This is in a series of literature reviews on InGaN solar cells, which supported the comprehensive review by D.V.P. McLaughlin & J.M. Pearce, "Progress in Indium Gallium Nitride Materials for Solar Photovoltaic Energy Conversion"Metallurgical and Materials Transactions A 44(4) pp. 1947-1954 (2013). open access
Others: InGaN solar cells| InGaN PV| InGaN materials| InGan LEDs| Nanocolumns and nanowires| Optical modeling of thin film microstructure| Misc.


Page data
Type Literature review
Authors Ankitvora
Chenlong Zhang
Published 2021
License CC-BY-SA-4.0
Impact Number of views to this page. Views by admins and bots are not counted. Multiple views during the same session are counted as one. 112


Nitrogen supply rate dependence of InGaN growth properties, by RF-MBE[1][1][1][1][edit | edit source]

Abstract: Nitrogen flow rate dependence of InGaN growth mechanism on sapphire (0 0 0 1) was studied. The indium content of the InGaN layer can be controlled by changing the nitrogen supply rate by structural and optical investigations. In case of lower nitrogen flow rate, the formation of phase separation due to dissociation of InN and condensation of indium on the growing surface were observed. In case of higher nitrogen flow rate (under N-rich condition), on the other hand, the indium content became uniform and the photoluminescence (PL) emission showed single peak. It is considered that the causes of phase separation formation were suppressed and nonequilibrium condition was enhanced by the increase of the nitrogen supply rate and the growth rate. Judging from the estimated thermal activation energies and degree of fluctuation in the indium content by PL line-width, however, the non-radiative recombination centers within the spontaneously formed potential wells as well as in the extended states were formed in case of excess nitrogen supply rate. The structural defects that were caused by the suppression of the surface migration of group-III atoms with excessive activated nitrogen might have worked as non-radiative recombination centers. Thus, a slightly N-rich condition is desirable for the realization of the uniform InGaN alloy composition and the high optical quality, especially at relatively high growth temperature.

  • indium content in the InGaN layer became uniform at N-rich condition. It is considered that the phase separation has been suppressed at the expense of the surface flatness due to the reduction of the surface migration of group-III adatoms by the excess nitrogen.
  • With increasing nitrogen flow rate, growth rate monotonically increased due to indium incorporation. This result shows that the dissociation of InN was prevented due to an increase of the activated nitrogen on the growing surface.
  • phase separation has been suppressed at high nitrogen flow rate is because of suppression of dissociation of InN due to excess activated nitrogen on the growing surface, and the enhancement of the non-equilibrium condition due to increase in growth rate. Also, it seems that they prevent indium condensation by its limited diffusion length on the growing surface.
  • It is considered that non-radiative recombination center is attributed to structural defects caused by the difference of bond length between In–N and Ga–N, and point defects caused by growths under the strongly off-stoichiometric conditions. impurities such as oxygen may be easily incorporated into defective high-indium-content region and they might work as non-radiative recombination centers.
  • slightly N-rich condition is desirable for realization of InGaN growth in terms of realization of a uniform alloy composition and high optical quality.
  • The thickness of all the InGaN layers was estimated by a subtraction of the buffer thickness from the total layer thickness, since InGaN/GaN interface was not clearly observed by SEM because the contrast of indium content between InGaN and GaN was small.
  • High nitrogen flow rate->excess activated nitrogen->1)reduction of group III adatoms migration(reduction of surface flatness); 2) increased growth rate push the growth process to nonequalibrium condition->suppressed phase separation-> increased indium uniformity.
  • From the unexpected behavior of the observed relationship between thermal activation energy Ea and the FWHM of PL spectra, the author concludes that non-radiative recombination center exists not only in the extended states but also in the localized states in high-indium-content region.

