PSI - Issue 42

Margo Cauwels et al. / Procedia Structural Integrity 42 (2022) 977–984 Author name / Structural Integrity Procedia 00 (2019) 000–000

978

2

Keywords: Hydrogen embrittlement; Pipeline steels; Charpy V-notch; Impact toughness; Fractography

1. Introduction Hydrogen gas will play an important role in the storage and transport of energy (produced by renewable sources or in conjunction with carbon capture and storage). From an economic standpoint, existing pipeline systems are an attractive infrastructure for transporting and distributing hydrogen gas. However, hydrogen entry into pipeline steels can cause a degradation in mechanical properties, generally referred to as hydrogen embrittlement (HE). Since present pipelines were not designed for high-pressure hydrogen transport, their suitability must be checked (Laureys et al. (2022)). Charpy impact testing is easy, cheap, and widely used to qualify the toughness of pipeline steels and their welds. This makes the test an attractive option for the qualification of hydrogen embrittlement or the impact toughness of a material in hydrogen charged conditions. However, it is unclear whether the results properly reflect the interaction of hydrogen with the material under practical circumstances. Hydrogen embrittlement severity depends not only on material properties and microstructure, but also on the type of test and the testing conditions. It is, for example, often reported that the degree of hydrogen embrittlement in tensile tests depends on the strain rate (Depover et al. (2016)). Since Charpy impact testing is a dynamic test, the time for hydrogen diffusion during the test is very limited. This high strain rate might make Charpy impact testing inappropriate for hydrogen embrittlement evaluation (Nagumo (2016)). Still, some authors have reported a change in ductile to brittle transition temperature (DBTT), and in impact energy when performing Charpy tests on hydrogen charged specimens. Fassina et al. (2012) tested two pipeline materials, an API 5L X65 and an ASTM A 182 F22 steel. They reported a diffusible hydrogen content of 0.6 – 2 ppm and found an increase in the DBTT and a decrease in upper shelf energy (USE) due to hydrogen for both steels. Additionally, the scatter on the results of the hydrogen-charged samples was larger. They observed no difference in fracture surface appearance. Mori et al. (2015) found that the Charpy impact energy of 4340 steel tempered at temperatures above 468°C was affected. Rosenberg and Sinaiova (2017) found no influence on the Charpy impact energy values after hydrogen charging an API 5L X70 pipeline steel as well as an S355 structural steel, although a higher scatter for hydrogen charged specimens was also reported. However, after subjecting the specimens to a static preload for 24h, allowing hydrogen to diffuse to the notch tip, they did note an influence on the upper shelf energy and the DBTT. Golisch et al. (2022) investigated base metal and weld specimens of a spiral welded grade L415ME steel pipe. They reported a diffusible hydrogen content of 3.43 ppm and 3.75 ppm for base metal and weld, respectively, and observed a significant reduction in CVN impact energy. The reduction in impact energy found by Golisch et al. (2022) (about 84 J to 99 J) was more than double the one seen by Fassina et al. (2012), who found the USE decreased about 20 J and 40 J for the X65 and F22 steel, respectively. This is possibly related to the difference in hydrogen content between the two studies, or the different materials studied. Previously mentioned authors all used electrochemical methods to charge the samples with hydrogen. In the present work, the influence of electrochemical hydrogen charging on the Charpy impact energy of an API 5L X70 pipeline steel is characterized. 2. Materials and methods The tested material is an API 5L X70 steel, produced in 1991. Specimens were extracted from the mid-thickness region of a pipe with diameter 1016 mm (40”) and wall thickness 15.8 mm that had been previously in service in a natural gas pipeline. Standard Charpy V-notch specimens (10 mm x 10 mm cross section) were machined according to EN-ISO 148-1 (International Organisation for Standardization (2016)), and had a longitudinal-transverse (L-T) orientation. Impact tests were performed for temperatures ranging from -80 °C to 20°C in steps of 20 °C. The specimens were cooled in a cooling bath at the testing temperature for 5 min, and the time between removing the specimen from the bath and the striking of the hammer was no more than 5 s. For hydrogen charged specimens, the time between the end of charging and submerging into the cooling bath was minimized and never exceeded 1 min. Specimens that were not fully fractured during the test were submerged in liquid N 2 and broken in order to examine the fracture surface. Hydrogen charging was done electrochemically using a 0.5 M H 2 SO 4 electrolyte with 1g/L thiourea (CH 4 N 2 S) added as a hydrogen recombination poison. A constant current density of 0.8 mA/cm 2 was applied. Table 1 gives the