Structure ofC60layers on theSi(111)−3

Abstract

The structure of a monolayer of C 60 molecules adsorbed on the Si(111)-)ϫ)-Ag surface has been investigated by scanning tunneling microscopy ͑STM͒ at room temperature and at low temperature ͑60 K͒. The C 60 molecules are arranged in a ͱ21ϫͱ21(RϮ10.9°) double domain structure and also in a 3) ϫ3)(R30°) structure in part. The intramolecular structures are observed only for the molecules adsorbed at step edges at room temperature, because the rotation of C 60 molecules is suppressed due to the strong interaction with the substrate at step edges, while the molecules adsorbed on terraces do not exhibit the internal structure because of their fast rotation. On the other hand, the internal structure of C 60 is resolved for every molecule adsorbed on terraces at low temperature because the rotation of C 60 is suppressed as in the C 60 bulk crystal. The orientations of the individual C 60 seem to be determined by the directions of underlying Si and Ag trimers of the)ϫ)-Ag surface. ͓S0163-1829͑99͒03140-9͔

Calculation of images of orientedC60molecules using molecular orbital theory

“…Furthermore, the appearance of the observed high-resolution STM image confirms that lower resolution experiments would indeed detect C 60 molecules with striped features. It is also interesting to note that the −1.5 V bias used to record the high-resolution STM images ͓on a Si͑111͒-͑7 ϫ 7͒ surface͔ is fairly consistent with the −2 V bias used 31 in the lower-resolution study. The HOMO EЈ component for the double bond-prone orientation appears to have fourfold symmetry and, at lower resolution, this orbital pair would produce an STM image resembling a four-petaled flower.…”

“…Certainly, the complex nature of the Si͑111͒-ͱ 3 ϫ ͱ 3-Ag surface used in Ref. 31 would lend some credence to this explanation. However, one would still expect the distribution of A-and B-type molecules to follow a regular pattern on this surface and not be randomly distributed as observed.…”

“…However, the authors of Ref. 31 believe the interaction between the Si͑111͒-ͱ 3 ϫ ͱ 3-Ag surface and adsorbed molecules to be weak because of the long diffusion length inferred for C 60 molecules on the surface. If this is the case, the surface-induced splitting is expected to be weak and the HOMO-derived components should all be close in energy and therefore imaged at nearly the same bias.…”

“…In fact, several interesting images recorded at this bias value are presented in Ref. 31 and are, therefore, potentially HOMO related. In one particular image ͑Fig.…”

Jahn–Teller Effects in Molecules on Surfaces with Specific Application to C60

“…Further internal electronic structure arising from the HOMO orbitals was observed by Tsuchie et al [35], who studied C 60 monolayers on a Si(111)-√ 3 × √ 3-Ag surface at room temperature and at 60 K using a fixed sample bias of -2 V. Fig. 9 shows the images obtained from a plain C 60 monolayer and one that has been doped with potassium.…”