samples failed so rapidly that there was insufficient times for the hydrides to nucleate and grow to any extent. DISCUSSION In general, hydrogen cracking is thought of as a brittle fracture mode. In steels, nickel-base alloys, and intermetallic compounds where atomic hydrogen is the embrittling species, the fracture mode is usually transgranular cleavage or intergranular fracture. Both of these fracture modes are associated with very limited ductility. In systems where hydrides form, the fracture mode is also usually reported to be brittle [1]. The most general models suggest that hydrides form at the crack tip in the tensile field of the crack. These hydrides then fracture in a brittle manner or a crack forms along the hydride-matrix interface. In our study of the grade 2 and grade 3 titanium, we found no evidence of brittle failure. The fracture mode remained ductile, even though a decrease in the elongation ratio was found for the grade 3 material. Our results suggest that hydrides nucleate in Figure 5: The elongation ratio plotted as a function of the electrochemical potential for Ti-5111. Results are shown for tests run in both pH=1 and pH=8 3.5%NaCl. the matrix in front of the crack tip. These hydrides then act as sights for microvoid formation. As the stress in front of the crack tip increases, these microvoids connect up, leading to crack advance. New hydrides then form in front of the crack tip and the process is repeated. In the grade 2 material, the hydrides do not form in front of the crack tip and this process does not occur. The reasons why the grade 3 material forms hydrides in front of the crack tip whereas the grade 2 material does not form these hydrides has been discussed in previous publications [4,6]. In short, we propose that the main reason for this difference is the lower solubility of hydrogen in grade 3 material as a result of its higher interstitial content. In addition, the higher yield strength of the grade 3 material may contribute to its greater susceptibility to hydride-induced ductile failure. When β-phase is present in the material but does not provide a continuous path through the material, the fracture mode should be similar to that reported above. This idea is consistent with the results of Nelson and Williams on Ti-6Al-4V [7,8]. Once the
RkJQdWJsaXNoZXIy MjM0NDE=