PSI - Issue 37

R.J. Mostert et al. / Procedia Structural Integrity 37 (2022) 763–770

764

2

Mostert et al/ Structural Integrity Procedia 00 (2021) 000 – 000

1. Introduction High Temperature Hydrogen Attack (HTHA) is the progressive degradation of process equipment fabricated from carbon and low-alloy steels at elevated temperatures (at least 204 °C ), when exposed to hydrogen-rich environments such as that found in petrochemical and refining industries (Askari and Das 2006; Ali et al. 2020; Defteraios et al. 2020; Manna et al. 2007) The HTHA occurs when hydrogen molecules dissociate to hydrogen atoms and diffuse into the vessel wall. The hydrogen atoms react with carbon in solution in the steel to form methane gas molecules thereby compromising the structural integrity of the component through the formation of decarburized zones, internal voids, microcracks and fissures (Manna et al. 2007; Wahab, Saba, and Raman 2004; Munsterman, Seijas, and Williams 2010). The methane molecules are trapped and accumulate over time within cavities on the grain boundaries. The gas pressure builds up within these cavities, inducing additional stress to the surrounding material. The cavities grow over time and eventually coalesce to form microcracks and fissures which, if left unchecked, can lead to unexpected and catastrophic (Askari and Das 2006; Ali et al. 2020; Manna et al. 2007; Wahab, Saba, and Raman 2004) The so-called Nelson Curves, which were first published in 1949, have been widely used in mitigating HTHA damage through selection of appropriate steel alloys for specific operational conditions (temperature and hydrogen pressure (Manna et al. 2007; Saba 2003) The Nelson Curves are limit lines, independent of time, based on plant experience, and used for steel selection during design. Due to ongoing changes in plant experience regarding carbon and carbon – 0.5 % molybdenum steels in HTHA service, the corresponding operational envelopes for vessels fabricated from these steels have changed over time. The current edition of API 941, for example, does not have a limit curve for C-0.5 Mo steels (American Petroleum Institute, Division of Refining 2016a). The assessment of the structural integrity of older vessels in hydrogen service is therefore of critical importance, especial ly in the case of C- and C-0.5 Mo vessels operating with conditions outside of the currently recommended envelopes. Current editions of the leading standard regarding structural integrity assessment, API 579-1/ ASME FFS-1 however do not have an assessment methodology regarding vessels with HTHA damage (American Petroleum Institute, Division of Refining 2016b) . The Welding Research Council (WRC) have therefore coordinated research efforts with a view to the publication of methodologies quantifying HTHA damage development and structural integrity assessment (Welding Research Council 2021a, 2021b) The WRC model regarding the development of HTHA damage is referred to as the “α - Ω HTHA model” and the “Ω” in the title refers to the MPC Omega creep model, which forms the basis of the creep damage life assessments in part 10 of API 579. The α Ω HTHA model facilitates the calculation of “time - dependent Nelson Curves” , called the “ Prager Curves ” and is a breakthrough in the structural integrity assessment of vessels in HTHA service. The model regards the HTHA damage process as a methane enhanced creep process. Central to the calculations and model is the “Methane Factor α”, which is a material -dependent parameter which is composition-dependent and probably influenced by metallurgical variables such as microstructure. To date, the parameter has been estimated using graphical interpolation by comparing the creep curves in inert and hydrogen environments with similar rupture times. The additional HTHA stress is first obtained as the difference between the applied stresses in the two environments and the additional HTHA stress is then divided by the equilibrium methane pressure at the test conditions, to yield the value of the Methane Factor. The values published for α in WRC 585 and 586 are regarded as “upper bound” values and have been established using fairly elaborate test methods which allow creep testing to be performed in pressurized hydrogen environments. Consequently, the available literature regarding values for α is fairly limited. Information regarding the variation of α -values through a welded joint is, for example, not presented in WRC 585 and 586. More research regarding the influence of metallurgical param eters on α values will clearly be beneficial and will lead to more accurate and less conservative assessments. The first objective of this paper is therefore to develop and describe an alternative and relatively simple methodology to experimentally determi ne α -values. The HTHA damage process is similar to the creep damage process, but more severe due to an additional driving force from the internal pressure of the methane inside the cavities. It is therefore regarded as a methane-enhanced creep process. Methane pressure is the root cause of the time dependent progressive damage mechanism of HTHA and the driving force for cavity growth. The methane pressure inside the cavities is calculated from the hydrogen pressure and temperature using time dependent chemical reaction equations for methane formation that involve both forward and reverse reaction rates. The value of carbon activity indicates the stability of the least stable carbides, a measure of how much carbon in the metal solution is available to react with hydrogen to form methane. Carbides only react with hydrogen when dissolved, therefore, alloys with stable carbides have lower carbon activity. Fabrication processes such as welding as well as heat treatment processes such as PWHT (Post Welding Heat Treatment) and tempering also decrease the carbon activity. It can therefore be expected that the different zones of a welded joint will give different responses to HTHA attack. A need clearly exists to accurately track and monitor the HTHA damage evolution process in various zones of structures. The second aim of this paper is therefore to develop constitutive equations for the kinetics of HTHA to aid in the monitoring of structural integrity. It will be shown how these equations, together with the application of high-temperature strain gauging, can be used to evaluate the progress of HTHA damage and remaining life in service. 2. Materials used and experimental procedure Specimens representing the fine-grained base metal (BM), the coarse grained weld-metal (WM) and the (HAZ) regions of a welded joint of a 69 mm thick ASME SA 302 Gr C plate (refer to Table 1 for chemistry), were removed by EDM. The specimens were polished to a 1-micron finish and high-temperature capsule-type strain gauges were attached to the specimens with