GaN Growth Rate Research Articles

In situ Pyrometric Interferometry For Molecular Beam Epitaxy of AlxGa1−xN on Si (111)

ABSTRACTWe apply pyrometric interferometery to control the growth of GaN, AlN, and AlxGa1−xN on Si (111) by molecular beam epitaxy. We show that the period of intensity oscillations in the temperature measured by the pyrometer is caused by optical interference in the growing nitride layer. The period of oscillations is inversely proportional to the growth rate. This effect is used for in-situ measurement of the growth rate of GaN, AlN, and AlxGa1−xN (for 0<×<1). We also show that pyrometric interferometery can be used for in-situ determination of the composition of the growing layer.

Influence of the Mg precursor on the incorporation of Mg in MOVPE grown GaN.

Incorporation of Mg in metalorganic vapour phase epitaxy (MOVPE) GaN has been investigated, using two different Mg precursors: bis-methylcyclopentadienyl magnesium [(MeCp)2Mg] and Solution bis-cyclopentadienyl magnesium [Solution Cp2Mg]. SIMS analysis reveals an increased (two fold) efficiency of Mg incorporation for Solution Cp2Mg as compared to (MeCp)2Mg. These results are attributed to the stronger interaction of (MeCp)2Mg with NH3, leading to the formation of alkylmagnesium amine adducts, and a reduced effective Mg surface concentration. A decreased GaN growth rate with increasing Mg fluxes is also reported for both precursors. This effect is more pronounced for Solution Cp2Mg indicating that incorporation of Mg in the lattice proceeds via the capture of Mg into group III sites, and that the supply of Mg from the surface is reduced in the case when (MeCp)2Mg is used.

Remote Plasma MOCVD Growth and Processing of GaN: A Study by Real Time Ellipsometry

In situ real time spectroscopic ellipsometry, X-ray photoelectron spectroscopy and optical emission spectroscopy are used to study the processes of sapphire substrate cleaning and nitridation upon exposure to rf plasma generated H- and N-atoms, respectively, at temperature as low as 200 °C. It is found that H-atoms are effective in removal carbon contamination and that the in-diffusion process of H-atoms can be controlled by ellipsometry. Moreover, it is found that remote plasma nitridation at 200 °C forms an homogeneous surface nitrided layer after ≈︂30 min of exposure. Finally, the study of the GaN growth rate for the low temperature N2 plasma assisted growth is reported.

Halogen-Transport Atomic-Layer Epitaxy of Cubic GaN Monitored by In Situ Gravimetric Method

In situ gravimetric monitoring (GM) of atomic-layer epitaxy (ALE) of cubic GaN on a GaN buffer layer/GaAs (001) is investigated using a halogen-transport system with GaCl and NH3 sources. The cubic GaN growth rate of one monolayer/cycle is obtained at a temperature ranging from 350 to 400°C. It is found that pure cubic GaN can be grown by halogen transport ALE. The growth rate decreases with increasing growth temperature, and a constant growth rate of about 0.45 is observed from 410 to 550°C. In this paper, it is shown that the in situ GM method is a powerful tool for understanding growth mechanism of the group III nitrides, as well as that of GaAs.

Coaxial rf-magnetron nitrogen activator for GaN MBE growth

Novel compact coaxial magnetron nitrogen activator with a radio frequency (rf) capacitively-coupled discharge has been used for the first time for GaN molecular beam epitaxial growth on different substrates, including GaAs(113). Optical emission spectra of the discharge have been studied as a function of nitrogen flow rate, rf power, and magnetic field, focusing on first negative (391 nm) and second positive (380 nm) lines associated with nitrogen molecular ions and excited molecules, respectively. Optimization of the activator parameters and distance of the discharge zone from the substrate resulted in GaN growth rate as high as 0.5 μm h −1 at a 350 l s −1 pumping speed.

