Recently it was discovered that the structure of the molecule-metal interface in alkane thiol-based self-assembled monolayers (SAMs) is more complex that first believed. Thiols have been shown not only to lift the herringbone reconstruction of Au(111) but remove a significant fraction of the Au surface atoms. The etch pits formed by these vacancies are thought to be one of the weakest areas of the SAM films in terms of susceptibility to and degradation by oxidizing species.

 

In an effort to slightly weaken the molecule-metal interaction and prevent the formation of etch pits, we have chosen the study the interaction and assembly of trimethylphosphine (PMe3) on Au(111) using a scanning tunnelling microscope (STM). This is to our knowledge the first atomic-scale characterization of a phosphine species adsorbed on a metal surface.

 

At full monolayer coverage PMe3, the molecules formed a hexagonally packed layer which exhibited a (√7 x √7)R19o unit cell. The interaction between PMe3 and Au caused Au atoms to be ejected from the herringbone reconstruction, but not the substrate surface itself (as in the thiol case) and led to the formation of small Au islands. This effect manifested in the herringbone spacing increasing from that of native gold (6.33 nm) to 11.2 ± 0.9 nm. As this system was exposed to various annealing treatments, a fraction of the molecules desorbed, the Au islands coalesced, and the herringbones disappeared entirely, indicating that the underlying Au surface adopted a 1 x 1 reconstruction. The data indicate that the PMe3/Au island formation is a kinetically limited process.

Finally, we have developed a mathematical equation that gives the theoretical island coverage ( q ) as a function of the maximum island coverage ( qmax), the native herringbone spacing (xo) and the experimental herringbone spacing (x): q=qmax(1-xo/x).This should be useful in future studies of many types of SAMs on Au(111), or any similarly reconstructed surface.