Growth of In-rich InGaN on InN template by radio-frequency plasma assisted molecular beam epitaxy[2][2][2][2][edit | edit source]

Abstract: High-quality epitaxial InN films have been employed as underlying templates for the growth of In-rich InxGa1−xN (0.71<x<0.90) films by radio-frequency plasma assisted molecular beam epitaxy. The epitaxial InN films (InN templates) with the thickness of 500 nm were grown on (0 0 0 1) sapphire substrates with a low-temperature deposited 10 nm thick InN (LT-InN) buffer layer, and then In-rich InxGa1−xN films with the thickness of 250 nm were grown on these InN templates. As compared with In-rich InxGa1−xN films grown directly on the LT-InN buffer layers, the In-rich InxGa1−xN films grown on the InN templates showed a decrease in full-width at half-maximum of the (0 0 0 2) X-ray rocking curves. It was found that the insertion of the InN template was very effective in improving the crystalline quality of In-rich InxGa1−xN. Optical properties were also investigated by photoluminescence. Lattice distortion of the In-rich InxGa1−xN films induced by the underlying InN templates is also discussed.

  • FWHM values for the InGaN films grown on the InN templates are much smaller than those for the InGaN films grown directly on the LT-InN buffer layers. These results indicate that the insertion of the InN template is very effective in improving the crystalline quality of In-rich InGaN.
  • For InGaN films grown on InN template with In-compositions >=0.8, an smooth surface shown in SEM images, which indicates improvements in the surface morphology. For InGaN films with In-composition <0.8, no distinct differences in the surface morphology were observed.
  • Author(s) use a model to determine the strain lattice constant and compare them with the observed results. It is found that In0.9Ga0.1N film is under considerable tensile strain induced by the underlying InN template, while In0.79G0.21N film is found to be nearly relaxed with slight tensile strain remaining.
  • Using PB model calculate the critical thickness of InGaN grown on strain-free InN. The thickness of critical layer is, 126, 19.8 and 8.19 for x=0.9, 0.79 and 0.71 respectively.

Epitaxial growth and optical properties of semipolar (11-22) GaN and InGaN/GaN quantum wells on GaN bulk substrates[3][3][3][3][edit | edit source]

Abstract: GaN and InGaN/GaN multiple quantum well (MQW) were grown on semipolar (11-22) GaN bulk substrates by metal organic vapor phase epitaxy. The GaN homoepitaxial layer has an atomically flat surface. Optical reflection measurements reveal polarization anisotropy for the A, B, and C excitons. Free A excitons dominate the photoluminescence (PL) spectrum at 10 K and are accompanied by a weaker, sharp doublet emission due to neutral donor-bound excitons. The InGaN/GaN MQW grown on a GaN homoepitaxial layer involves fast radiative recombination processes. The PL decay monitored at 428 nm can be fitted with a double exponential curve, which has lifetimes of 46 and 142 ps at 10 K. These values are two orders of magnitude shorter than those in conventional c-oriented QWs and are attributed to the weakened internal electric field. The emissions from GaN and MQW polarize along the [1-100] direction with polarization degrees of 0.46 and 0.69, respectively, due to the low crystal symmetry.

  • Lower temperature of 975C provides atomically flat surfaces for (11-22) GaN due to conventional regrowth techniques on patterned c-oriented GaN where the (11-22)facets tend to appear at low temperatures.
  • Radiative lifetime, which is inversely proportional to the transition probability, is a good index of the strength of the internal electric field.
  • PL lifetime of (0001)QW is two orders of magnitude larger than in the (11-22) MQW, because the PL linewidth of the (0001) QW is narrower than that of the (11-22) MQW, potential fluctuations is less in the (0001) QW and thus the radiative recombination process should be faster. Shorter lifetime in (11-22) MQW indicates higher transition probability due to weakened internal electric field.