プラズマ発光分光分析によるＧａＮ電子サイクロトロン共鳴プラズマ励起分子線エピタキシャル成長の観察

We observed excited species near a substrate during GaN growth by optical emission spectroscopy (OES). Both GaN growth rate and excited Ga emission intensity were studied by varying substrate bias conditions. Decrease in Ga emission intensity under high positive bias condition can be explained by suppression of Ga desorption. Thus, OES is considered to be a useful tool to understand GaN growth mechanism in ECR plasma-excited MBE.We also demonstrated GaN growth using helium-nitrogen mixed gas plasma. It was found that, compared with the GaN growth using nitrogen plasma without helium, the GaN growth rate was increased by using the mixed gas plasma.

Cubic GaN Heteroepitaxy on Thin-SiC-Covered Si(001)

We have investigated the growth conditions of cubic GaN (β-GaN) layers on very thin SiC-covered Si(001) by using gas-source molecular beam epitaxy as functions of SiC layer thickness, Ga-cell temperature and substrate temperature. Under the present SiC formation conditions on Si substrates by carbonization using C2H2 gas, the SiC layers with the thickness between 2.5 and 4 nm result in the epitaxial growth of β-GaN on thus SiC-formed Si substrates. At the highest GaN growth rate of 110 nm/h ( a Ga-cell temperature of 950 °C), β-GaN layers grown at a substrate temperature of 700 °C show a nearly flat surface morphology and the fraction of included hexagonal GaN becomes negligible when compared to the results of β-GaN layers grown under other conditions of Ga-cell and substrate temperatures. Thus obtained β-GaN films have good performance in photoluminescence intensity although the FWHM of band-edge recombination peak is still wider (137 meV) than the reported values for the β-GaN on 3C-SiC and GaAs.

The growth rate evolution versus substrate temperature and V/III ratio during GaN MBE using ammonia

The growth rate evolution versus V/III ratio and substrate temperature was studied by means of optical reflectivity during MBE of GaN layers using NH3 as nitrogen source. The GaN desorption becomes observable at temperatures above 800°C and causes the reduction of growth rate accompanied with the surface roughening at temperatures above 850-870°C. Unlike GaAs, which evaporates in accordance with the action mass law, the desorption rate of GaN is found to be almost independent of V/III ratio within the N-rich growth conditions. The activation energy for GaN desorption during the growth is found to be (3.2±0.1)eV. This value is very close to the activation energy for free evaporation. At V/III ratio values exceeding 200 the GaN growth rate reduction caused by violation of the molecular flow regime is observed. The Mg-doped samples grown under these extreme conditions tend to have improved acceptor activation and thus p-type conductivity.

Comparison between monomethyl hydrazine and ECR plasma activated nitrogen as a nitrogen source for CBE growth of GaN

Monomethyl hydrazine (MMHy) was investigated as a nitrogen source for GaN CBE growth in comparison to ECR plasma excited nitrogen sources. The growth rate of GaN on (0 0 1)GaAs was comparable with the highest growth rate by the ECR plasma excited nitrogen source, though the growth rate depends on the plasma gun structure. Unexpectedly, the carbon and oxygen contamination was more serious for the ECR plasma nitrogen source than for MMHy, especially for the ECR plasma without an ion removal magnet.

