Library

Notes: Thick p-type porous 6H SiC layers were fabricated by anodization of p-type 6H SiC bulk crystals in dilute HF. Striking differences are observed in the reststrahl region room-temperature reflectance of these porous layers compared to that of bulk 6H SiC crystals. Instead of the single broad band reflectance spectrum typically observed in bulk 6H SiC, a two-band reflectance spectrum is observed. Several effective medium models, based on different morphologies of the component materials, 6H SiC and air, are used to obtain the frequency-dependent dielectric function of porous SiC from which calculated reflectance spectra are generated. The best match between measured and calculated spectra is obtained for a Maxwell–Garnett model with SiC acting as the host material and air cavities acting as the inclusion material. The model reproduces the two reflectance band structure observed in the measured reflectance of the porous SiC layers. The differences in the reststrahl region reflectance spectra of the porous SiC layers, compared to bulk SiC, are associated with polarization effects introduced by the cavities combined with a mean field average of interactions among the cavities. © 1996 American Institute of Physics.

Notes: The room-temperature infrared reflectance of AlN–GaN short period superlattice films has been measured. These superlattice films were deposited by switched atomic layer metalorganic chemical vapor deposition onto GaN or AlN buffer layers deposited on basal plane sapphire substrates. The measured reflectance spectra are compared to calculated spectra using an effective medium theory to model the dielectric function of the superlattice. The optical properties of the individual materials making up the samples are modeled with Lorentz oscillators using only bulk input parameters. The effects of film and substrate anisotropy and off-normal incidence are included in the calculation. Using this modeling technique, it is possible to obtain thickness estimates for the superlattice film and the buffer layer. The complicated structures seen in the reststrahl region reflectance of these films are also analyzed by comparison to the calculated spectra. © 1996 American Institute of Physics.

Notes: The room temperature electroluminescence spectra of moderately nitrogen doped (1×1017 cm−3) 3C and 6H SiC p+–n junctions are studied as a function of forward current. The free to bound transition due to the unintentional, deep boron (defect) center is dominant at low forward bias, while the free to bound transitions due to nitrogen donors and aluminum acceptors are dominant at higher forward biases. These results can be explained using Shockley–Read–Hall analysis of the recombination rate as a function of bias. The origin of boron related electroluminescence is suggested to be primarily from recombination within the depletion region, while aluminum and nitrogen related electroluminescence is attributed to recombinations in the minority carrier injection regions. © 1996 American Institute of Physics.

Notes: Cathodoluminescence of AlN–GaN short period superlattice films was measured at 6 K, 77 K, and room temperature. The superlattice films were deposited using a switched atomic layer metalorganic chemical vapor deposition process onto a buffer layer of either AlN or GaN, which was deposited on basal plane sapphire substrates. The individual AlN and GaN layers of the superlattice films ranged in thickness from 2.6 to 20.8 A(ring). The cathodoluminescence of these samples was measured at several electron acceleration voltages to allow depth profiling of the samples. This allows the region of the sample (superlattice film, buffer layer, and substrate) from which the spectral features originate to be determined. A spectral peak in the ultraviolet region above the 3.5 eV band gap of GaN has been observed in all the superlattice samples studied to date. Our results indicate that the location of this peak is determined by quantum confinement in the GaN layers. © 1996 American Institute of Physics.

Notes: It is shown that n+ and/or p+ contacts on p-i-n diodes can function as solid-state photoemitters at temperatures (approximately-less-than)20 K. Infrared radiation can excite electrons or holes over small n-i or p-i interfacial barriers and into the intrinsic region when the diode is forward biased. Photoelectric thresholds in the far infrared corresponding to 37 and 61 μm cutoffs have been observed for silicon devices using a Fourier transform spectrometer. Suggestions are made to tailor the cutoff wavelengths using different concentrations of various impurities near the metal-insulator transition. Epitaxially grown multilayered (superlattice) detectors are proposed.

