Mavrikakis and co-workers have pioneered the theory and application of near-surface alloys (NSAs) in heterogeneous catalysis and electrocatalysis with great success.1–3 NSAs are idealized structures with a pure solute monolayer on the surface or the first subsurface layer and can be experimentally approximated with core-shell particles. Previously, Greeley and Mavrikakis have also predicted that NSAs with defect sites may have unusual catalytic properties,4 which has been experimentally confirmed for single atom defects, i.e. single atom alloys, for hydrogen dissociation.5 A key advantage of NSAs is the fact that the amount of expensive precious metals can be minimized without sacrificing catalytic efficiency.
Here, we will discuss NSAs where the solute monolayer is present in a sub-monolayer (sML) configuration on the host metal surface. The main distinguishing feature of the sML system is the presence of finite size effects, which cause compressive strain in the sML. The resulting effective strain is a superposition of epitaxial strain caused by lattice constant mismatches and the compressive strain induced by finite size effects. We have used underpotential deposition (UPD) and surface limited red-ox replacement to synthesize Pt(sML)/Pd(100) and RuPt(core-edge)/Au(111) sML systems and investigated their properties using CO adsorption, CO electro-oxidation, and density functional theory (DFT) calculations.6 Epitaxial strain is essentially absent in the Pt(sML)/Pd(100) system, but the effective strain quantified by DFT simulations shows a radial dependence being always more compressive at the periphery than in the center of the Pt nanocluster. The scenario is reminiscent of surface tension, but with reduced dimensionality. To demonstrate the unusual catalytic properties of sML systems, we synthesized and tested 2D RuPt nanoclusters on Au(111) with a unique core-edge structure having a Ru core and Pt at the edge along the perimeter. Cyclic voltammetry demonstrates that the 2D RuPt core-edge catalyst morphology is an order of magnitude more active for CO electro-oxidation than either Pt or Ru sML catalysts. DFT calculations in combination with infra-red spectroscopy data point towards oscillating variations (ripples) in the adsorption energy landscape along the radial direction of the Ru core and an active Ru-Pt interface as the origin of the observed behavior. The ripples, in turn, are caused by strong finite size effects causing compressive strain.
Our combined theoretical and experimental results provide clear evidence that the morphology of sML catalyst systems is important in determining their overall catalytic activity. Our work extends beyond earlier approaches using overlayer NSAs, in which the strain level is solely dictated by the epitaxial relationship.
(1) Greeley, J.; Mavrikakis, M. Nat. Mater. 2004, 3, 810–815.
(2) Alayoglu, S.; Nilekar, A. U.; Mavrikakis, M.; Eichhorn, B. Nat. Mater. 2008, 7, 333–338.
(3) Zhang, J.; Vukmirovic, M. B.; Sasaki, K.; Nilekar, A. U.; Mavrikakis, M.; Adžić, R. R. J. Am. Chem. Soc. 2005, 127, 12480–12481.
(4) Greeley, J.; Mavrikakis, M. Catal. Today 2006, 111, 52–58.
(5) Kyriakou, G.; Davidson, E. R. M.; Peng, G.; Roling, L. T.; Singh, S.; Boucher, M. B.; Marcinkowski, M. D.; Mavrikakis, M.; Michaelides, A.; Sykes, E. C. H. ACS Nano 2014, 8, 4827–4835.
(6) Grabow, L. C.; Yuan, Q.; Doan, H. A.; Brankovic, S. R. Surf. Sci. 2015 DOI: 10.1016/j.susc.2015.03.021.