Adsorption of N@C60 on Si(100)

Abstract

The interactions between C 60 molecules and the Si(100) surface, as well as the interactions between the endohedrally doped N@C 60 molecules and the Si(100) surface have been explored via ab initio total energy calculations. Configurations which have the cage located upon the dimer row bonded to two dimers (r2) and within the dimer trench bonded to four dimers (t4) have been investigated, as these have previously been found to be the most stable for the C 60 molecule. We show that our results for the adsorption of the C 60 molecule upon the Si(100) surface are comparable with previous studies. We have investigated the differences between the adsorption of the C 60 and N@C 60 molecules upon the Si(100) surface and found that there are only minimal differences. Two interesting cases are the r2g and t4d configurations, as they both exhibit differences that are not present in the other configurations. These subtle differences have been explored in-depth. It is shown that the effects on the endohedral nitrogen atom, due to its placement within the fullerene cage, are small. Bader analysis has been used to explore differences between the C 60 and N@C 60 molecules.

Interaction of molecules on Si(100)

“…The binding energies of these systems are presented in Table 1, alongside the corresponding energies from our previous study [6]; the previous study used a smaller 128 atom Si (100) surface which has been used extensively in other studies [7][8][9][10][11]. As Table 1 illustrates, there appears to be a shift in not only the magnitude of the binding energies, with every configuration becoming less favourable on the larger surface, but the ordering of the favourability hierarchy has also changed.…”

“…Thus our model consists of six layers of Si atoms, with the bottom two layers of Si atoms being pinned in their respective bulk; and the bottom layer's dangling bonds are saturated with H atoms (which in turn have been pinned in their own relaxed positions). As with our previous study [6], we have chosen a 2 × 2 tilted dimer reconstruction, as this has been shown to be energetically most favourable within our approach [9]. A large vacuum gap has been included above the Si surface; this vacuum gap ensures a sufficient distance between the top of the fullerene cage(s) and the bottom of the surface for the periodic repeat, so as to minimise the interaction between the two.…”

“…The differing atom species present in the simulation cell are described by a series of basis sets that vary greatly in complexity (the same basis sets as used in our previous study [6]). The simplest description is for the H atoms, is described with a single s function; as the only purpose of the H atoms in our simulation cell is to saturate the dangling bonds from the final Si layer, so a more complete description is not necessary.…”

“…The use of both the electronic and nuclear spin as qubits has also been discussed [5]. We explored the suitability of N@C 60 molcules in our previous study [6], and concluded that they met the required criteria for use as qubits. There are still some fundamental issues that need to be addressed with an architecture of this type, for example the qubit readout, however various solutions have been suggested [5].…”

Plato: A localised orbital based density functional theory code

“…Plato was used in the first ab-initio studies of the bonding of C 60 molecules to the Si (100) surface that allowed for atomic relaxation [44,45]. Subsequently it has been used to study the adhesion of C 82 molecules and endohedral C 60 molecules and the interaction between C 60 molecules on the Si (100) surface [46,47,48]. In these studies TNDP basis sets were employed, which massively reduced the basis set superposition error to the point where such corrections were unnecessary.…”