PSI - Issue 68

Robert Sundström et al. / Procedia Structural Integrity 68 (2025) 1081–1090 Robert Sundström / Structural Integrity Procedia 00 (2025) 000–000

1083

3

considerations (Jürgensen and Pohl 2023). Moreover, it seems uncertain that the current number of testing facilities can meet the demand for testing. This has led to interest in an alternative method where the pressurised hydrogen gas is contained within a hollow in the specimen itself while being subjected to a mechanical test. This hollow specimen method is described in this paper. Based on an extensive review of the literature, the history of this method, its use in other research fields, hole manufacturing methods and their effects on surface characteristics, gas purity and comparisons to solid specimens are discussed to provide a view of the benefits, drawbacks, status and future developments of this testing method. 2. Mechanical testing in gaseous hydrogen 2.1. History Hydrogen embrittlement depends on many parameters. Among them are temperature, pressure, strain rate, loading frequency, type of embrittlement (HEE or IHE) and hydrogen purity (Nibur and Somerday 2012, Lee 2016). These testing parameters determine the difficulty, cost and feasibility of the testing method that needs to be used. An aqueous electrochemical or gaseous thermal charging process can be used to mimic IHE conditions, while an active gas environment is required to mimic HEE conditions. Assessing mechanical properties of metallic materials in pressurized gaseous hydrogen environments requires mechanical testing to be performed in pressure chambers. This type of testing can be technically difficult (Ogata and Ono 2019), costly (Vesely et al. 2002) and involves safety issues with high pressure and high volumes of hydrogen (Michler et al. 2022). A mechanical test in a gaseous hydrogen environment might cost tens or hundreds of times more than a conventional test under ambient conditions (Vesely et al. 2002) and fewer tests can be made in a given time span (Jürgensen and Pohl 2023). Technical difficulties include load transmission, contamination of high-purity hydrogen gas and heating and cooling of specimen (Chandler et al. 1974, Liu et al. 2015). Testing in strain control involving compressive loads is also difficult (Bradley et al. 2023). While high and low temperatures can be achieved in these chambers, they are usually limited. These various difficulties prohibit the rapid evaluation of metallic materials for use in gaseous hydrogen applications. These difficulties were recognized decades ago (Chandler et al. 1974) so an alternative technique was developed: a test specimen with a hollow containing high-pressure gaseous hydrogen (Chandler et al. 1974), called the hollow or tubular specimen. This type of specimen does not require a pressure chamber so the volume of hydrogen used can be as little as a few millilitre (Boot et al. 2021a) which increases test safety and reduces costs (Chandler et al. 1974). Load transmission is no different than when using a conventional solid specimen in lab air, as there is no pressurised chamber with seals that impart frictional forces on the load train. Use of extensometers and heating or cooling of this specimen is easier. This makes it possible to reach up to 1000 °C in a resistance furnace or induction coil, or –269 °C using a liquid helium cooling bath, as can be done for a conventional solid specimen. However, the inner surface is difficult to access which makes it impossible to perform testing with notches, pre-cracks or coatings (Chandler et al. 1974). Conventional, solid specimens for tensile testing, fatigue, creep or fracture toughness became the norm for mechanical testing in gaseous hydrogen environments and most of the mechanical testing in gaseous hydrogen has been done using these conventional specimen geometries. Hence, there is limited historical data published using hollow specimens with gaseous hydrogen environments. Oxygen inhibits embrittlement and for testing in autoclaves acceptable oxygen levels are less than one ppm by volume (Nibur and Somerday 2012). The hollow specimen method is no less sensitive to oxygen contamination than the autoclave. In fact, some have argued that it might be more sensitive to oxygen contamination because the gas volume is larger in an autoclave than in a hollow specimen in relation to the exposed surface area (Freitas et al. 2024). A lesser absolute amount of oxygen would then have a larger effect on embrittlement in a hollow specimen than for a solid specimen in an autoclave. It has been shown that the ductility of pipeline steel is affected by estimated oxygen levels above two ppm when using a hollow specimen (Michler et al. 2022). The effect was seen by variation of the number of purging cycles prior to slow strain rate (SSR) testing and the oxygen content was calculated using an equation. The actual composition of the gas inside the hollow is too small in volume to be measured unless hydrogen gas is actively flowing through the test piece during measurement. In such a setup, the concentration of oxygen and water vapour could be measured with a suitable oxygen analyser and dew point meter, as is commonly done for testing in autoclaves. Such measurements have not been demonstrated in the published literature for the hollow specimen.