Surface atomic relaxation and magnetism on hydrogen-adsorbed Fe(110) surfaces from first principles

Urslaan Chohan, Enrique Jimenez-Melero, Sven Koehler

    Research output: Contribution to journalArticlepeer-review

    271 Downloads (Pure)

    Abstract

     We have computed adsorption energies, vibrational frequencies, surface relaxation and buckling for hydrogen adsorbed on a body-centred-cubic Fe(110) surface as a function of the degree of H coverage. This adsorption system is important in a variety of technological processes such as the hydrogen embrittlement in ferritic steels, which motivated this work, and the Haber–Bosch process. We employed spin-polarised density functional theory to optimise geometries of a six-layer Fe slab, followed by frozen mode finite displacement phonon calculations to compute Fe–H vibrational frequencies. We have found that the quasi-threefold (3f) site is the most stable adsorption site, with adsorption energies of ∼3.0 eV/H for all coverages studied. The long-bridge (lb) site, which is close in energy to the 3f site, is actually a transition state leading to the stable 3f site. The calculated harmonic vibrational frequencies collectively span from 730 to 1220 cm−1, for a range of coverages. The increased first-to-second layer spacing in the presence of adsorbed hydrogen, and the pronounced buckling observed in the Fe surface layer, may facilitate the diffusion of hydrogen atoms into the bulk, and therefore impact the early stages of hydrogen embrittlement in steels.
    Original languageEnglish
    Pages (from-to)385–392
    JournalApplied Surface Science
    Volume387
    Early online date21 Jun 2016
    DOIs
    Publication statusPublished - 30 Nov 2016

    Keywords

    • ferritic steels
    • hydrogen embrittlement
    • Density Functional Theory
    • adsorption
    • Haber-Bosch process

    Fingerprint

    Dive into the research topics of 'Surface atomic relaxation and magnetism on hydrogen-adsorbed Fe(110) surfaces from first principles'. Together they form a unique fingerprint.

    Cite this