Muduli, Manisha
Email Address
ORCID
Disciplines
Search Results
Now showing 1 - 2 of 2
Publication Stress in Silicon Oxide Thin Films Grown by Dry Thermal Oxidation(2020-11-13) Muduli, Manisha; Yamamoto, HiromichiDry thermal oxidation is performed at 900, 950, 1000, and 1050 C in fused silica tube furnace (Sandvik)for 10, 20, 50, 100 and 200 min. The properties of the thin-films such as refractive index and stress arestudied in this paper. Refractive indexes of the thin-films are obtained using ellipsometer and decreasesfrom 1.78 to 1.46 as the film thickness increases.Stress of the thin-films is theoretically calculated,experimentally measured and then compared.Stress is calculated theoretically using Stoney’s stressequation and Goklaney’s stress equations. Stress is measured experimentally using the profilometer. Ex-perimentally measured stress and refractive indexes are then compared to discuss the density of the thin-films.Publication Thin Dry Silicon Oxide Films Grown by Thermal Oxidation(2020-03-18) Muduli, Manisha; Yamamoto, HiromichiDry thermal oxidation is performed at 900, 950, 1000, and 1050 C in fused silica tube furnace (Sandvik) for 10, 20, 50, 100 and 200 min. The results are analyzed by both of Deal-Grove model and the method of Gerlach, Maser, and Saad. In the analysis using the Deal-Grove model, the parabolic rate constants are obtained to be 2.1, 7.0, 17.5, and 42.4 nm^2/min at 900, 950, 1000, and 1050 C, respectively, and the linear rate constants are acquired to be 0.2, 0.4, 0.8, and 1.2 nm/min at 900, 950, 1000, 1050 C, respectively. Activation energies of the parabolic rate constant and the linear rate constant are obtained to be 2.67 and 1.55 eV, respectively. In the analysis using the method of Gerlach, Maser, and Saad, the activation energy of the oxide growth rate for the 5 nm thick oxide is obtained to be 2.49 eV, whereas that for the 40 nm thick oxide to be 1.93 eV. The film uniformity for 30 nm thickness shows more than 10%, while that for the thickness of more than 50 nm indicates less than 5%, suggesting that measurement on 30 nm thick film using Filmetrics F50 is not still accurate enough. Standard deviation and coefficient of variation are also examined to discuss the film uniformity.