Search

Enhancement of light extraction in GaN light-emitting diodes (LEDs) employing omnidirectional reflectors (ODRs) is presented. The ODR consists of GaN, ITO nanorod low-refractive-index layer, and an Ag layer. An array of ITO nanorods is deposited by oblique-angle deposition using e-beam evaporation. The refractive index of the ITO nanorods is 1.34 at 461 nm, significantly lower that that of dense ITO, which is n = 2.06 at 461 nm. It is experimentally shown that the GaN LED with GaN/ITO nanorods/Ag ODR show much better electrical properties and higher light-extraction efficiency than LEDs with Ag contact. This is attributed to enhanced reflectivity of the ODR by using an ITO low-refractive-index layer with high transparency, high conductivity, and low refractive index.

An optimized packaging configuration for high-power white-light-emitting diode (LED) lamps that employs a diffuse reflector cup, a remote phosphor and a hemi-spherically-shaped encapsulation is presented. Ray tracing simulations for this configuration show that the phosphor efficiency can be enhanced by up to 50% over conventional packages. It is experimentally shown that dichromatic LED lamps with remote phosphor and diffuse reflector cup configuration have higher phosphor efficiency by 15.4% for blue-pumped yellow phosphor and by 27% for ultraviolet-pumped blue phosphor over conventional packages. Those improvements are attributed to reduced absorption of the phosphorescence by the LED chip and the reduction of deterministic optical modes trapped inside the encapsulant.

In this paper, three different integer frequency offset estimators for orthogonal frequency-division multiplexing systems are derived: maximum-likelihood, blind, and correlation estimators. These are based on one pilot symbol with adjacent subcarriers differentially modulated. Specially, the correlation estimator attains a good performance even if there exists residual symbol timing offset. We present the method used to compose the pilot symbol and derive three estimators based on it. Results from simulations are presented to verify the performance of the estimators.

This paper proposes a burst synchronization method with a preamble for high data rate wireless asynchronous transfer mode (WATM) systems that use orthogonal frequency division multiplexing (OFDM). The preamble for synchronization is composed of four identical periods. Using this signal repetition, the correlation for symbol timing detection and frequency offset estimation are calculated. The proposed method improves the performance of synchronization by increasing the size of data used in the correlation operation. The analytic results show that by using the proposed method a gain of 1 dB in SIR (signal to interference ratio) is possible.

The increase of the capacity of the digital magnetic recording systems inevitably causes severe intersymbol interference (ISI) and nonlinear distortions in the magnetic channel. Thus to achieve large capacity magnetic recording systems it is necessary to compensate these problems. In order to cope with the severe ISI and nonlinear distortions a neural decision feedback equalizer (NDFE) is applied to the digital magnetic recording channel. In the performance comparison between the NDFE and the conventional decision feedback equalizer (DFE) via simulations, it is found that as nonlinear distortions increase the NDFE has a greater SNR (signal to noise ratio) advantage over the DFE.

When convolution code is used in wireless coded orthogonal frequency division multiplexing (COFDM) systems, effective maximum likelihood (ML) decoding requires information on the frequency response of each subcarrier. This translates into increased hardware complexity, for instance, in the form of an additional deinterleaving block. We propose an optimal weighting scheme that combines the metrics for soft decision and channel influence into a single metric, can be performed before the deinterleaving block, and makes ML decoding possible. It shows comparable performance to ML decoding and better performance over suboptimal decoders.

An OFDM signal with large peak-to-average power ratio (PAPR) can cause power degradation (inband distortion) and spectral spreading (out-of-band radiation) by being clipped passing through a power amplifier. The PAPR characteristics are analyzed in two different aspects. From the foundation of the analysis, we propose Hadamard transform, which is ascribed to the relationship between the autocorrelation function and PAPR. Extensive computer simulations show that Hadamard transform is an effective technique to reduce PAPR.

