Spontaneous Activation of CO₂and Possible Corrosion Pathways on the Low-Index Iron Surface Fe(100)

Abstract

Spin-polarized density functional theory (DFT) calculations and periodic slab models were used to study the reactive pathways leading to corrosion products on the low-index (100) surface of iron. We determined the binding energies of CO 2 and the barrier to decomposition of adsorbed CO 2 to O + CO, as well as to the formation of adsorbed CO 3 2and H 2 CO 3 on the Fe(100) surface. The barriers of these pathways were determined with nudged elastic band (NEB) calculations. Short trajectories with DFT-based dynamics were employed to identify the most important species. These simulations (up to 0.5 ML coverage) show that CO 2 is spontaneously activated and can bind with two or all three atoms assuming bent configurations strongly reminiscent of the radical CO 2 -. This spontaneous activation of CO 2 is possible through charge rearrangement of the slab density. The CO 2 decomposition to O + CO has a barrier of 5.0 kcal/mol. The subsequent formation of CO 3 2by reaction with an incoming CO 2 is strongly favored thermodynamically. Interaction of H 2 O with the adsorbed CO 2 forms a loosely bound complex that leads to the formation of surface-bound carbonic acid, with a barrier of approximately 35.0 kcal/mol.

CO2 activation and dissociation on the low miller index surfaces of pure and Ni-coated iron metal: a DFT study

“…The adsorption of both CO and O at the hollow sites released an adsorption energy of -179.9 kJmol -1 , whereas for the co-adsorption of oxygen at the bridge site with CO at the hollow site, an adsorption energy of -143.7 kJmol -1 was released. Our results are consistent with those of Glezakov et al, who reported a co-adsorption energy of -126.0 kJmol -1 for CO and O at the hollow sites using the projected augmented wave pseudopotentials 28 . On the nickel-deposited Fe(100) surface (Figure 6a), the CO moves from the most table hollow site to the bridge site.…”

“…For the Fe (100) surface calculations, a p(3 x 3) super-cell was employed for the CO2 adsorption and dissociation reactions, similar to earlier calculations 28,29 . 28, 29 Glezakou et al 28 also obtained the hollow-C2v structure as the preferred CO2 activated mode on the Fe (100) surface with an adsorption energy of -69.9 kJmol -1 , when they studied CO2 corrosion pathways using spin-polarized DFT and norm-conserving pseudopotentials.…”

“…28, 29 Glezakou et al 28 also obtained the hollow-C2v structure as the preferred CO2 activated mode on the Fe (100) surface with an adsorption energy of -69.9 kJmol -1 , when they studied CO2 corrosion pathways using spin-polarized DFT and norm-conserving pseudopotentials. Later, Liu et al 29 , employing spin-polarized DFT and projected augmented wave pseudopotentials, also reported the hollow-C2v structure as the preferred structure with an adsorption energy of -61. strength from -88.7 kJmol -1 to +47.2 kJmol -1 could at least in part be attributed to the higher electronegativity of nickel, resulting in less electron transfer into the CO2 moiety, as nickel is observed to bind adsorbates less strongly compared to iron 13,30 .…”

“…Unlike the previous studies of CO2 dissociation energies on Fe (100) 28 and Fe (111) 32 that focused on the barriers leading to dissociation, we have calculated both the activation energies required for CO2 adsorption (activation) and dissociation by locating the transition state structures TS1 and TS2 between the stable states of the isolated, adsorbed and dissociated CO2 molecule on the various iron facets. The reaction energy profiles for CO2 dissociation on both the bare Fe and Ni-deposited surfaces is shown in Figure 7, whereas the geometries of stationary point structures along the reaction coordinate are shown in SI Figure It can be seen from the energy profile that the CO2 dissociation step on the facets is the ratedetermining step.…”

“…Earlier theoretical works have reported spontaneous CO2 activation on the low Miller indices of iron and the energetics to dissociate CO2 species on the (100) and (111) surfaces but not the (110) surface [28][29][30][31][32][33] . A recent DFT study has shown that nickel deposition on iron surfaces increased the work function of the (100) and (111) facets but reduced the work function of the most stable and close-packed (110) surface 34 .…”