Efficient radiative recombination from 〈11-22〉 -oriented InxGa1−xN multiple quantum wells fabricated by the regrowth technique[4][4][4][4][edit | edit source]

Abstract: InxGa1−xN multiple quantum wells (QWs) with [0001], <11-22>, and <11-20> orientations have been fabricated by means of the regrowth technique on patterned GaN template with striped geometry, normal planes of which are (0001) and {11-20}, on sapphire substrates. It was found that photoluminescence intensity of the {11-22} QW is the strongest among the three QWs, and the internal quantum efficiency of the {11-22} QW was estimated to be as large as about 40% at room temperature. The radiative recombination lifetime of the {11-22} QW was about 0.38 ns at low temperature, which was 3.8 times shorter than that of conventional [0001]-oriented InxGa1−xN QWs emitting at a similar wavelength of about 400 nm. These findings strongly suggest the achievement of stronger oscillator strength owing to the suppression of piezoelectric fields.

  • Two competitive factors determining the internal quantum efficiency of the present LEDs; one is carrier/exciton localization and the other is the quantum confinement Stark effect (QCSE). The former suppresses nonradiative processes to improve internal quantum efficiency, and the latter prevents the radiative recombination to degrade the efficiency.
  • QSCE becomes remarkable with increasing Indium composition in InGaN/GaN quantum walls and , so, can be a major drawback for realzing LEDs operating at longer wavelengths.
  • Use of nonpolar planes such (1010), (1120) and (11-22) can avoid the QCSE and therefore improve the quantum efficiency of LEDs.
  • STEM images shows that regrowth methods transform the square like GaN surface into large shape with two inclined facets, of which the angle was estimated to be 56 degree, indicating that the planes are (11-22)
  • Transition energies calculated from In composition are 2.22 and 3.15 eV for (0001) and (11-22) respectively. If the electric field are totally screened, the transition energies could be 2.96 and 3.23 eV for (0001) and (11-22) respectively.
  • In photoluminescent experiment, (11-22) facets has the highest PL intensity due to its rather weak PFs. (0001) QWs has the weakest intensity for its strong PFs and, furthermore, the widest well width, which leads to a lower transition probability.
  • Temperature dependent PL reveals that efficiency of (11-22) QWs was estimated about 40% at RT, which is about three times as high as that of the conventional C-oriented QWs.

Spatial and temporal luminescence dynamics in an InxGa1−xN single quantum well probed by near-field optical microscopy[5][5][5][5][edit | edit source]

Abstract: Spatial distribution of photoluminescence (PL) with spectral, spatial, and/or time resolution has been assessed in an InxGa1−xN single-quantum-well (SQW) structure using scanning near-field optical microscope (SNOM) under illumination-collection mode at 18 K. The near-field PL images revealed the variation of both intensity and peak energy in PL spectra according to the probing location with the scale less than a few hundredths of a nanometer. PL linewidth, the value of which was about 60 meV in macroscopic PL, was as small as 11.6 meV if the aperture size was reduced to 30 nm. Clear spatial correlation was observed between PL intensity and peak wavelength, where the regions of strong PL intensity correspond to those of long wavelength. Time-resolved SNOM–PL study showed the critical evidence that supports the model of diffusion of carriers to potential minima.

Electrical and optical properties of p-type InGaN[6][6][6][6][edit | edit source]

Abstract: Mg-doped InxGa1−xN alloys were grown by metal organic chemical vapor deposition on semi-insulating c-GaN/sapphire templates. Hall effect measurements showed that Mg-doped InxGa1−xN epilayers are p-type for x up to 0.35. Mg-acceptor levels (EA) as a function of x, (x up to 0.35), were experimentally evaluated from the temperature dependent hole concentration. The observed EA in Mg-doped In0.35Ga0.65N alloys was about 43 meV, which is roughly four times smaller than that in Mg doped GaN. A room temperature resistivity as low as 0.4 Ω cm (with a hole concentration ~ 5×1018 cm−3 and hole mobility ~ 3 cm2/V s) was obtained in Mg-doped In0.22Ga0.78N. It was observed that the photoluminescence (PL) intensity associated with the Mg related emission line decreases exponentially with x. The Mg energy levels in InGaN alloys obtained from PL measurements are consistent with those obtained from Hall-effect measurements.

