Phase-separation suppression in GaN-rich side of GaNP alloys grown by metal-organic chemical vapor deposition

The GaN-rich side of GaNP ternary alloys has been successfully synthesized by light-radiation heating and low-pressure metal-organic chemical vapor deposition. X-ray diffraction (XRD) rocking curves show that the ( 0002) peak of GaNP shifts to a smaller angle with increasing P content. From the GaNP photoluminescence (PL) spectra, the red shifts from the band-edge emission of GaN are determined to be 73, 78 and 100 meV, respectively, in the GaNP alloys with the P contents of 1.5%, 5.5% and 7.5%. No PL peak or XRD peak related to GaP is observed, indicating that phase separation induced by the short-range distribution of GaP-rich regions in the GaNP layer has been effectively suppressed. The phase-separation suppression in the GaNP layer is associated with the high growth rate and the quick cooling rate under the given growth conditions, which can efficiently restrain the accumulation of P atoms in the GaNP layer.

Optical properties of phase-separated GaN1-xPx alloys grown by light-radiation heating metal-organic chemical vapour deposition

Based on the results of the temperature-dependent photoluminescence (PL) measurements, the broad PL emission in the phase-separated GaNP alloys with P compositions of 0.03, 0.07, and 0.15 has investigated. The broad PL peaks at 2.18, 2.12 and 1.83 eV are assigned to be an emission from the optical transitions from several trap levels, possibly the iso-electronic trap levels related to nitrogen. With the increasing P composition (from 0.03 to 0.15), these iso-electronic trap levels are shown to become resonant with the conduction band of the alloy and thus optically inactive, leading to the apparent red shift (80-160meV) of the PL peak energy and the trend of the red shift is strengthened. No PL emission peak is observed from the GaN-rich GaNP region, suggesting that the photogenerated carriers in the GaN-rich GaNP region may recombine with each other via non-radiation transitions.

Effects of the wave function localization in AlInGaN quaternary alloys

Using the first-principles band-structure method and the special quasirandom structures approach, the authors have investigated the band structure of random AlxInyGa1-x-yN quaternary alloys. They show that the wave functions of the band edge states are more localized on the InN sites. Consequently, the photoluminescence transition intensity in the alloy is higher than that in GaN. The valence band maximum state of the quaternary alloy is also higher than GaN with the same band gap, indicating that the alloy can be doped more easily as p-type. (c) 2007 American Institute of Physics.

The growth of high-Al-content InAlGaN quaternary alloys by RF-MBE

High-Al-content InxAlyGa1-x-yN (x = 1-10%, y = 34-45%) quaternary alloys were grown on sapphire by radio-frequency plasma-excited molecular beam epitaxy. Rutherford back-scattering spectrometry, high resolution x-ray diffraction and cathodoluminescence were used to characterize the InAlGaN alloys. The experimental results show that InAlGaN with an appropriate Al/In ratio (near 4.7, which is a lattice-match to the GaN under-layer) has better crystal and optical quality than the InAlGaN alloys whose Al/In ratios are far from 4.7. Some cracks and V-defects occur in high-Al/In-ratio InAlGaN alloys. In the CL image, the cracks and V-defect regions are the emission-enhanced regions.

Photoluminescence from the nitrogen-perturbed above-bandgap states in dilute GaAs1-xNx alloys: A microphotoluminescence study

Using microphotoluminescence (mu-PL), in dilute N GaAs1-xNx alloys, we observe a PL band far above the bandgap E-0 with its peak energy following the so-called E+ transition, but with contribution from perturbed GaAs host states in a broad spectral range (> 100 meV). This finding is in sharp contrast to the general understanding that E+ is associated with a well-defined conduction band level (either L-1c or N-x). Beyond this insight regarding the strong perturbation of the GaAs band structure caused by N incorporation, we demonstrate that a small amount of isoelectronic doping in conjunction with mu-PL allows direct observation of above-bandgap transitions that are not usually accessible by PL.

