Kikurou Takemoto

Preparation of Large Freestanding GaN Substrates by Hydride Vapor Phase Epitaxy Using GaAs as a Starting Substrate

A freestanding GaN substrate over 2 inches in size was successfully prepared for the first time by hydride vapor phase epitaxy (HVPE) using GaAs as a starting substrate. In the experiment, a GaAs (111)A substrate with a SiO2 mask pattern on its surface was used. A thick GaN layer was grown on the GaAs substrate at 1030°C through the openings in the SiO2 mask. By dissolving the GaAs substrate in aqua regia, a freestanding GaN substrate about 500 µm thick was obtained. The full-width at half maximum (FWHM) in the ω-mode X-ray diffraction (XRD) profile of GaN (0002) plane was 106 arcsec. The dislocation density of the GaN substrate obtained was determined to be as low as 2×105 cm-2 by plan-view transmission electron microscopy (TEM). Hall measurements revealed the n-type conductivity of the GaN substrate with typical carrier concentration and carrier mobility of 5×1018 cm-3 and 170 cm2V-1s-1, respectively.

Preparation of large GaN substrates

Abstract A freestanding GaN substrate of over 2-in. size with low dislocation density was prepared by hydride vapor phase epitaxy (HVPE) using GaAs (111)A as a starting substrate. A SiO 2 mask pattern with round openings was formed directly onto the GaAs (111)A substrate. Then, a thick GaN layer was grown with numerous large hexagonal inverse-pyramidal pits constructed mainly by {11–22} facets maintained on the surface. After removing the GaAs substrate and subsequent lapping and polishing, a freestanding GaN about 500 μm in thickness was obtained. Etch pit observation reveals that etch pit groups with etch pit density 2×10 8 cm −2 at the center exist in the matrix area with etch pit density as low as 5×10 5 cm −2 . This distribution is due to the effect of large hexagonal pits on collecting dislocations at the bottom of the hexagonal pit. Dislocations propagate into the bottom of the pit mainly in the 〈11–20〉 or 〈1–100〉 direction parallel to (0001).

Thermodynamics on tri-halide vapor-phase epitaxy of GaN and InxGa1-xN using GaCl3 and InCl3

Abstract Thermodynamic analyses on tri-halide vapor-phase epitaxy (THVPE) of GaN and In x Ga 1− x N using GaCl 3 and InCl 3 are described. The partial pressures of gaseous species in equilibrium with GaN and In x Ga 1− x N and the driving force for the deposition are calculated for growth temperature and the hydrogen mole fraction in the carrier gas. It is shown that the deposition of GaN and In x Ga 1− x N is significantly influenced by the hydrogen mole fraction in the carrier gas. The vapor-solid distribution relationship of In x Ga 1− x N alloy deposition using GaCl 3 and InCl 3 is discussed in comparison with the experimental data reported in the literature. It is shown that the solid composition x in In x Ga 1− x N is thermodynamically controlled.

Thermodynamics on halide vapor-phase epitaxy of InN using incl and InCl3

AbstractA thermodynamic analysis on halide vapor-phase epitaxy of InN using InCl or InCl 3 as In sources is described. Theequilibrium partial pressures of gaseous species and the driving force for the deposition are calculated for growthtemperature. It is shown that the deposition of InN is very di†cult using InCl. On the contrary, by using InCl 3 , thedeposition is possible under a condition of inert carrier gas or mixed carrier gas of hydrogen and inert gas. In the InCl 3 system, hydrogen in the carrier gas plays a crucial role for the deposition of InN. These results agree well with theexperimental data reported in the literature. # 2001 Elsevier Science B.V. All rights reserved. PACS: 81.05.Ea; 81.10.Bk; 81.15.Kk; 05.70.yaKeywords: InN; Halide VPE; InCl; InCl 3 ; Thermodynamic analysis 1. IntroductionGroup III nitrides have attracted much atten-tion as materials for optoelectronic and high-temperature device applications. Among thesenitrides, In x Ga 1yx N is a key material for thefabrication of light emitters covering the wave-length from ultraviolet to red range. To date,however, the high-quality material of red rangehas not been obtained, because the epitaxialgrowth of InN is very di†cult probably due toits poor thermal stability and the lack of a suitablesubstrate material.In the previous papers [1,2], we showed experi-mentally that the vapor-phase epitaxy (VPE) ofInN was possible at the growth temperature ashigh as 7508C using InCl

Growth of GaN directly on Si(111) substrate by controlling atomic configuration of Si surface by metalorganic vapor phase epitaxy

The direct growth of a GaN epitaxial layer on a Si(111) substrate by metalorganic vapor phase epitaxy (MOVPE) was performed using a low-temperature (LT)-GaN buffer layer with no Al-containing intermediate layer (e.g., AlN or AlGaN). No deterioration in the Si surface caused by the reaction between Si and Ga vapor was observed. However, when there were Ga droplets on the surface, Ga and Si formed a Ga?Si alloy, which caused the generation of numerous holes on the surface by melt-back etching at high temperatures. In addition, it was revealed that the coverage of the LT-GaN buffer layer on Si was strongly affected by the hydrogen (H2) partial pressure in the carrier gas. Using nitrogen (N2) carrier gas, a complete coverage of the LT-GaN buffer layer could be achieved directly over the Si surface. These features can be explained by the facts that the Si surface is partially terminated by hydrogen atoms and the coverage of hydrogen on Si surface depends on H2 partial pressure.