N-Limited Versus Ga-Limited Growth on ${\rm GaN}(000\bar{1})$ by MBE Using NH3

GaN was grown on [Formula: see text] by MBE using NH 3 and a Ga Knudsen cell. The growth kinetics on samples of this polarity were investigated with desorption mass spectroscopy (DMS) and reflection high energy electron diffraction (RHEED). Both techniques were used to observe and control surface termination, Ga condensation and surface temperature. GaN growth and decomposition rates were measured by DMS. Two stable surface terminations were found to exist — N-terminated and Ga-terminated [Formula: see text]. The N-terminated surface also contained hydrogen which desorbed during growth at a rate proportional to the growth rate. Low temperature reconstructions were only observed by adding weakly adsorbed Ga on top of the Ga-terminated surface. During growth two distinct growth regimes were identified: growth under excess NH 3 and growth under excess Ga. Growth is limited in both regimes by GaN decomposition at high temperatures with an activation energy of 3.4 eV. Growth in the excess Ga regime ceased below the Ga condensation temperature. Under conditions of excess NH 3, strong but damped oscillations in the specular RHEED intensity were observed on smooth surfaces. Contrary to previous suggestions, the period of these oscillations did not correspond exactly to integral layer deposition and was not characteristic of a narrow growth front. Further, the growth mode changed from island nucleation to step flow with an activation energy of 1.2 eV. Under conditions of excess Ga, the diffraction was 2-D but RHEED intensity oscillations were not observed, indicating a step flow growth mode. In this latter regime RHEED measurements were very sensitive to termination changes on the [Formula: see text] surface, and the growth rate was found to decrease linearly with increasing Ga flux. This reduction is explained by a model in which weakly adsorbed Ga blocks reaction at strongly bound Ga. A map is presented to provide a framework for categorizing the overall growth.

Dependence of GaN MOMBE growth on nitrogen source: ECR plasma gun structure and monomethyl-hydrazine

GaN layers were grown on (001)GaAs substrates by metal-organic molecular beam epitaxy (MOMBE), using electron cyclotron resonance (ECR) plasma activated nitrogen and monomethyl hydrazine (MMHy) as a nitrogen (N) source. We investigated dependence of GaN crystalline quality, growth rate and contamination, on the N source. It was found that the growth rate for a borosilicate glass guide tube was higher than that for quartz guide tube, suggesting that the quartz guide tube kills N radicals. Unexpectedly, higher oxygen and carbon contamination was detected for the layer grown with the ECR plasma gun without the ion extraction magnet and quartz, than those for the layer grown with MMHy.

Molecular beam epitaxy growth and properties of GaN, InGaN, and GaN/InGaN quantum well structures

Growth of III–V nitrides by molecular beam epitaxy (MBE) was studied using rf nitrogen plasma sources. Plasma sources from three different vendors have been tested. All three of the sources have been used to grow high quality GaN. However, the EPI rf source produces an optical emission spectrum that is very rich in the active nitrogen species of 1st-positive excited nitrogen molecules and nitrogen atoms. GaN growth rates at 800 °C of 1 μm/h have been achieved using this source. The MBE-grown GaN films are deposited homoepitaxially on high quality metalorganic vapor phase epitaxy-grown GaN/SiC substrates. With the growth conditions for high quality undoped GaN as a base line, a detailed study of Mg doping for p-type GaN was performed. An acceptor incorporation of 2×1019 cm−3 was measured by both capacitance–voltage and secondary ion mass spectroscopy for a doping source temperature of 290 °C. However, a faceted three-dimensional growth mode was observed by reflection high energy electron diffraction during Mg doping of GaN. Additional studies suggest an interdependence between Mg incorporation and growth surface morphology. Quantum well structures made from the InGaN ternary alloy were grown using a modulated beam MBE method. With this technique, quantum well compositions were controllable, grown with visible luminescence ranging from 400 to 515 nm depending on indium mole fraction. Light emitting diode test structures, combining Mg p-type doping with InGaN quantum wells, were fabricated and tested.

Analysis of reactor geometry and diluent gas flow effects on the metalorganic vapor phase epitaxy of AIN and GaN thin films on α(6H)-SiC substrates

