PSI - Issue 13

A. Laureys et al. / Procedia Structural Integrity 13 (2018) 1330–1335 Author name / Structural Integrity Procedia 00 (2018) 000–000

1331

2

built-up by the formation of hydrogen gas bubbles in internal voids and micro-cracks in the metal’ microstructure. Atomic hydrogen recombines to form molecular gaseous hydrogen, which cannot migrate further and thus locally induces an increased internal pressure. Also grain boundaries, second phase particles and tangled dislocations are suitable heterogeneities to initiate crack nucleation during hydrogen charging, according to Lee and Lee (1987). When this is taking place close to the sample surface, hydrogen blistering is observed, as the material is pushed upwards. Hydrogen charging experiments as a function of time illustrated that a critical hydrogen amount should be reached prior to the initiation of blisters, according to e.g. Laureys et al. (2017) and Jin et al. (2010). Jin et al. (2010) and Dong et al. (2009) found that the hydrogen concentration in the metal increased with increasing charging current density. Also the charging kinetics depended on the applied charging current density. They explained this by a higher surface coverage with hydrogen atoms due to the higher current density, resulting in a sample charged with more hydrogen. Several characterization techniques are available to evaluate to hydrogen/material interaction. At first, the hydrogen content can be determined by hot and melt extraction, revealing the diffusible and total hydrogen amount, respectively. Secondly, thermal desorption spectroscopy (TDS) is used to evaluate trapping at the potential hydrogen traps in the material. An adequate deconvolution and analysis procedure allows determining the corresponding activation energies for the different traps. Consequently, a correlation between microstructural features and hydrogen trapping at these sites can be made. Depover et al. (2016a, 2016b, 2018) performed detailed TDS analysis on the hydrogen trapping ability of several carbides in ternary Fe-C-X alloys. They demonstrated the efficient trapping of TiC and V 4 C 3 precipitates, within a certain carbide size range, and confirmed their beneficial effect on the possible hydrogen induced ductility loss, as assessed by in-situ hydrogen charged tensile tests. Hydrogen trapped at dislocations was mainly set responsible for the hydrogen embrittlement susceptibility. This was a clear confirmation of the hydrogen enhanced localized plasticity (HELP) theory, which proposes an enhanced dislocation mobility due to hydrogen. Beside the HELP theory, also the HEDE mechanism is reported to be responsible for hydrogen induced failure. Laureys et al. (2015, 2016) studied hydrogen assisted cracking in a transformation induced plasticity (TRIP) steel and demonstrated that the dominating mechanism for crack initiation occurred at the martensite-martensite interface, implying a hydrogen enhanced interface decohesion (HEIDE) mechanism, being a variant of the often cited HEDE mechanism. The diffusible of mobile hydrogen fraction is reported by Depover et al. (2018) to play a crucial role in terms of the HELP theory. Permeation experiments can be done to obtain the hydrogen diffusion coefficient of the material, whereas it can also serve to evaluate the effect of cold deformation on the hydrogen diffusion. The application of cold deformation increases the density of hydrogen traps. This resulted in a decreased diffusivity, as revealed by Kumnick and Johnson (1980) and Laureys et al. (2017) by these permeation transient measurements. This work aims to determine if hydrogen induced damage can be detected by melt extraction or TDS. As indicated above, melt extraction is commonly used to determine the total amount of hydrogen inside materials. As there is a correlation between hydrogen damage and hydrogen in the material, it could be expected that samples with damage might contain a higher hydrogen content, if the hydrogen was not released when the blister reached the sample surface. On the other hand, TDS can provide information on the available hydrogen trapping sites in a material including their trap density and corresponding desorption activation energy. This analysis is complementary to the melt extraction and will inform on the potential traps induced by hydrogen induced damage. Therefore, these two techniques were used on cold deformed ultra-low carbon (ULC) steel with and without hydrogen induced damage. 2. Experimental procedure 2.1. Material ULC steel was chosen as material of study to avoid the effect of complex microstructural characteristics on blister or internal crack characterization. The steel contains 214 wppm C, 88 wppm N, 38 wppm S, 73 wppm P, 0.25 wt% Mn, 0.002 wt% Ti, and 0.047 wt% Al. The material was cold deformed with 60% reduction to a thickness of 1.7 mm. 2.2. Electrochemical hydrogen charging Hydrogen was introduced in the samples by cathodic charging in a 0.5 M H 2 SO 4 electrolyte containing 1 g/l of thiourea. Charging occurred in a polycarbonate cell where the sample, connected as cathode, was positioned symmetrically in between two platinum anodes. All samples were charged at room temperature using different