  • Achieving highly conductive p-type GaN and AlGaN is very difficult due to high activation energies (EA) of the Mg-acceptor, which is a universally accepted p-type dopant for GaN and related alloys.
  • Since EA decreases with a decrease in band gap energy, Mg-doped InGaN (InGaN:Mg) is expected to have a higher hole concentration (p) than Mg-doped GaN.
  • P-type doping in relatively high In content InGaN alloys is highly challenging due to the presence of high background electron concentrations, which is believed to originate from defects such as oxygen and hydrogen impurities or nitrogen vacancies.
  • presence of high background electron concentration is main hindrance for obtaining p-type conductivity and p-type InxGa1−xN alloys with In content x>0.35.
  • P-type resistivity in Mg-doped InGaN alloys was found to be lower than that of Mg-doped GaN, that indicates a high hole concentration.
  • It was found that EA continuously decreases with an increase in x. Lower values of EA are the main physical reason for higher values of p in InxGa1−xN:Mg alloys of higher x.
  • InxGa1−xN materials with x>0.35 are generally highly n-type and conversion of these materials to p-type by Mg doping is still very difficult. Author(s) results indicate that p-type conductivity in InGaN:Mg could be further improved if a better control of the background electron concentration could be achieved.
  • In PL spectra, deep donors appeared in Mg doped GaN but disappeared in all InGaN:Mg alloys. Author(s) suggest that the lower growth temperatures employed for InGaN alloys somehow suppresses the formation of these deep donors.
  • Mg impurity related PL emission intensity is found to decrease exponentially with In-content, reduction in PL intensity may be related with the incorporation of impurities, which are also responsible for the high background electron concentrations in high In content InGaN alloys.

Investigation on the Correlation Between the Crystalline and Optical Properties of InGaN Using Near-Field Scanning Optical Microscopy[7][7][7][7][edit | edit source]

Author(s) have performed the polarization-modulation near-field scanning optical microscopy (PM-NSOM) and photoluminescence NSOM (PL-NSOM) measurements on the InGaN alloy epitaxial layer. Spatial variations in the crystalline quality of nanoscale domains in InGaN film were found by PM-NSOM. It was found that the luminescent property of InGaN correlates closely with the local crystalline quality. Regions with better crystallinity have higher luminescence intensity and longer emission wavelength, while regions with poorer crystallinity exhibit a luminescence of lower intensity and shorter emission wavelength. Author(s) show that the combination of PM-NSOM and PL-NSOM is a useful diagnostic tool to the correlation between crystalline and optical properties of the nanostructures.

  • Near-field scanning optical microscopy(NSOM) is well suited for studying the correlation between the optical and structural properties because it simultaneously provides optical and topographic images at subwavelength scale.
  • Phase separated InGaN film comprises nanodomains with different crystalline quality. The better crystalline region exhibits a higher intensity and longer wavelength emission, while the poorer crystalline region shows a lower intensity and shorter wavelength emission.
  • nanosized structures with the typical size of ~100nm are attributed to the island formation due to lattice mismatch with GaN.
  • The spatial variation of dichroic ratio is associated with the spatially inhomogeneous crystallinity of the InGaN film. Brighter area with higher dichroic ratio in the dichroic ratio image corresponds to the higher crystalline quality region, while the darker area represents the more defective region.
  • A smaller bandgap is expected to exist in the In-rich region, where the photoexcited electron-hole pairs are easily accumulated and confined in In-rich nanoclusters.