The influence of diluent gas on the metalorganic vapor phase epitaxy of AlN and GaN thin films has been investigated. A computational fluid dynamics model using the finite element method was employed to improve film uniformity and to analyze transport phenomena. The properties of AlN and GaN thin films grown on α(6H)-SiC(0001) substrates in H2 and N2 diluent gas environments were evaluated. Thin films of AlN grown in H2 and N2 had root mean square (rms) roughness values of 1.5 and 1.8 nm, respectively. The surface and defect microstructures of the GaN thin films, observed by scanning and transmission electron microscopy, respectively, were very similar for both diluents. Low temperature (12K) photoluminescence measurements of GaN films grown in N2 had peak intensities and full widths at half maximum equal to or better than those films grown in H2. A room temperature Hall mobility of 275 cm2/V·s was measured on 1 µm thick, Si-doped, n-type (1×1017 cm−3) GaN films grown in N2. Acceptor-type behavior of Mg-doped GaN films deposited in N2 was repeatably obtained without post-growth annealing, in contrast to similar films grown in H2. The GaN growth rates were ∼30% higher when H2 was used as the diluent. The measured differences in the growth rates of AlN and GaN films in H2 and N2 was attributed to the different transport properties of these mixtures, and agreed well with the computer model predictions. Nitrogen is shown to be a feasible alternative diluent to hydrogen for the growth of AlN and GaN thin films.

Thermodynamic Analysis of Hydride Vapor Phase Epitaxy of GaN

A thermodynamic analysis of hydride vapor phase epitaxy (HVPE) is described for GaN. The partial pressures of gaseous species in equilibrium with GaN are calculated for temperatures, input GaCl partial pressures, input V/III ratios and mole fractions of hydrogen relative to the inert gas atoms. It is shown that the deposition of GaN is significantly influenced by the hydrogen mole fraction in the carrier gas. The growth rate is discussed in comparison with the experimental data reported in the literature. It is shown that the growth rate of GaN grown using HVPE is thermodynamically controlled.

Magnesium induced changes in the selective growth of GaN by metalorganic vapor phase epitaxy

Atmospheric pressure metalorganic vapor phase epitaxy is used to perform selective regrowth of GaN on a silicon nitride patterned mask, capping a GaN epitaxial layer deposited on (0001) sapphire substrate. The basic pattern is constituted by a 15 μm period hexagonal array of hexagonal openings in the mask, these openings being circumscribed into 10 μm diam circles. We investigate the effect of Mg incorporation on the growth anisotropy of the localized GaN islands by varying the ratio [Mg]/[Ga] of bis-methylcyclopentadienyl-magnesium and trimethylgallium partial pressures. Both undoped and Mg-doped GaN hexagonal pyramids, delimited by C (0001) and R (1 1̄01) facets, are achieved with a good selectivity. It is found that the GaN growth rates VR and VC, measured in the R〈11̄01〉 and C 〈0001〉 directions respectively, are drastically affected by the Mg incorporation. By adjusting the Mg partial pressure in the growth chamber, the VR/VC ratio can be increased so that the delimiting top C facet does not vanish as usually observed in undoped GaN localized growth, but by contrast expands.

Effect of Magnesium and Silicon on the lateral overgrowth of GaN patterned substrates by Metal Organic Vapor Phase Epitaxy

Metalorganic vapor phase epitaxy was used to achieve selective regrowth of undoped, Mg- and Si-doped GaN on a silicon nitride patterned mask, capping a GaN epitaxial layer deposited on (0001) sapphire substrate. Hexagonal openings in the mask defined into 10 µm diameter circles separated by 5µm were used as a pattern for the present study. Uniform undoped and Mg-doped GaN hexagonal pyramids, delimited by C (0001) and R {101} facets, were achieved with a good selectivity. Si-doped GaN hexagonal pyramids delimited by vertical {100} facets and (0001) top facet were obtained for a high SiH4 flow rate in the vapor phase. We found that the GaN growth rates VR and VC, measured in the R <101> and C <0001> directions respectively, were drastically affected by the Mg and Si incorporation. By adjusting the Mg partial pressure in the growth chamber, the VR/VC ratio can be increased. Hence, the delimiting top C facet do not vanish as usually observed in undoped GaN selective regrowth but conversely expands. On the other hand, under proper growth conditions, 20µm-high Si-doped GaN columns were obtained.