Notes: Two distinct boron-related centers are known in silicon carbide polytypes, one shallow (ionization energy ∼300 meV) and the other deep (∼650 meV). In this work, 4H SiC homoepitaxial films are intentionally doped with the shallow boron center by controlling the silicon to carbon source gas ratio during chemical vapor deposition, based on site competition epitaxy. The dominance of the shallow boron center for samples grown with a low Si/C ratio, favoring the incorporation of boron onto the silicon sublattice, is verified by the temperature dependent Hall effect, admittance spectroscopy and deep level transient spectroscopy. In these samples a peak near 3838 Å appears in the low temperature photoluminescence spectrum. Further experiments support the identification of this peak with the recombination of a four particle (bound exciton) complex associated with the neutral shallow boron acceptor as follows: (1) The intensity of the 3838 Å peak grows with added boron. (2) Momentum conserving phonon replicas are observed, with energies consistent with other four particle complexes in SiC. (3) With increasing temperature excited states are observed, as for the neutral aluminum and gallium acceptor four particle complexes. However, the intensity of the shallow boron spectrum is quenched at lower temperatures than the corresponding spectra for Al and Ga, and the lineshapes are strongly sample dependent. These results may be related to the unusual configurational and electronic structure of this center inferred from recent spin resonance experiments by other groups. When the Si/C ratio is high, the optical signatures of the deep boron center, nitrogen-boron donor-acceptor pairs and conduction band to neutral acceptor free-to-bound transitions, are observed in the photoluminescence. At T=2 K well resolved, detailed nitrogen-boron pair line spectra are observed in addition to the peak due to distant pairs. As the temperature is raised, the donor-acceptor pair spectrum decreases in intensity while the free-to-bound no-phonon peak appears. Extrapolation of the temperature dependence of the free-to-bound peak to T=0 K, after correction for the temperature dependence of the exciton energy gap, leads to the value EA(B)−EX=628±1 meV, where EA(B) is the ionization energy of the deep boron center and EX is the binding energy of the free exciton which, for 4H SiC, can only be estimated at this time. © 1998 American Institute of Physics.

Notes: We report the values of the absorption coefficient of 4H SiC at room temperature, in the wavelength range from 3900 to 3350 Å and at 3250 Å. By using the known shift in the band gap with temperature, we also present an estimate of the absorption coefficient of 4H SiC at 2 K. © 1998 American Institute of Physics.

Notes: The Hall scattering coefficient rH determines the relationship between the measurable Hall coefficient RH and the free carrier concentration. Reliable knowledge of rH is necessary for the precise interpretation of Hall measurements and to validate theoretical transport calculations. We have measured the Hall scattering factor in nitrogen doped 4H and 6H epitaxial layers from 35 to 290 K in magnetic fields up to 9 T. The effective Hall scattering factor varies between 0.91 and 1.21 in 4H SiC and 0.84 and 1.02 in 6H SiC. The effect of the Hall scattering factor on dopant activation energies obtained from Hall measurements is not large enough to explain the difference between dopant activation energies obtained from Hall effect and infrared absorption measurements. © 1998 American Institute of Physics.

Notes: A long minority-carrier diffusion length and the transmission of Alx Ga1−x As luminescence through Alx Ga1−x As layers are identified as two processes causing the excitation of GaAs spectra through thick Alx Ga1−x As layers, as well as contributing to enhancements in low-temperature photoluminescence intensity observed in Alx Ga1−x As layers without GaAs substrates. A simple model for intensity enhancement due to below-band-gap photon recycling is introduced to explain the observed enhancements. Some features that uniquely distinguish below-band-gap photon recycling from the better known room-temperature process are presented.

Notes: Hall effect measurements in a Hall-bar configuration are performed on nitrogen-doped n-type bulk 4H, 6H, and 15R SiC single crystals cut into small parallelepipeds with their longest edges either parallel or perpendicular to the cˆ axis. In the temperature range investigated (40–700 K), an anisotropy of the electron Hall mobility is observed in all three polytypes. While the mobility perpendicular to the cˆ axis—with magnetic field perpendicular or parallel to the cˆ axis—is greater than the mobility parallel to the cˆ axis for 6H and 15R SiC, 4H SiC shows the opposite behavior. © 1994 American Institute of Physics.