Compositional dependence of the strain-free optical band gap in InxGa1−xN layers[edit | edit source]

Abstract: The effect of strain on the compositional and optical properties of a set of epitaxial single layers of InxGa1−xN was studied. Indium content was measured free from the effects of strain by Rutherford backscattering spectrometry. Accurate knowledge of the In mole fraction, combined with x-ray diffraction measurements, allows perpendicular strain (ϵzz) to be evaluated. Optical band gaps were determined by absorption spectroscopy and corrected for strain. Following this approach, the strain free dependence of the optical band gap in InxGa1−xN alloys was determined for x < 0.25. Our results indicate an "anomalous," linear, dependence of the energy gap on the In content, at room temperature: Eg(x) = 3.39–3.57x eV. Extension of this behavior to higher concentrations is discussed on the basis of reported results.

Luminescences from localized states in InGaN epilayers[edit | edit source]

Abstract: Optical spectra of the bulk three-dimensional InGaN alloys were measured using the commercially available light-emitting diode devices and their wafers. The emission from undoped InxGa1−xN(x<0.1) was assigned to the recombination of excitons localized at the potential minima originating from the large compositional fluctuation. The emission from heavily impurity-doped InGaN was also pointed out related to the localized states.

  • InGaN alloys exhibit intense electroluminescence(EL) or photoluminescence(PL) peaks in spite of their large threading dislocation densities.
  • Stokes shift of about 85meV is observed from PL and PLE spectra of In0.09Ga0.91N 3D multi-quantum-well(MQW), indicating the localized states exist in the MQW. The wide FWHM also shows the large alloy potential fluctuation due to the compositional inhomogeneity of Indium.
  • EL peak of the undoped 3D In0.06Ga0.94N locates in the lower-energy tail of the FE resonance even at RT, showing the Stokes shift of 40meV.
  • Temperature dependences of FE energies and EL (or PL) peak energies in undoped samples show that the EL peaks in undoped InGaN are originated from bound states.
  • The author concludes from their observations that the emssion peak in the undoped 3D In0.06Ga0.94N is assigned to the recombination of localized excitons.
  • The origin of the localized states is assigned to the phase-separated In-rich regions of the large alloy compositional fluctuation.
  • The emission from heavily impurity-doped 3D InGaN is also related to the localized states

Photoluminescence from quantum dots in cubic GaN/InGaN/GaN double heterostructures[8][8][8][8][edit | edit source]

Abstract: Author(s) have measured photoluminescence spectra of molecular-beam-epitaxy-grown cubic GaN/InxGa1−xN/GaN double heterostructures with x between 0.09 and 0.33. Author(s) observe a luminescence peak at about 2.3–2.4 eV which is almost independent of the InGaN layer composition. High-resolution x-ray diffraction measurements revealed a pseudomorphic In-rich phase with x = 0.56±0.02 embedded in the InGaN layers. Including strain effects Author(s) calculate a gap energy Eg = 2.13 eV of this phase. In cubic InGaN, spontaneous polarization and strain-induced piezoelectric fields are negligible. Therefore, the observed difference between the luminescence energy and the gap of the In-rich phase is assumed to be due to the localization of excitons at quantum-dot-like structures with a size of about 15 nm.

  • gap energy is different for strained and strain free InGaN.
  • In InGaN with cubic crystal structure (c-InGaN) spontaneous polarization does not exist due to the higher crystal symmetry, and due to the (001) growth direction, strain-induced piezoelectric fields are negligible.
  • Author(s) also considered the effect of a biaxial compression on the gap energy by taking into account the variation Eg of gap energy due to an biaxial in-plane strain.
  • Since in c-III-nitrides no spontaneous polarization- or piezoinduced electric fields exist, the difference between the InGaN gap and the luminescence peak energy is equal to the localization energy Eloc of the excitons. (need to check the validity of this argument)
  • In-rich inclusions have also been found in thick c-InGaN layers. Since micro-Raman experiments revealed clear evidence that these inclusions form nanometer sized QD-like structures, author(S) assume that PL from their DH structures is due to the recombination of excitons localized at In-rich QD-like structures which are embedded in the InGaN layers. The existence of QD-like structures in c-GaN/InGaN/GaN DH structures is further supported by the fact that the PL and EL from c-InGaN/GaN DH and QW structures grown by MOCVD is also observed close to the energy gap.
  • The shift of PL peak energy is mainly due to the decrease of the QD size with decreasing In content of the layers.

