Chemical Identification of Single Atoms in Heterogeneous III–IV Chains on Si(100) Surface by Means of nc-AFM and DFT Calculations

Text

We have demonstrated a new way to resolve the chemical identity of individual atoms in surface nanostructures. Our method is based on a combination of atomic force microscopy (AFM) measurements with density functional theory (DFT) calculations. This approach significantly enlarges the available means of surface and nanostructure analysis. It allows us to understand semiconductor nanostructures formed on surfaces from many aspects. Namely we can study the process of their formation, their stability, their physical properties as well as chemical composition. The chemical composition, in particular, has a great impact on electron transport properties of one-dimensional atomic chains, which are expected to become basic building elements in the emerging field of nanoelectronics.

We focused on chemical identification of individual constituent atoms in mixed In-Sn atomic chains grown on the Si(100)-(2x1) surface. The study was carried out at room temperature using dynamical AFM measurements supported by DFT simulations. We have shown that the chemical identification is possible by exploiting the distance dependence of short-range forces acting between the AFM tip and the probed atom. Beside other results, our new method allowed us to find out that Si atoms from the substrate incorporate into the metallic chains deposited on the silicon surface. The presence of the Si atoms, which was completely ignored thus far, affects in an important way the formation and stability the atomic chains. Furthermore, our data even indicate the possibility of discerning different ways of binding chemically identical atoms according to their atomic-orbital hybridization.

4-2012_0.png
Description
Left: Atomically-resloved AFM image of Sn-In chains on the Si(100) surface. Middle: Comparison between characteristic experimental and simulated force curves on In and Sn atoms. Right: Ball and stick model for the simulations.