R. Kosiba

Growth of cubic InN on r-plane sapphire

InN has been grown directly on r-plane sapphire substrates by plasma-enhanced molecular-beam epitaxy. X-ray diffraction investigations have shown that the InN layers consist of a predominant zinc blende (cubic) structure along with a fraction of the wurtzite (hexagonal) phase which content increases with proceeding growth. The lattice constant for zinc blende InN was found to be a=4.986 A. For this unusual growth of a metastable cubic phase on a noncubic substrate an epitaxial relationship was proposed where the metastable zinc blende phase grows directly on the r-plane sapphire while the wurtzite phase arises as the special case of twinning in the cubic structure.

MC simulations of depth profiling by low energy ions

Abstract Surface analytical methods like AES, XPS or SIMS mostly use ion beam sputtering as a tool for depth profile analysis. The depth resolution of sputter depth profiles depends strongly on the ion–solid interactions. On this background, Monte Carlo (MC) simulations of sputter depth profiles have been carried out by means of T-DYN. After introducing the simulation model special emphasis has been paid to the input parameters such as the surface binding energy, bulk binding energy, cut-off and displacement energy, atomic density and their influence on the simulation results. A silicon carbide layer on Si and a Pt delta-doped silicon serve as targets for simulation tests. Although the influence of the surface binding energy is mainly important for the sputtering yield, this parameter together with the cut-off energy prove to be important for the resultant depth profiles and their depth resolution. The uniformity of the atomic density over the whole target in T-DYN limits the applicability of this simulation code.

Etching of SiC with Fluorine ECR Plasma

Electron cyclotron resonance (ECR) dry etching of 3C-SiC with different fluorinated gases, namely, sulfurhexafluoride (SF6) and tetrafluoromethane (CF4), was carried out. The influence of the gas flow, the etch gases and the applied bias voltages on the etch rate was studied. The maximum etch rates in the case of SF6 achieved were 1570 Å/min and 260 Å/min for Si and 3CSiC, respectively. In the case of CF4 the 260 Å/min (Si) and 160 Å/min (3C-SiC) were obtained. Furthermore, we investigate the selectivity of this dry etching process of SiC against Si. The residue free surface conditions were determined with Auger electron spectroscopy.

Low Energy Ion Modification of 3C-SiC Surfaces

The effects of argon and nitrogen bombardment of 3C-SiC surfaces at acceleration voltages below 2 keV were studied by stylus profilometry, reflectometry, reflection high energy electron diffraction and Auger electron spectroscopy (AES). The erosion rate of the SiC surface was determined. It was found that the sputtering rate for argon was three times higher compared to nitrogen. AES measurements revealed argon and nitrogen incorporation at a depth of a few nanometers as well as stoichiometric changes at the same depth scale independent of the acceleration voltage. In the case of the interaction of nitrogen ions with the 3C-SiC surface the formation of a SiCNalloy was detected.

Sputtering of SiC with low energy He and Ar ions under grazing incidence

The effect of low energy sputtering under grazing incidence upon the surface composition of SiC was investigated by Auger electron spectroscopy. The energy of the sputtering projectiles (He, Ar) varied from 200 to 1500 eV. Peak shifts to the higher energies with increasing argon ion energy were observed for all silicon and carbon Auger transitions. These shifts were explained by enhanced damage of the surface region within the sampling depth of the Auger electrons. The insensitivity of the Auger peak position to the energy of helium ions indicates that the damage state in the surface region does not change with the increasing energy of helium ions. An increase of the carbon concentration with the decrease of the argon energy was observed. The experiments were accompanied by dynamic Monte Carlo simulations by the TRIDYN code.