Photoluminescence associated with quantum dots in cubic GaN/InGaN/GaN double heterostructures[edit | edit source]

Abstract: Author(s) report on investigations of the photoluminescence of cubic GaN/ InxGa1−xN/ GaN double heterostructures with x between 0.09 and 0.33. The room temperature emission of all samples is found at about 2.3-2.4 eV. High resolution X-ray diffraction measurements reveal an In-rich phase with x=0.56. Luminescence line narrowing in resonant excitation experiments indicate that the photoluminescence stems from quantum-dot-like structures of the In-rich phase. Postgrowth annealing at temperatures up to 700°C demonstrates an obvious stability of the quantum dots.

InxGa1−xN refractive index calculations[edit | edit source]

Abstract: The growth of InxGa1−xN Wurtzite structure is a well established fact. It permits to design optoelectronic devices such as laser diodes or LEDs, from the near ultraviolet to the infrared light spectrum. This sweeps indeed, the whole of the visible spectrum and, hence, appears to be very useful to the recent development of liquid crystal display screens, or designing photodiodes and perhaps solar cells (after studying their energetical efficiencies). Nevertheless, refractive indices of InxGa1−xN structure have not been studied. The refractive index of such structures is increasing from the GaN refractive index to the InN one, with therefore, a bowing of the curve due to the lattice mismatch between these two constituting binary alloys. The index is, in a certain range of the "n(x)" characteristic, less than the GaN one. This seems to be particularly interesting in the integrated optics domain or optical waveguides realization, because the growth of GaN is easier than the growth of InxGa1−xN.

  • GaN is commonly grown on sapphire, although its so high 16% mismatch to GaN, and the fact that lattice mismatch of even 10^-3% are sufficient to generate misfit dislocations, which degrade the luminescence efficiency of the material.
  • SiC substrates are too expensive despite of their so little 3% mismatch on GaN.
  • The bowing is due to the virtual crystal average of the pseudopotential. It's not constant but itselft composition-dependent.
  • The importance of the refractive index is due to its direct reverse proportional relation with gap of a material. Thus wavelength and refractive index must be directly proportional because of the proportional relationship between the bandgap of a material and wavelength.
  • The refractive index of InGaN obeys to the same law which governs the bandgap evolutions, the Vegard's Law!
  • The decreasing tendency of n of the first quarter of the molar fraction is due to the lattice mismatch. However, when the alloy begins to contain more and more Indium, the natual tendency of decreasing gap takes again its real place and the increase of the refractive index becomes normal and consequently greater.
  • This is just a theoretical study with no convincing experimental data and observation!

Refractive index and gap energy of cubic InxGa1−xN[edit | edit source]

Abstract: Spectroscopic ellipsometry studies have been carried out in the energy range from 1.5 to 4.0 eV in order to determine the complex refractive indices for cubic InGaN layers with various In contents. The films were grown by molecular-beam epitaxy on GaAs(001) substrates. By studying GaN films, we prove that for the analysis of optical data, a parametric dielectric function model can be used. Its application to the InGaN layers yields, in addition, the composition dependence of the average fundamental absorption edge at room temperature. From the latter, a bowing parameter of 1.4 eV is deduced.

InxGa1−xN refractive index calculations[edit | edit source]

Abstract: The growth of InxGa1−xN Wurtzite structure is a well established fact. It permits to design optoelectronic devices such as laser diodes or LEDs, from the near ultraviolet to the infrared light spectrum. This sweeps indeed, the whole of the visible spectrum and, hence, appears to be very useful to the recent development of liquid crystal display screens, or designing photodiodes and perhaps solar cells (after studying their energetical efficiencies). Nevertheless, refractive indices of InxGa1−xN structure have not been studied. The refractive index of such structures is increasing from the GaN refractive index to the InN one, with therefore, a bowing of the curve due to the lattice mismatch between these two constituting binary alloys. The index is, in a certain range of the "n(x)" characteristic, less than the GaN one. This seems to be particularly interesting in the integrated optics domain or optical waveguides realization, because the growth of GaN is easier than the growth of InxGa1−xN.

