Date of Award

1-1-1981

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Physics & Astrophysics

Abstract

Experimental measurements of the optical reflectance of bismuth tellurium sulfide, Bi(,2)(Te(,1-x)S(,x))(,3); x = 0.48 were carried out both at normal and oblique incidence angles over a range of wavelengths. Bismuth tellurium sulfide is a uniaxial crystal that grows as a layered structure in sets of five atomic layers called quintuple layers. These layers are perpendicular to the optic axis of the crystal. Due to week Van der Waals bonds between the adjacent quintuple layers, the crystal cleaves easily along these layers and presents atomically smooth surfaces for reflectance studies. The measured reflectance spectra were analyzed by two different methods to determine the optical constants n and k.Two sets of optical constants, (n(,a),k(,a)) and (n(,c),k(,c)) are associated with a uniaxial crystal depending upon the polarization of the incident electromagnetic wave. Both the sets as functions of wavelength were determined. The set (n(,a),k(,a)) was obtained from the normal incidence spectrum whereas the set (n(,c),k(,c)) was obtained from the oblique incidence spectrum. The normal incidence measurements were performed from 1(mu) to 17(mu) with unpolarized beam and were supplemented with the available data from 0.2(mu) to 1(mu) . The extended spectrum was analyzed by both the classical oscillator method and the Kramers-Kronig method. The oblique incidence reflectance measurements were carried out at 45(DEGREES) with plane-polarized beam with the electric vector in the plane of incidence. The corresponding wavelengths ranged from 1(mu) to 16.5(mu) . The oblique angle reflectance spectrum was analyzed by the oscillator method.A set of fifteen oscillators was found to be necessary, along with the Drude parameters and the frequency independent dielectric constant, to fit the normal incidence reflectance spectrum. In case of the oblique incidence spectrum however seven oscillators were found to be sufficient. The normal incidence spectrum was also analyzed by the Kramers-Kronig method. The results obtained by the two methods were compared. Each method was found to have its own limitations. The oscillator method suffers from uncertainties of an acceptable match between the theoretical fit and the experimental spectrum, whereas the Kramers-Kronig method suffers from the uncertainties of realistic extrapolations to the low and high frequency regions of the measured spectral range. Either method alone did not guarantee a unique pair of optical constants at each wavelength. Thus it was found necessary to use both the methods.

Download

Share

COinS