Reprint of: Surface reconstruction of pure-Cu single-crystal electrodes under CO-reduction potentials in alkaline solutions: A study by seriatim ECSTM-DEMS

Quasi-operando electrochemical scanning tunneling microscopy (ECSTM) recently showed that a polycrystalline Cu electrode kept in 0.1 M KOH at − 0.9 V (SHE), a potential very close to that for electrochemical CO reduction, underwent a two-step surface reconstruction, initially to Cu(111), or Cu(pc)-[Cu(111)], and terminally to Cu(100), or Cu(pc)-[Cu(100)]. When subjected to monolayer-limited Cu(s) ↔ Cu2O(s) oxidation-reduction cycles (ORC), the Cu(pc)-[Cu(100)] surface was further transformed to Cu(pc)-[Cu(511)] that produced C2H5OH exclusively, as detected by differential electrochemical mass spectrometry, at an overvoltage lower by 645 mV relative to that for the formation of hydrocarbons. In this paper, results are presented from studies with the native monocrystalline surfaces Cu(111), Cu(100) and Cu(110). Whereas the intermediate Cu(pc)-[Cu(111)] layer was eventually converted to Cu(pc)-[Cu(100)], the surface of a pristine Cu(111) single crystal itself showed no such conversion. The surface of an original Cu(100) electrode likewise proved impervious to potential perturbations. In contrast, the outer plane of a Cu(110) crystal underwent three transformations: first to disordered Cu(110)-d[Cu(110)], then to disordered Cu(110)-d[Cu(111)], and finally to an ordered Cu(110)-[Cu(100)] plane. After multiple ORC, the converted [Cu(100)] lattice atop the Cu(110) crystal did not generate ethanol, in contrast to the [Cu(100)] phase above the Cu(pc) bulk. Quasi-operando ECSTM captured the disparity: Post-ORC, Cu(110)-[Cu(100)] was converted, not to Cu(110)-[Cu(511)], but to an ordered but catalytically inactive Cu(110)-[Cu(111)]; hence, no C2H5OH production upon reduction of CO, as would have been the case for a stepped Cu(511) surface.