Cubic GaN Heteroepitaxy on Thin-SiC-Covered Si(001)

We have investigated the growth conditions of cubic GaN (β-GaN) layers on very thin SiC-covered Si(001) by using gas-source molecular beam epitaxy as functions of SiC layer thickness, Ga-cell temperature and substrate temperature. Under the present SiC formation conditions on Si substrates by carbonization using C2H2, gas, the SiC layers with the thickness between 2.5 and 4 nm result in the epitaxial growth of β-GaN on thus SiC-formed Si substrates. At the highest GaN growth rate of 110 nm/h (a Ga-cell temperature of 950 °C), β-GaN layers grown at a substrate temperature of 700 °C show a nearly flat surface morphology and the fraction of included hexagonal GaN becomes negligible when compared to the results of β-GaN layers grown under other conditions of Ga-cell and substrate temperatures. Thus obtained β-GaN films have good performance in photoluminescence intensity although the FWHM of band-edge recombination peak is still wider (137 meV) than the reported values for the β-GaN on 3C-SiC and GaAs.

Plasma characteristics and the growth of group III-nitrides by metalorganic molecular beam epitaxy

Improved quality and controllability of growth processes are key issues for the maturation of III-N technologies. One of the most important concerns for the growth of III-N materials in ultra high vacuum is the ability to provide an effective nitrogen flux to the growth surface. This work has sought to correlate radio frequency (rf) plasma parameters and their impact on the growth of GaN by plasma-assisted metalorganic molecular beam epitaxy. Utilizing optical emission spectrometry, the atomic nitrogen production has been optimized as a function of rf power and N2 flow rate. Growth experiments indicate that the abundance of atomic nitrogen alone does not control growth. Increasing energy per molecule in the rf source, with a constant level of atomic nitrogen, dramatically decreases the GaN growth rate. High levels of atomic nitrogen with a low energy per molecule resulted in restoration of the growth rate to ∼0.5 µm/h.

Arcjet plasma enhanced vapor phase epitaxy of GaN

This Letter reports on the vapor-phase epitaxy of GaN on basal plane oriented sapphire using a nitrogen arcjet plasma expanding into low pressure. A hydrogen-free and carbon-free growth environment was achieved by use of a pure Ga vapor source introduced downstream of the expanding plume by electron-assisted evaporation. Heteroepitaxial GaN growth rates of 8 μm/h were obtained in the studies reported. We believe these growth rates are the highest reported for a non-chemical decomposition process. The growth rates are attributed to the very high nitrogen atom fluxes achieved by arc dissociation of the molecular nitrogen source gas. The initial growth rates are believed to be well below those possible with a fully optimized system. The GaN films are characterized by UV-visible transmission spectroscopy and X-ray diffraction.

Studies of GaN layers grown on sapphire using an RF-source

Abstract We have studied the relation between MBE growth parameters and film quality after growth of GaN on the c -plane of sapphire using a conventional Ga-source and an RF-source for the excitation of N 2 . The growth rate was varied using Ga-flux, determined by RHEED oscillations on GaAs, r GaAs = 0.1–0.8 μ m/h. The N 2 -flux was 0.5–1.5 sccm. Most layers of GaN were grown at 760°C as measured by a pyrometer, while higher temperatures were utilised to determine the dismissal of growth. The RF-power and AlN buffer layer parameters were kept constant. The GaN-thicknesses were 0.2–1 μm/h. For low growth rates of ∼ 50 nm/h, there was a preferential growth of microcrystals in the growth direction which therefore overemphasised the measured film thickness. No film could be grown above 800°C as the desorption rate was too high. For a GaN growth rate of 300 nm/h the sticking coefficient was ∼ 85%. By varying the Ga- and N 2 -fluxes it was evident that the film quality, as provided by the photoluminescence spectra, strongly depended on the growth parameters. Photoluminescence peak intensities generally improved with film thickness below 1 μm. In 1 μm thick films, we observed excitonic related peaks close to the band gap as well as peaks 50–60 meV below due to the presence of defects or impurities. We found a strong correlation between growth parameters and optimum growth and therefore the highest layer quality could be obtained only when all parameters were carefully optimised.