PSI - Issue 42

Hiroshi Nishiguchi et al. / Procedia Structural Integrity 42 (2022) 1442–1448 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

1443

2

1. Introduction A global effort is currently underway to achieve a carbon-neutral society with overall zero emissions of carbon dioxide and other greenhouse gases by 2050. In a carbon-neutral society, a supply chain that enables the production, transportation, and consumption of hydrogen in each region is planned to be established through collaboration between industry, academia, and government. However, to enable a hydrogen-based economy, it is necessary to solve the problem of hydrogen embrittlement, which degrades the mechanical properties of metallic materials used in hydrogen using equipment. To date, many studies have been conducted to elucidate the mechanism of hydrogen embrittlement (Troiano 1960; Oriani and Josephic 1974; Beachem 1972; Birmbaum and Sofronis 1994; Nagumo 2001; Murakami et al. 2012; Matsuoka et al. 2011, 2016). Nevertheless, to realize a hydrogen-energy-based society that is not only environmentally friendly but also economically efficient, it is necessary to solve the problem of high cost through studies on the use of less expensive metallic materials such as low-alloy and carbon steels (Ogawa et al. 2018). Several studies on hydrogen embrittlement focused on tensile properties (Matsuoka et al. 2017; Yamabe 2016), fatigue and fatigue crack growth properties (Ogawa et al. 2018; Matsuoka et al. 2011, 2017), and hydrogen enhancement of the dislocation mobility (Ferreira 1998). Narita et al. (1994) focused on the relationship between dislocations and impurity atoms to understand the elastic interaction with the cracks, and they explained that hydrogen distribution near the crack tip gives rise to an anti-shielding effect, resulting in a decrease in the fracture toughness. According to the results of the calculations on crack-tip shielding by Narita et al. (1989) and Higashida et al. (1990, 2002, 2004, 2008), the internal stress field due to the dislocations introduced around the crack tip gives rise to compressive stress concentration at the crack tip. Consequently, the tensile stress concentration due to the external force at the crack tip is relaxed (shielded). Therefore, the tensile stress concentration at the crack tip must be decreased to suppress crack propagation and obtain improved fracture toughness. In this study, the effects of hydrogen on the internal stress fields due to the dislocations and crack propagation in silicon crystals were investigated in a fundamental investigation of the mechanism of hydrogen embrittlement. Silicon crystals were employed because (1) the residual stress field due to the dislocations can be visualized by infrared Raman spectroscopy, (2) the changes in the elastic field of the dislocations due to the introduction of hydrogen can be directly observed, and (3) dislocation-free crystals can be obtained more easily than when using other materials. Vickers indentations were formed on {100} silicon wafers, and cracks were introduced from the four corners of the indent along the {110} plane. In addition to the introduction of the cracks, strong internal stress fields were formed around the indent. Then, the indented wafers were exposed to hydrogen gas. It was found that high-pressure hydrogen gas exposure has a significant effect on residual stress and crack propagation. Based on these experimental results, the mechanism of hydrogen embrittlement is discussed from the viewpoint of the effects of hydrogen on crack propagation and crack-tip shielding by dislocations. 2. Experimental procedure The specimens were {100} silicon wafers. A 10 mm × 10 mm square chip specimen was cut from a silicon wafer disk with a diameter of 300 mm. The base silicon wafers were grown using the Czochralski (CZ) method, sliced, and mirror-finished. The surface was in the (001) plane, and the resistivity was approximately 15 Ωcm. Vickers indentations were introduced on the silicon surface. The load was 100 gf and 200 gf, and the holding time was 30 s. The indentation and cracks were introduced so that the cracks could propagate in four directions along the {110} plane from the corners of the indentation. Residual stress distributions were measured using a Nanophoton Raman laser microscope (RAMANtouch VIS-NIR-SWMex, Nanophoton). The accuracy of the infrared Raman method for stress measurement was confirmed to be within 30 MPa based on the results of the direct stress measurements conducted at the same point. After the indentation, the specimens were exposed to high-pressure hydrogen gas at 270 o C for 24 h, with the pressure varying from 10 MPa to 100 Mpa, and hydrogen-charged specimens were prepared. Additionally, uncharged specimens were prepared by exposing the specimens to Ar gas at the same temperature and holding time as those for the hydrogen gas exposure. After the internal stress distribution was visualized by infrared Raman, the specimens were exposed to high-pressure hydrogen gas and then hydrogen-charged in silicon. The internal stress distribution around the Vickers indentation