Charles C. Kim

Modeling the optical dielectric function of II‐VI compound CdTe

In previous papers, a new model was proposed for the optical dielectric function of zinc blende semiconductors and was applied successfully to the alloy series AlxGa1−xAs. It was found to be more generally valid than any previous model. In this paper, it is used to obtain an analytic expression for the dielectric function of the II‐VI compound CdTe at room temperature. Unlike the previous application, additional polynomial terms in the real part of the optical dielectric function are not introduced, so that the Kramers‐Kronig relationship between the real and imaginary parts of the dielectric function is satisfied exactly. As demonstrated before, the model covers not only the entire photon energy range of the given spectral data, but also is valid below and somewhat above the given spectral range. This is advantageous especially when spectral data are not provided in the range of interest, or when joining two separate data causes an artificial discontinuity. The model determines the optical dielectric fun...

Determination of accurate critical-point energies, linewidths and line shapes from spectroscopic ellipsometry data

Abstract The usual procedure for obtaining critical-point energies and linewidths from spectroscopic ellipsometry (SE) data is numerically to smooth and differentiate those data and then to fit them to parametrized theoretical functional forms. That procedure introduces systematic distortion and broadening of the differentiated line shape, which affects the line shape and the values found for critical-point energies and linewidths. In this paper we present a simple new procedure which does not introduce such systematic errors. We present several examples which show the accuracy of this new procedure and we present arguments for the use of SE rather than the electroreflectance or photoreflectance for the determination of critical-point energies, linewidths and line shapes.

Molecular beam epitaxial growth of ZnSe on GaAs substrates : influence of precursors on interface quality

The initial growth of thin ZnSe layers, grown on GaAs(0 0 1) using molecular beam epitaxy (MBE), has been studied using real-time reflection high-energy electron diffraction (RHEED), as well as X-ray double-crystal rocking curves (DCRC), reflectivity difference spectroscopy (RDS) and transmission electron microscopy (TEM). A noticeable difference in interface quality and growth was observed depending upon the initial precursor. Exposure to Se flux before growth led initially to three-dimensional growth, whereas exposure to Zn flux or no precursor led immediately to two-dimensional growth. X-ray DCRC and RDS confirmed that the overall sample quality was better with two-dimensional growth, and TEM observations were consistent with previous studies which have shown that initial three-dimensional growth is related to the formation of Ga2Se3 at the interface, leading to high densities of stacking faults and edge dislocations.

New Theoretical Treatment of Electroreflectance from Surfaces and Interfaces Having Large Built-in Fields

We have constructed a completely quantum-mechanical theoretical treatment of the modification of the local dielectric function by intense electric fields. It includes the numerical calculation of depletion-region bound-state energies and wavefunctions as well as band wavefunctions. We use this treatment to construct electroreflectance (ER) lineshapes. Our numerical results differ significantly from those of previously proposed ER lineshapes for GaAs with N d >> 10 16 . They yield excellent fits to ER data and allow the accurate determination of doping levels and band bendings from ER data.