4c,7 Our own analysis of the electronic structure of this family suggests that residual Si–Si π bonding within the Si 12 cage is an important stabilising component, 8 and this unsaturation may play an important role in binding to a surface. 6b Nevertheless, the structure and spectroscopy of the 12 family of clusters has been the subject of numerous experimental and computational studies which have established the stability of a hexagonal prismatic geometry for 12 and its isoelectronic analogues 12 and 12] +. Lu and Nagase's study of neutral clusters has also emphasised the relative thermodynamic stability of 16 over 12. 6a The smaller 12] + clusters, in contrast, owe their stability to kinetic factors, because further growth would require direct contact between the transition metal and additional molecules of SiH 4, and this is prevented by the adoption of the endohedral geometry.
![numpy.arange quantumwise numpy.arange quantumwise](https://www.e-cam2020.eu/wp-content/uploads/2019/03/Events19-posterboard-website.png)
Computational studies on cationic WSi n clusters ( n = 6–16) suggest that the species such as the 15] + ion observed in Beck's experiments are truly ‘magic’ in the sense that they represent a thermodynamic sink. In the case of W, the dominant peak with n = 12 is completely devoid of hydrogens ( x = 0), an observation that was interpreted to indicate that the metal atom is endohedrally encapsulated, such that it saturates the valence requirements of all twelve silicon atoms 5 (the is used from hereon in to indicate an endohedral species). For example, n varies from 14 for early transition metals such as Hf to 9 in later metals such as Ir.
![numpy.arange quantumwise numpy.arange quantumwise](https://docs.quantumatk.com/_images/si_nanowire_8a.png)
These experiments generate clusters with general formula +, which typically contain smaller numbers of Si atoms ( n) than observed in Beck's experiments. Of more direct relevance to this paper are the complementary experiments carried out by Kanayama, Hiura and co-workers using SiH 4, rather than a silicon wafer, as a source of silicon. The dominant peaks in the mass spectra typically have values of n in the range 14–17, with + and + particularly prominent in the experiments performed with W(CO) 6. Beck's experiments used laser vaporization of a silicon wafer to generate clusters in a molecular beam which is then quenched in the presence of metal carbonyl species to generate cationic clusters with formula n] +. 4 The structures and magnetic properties of these clusters depend critically on the identity of the transition metal, potentially offering a means of tuning the properties of a surface upon which they might be absorbed.
![numpy.arange quantumwise numpy.arange quantumwise](https://i1.rgstatic.net/publication/312194719_ATK-Classical_A_New_Generation_Molecular_Dynamics_Software_Package/links/60d466f6a6fdcc75a2502e7d/largepreview.png)
The fact that some of these clusters are particularly stable in the gas phase for certain ‘magic’ values of n is well established in the literature, courtesy of Beck's original mass spectrometric investigations 2,3 and a range of subsequent spectroscopic probes. Molecules that themselves contain silicon are obvious candidates in this regard, and there is a growing body of data on the gas-phase chemistry of endohedral silicon clusters, n, containing a wide range of transition elements. The integrated chip industry is, however, based on silicon, and there is clearly merit in studying molecules that are compatible with existing CMOS architectures. This assumption is somehow intrinsic to the concept of ‘molecular electronics’ because only in that limit can we hope to interpret current/voltage characteristics based on the properties of the isolated molecule. 1 In such circumstances, the boundary between molecule and surface is clearly defined and the effects of surface binding can be viewed as a small perturbation to the electronic structure of the molecule. Introduction Many of the recent advances in the field of molecular electronics have been driven by studies of molecules with appropriate linker groups (thiolate, for example) absorbed on metallic surfaces. The STM images therefore provide a very direct probe of the W–Si surface bond. The W 5d z 2 orbital, the LUMO of the isolated cluster, plays a critical role in all aspects, forming a covalent bond between the metal and the silicon surface, and then providing an effective transmission channel that allows current to flow from the surface to STM tip. The link between the intrinsic electronic properties of an endohedral metallo-silicon cluster, 12, its ability to bind to a Si(111)-(7 × 7) surface and the impact on transmission properties is explored using periodic density functional theory.