Optical Properties of Strained AlGaN and GaInN on GaN[edit | edit source]

Abstract: The composition of alloys in strained ternary alloy layers, Al xGa1- xN (0<x<0.25) and Ga1- xIn xN (0<x<0.20), on thick GaN was precisely determined using the high-resolution X-ray diffraction profile. The band gap of strained AlGaN is found to increase almost linearly according to the AlN molar fraction, while that of strained GaInN has a large bowing parameter of 3.2 eV.

Optical and microstructural properties versus indium content in InxGa1−xN films grown by metal organic chemical vapor deposition[edit | edit source]

Abstract: Author(s) present comparative investigations of single phase InxGa1−xN alloys for a varying In content (x = 0.07 to 0.14) grown by metal organic chemical vapor deposition (MOCVD) technique. While the composition was determined using secondary ion mass spectroscopy, Author(S) have investigated the microstructures in InxGa1−xN/GaN films by using transmission electron microscopy and correlated these with the refractive index of the material. Based on ellipsometric analysis of the films, the dispersion of optical indices for InxGa1−xN films is determined by using Tauc–Lorentz dispersion equations.

  • In this study, secondary ion mass spectroscopy was used to determined the In and Ga atomic fractions. TEM analysis was conducted in order to study the microstructural defects present in the layers. Optical properties of the InGaN film was investigated by spectroscopic ellipsometry.
  • TEM images shows threading dislocations originate in the GaN layer and extend further into InGaN layer with some of these defects eventually terminating into inverted pyramidal pits (V-pits). the higher the In content, the larger the number of such V-pits.
  • Author(s) use SE to determine the refractive indices of InGaN sample. The thickness of each layer obtained by using the model agrees well with that observed in SEM, however, it is difficult to extract optical indices using the same methods. The author(s) include the possible reasons in the article.
  • Decrease in refractive index in high In-content layer is observed possibly due to the inhomogeneity in high In-content layer. The higher is the In composition, the higher will be the V-pits density, resulting in a lower refractive index value.

Time-Resolved Photoluminescence Studies of Indium-Rich InGaN Alloys[edit | edit source]

Abstract: Time-resolved photoluminescence (PL) spectroscopy has be used to investigate indium-rich InGaN alloys grown on sapphire substrates by metal organic chemical vapor deposition. Photoluminescence measurement indicates two dominant emission lines originating from phase-separated high- and low-indium-content regions. Temperature and excitation intensity dependence of the two main emission lines in these InGaN alloys have been measured. Temperature and energy dependence of PL decay lifetime show clearly different decay behaviour for the two main lines. Author(s) results show that photo-excited carriers are deeply localized in the high indium regions while photo-excited carriers can be transferred within the low-indium-content regions as well as to high-content regions.

  • two PL peaks can be observed in PL emission spectra of InGaN sample, which originate from the two phase segregation or different size of quantum dots. The 464nm peak is due to small size QDs or low-indium region and the 678nm peak is from larger-sized QDs or high-indium-content region.
  • The energy positions increase linearly with the increase of the excitation intensity. The fact is weak Coulomb screening of the quantum confined Stark effect induced by the weak piezoelectric field due to the thickness of the author's sample in thickness 100nm.
  • PL decay temporal responses at the peak 464nm can be fitted well by a single exponential function, but that at 678nm must be fitted by two exponential functions. This shows that the recombination channels of the two lines are different.
  • The emission energy dependence of the PL decay is a characteristic of a distribution of localized excitons. The fitted value Em for line at 678nm is 40meV above the 10K PL emission peak, which represents the behaviour of strongly localized excitons.
  • The Pl decay lifetime behaviour shows that the photo-excited carriers in these materials can be transferred within the low indium regions as well as to high-content regions.

