Spin-polarized density functional theory (DFT) has been used to characterize hydrogen atom adsorption and diffusion energetics on the Fe(I 10) surface. The Kohn-Sham equations are solved with periodic boundary conditions and within the all-electron projector-augmented-wave (PAW) formalism, using a generalized gradient approximation (GGA) to account for electron exchange and correlation. We investigate the site preference of H on Fe(I 10) for 0.25, 0.50, and 1.0 ML coverages and find that the quasi three-fold site is the only stable minimum (in agreement with experiment). We find the long and short bridge sites to be transition states for H diffusion on Fe(I 10), while the on top site is a rank-2 saddle point. The preference of the three-fold site is rationalized via an analysis of the site- and orbital-resolved density of states. An analysis of charge density differences suggests that the H-Fe interaction is quite covalent, with only similar to0. 1 electron transferred from Fe atoms to H in the three-fold site of Fe(I 10). We also compare two experimentally observed 0. 50 ML phases for H/Fe(I 10): a graphitic (2 x 2)-2H and a (2 x 1) phase. We confirm the LEED data that the Fe(I 10)-(2 x 2)-2H superstructure is more stable at low temperature. The predicted adsorption structure and weak substrate reconstruction for the Fe(I 10)-(2 x 2)-2H phase roughly agree with experiment, though discrepancies do exist regarding the H-surface height and the H-H distance. Moreover, trends in work function with coverage are predicted to be qualitatively different than older measurements, with even the sign of the work function changes in question. Lastly, a zig zag diffusion path for H atoms on Fe(I 10) is proposed, involving a very low (<0.2 eV) barrier. (C) 2003 Elsevier B.V. All rights reserved.