Papers by Author: Fei Yan

Abstract: The C-C stretch vibration associated with the dicarbon antisite in 4H SiC has been observed out to the fifth harmonic in the low temperature photoluminescence spectrum. The anharmonicity is accounted for reasonably well by fits to the data based on the Morse potential. We combine the observations from experiment, the analytically tractable Morse potential, and results obtained from first principles calculations on this defect to obtain an estimate of the thermal expansion coefficient of the C-C bond. This local thermal expansion coefficient is considerably smaller than the linear thermal expansion coefficient of bulk 4H SiC, in striking contrast with the recent result for the nitrogen-vacancy center in diamond that the local thermal expansion coefficient is larger than the bulk value.

Abstract: The analysis of the donor-acceptor pair luminescence of P-Al and N-Al pairs obtained recently for the cubic 3C polytype of SiC is viewed in some detail. A detailed consideration is given to the fitting procedure applied to the P-Al and N-Al spectra. Fit with theoretical models of spectra of type I and type II are applied to both N-Al and P-Al experimental spectra, and it is demonstrated that only contribution from P on Si site is observable in the presented samples. The accuracy of the obtained phosphorus ionization energy of 48.1 meV is also discussed.

Abstract: Low doped epitaxial films of 4H SiC irradiated at 1 MeV and electron fluences of 1×1015 cm-2, 3×1015 cm-2 and 1×1016 cm-2 have been used to measure temperatures in two temperature intervals: 580°C to 640°C and 1220°C to 1320°C. Possible accuracy of the temperature measurements is judged to be better than 10 degrees Centigrade. Similar measurements should be possible from 100°C to 1500°C.

Abstract: Ultra-precision machining is dominated by single-crystal diamond cutting tools, and is typically applied to a narrow range of materials, particularly aluminum and copper. Single-crystal SiC can be comparable to some diamonds in hardness and thermal conductivity, while potentially having superior chemical and thermal stability, yet it has not been explored as a cutting tool for ultra-precision machining. We made two cutting tools with single-crystal SiC, one with sharp corners and one with a large circular radius, and used them to cut flat surfaces on two materials, 316 stainless steel and nickel. These materials generally cause unacceptably rapid diamond tool wear. We report the average roughness of the resulting surfaces cut with single-crystal 4H and 6H SiC tools.

Abstract: Aluminum-doped 4H-SiC samples were either irradiated with high-energy electrons (170 keV or 1 MeV) or implanted with a box-shaped He+-profile. Admittance spectroscopy (AS) and deep level transient spectroscopy (DLTS) were employed to search for defect centers. AS spectra of as-grown as well as of electron-irradiated (170 keV or 1 MeV) 4H-SiC epilayers reveal the Al acceptor (ΔE(Al) = 200 meV) and an unknown defect (ΔE(SB) = 177 meV), while AS spectra of the He+-implanted and annealed sample show in addition to the Al-acceptor two energetically deeper acceptor-like defect centers (ΔE(RE3) = 255 meV and ΔE(KR3) = 375 meV). The KR3-center is not directly formed by the He+-implantation, it requires an annealing process. The DLTS spectra of the He+-implanted and annealed sample resolve a double-peak structure of the KR3-defect (ΔE(KR3A) = 380 meV and ΔE(KR3B) = 410 meV).

Abstract: This paper comprises a systematic study of the thermal stability of defect centers observed in n- and p-type 4H-SiC by deep level transient spectroscopy (DLTS); the defects are generated by irradiation with high-energy electrons of 170 keV or 1 MeV.

Abstract: In this paper we describe an effort to find correlations between low temperature photoluminescence spectroscopy (LTPL) and deep level transient spectroscopy (DLTS) of electron irradiated samples annealed from 25 °C to 1700 °C in 100 °C steps. We report on thermal histories of defect centers created by 170 keV and 1 MeV electron irradiation, as observed by LTPL only. The DLTS results on "twin" samples are presented in a separate paper. Our results indicate that in n-type 4H SiC there is no correlation between the Z1/Z2 center in DLTS and the L1 peak of the DI center seen in LTPL. In p-type 4H SiC we do not find a correlation between a 350 meV DLTS peak above the valence band and the LTPL L1 peak of the DI center. Consequently, we cannot find evidence for a 350 meV ground state postulated in the “Pseudo–Donor” model [3].

Abstract: In this paper we revisit sharp low temperature luminescence lines (LTPL) previously generated by high dose 1018 to 1020 cm-2 electron beams in an electron microscope and now produced by low dose 1015 cm-2 electron, 5x1010 cm-2 proton and helium ion irradiation. New no phonon lines E0, F0, θ0, Φ0, K0, G0, J0, M0 and phonon replicas are found. Phonon replicas up to the fifth harmonic are well accounted for by theory giving convincing new evidence that the di-carbon antisite is responsible for these deep defect lines.

Abstract: This article discusses the electrochemical polishing of silicon-face p-type 4H SiC using diluted aqueous HF solution. Etchings on the silicon and carbon faces of SiC samples are performed and compared. The experimental results show that the RMS surface roughness of the electrochemically polished Si-face could be as low as about 2 nm. Carbon-face electrochemical polishing gives a rougher surface. Therefore, silicon-face 4H SiC is a better candidate for MEMS processing. The underlying mechanism is also discussed.

Abstract: The effect of the Schottky barrier height on the detection of the midgap defects EH6 and EH7 in 4H-SiC by deep level transient spectroscopy (DLTS) is systematically studied. The results show that the DLTS peak height - and as a consequence the observed defect concentration - increases with the increasing barrier height and saturates above 1.5 eV for EH6 and above 1.7 eV for EH7, while below 1.1 eV the DLTS peak height completely disappears. A model is applied, which determines the position of the quasi-Fermi level in the space charge region as a function of the barrier height and of the reverse bias applied and which explains the variation of the DLTS peak height.