Correlation of crystalline defects with photoluminescence of InGaN layers[edit | edit source]

Abstract: Author(s) report structural studies of InGaN epilayers of various thicknesses by x-ray diffraction, showing a strong dependence of the type and spatial distribution of extended crystalline defects on layer thickness. The photoluminescence intensity for the samples was observed to increase with thickness up to 200 nm and decrease for higher thicknesses, a result attributed to creation of dislocation loops within the epilayer. Correlation of physical properties with crystalline perfection open the way for optimized designs of InGaN solar cells, with controlled types and dislocation densities in the InGaN epilayers, a key requirement for realizing high photocurrent generation in InGaN.

MOVPE growth and Mg doping of InxGa1−xN (x∼0.4) for solar cell[edit | edit source]

Abstract: MOVPE growth and Mg doping of InGaN films are studied to develop technologies for the InGaN-based solar cell. By optimizing growth temperature and the TMI/(TMI+TEG) molar ratio, InGaN films with an In content up to 0.37 are successfully grown without phase separation and metallic In incorporation. It is found that the In composition in the InGaN films is governed by growth temperature, and the TMI/(TMI+TEG) molar ratio has very small effect on the composition change. InGaN films doped with Mg using CP2Mg show the compensation effect of carriers and those with an In content up to 0.2 show p-type conduction. The film with an In content of 0.37 shows phase separation when the CP2Mg/(TMI+TEG) molar ratio exceeds 0.05, indicating that Mg atoms incorporated have a significant effect on the crystal growth of InGaN.

Cathodoluminescent investigations of In x Ga1− x N layers[edit | edit source]

Abstract: The aim of this work was the investigation of the InGaN epilayers of various contents and various thickness; namely the influence of these two factors upon the cathodoluminescent (CL) properties. The studied epilayers were grown by plasma assisted molecular beam epitaxy. The samples were studied by electron probe microanalysis, CL, X-ray diffraction (XRD), and scanning electron microscopy (SEM). Some interesting peculiarities of CL spectra were obtained; the nature of the CL bands is discussed.

  1. Komaki, H. et al., 2007. Nitrogen supply rate dependence of InGaN growth properties, by RF-MBE. Journal of Crystal Growth, 305(1), pp.12-18
  2. Kurouchi, M. et al., 2005. Growth of In-rich InGaN on InN template by radio-frequency plasma assisted molecular beam epitaxy. Journal of Crystal Growth, 275(1-2), p.e1053-e1058
  3. Ueda, M. et al., 2006. Epitaxial growth and optical properties of semipolar (112) GaN and InGaN/GaN quantum wells on GaN bulk substrates. Applied Physics Letters, 89(21), p.211907
  4. Nishizuka, K. et al., 2004. Efficient radiative recombination from 〈112〉 -oriented InxGa1−xN multiple quantum wells fabricated by the regrowth technique. Applied Physics Letters, 85(15), p.3122
  5. Kaneta, A. et al., 2002. Spatial and temporal luminescence dynamics in an InxGa1−xN single quantum well probed by near-field optical microscopy. Applied Physics Letters, 81(23), p.4353
  6. Pantha, B.N. et al., 2009. Electrical and optical properties of p-type InGaN. Applied Physics Letters, 95(26), p.261904
  7. Lin, T.Y. et al., 2007. Investigation on the Correlation Between the Crystalline and Optical Properties of InGaN Using Near-Field Scanning Optical Microscopy. Electrochemical and Solid-State Letters, 10(7), p.H217-H219
  8. O. Husberg et al., "Photoluminescence from quantum dots in cubic GaN/InGaN/GaN double heterostructures," Applied Physics Letters, vol. 79, no. 9, pp. 1243-1245, Aug. 2001