptake within the material is critical.In addition to the time dependence of the initiation and propagation process, the morphology of the transition from pit to crack has proven to be very complicated.The above that an understanding of the pit-to-crack initiation process requires the simultaneous application of high resolution detection, analysis and modelling techniques. Fortunately, such techniques are now becoming available. The application of these techniques will be discussed further below. In this project we propose to bring to bear state of the art analytical techniques for the detection, analysis and modelling of pit initiation and pit-to-crack initiation process. The research is being carried out in two phases: Phase I: Assessing pre-cursors to pit initiation and the pit-crack transition. Phase II: Development of Hybrid modelling methods for predicting early stage damage. Phase I: Understanding Material Behaviour -- Critical to the success of the project will be the development of a detailed understanding of the pitting/crack initiation process. To this end high resolution experiments will be performed to study the initiation and propagation of pits as associated crack initiation from pits. Pitting, short crack growth and crack link up studies will be performed. Analysis of pits/cracks will be conducted to develop an understanding of the relationship between the environment, microstructure and mechanical parameters. The pitting and crack initiation process will be evaluated in aqueous environments containing CO2 and CO2/H2S containing environments over the temperature range 25-150C using high resolution Multi-Frequency Alternating Current Potential Drop (MFACPD) techniques that have been developed in the MIT Uhlig Corrosion Laboratory. Digital Image Correlation will be used to determine displacements (strains) around individual and multiple pit sites, the latter representing ‘multi-site damage’. Pits/cracks will then be analyzed using multiple high resolution analytical techniques the goals of which will be to relate local microstructure and strains/stresses to the initiation process. These results will then be used in a modeling effort in the overall process modeling effort.Phase II: Hybrid Modelling -- Historically models for quantitatively predicting environmental-assisted crack growth have been based upon the two primary corrosion mechanisms of anodic dissolution and hydrogen embrittlement. In the case of material /environment systems that give rise to stress corrosion cracking susceptibility, a superposition model is often invoked to account for ‘Stress Corrosion Fatigue’ (SCF). The majority of these models have been developed based upon structures or components containing pre-existing defects and therefore address the ‘long-crack’ propagation regime. However, more recently studies concerned with the prediction of lifetime for structures/components nominally ‘free’ from defects has received increasing attention. These studies have shown that some 70-90% of lifetime may be taken up in growing defects of a size less than 0.5 mm, i.e., in the short-crack growth regime. Many of the models adopted to predict EAC, in particular pitting corrosion fatigue models, invoke a Linear Elastic Fracture Mechanics (LEFM) approach, whereby a pit is considered as a notch and the initiation of a crack from such a defect is deemed to occur when the Stress Intensity Factor (delta-K) is greater than the Threshold Stress Intensity Factor (delta-KTH). This approach has many limitations, notwithstanding the fact that LEFM does not apply to small defects. To overcome this problem we propose to develop a model that applies over several length scales from the microstructural region to the continuum region. The modelling approach that we propose to use is one of the ‘site bond’ methods in which assessment of the state of damage, cracking, or corrosion of a material at a microstructural scale is dealt with separately, from the continuum response of the engineering structure.