Temperature dependence of the formation of nano-scale indium clusters in InAlGaN alloys on Si(111) substrates

The temperature dependence of the formation of nano-scale indium clusters in InAlGaN quaternary alloys, which are grown by metalorganic chemical vapour deposition on GaN/Si(111) epilayers, is investigated. Firm evidence is provided to support the existence of phase separation, or nano-scale In-rich clusters, by the combined results of high-resolution transmission electron microscopy (HRTEM), high-resolution x-ray diffraction (HRXRD) and micro-Raman spectra. The results of HRXRD and Raman spectra indicate that the degree of phase separation is strong and the number of In clusters in the InAlGaN layers on silicon substrate is higher at lower growth temperatures than that at higher growth temperatures, which limits the In and Al incorporated into the InAlGaN quaternary alloys. The detailed mechanism of luminescence in this system is studied by low temperature photoluminescence (LT-PL). We conclude that the ultraviolet (UV) emission observed in the quaternary InAlGaN alloys arises from the matrix of a random alloy, and the second emission peak in the blue-green region results from the nano-scale indium clusters.

A study of the growth and optical properties of AlInGaN alloys

We have studied the growth and optical properties of AlInGaN alloys in this article. By the measurement of three samples, we found that the incorporation of In decreases with the increase of temperature, while there is nearly no change for the incorporation of Al. The sample grown at the lowest temperature had the best material and optical properties, which owes to the high In component, because the In component can reduce defects and improve the material quality. We also used the time-resolved photoluminescence(PL) to study the mechanism of recombination of carriers, and found that the time dependence of PL intensity was not in exponential decay, but in stretched-exponential decay. Through the study of the character of this decay, we come to the conclusion that the emission comes from the recombination of localized excitons. Once more, this localization exhibites the character of quantum dots, and the stretched, exponential decay results from the hopping of carriers between different localized states. In addition, we have used the relation of emission energy dependence of carrier's lifetime and the character of radiative recombination and non-radiative combination to confirm our conclusion.

Time-resolved photoluminescence studies of AlInGaN alloys

We study the two samples of AIInGaN, i.e., 1-mum GaN grown at 1030degreesC on the buffer and followed by a 0.6-mum-thick epilayer of AIInGaN under the low pressure of 76 Torr and the AIInGaN layer deposited directly on the buffer layer without the high-temperature GaN layer, by temperature-dependent photoluminescence (PL) spectroscopy and picosecond time-resolved photoluminescence (TRPL) spectroscopy. The TRPL signals of both the samples were fitted well as a stretched exponential decay at all temperatures, indicating significant disorder in the material. We attribute the disorder to nanoscale quantum dots or discs of high indium concentration. Temperature dependence of dispersive exponent beta shows that the stretched exponential decay of the two samples comes from different mechanisms. The different depths of the localization potential account for the difference, which is illustrated by the results of temperature dependence of radiative recombination lifetime and PL peak energy.

GaN1-xPx ternary alloys with high P composition grown by metal-organic chemical vapor deposition

GaN1-xPx ternary alloys with high P compositions were deposited on sapphire substrates by means of metal-organic chemical vapor deposition. Depth profiles of the elements indicate that the maximum P/N composition ratio is about 17% and a uniform distribution of the P atoms in the alloys is achieved. 2theta/omega XRD spectra demonstrate that the (0002) peak of the GaN1-xPx alloys shifts to smaller angle with increasing P composition. From the photoluminescence (PL) spectra, the red shifts to the bandedge emission of GaN are determined to be 73, 78, 100 and 87 meV for the GaN1-xPx alloys with the P/N composition ratios of 3%, 11%, 15% and 17%, respectively. No PL peak related to GaP is observed, indicating that the phase separation between GaN and GaP is well suppressed in our GaN1-xPx samples. (C) 2003 Elsevier Science B.V. All rights reserved.

Alloy states in dilute GaAs1-xNx alloys (x < 1%)

A set of GaAs1-xNx samples with small nitrogen composition (x<1%) were investigated by continuous-wave photoluminescence (PL), pulse-wave excitation PL, and time-resolved PL. In the PL spectra, an extra transition located at the higher-energy side of the commonly reported N-related emissions was observed. By measuring the PL dependence on temperature and excitation power, the PL peak was identified as a transition of alloy band edge-related recombination in GaAsN. The PL dynamics further confirms its intrinsic nature as being associated with the band edge rather than N-related bound states. (C) 2003 American Institute of Physics.

Photoluminescence from ZnS1-xTex alloys under hydrostatic pressure

ZnS1-xTex (0.02less than or equal toxless than or equal to0.3) alloys are studied by photoluminescence under hydrostatic pressure at room temperature. Only a wide emission band is observed for each sample. Its peak energy is much lower than the corresponding band gap of alloys. These bands are ascribed to the radiative annihilation of excitons bound at Te-n(ngreater than or equal to2) isoelectronic centers. The pressure coefficients of the emission bands are smaller than those of alloy band gaps from 48% to 7%. The difference of the pressure coefficient of the emission band and the band gap increases when the binding energy of Te-n centers decreases. It seems contrary to our expectation and needs further analysis. The integrated intensities of emission bands decrease with increasing pressure due to the decreasing of the absorption coefficient associated with the Te-n centers under pressure. According to this model the Stokes shifts between the emission and absorption bands of the Te-n centers are calculated, which decrease with the increasing Te composition in alloys.