In Situ Gravimetric Monitoring of Decomposition Rate from GaN Epitaxial Surface

Decomposition of GaN on its surface was investigated under atmospheric pressure using the in situ gravimetric monitoring (GM) method. Weight change of the GaN substrate both in the H2 carrier gas and in the He carrier gas ambient was monitored at temperatures ranging from 600°C to 950°C with and without the presence of NH3 flow. It was found that the GaN decomposition did not occur in the presence of NH3 flow both in the H2 carrier gas and in the He carrier gas. However, without NH3 flow, the decomposition rate of GaN drastically increased as the temperature increased in the H2 carrier gas, whereas the decomposition of GaN was negligible in the He carrier gas. Dependence of the decomposition rate on the H2 partial pressure in the carrier gas (PH2 ) was also investigated, and it was found that the decomposition rate is proportional to the PH2 3/2. These results indicate that the decomposition is governed by the reaction of GaN(surface)+3/2H2(g)→Ga(surface)+NH3(g).

Growth of Thick Hexagonal GaN Layer on GaAs (111)A Surfaces for Freestanding GaN by Metalorganic Hydrogen Chloride Vapor Phase Epitaxy

Thick hexagonal GaN was grown on GaAs (111)A surfaces by metalorganic hydrogen chloride vapor phase epitaxy (MOHVPE) in the temperature range from 920°C to 1000°C. Both the surface morphology and the photoluminescence (PL) property of the grown layer were greatly improved with increase of the growth temperature up to 1000°C. However, the full-width at half maximum (FWHM) in the ω mode X-ray diffraction (XRD) of the GaN (0002) plane increased with increasing growth temperature above 960°C, due to the bending of the grown layer. The bending could be suppressed by growing a thicker layer, even at 1000°C. A mirror-like GaN layer with the FWHM value of 4.7 min was obtained by growing a 100-µm-thick layer at 1000°C, which indicates that the growth of a thick GaN layer on the GaAs (111)A surface is a promising method for the preparation of freestanding GaN substrates.

Influence of Polarity on Surface Reaction between GaN{0001} and Hydrogen

The influence of polarity on GaN decomposition has been investigated by an in situ gravimetric monitoring (GM) method using freestanding GaN(0001). The decomposition rate of the GaN was measured as a function of P H2 in the temperature ranging from 800 to 950 °C. In the low-temperature region, the decomposition rate of GaN(0001) is faster than that of GaN(0001), where the decomposition rates of both surfaces are proportional to P 3/2 H2 . On the other hand, the decomposition rate of GaN(0001) is faster than that of GaN(0001) in the high-temperature region. In this case, the decomposition rates of both surfaces are proportional to P 1/2 H2 These results indicate that the rate-limiting reactions of GaN decomposition can be written as follows: N(surface) + 3/2H 2 (g) → NH 3 (g) at lower temperatures, and Ga(surface) + 1/2H 2 (g) → GaH(g) at higher temperatures.

Growth of Fe-Doped Thick GaN Layers for Preparation of Semi-Insulating GaN Substrates

Vapor-phase epitaxy of Fe-doped thick GaN layers was performed using GaCl, NH3 and FeCl2 as source gases on (0001) sapphire and (111)A GaAs substrates with the aim of preparing semi-insulating (SI) GaN substrates. On sapphire, the resistivity of the GaN layer increased with increasing FeCl2 input partial pressure. A 12-µm-thick SI GaN layer showing a resistivity of 3.0×109 Ω cm at room temperature was successfully grown by compensating for background donors. In contrast, the resistivity of the GaN layers grown on the GaAs substrate remained low (on the order of 10-2 Ω cm) even though the same growth conditions were used as on the sapphire substrate. Secondary ion mass spectroscopy (SIMS) measurements suggested that the presence of As vapor-species caused by the degradation of the substrate hindered Fe from becoming incorporated into GaN.

Fabrication of Semi-Insulating GaN Wafers by Hydride Vapor Phase Epitaxy of Fe-Doped Thick GaN Layers Using GaAs Starting Substrates

The hydride vapor phase epitaxy of a 400-µm-thick GaN layer doped with Fe was performed on a (111)A GaAs starting substrate covered by a NiTi protective layer on its back. By removing the GaAs substrate after GaN growth, a semi-insulating (SI) (0001) GaN wafer showing a resistivity of 8.8×1012 Ω cm at room temperature was successfully fabricated. A secondary ion mass spectrometry (SIMS) measurement revealed that doping with an Fe concentration of 1.5×1019 cm-3 was achieved. The full-width at half-maximum (FWHM) of X-ray diffraction (XRD) rocking curves of the (0002) and (1010) planes of GaN did not vary with Fe doping. The etch-pit density (EPD) of a SI GaN wafer was evaluated to be 8×106 cm-2.