Abstract:The key issue in GaN growth by radio‐frequency plasma‐assisted molecular beam epitaxy is the low growth rate compared with that obtained using an ammonia source. To reduce the processing time and to improve the crystalline quality of the epilayer, a high‐density radical source (HDRS) with high stability has been developed. The growth rate of the GaN epilayer was improved using the HDRS rather than a conventional radical source. During the growth, a sharp streak pattern obtained by reflection high‐energy electr… Show more
“…Though we have presented mainly pulsed metal deposition data, the observations made also have implications for conventional film growth with plasma sources. For instance, our results may explain an apparent inconsistency observed by Kawai et al 43 who, for a 1000 fold increase in active nitrogen species from a new MBE plasma source, were only able to achieve a factor of 3 increase in GaN growth rate. Beyond that increase they also observed gallium droplet formation.…”
It is shown that attractive electrostatic interactions between regions of positive charge in RF plasmas and the negative charge of metal wetting layers, present during compound semiconductor film growth, can have a greater influence than substrate temperature on film morphology. Using GaN and InN film growth as examples, the DC field component of a remote RF plasma is demonstrated to electrostatically affect metal wetting layers to the point of actually determining the mode of film growth. Examples of enhanced self-seeded nanopillar growth are provided in the case where the substrate is directly exposed to the DC field generated by the plasma. In another case, we show that electrostatic shielding of the DC field from the substrate can result in the growth of Ga-face GaN layers from gallium metal wetting layers at 490 C with root-mean-square roughness values as low as 0.6 nm. This study has been carried out using a migration enhanced deposition technique with pulsed delivery of the metal precursor allowing the identification of metal wetting layers versus metal droplets as a function of the quantity of metal source delivered per cycle. It is also shown that electrostatic interactions with the plasma can affect metal rich growth limits, causing metal droplet formation for lower metal flux than would otherwise occur. Accordingly, film growth rates can be increased when shielding the substrate from the positive charge region of the plasma. For the example shown here, growth rates were more than doubled using a shielding grid.
“…Though we have presented mainly pulsed metal deposition data, the observations made also have implications for conventional film growth with plasma sources. For instance, our results may explain an apparent inconsistency observed by Kawai et al 43 who, for a 1000 fold increase in active nitrogen species from a new MBE plasma source, were only able to achieve a factor of 3 increase in GaN growth rate. Beyond that increase they also observed gallium droplet formation.…”
It is shown that attractive electrostatic interactions between regions of positive charge in RF plasmas and the negative charge of metal wetting layers, present during compound semiconductor film growth, can have a greater influence than substrate temperature on film morphology. Using GaN and InN film growth as examples, the DC field component of a remote RF plasma is demonstrated to electrostatically affect metal wetting layers to the point of actually determining the mode of film growth. Examples of enhanced self-seeded nanopillar growth are provided in the case where the substrate is directly exposed to the DC field generated by the plasma. In another case, we show that electrostatic shielding of the DC field from the substrate can result in the growth of Ga-face GaN layers from gallium metal wetting layers at 490 C with root-mean-square roughness values as low as 0.6 nm. This study has been carried out using a migration enhanced deposition technique with pulsed delivery of the metal precursor allowing the identification of metal wetting layers versus metal droplets as a function of the quantity of metal source delivered per cycle. It is also shown that electrostatic interactions with the plasma can affect metal rich growth limits, causing metal droplet formation for lower metal flux than would otherwise occur. Accordingly, film growth rates can be increased when shielding the substrate from the positive charge region of the plasma. For the example shown here, growth rates were more than doubled using a shielding grid.
“…Recent report suggested that the surface structure of the GaN films are highly affected by the Ga/N flux (i.e. plasma power at constant Ga flux) [14]. Fig.…”
a b s t r a c tInfluence of active nitrogen species on surface and optical properties of homoepitaxial GaN films grown on GaN epilayers has been investigated. The epitaxial GaN films were grown at varying plasma powers (350e500 W) under identical growth conditions. High resolution X-Ray diffraction, Field Emission Scanning Electron Microscopy, Atomic Force Microscopy and Photoluminescence measurements were employed to characterize the structural, morphological and optical properties of the grown GaN films. High plasma power (500 W) lead to an increment in active nitrogen radicals and yielded high crystalline quality with reduced dislocations compared to low plasma power (350, 400 W) which divulge the presence of metallic gallium on the surface and low roughness. The valence band maximum position, electron affinity and ionization energy of the films increased with increment in plasma power. PL measurements revealed narrow and intense band to band edge emission with negligible defect related yellow band peak for the sample grown at 500 W. The analysis conveyed that higher amount of active nitrogen species encouraged good optical properties with insignificant defect states which can be employed for the fabrication of high performance optoelectronic & photovoltaic devices.
“…[7]) for transition 3 P2à 3 S1 at l0 =130.2 nm can be converted in s (1) (n0) = 1.83×10 -13 cm 2 that gives R (1) = 1.20x10 12 Hz. For the unresolved two-photon transition 3 P2à 3 P2,1,0 at l0 = 225.6 nm the (includes G (2) =2, see next section) [29] can be converted to s (2) (v0) = 4.84×10 -46 cm 4 s that gives R (2) = 6.24x10 4 Hz. Although the two-photon absorption is much less probable than the one-photon absorption, fluorescence signals are easily detectible.…”
Two-photon Absorption Laser Induced Fluorescence (TALIF) technique employing nanosecond lasers is often used to measure space-and time-resolved distributions of key atomic and molecular radicals in reactive environments such as plasmas and combustions.Although the technique was applied for about four decades, particularly in high pressure nonequilibrium plasmas accurate measurements of species densities remain challenging. With atomic oxygen as an example, central aspects of the technique including the role of photon statistics and line profiles on the two-photon absorption rates, selection rules, spatial and temporal resolutions, photolytic and quenching effects, and absolute calibration methods are discussed. Simulations using rate equations which include non-depletion regime, 3-level and 6-level models are compared and criteria for non-saturation regimes are given for low-and high-pressure plasmas. Solutions of the density-matrix model, which include coherent excitation and Stark detuning phenomena, and the rate equation model are compared. The validity criteria for non-depletion and photolytic-free regimes and rate models are given. The nanosecond TALIF quench-free regime at high laser intensities is investigated using the density-matrix model. The two-photon cross-sections for O, H and N atomic radicals and their ratio with Kr and Xe rare gases used for calibrations are revisited and recommendations are proposed. For TALIF applying ultrafast lasers, the appropriate model for the fluorescence probability is discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.