Rotating compensator ellipsometry applied to OMCVD growth

rotating compensator;gallium arsenide;epitaxial growth;ellipsometry;omcvd

Flock, Klaus ; Aspnes, D. E., Committee Chair,Kolbas, R., Committee Member,Hallen, H., Committee Member,Rowe, J. E., Committee Member,Flock, Klaus ; Aspnes ; D. E. ; Committee Chair ; Kolbas ; R. ; Committee Member ; Hallen ; H. ; Committee Member ; Rowe ; J. E. ; Committee Member

University:North Carolina State University

关键词: rotating compensator;    gallium arsenide;    epitaxial growth;    ellipsometry;    omcvd;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/4385/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

The objective of this research was to develop a better understanding of epitaxial growth through non-destructive real-time optical measurements. This required us to develop a rotating-compensator ellipsometer/polarimeter (RCE/RCP) integrated into a modified commercial organometallic chemical vapor deposition (OMCVD) system. The new instrument obtains spectra of 1024 wavelengths (pixels) from 230 to 840 nm at a rate of 5 per second. The observables are the intensity of p- and s-polarized light, respectively, and the sine and cosine of the relative phase Delta. The newer RCE technology removes a serious limitation of the previously employed rotating-polarizer (RPE) system, where the data is lost in the vicinity of Delta = 0 and 180 degrees.By synchronizing the rotation of the substrate to the rotation of the compensator, we simultaneously measure the signal induced due to optical anisotropy of the sample. The connection between our non-normal-incidence anisotropy spectroscopy and normal-incidence reflectance difference spectroscopy (RDS) is established. As an example, the additional information is used to analyze a thin layer of Ga on a GaAs(001) substrate. One of the major problems of optical real-time diagnostics is the simultaneous determination of the parameters n, k, and d, i.e., the complex refractive index and the thickness, respectively, of the most recently deposited material. I solve the problem by fitting an analytical representation to the lineshape of n and k as a function of d in a region of the spectrum where n and k vary slowly with wavelength, i.e., in the region where the overlayer is transparent. The actual layer thickness is obtained from the minimum of the merit function chi-square of the fit. This variation to lineshape analysis is applied to extract overlayer information post-acquisition from real-time data to characterize (1) a 0.11 nm layer of GaAs on an AlGaAs substrate, and (2) a sequence of spectra recorded during the transition from AlGaAs to AlAs, where the deposition rate varied from 0.25 nm/s to 1.35 nm/s with increasing aluminum flux.