ICAMS / Interdisciplinary Centre for Advanced Materials Simulation

Chemical trends for hydrogen-vacancy complexes and resulting self-diffusion parameters

The interaction of H with vacancies in metals can be the reason for several hydrogen-related processes. Since vacancies serve as trapping centers for dissolved H, they locally yield a critical H concentration sufficient for initiating mechanisms of hydrogen embrittlement. Our previous ab initio calculations for fcc Fe revealed that up to 6 H atoms can be trapped by a mono-vacancy, where the formation energy of such vacancy-hydrogen complexes is lower than the sum of the energies for incorporating a pure vacancy and isolated interstitial H atoms into the bulk material. Therefore, the formation of H-vacancy complexes simultaneously increases both the concentration of vacancies and the total hydrogen content in the metal. This phenomenon called as the superabundant vacancy (SAV) formation.

Formation energies of point defects in fcc Fe. The dashed line represents interstitial H, solid lines – H-vacancy complexes.

That such a phenomenon can have a very significant effect, became apparent experimentally, e.g., in the case of Pd, where it leads to the formation of up to 23% of vacancies in a hydrogen-rich atmosphere. There are indications that similar effects can be observed, e.g., in Ni, Ti, Pd-Rh alloys, Al, Mn, Cr and Co, some hydrogen storage materials and stainless steels. In many cases it is assumed that due to SAV formation established metal-hydrogen phase diagrams are actually metastable and need to be further investigated to account for vacancy-ordered phases. Large vacancy concentrations further influence diffusion rates and can therefore accelerate several kinetic processes such as, e.g., phase separation, ordering, inter-diffusion, and creep. Moreover, hydrogen-stabilized vacancies play a central role in the HESIV (Hydrogen-enhanced stress-induced vacancies) mechanism of hydrogen embrittlement.

One aim of the present study is to extend our previously obtained knowledge on SAV formation in Fe to a large set of fcc metals. Both the chemical screening and the much more pronounced effects in other metals as compared to Fe, allows for a systematic understanding of the phenomenon.

An important prerequisite for the study of SAV with ab initio methods is the correct determination of vacancy formation energies. The latter are known to sensitively depend on the choice of the exchange-correlation functional (LDA and GGA). Post-processing correction schemes have recently been suggested (by Carling et al.) to overcome these difficulties. Their performance for the investigated metals will be evaluated and the importance of (an)harmonic contributions to the formation energy will be tested. In a second step the interaction energies of (several) hydrogen atoms with vacancies will be determined by high-throughput DFT calculations. These parameters are in a third step used in thermodynamic models, which allow the prediction of SAV formation and provides the dependence of this effect on external conditions (including hydrogen chemical potential, temperature and pressure). The chemical trends are used to identify systems which are particularly sensitive to SAV formation. A comparison with tight-binding studies for hydrogen-vacancy complexes in the AMS department is also forseen.

A second aim of the project is to identify the importance of hydrogen-vacancy complexes for kinetic processes in the material and in particular to establish the impact of SAV on increased self-diffusions in metals. For this purpose, the mobility of hydrogen-vacancy complexes will be investigated systematically. The obtained activation barriers will be provided for kinetic Monte-Carlo simulations performed in the AMS group.

« back