Hydrogenation of CO and CO₂ Catalyzed by Potassium Chloride F Centers

Abstract

The role of F centers on the KCl (100) surface in the hydrogenation of carbon monoxide and carbon dioxide was investigated theoretically at density-functional theory level. The surface defect was found to reduce all carbon-containing species involved in the reactions. In all cases, the defect electron occupies the lowest antibonding orbital of the molecules which weakens at least one of the C–O bonds. All reaction barriers of the hydrogenation on the defect are significantly reduced in comparison with the gas-phase reactions. The thermodynamic and kinetic parameters of the reaction pathways were computed, allowing for a comparison with the industrially established Cu/ZnO catalyst for methanol synthesis. It was found that methanol synthesis from CO2 is thermodynamically and kinetically hindered, yielding formate as a product that blocks the F center. The reaction rate for the rate limiting step of the methanol synthesis from CO on the F center is about 10 times higher than that for the conventional catalyst. However, the end product methanol further dissociates into methanolate and free radical hydrogen with only a small activation barrier. Furthermore, it was found that water dissociates into OH– and H• on the defective surface, so that even trace amounts can deactivate the F center.

Methanol Synthesis on Copper-Doped F Centers

“…At the same time, the reaction displaces the neighboring cation as the copper atom is moving into the surface. At 500 K, the free activation energy for the LH hydrogenation is normalΔ G doped , LH ‡ = 1.19 eV, which matches the barrier for the same reaction on the empty defect ( normalΔ G empty , LH ‡ = 1.20 eV). Likewise, the adsorption of hydrogen on the Cu-doped defect, normalΔ G ads , normalH 2 doped = 0.40 eV, matches the adsorption on the empty defect ( normalΔ G ads , normalH 2 empty = 0.36 eV).…”

“…One focus of this work is the dissociation of methanol on the doped defect because this was an important side reaction on the undoped F center. 26 First, we modeled the dissociation to methane and CuO 4 → 5 on the Cu-doped defect (Figure 7). While the free reaction energy ΔG 4,5 = −1.05 eV for the formation of CuO and CH 4 (Figure 7b) is strongly negative at 500 K, the free activation energy G 4,5…”

“…PBE 28 with D3(BJ) dispersion correction 43,44 was employed for all calculations, in accordance with our previous investigations of the vacant F center. 25,26 The defective KCl (100) surface was modeled as a 4 × 4 supercell of a five-layer slab with 15 Å vacuum between the repeating slabs and a central mirror plane to prevent artificial dipole moments. For this model, a Γ-point Monkhorst−Pack grid was sufficient.…”

“…In the first step, we examine the adsorption of copper on the F center and the hydrogenation analogous to the reaction pathways calculated in our previous investigation to compare the reactivity of doped and undoped surfaces. Next, we explore the hydrogenation of the adsorbed copper atom and its ramifications on the reaction diagram.…”

“…The methodology applied in this work is based on our previous theoretical investigations. − We used the PBE functional to calculate the energetic properties of our models, which as a computationally efficient GGA functional provides a comparable charge transfer behavior as self-consistent hybrid functionals for various defective alkali halide surfaces. ,, For calculating kinetic properties such as reaction paths and activation barriers, we used the nudged elastic band method …”