Surface termination of ${\mathrm{CePt}}_5/\mathrm{Pt}\left(111\right)$: The key to chemical inertness

The surface termination of ${\mathrm{CePt}}_5/\mathrm{Pt}\left(111\right)$ is determined experimentally by LEED-IV. In accordance with recent theoretical predictions, a dense Pt terminated surface is being found. Whereas the ${\mathrm{CePt}}_5$ volume lattice comprises Pt kagome layers, additional Pt atoms occupy the associated hole positions at the surface. This finding provides a natural explanation for the remarkable inertness of the ${\mathrm{CePt}}_5$ intermetallic. Implications of the structural relaxations determined by LEED-IV analysis are discussed with regard to observations by scanning tunneling microscopy and electron spectroscopies.

Figure 1

(a) Hexagonal surrounding of a Ce atom (large green sphere) in ${\mathrm{CePt}}_5$. The bottom layer exhibits the kagome hole characteristic of the bulk structure. The top layer includes an extra Pt atom as suggested for the surface layer. Its vertical displacement indicates the possibility of relaxation. (b) LEED image of ${\mathrm{CePt}}_5$/Pt(111) at $t_{\text{nom}}=4$ u.c. taken at $E_{\text{kin}}=60$ eV. Basic LEED beam indexation is indicated. The sample is tilted by approx. $1^{\circ }$ with respect to normal electron incidence.

Figure 2

Illustration of the three different termination layers considered. (a) ${\mathrm{CePt}}_2$ layer (model C). (b) Pt kagome layer (model K). (c) Pt layer with filled kagome holes (model P).

Figure 3

Illustration of the three structural models for ${\mathrm{CePt}}_5$/Pt(111). The two ${\mathrm{CePt}}_2$ layers nearest to the Pt(111) interface are represented by a rigid ${\mathrm{CePt}}_5$ lattice with full bulklike symmetry and equidistant atomic layer spacing $c_{\text{bulk}}$. The following two layers were allowed to independently relax along the surface normal (parameters $c_1, c_2$). Pt and Ce atoms are allowed to individually relax (parameters $c_3, \mathrm{\Delta }{c}_3$) in the final ${\mathrm{CePt}}_2$ layer which terminates model C. Model K assumes the presence of an additional kagome layer at distance $c_4$, the holes of which are filled by additional Pt atoms at position $\mathrm{\Delta }{c}_4$ in model P.

Figure 4

Full LEED-IV results for model P. Intensities of calculated curves (in red) have been scaled such as to yield the same integrated intensities as the experimental ones (in black).