Metalorganic chemical vapor deposition of GaNAs alloys using different Ga precursors

The GaNAs alloys have been grown by metalorganic chemical vapor deposition (MOCVD) using dimethylhydrazine (DMHv) as the nitrogen precursor, triethylgallium (TEGa) and trimethylgallium (TMGa) as the gallium precursors, respectively. Both symmetric (004) and asymmetric (1 1 5) high-resolution X-ray diffraction (HRXRD) were used to determine the nitrogen content in GaNAs layers. Secondary ion mass spectrometry (SIMS) was used to obtain the impurity content. T e influence of different Ga precursors on GaNAs quality has been investigated. Phase separation is observed in the < 1 1 5 > direction when using TMGa as the Ga precursor but not observed when using TEGa. This phenomenon should originate from the parasitic reaction between the Ga and N precursors. Furthermore. samples grown with TEGa have better quality and less impurity contamination than those with TMGa. Nitrogen content of 5.742% has been achieved using TEGa and no phase separation observed in the sample. (C) 2002 Elsevier Science B.V. All rights reserved.

Thermal redistribution of localized excitons and its effect on the luminescence band in InGaN ternary alloys

Temperature-dependent photoluminescence measurements have been carried out in zinc-blende InGaN epilayers grown on GaAs substrates by metalorganic vapor-phase epitaxy. An anomalous temperature dependence of the peak position of the luminescence band was observed. Considering thermal activation and the transfer of excitons localized at different potential minima, we employed a model to explain the observed behavior. A good agreement between the theory and the experiment is achieved. At high temperatures, the model can be approximated to the band-tail-state emission model proposed by Eliseev et al. [Appl. Phys. Lett. 71, 569 (1997)]. (C) 2001 American Institute of Physics.

The effects of pre-irradiation on the formation of Si1-xCx alloys

Carbon ions were implanted into crystal Si to a concentration of (0.6-1.5)at% at room temperature. Some samples were pre-irradiated with S-29(i)+ ions, while others were not pre-irradiated. Then the two kinds of samples were implanted with C-12(+) ions simultaneously, and Si1-xCx alloys were grown by solid phase epitaxy with high-temperature annealing. The effects of preirradiation on the formation of Si1-xCx alloys were studied. If the dose of implanted C ion was less than that for amorphizing Si crystals, the implanted C atoms would like to combine with defects produced during implantation, and then it was difficult for Si1-xCx alloys to form after annealine, at 950 degreesC. Pre-irradiation was advantageous for Si1-xCx alloy formation. With the increase of C ion dose, the damage produced by C ions increased. Pre-irradiation was unfavorable for Si1-xCx, alloy formation. If the implanted C concentration was higher than that for solid phase epitaxy solution, only part of the implanted C atoms form Si1-xCx alloys and the effects of pre-irradiation could be neglected. As the annealing temperature was increased to 1050 degreesC, Si1-xCx alloys in both pre-irradiated and unpreirradiated samples of low C concentration remained, whereas most part of Si1-xCx alloys in samples with high C concentration vanished.

The formation and stability of Si1-xCx alloys in Si implanted with carbon

Si1-xCx alloys of carbon (C) concentration between 0.6%-1.0% were grown in Si by ion implantation and high temperature annealing. The formation of Si1-xCx alloys under different ion doses and their stability during annealing were studied. If the implanted dose was less than that for amorphizing Si crystals, the implanted C atoms would like to combine with defects produced during implantation and it was difficult to form Si1-xCx alloys after being annealed at 850 degreesC. With the increment of implanted C ion doses, the lattice damage increased and it was easier to form Si1-xCx alloys. But the lattice strain would become saturate and only part of implanted carbon atoms would occupy the substitutional positions to form Si1-xCx alloys as the implanted carbon dose increased to a certain degree. Once Si1-xCx alloys were formed, they were stable at 950 degreesC, but part of their strain would release as the annealing temperature increased to 1 000 degreesC. Stability of the alloys became worse with the increment of carbon concentration in the alloys.