Overall Mechanisms of High pH and Near-Neutral pH SCC, Models for Forecasting SCC Susceptible Locations, and Simple Algorithms for Predicting High pH SCC Crack Growth Rates

The current practice of determining pipeline stress corrosion cracking (SCC) susceptible locations, for pipe segments where In-Line Inspection (ILI) tools or hydrotests are not applicable, is based on known characteristics of past SCC sites. These characteristics include the type, age, and conditions of the coating; the type of steel and weld; the surface treatment, age, manufacturer, corrosion status and past SCC experience of the pipe; the type of soil, wet/dry cycles, and ground water chemistry including pH, ionic species such as Na⁺, Ca²⁺ and Cl⁻, molecular species such as CO₂ and O₂, and bacteria types and number; the performance of cathodic protection (CP); the operating fluid temperature, pressure and loading conditions, etc. As the range of SCC conditions continues to expand with time due to new SCC incidents, the number of SCC characteristic variables will continue to grow. These variables and their complex and often unknown interactions should, as much as possible, be accounted for when SCC susceptible sites are determined. These sites can either be locations where the ILI data cannot provide distinctive determination of whether the defects are actually cracks due to resolution limitations, or locations where ILI tools are not applicable while complying with federal regulations, verification of no SCC threat in those sites, particularly in high consequence areas, must be performed through the use of direct assessment (DA) methodology or excavations. Due to the lack of a clear understanding of the roles of the above variables and their interactions in SCC nucleation and propagation, the current practice often relies on empirical or expert judgment of how the above numerous variables and their interactions relate to SCC. Although useful, such an approach can lead to either crude or non-reliable determination of SCC sites due to the complexity of the cracking process,

An alternative approach would be to use a modeling tool, which would incorporate many, if not all, of the above variables and their interactions in a comprehensive package, either as a software code, or preferably, in some simplified mathematical formula or spreadsheet, that would allow for the judgment of SCC susceptible sites based on a single or no more than a few comprehensive outputs (from the tool). Soil models are an example of such an attempt. Although proven to be useful in the field for near-neutral pH SCC, such a model is limited to predict bulk soil chemistry while it is the chemistry and potential local in coating-disbonded regions that are most relevant to SCC susceptibility, initiation and propagation. For that reason, laboratory and field tests use field-sampled or -simulated chemistry in coating-disbonded regions near cracks to study SCC characteristics, such as cracking potential and/or crack growth rates (CGRs). Since both soil models and the current SCC DA methodology rely on conditions in bulk soils, clearly, there is a gap between the SCC sites determined from the current practice and the real SCC sites.

This gap must be filled in order to achieve a more reliable prediction of SCC locations. A project for developing a comprehensive predictive model to fill that gap is still on-going. Built on solid fundamental principles, this model would permit chemistry and potential in the coating-disbonded region to be predicted through the use of known or measurable bulk soil chemistry and potential near the holiday. By comparison to SCC characteristics known from previous lab or field studies or from field failure analyses or experience, a more reliable determination of SCC susceptible sites can be realized. Since some of the modeling framework and results were reported elsewhere^[1-2] and new modeling results are yet to be summarized, the concern of this work is to: (1) provide a comprehensive and new understanding of the fundamentals behind the process of both high pH SCC and near-neutral pH SCC, (2) discuss example results obtained from a recent model that is capable of predicting certain conditions in a coating-disbonded region relevant to SCC, and (3) derive simple algorithms for predicting high pH SCC growth rates.