A new coarse frequency offset estimation algorithm is proposed which performs robust operation in the presence of a symbol timing offset in an allowed range. The effect of a symbol timing offset on estimating a coarse frequency offset is analyzed and the coherence phase bandwidth according to the allowed symbol timing offset is introduced. In addition, a new technique for performing the coarse frequency estimation using the coherence phase bandwidth is described. Simulation results show that the proposed algorithm estimates a coarse frequency offset well under both AWGN and multipath channels even in the presence of the symbol timing offset within an allowed range.

The use of liquid crystal displays (LCDs) has become prevalent in our modern, technology driven society. We demonstrate a linearly polarized GaInN light-emitting diode (LED) embedded with a wire-grid polarizer (WGP). A derivation of rigorous coupled-wave analysis is given; starting from Maxwell?s equations and finishing by matching the boundary conditions in the grating and other regions of interest. Simulated results are shown for various grating parameters, including different metals used for the grating and the metal-line dimensions. An LED fabrication process is developed for demonstrating WGP-LEDs. A clear polarization preference for the light coming out of the WGP-LED is experimentally demonstrated with a polarization ratio over 0.90, which is in good agreement with simulation results.

BackgroundInterest in smartphone health apps has been increasing recently. However, we have little understanding of the cognitive and motivational factors that influence the extent of health-app use. ObjectiveThis study aimed to examine the effects of four cognitive factors—health consciousness, health information orientation, eHealth literacy, and health-app use efficacy—on the extent of health-app use. It also explored the influence of two different use patterns—information and information-behavior use of health apps—with regard to the relationships among the main study variables. MethodsWe collected and analyzed 765 surveys in South Korea. According to the results, there was a negligible gender difference: males (50.6%, 387/765) and females (49.4%, 378/765). All participants were adults whose ages ranged from 19 to 59. In order to test the proposed hypotheses, we used a path analysis as a specific form of structural equation modeling. ResultsThrough a path analysis, we discovered that individuals’ health consciousness had a direct effect on their use of health apps. However, unlike the initial expectations, the effects of health information orientation and eHealth literacy on health-app use were mediated by health-app use efficacy. ConclusionsThe results from the path analysis addressed a significant direct effect of health consciousness as well as strong mediating effects of health-app use efficacy. These findings contribute to widening our comprehension of the new, digital dimensions of health management, particularly those revolving around mobile technology.

Thermally reflowed photoresist is used as an etch mask in inductively coupled plasma reactive ion etching of dielectric graded-refractive-index (GRIN) coatings. The coatings have varying compositions of TiO2 and SiO2 and are used to fabricate GRIN micropillars with tapered sidewalls. The effects of ion implantation on the dry-etch-resistance of photoresist are investigated for Si, N, and Ar ion implantation. Compared with the unimplanted photoresist, the implanted photoresists show enhanced dry-etch-resistance under fluorine chemistry (CHF3 and O2). The etch rate of the Si-implanted photoresist is 72% lower than that of the unimplanted photoresist. The measured depth of modification of the photoresist is in good agreement with the trend predicted by ion-implantation-simulation software. Using Si-implanted photoresist as an etch mask, five-layer GRIN micropillars with tapered sidewalls are fabricated.

Researchers in Japan have used electron-beam excitation in an AlGaN/AlGaN quantum-well structure to demonstrate the emission of ultraviolet light at record-breaking efficiency, giving hope for the realization of high-power efficient semiconductor sources of deep-ultraviolet light.

A GaInN light-emitting diode (LED) structure is analyzed that employs a separate epitaxial growth for the p-type region, i.e., the AlGaN electron-blocking layer (EBL) and p-type GaN cladding layer, followed by wafer or chip bonding. Such LED structure has a polarization-inverted EBL and allows for uncompromised epitaxial-growth optimization of the p-type region, i.e., without the need to consider degradation of the quantum-well active region during p-type region growth. Simulations show that such an LED structure reduces electron leakage, reduces the efficiency droop, improves hole injection, and has the potential to extend high efficiencies into the green spectral region.

The forward voltage characteristics of GaInN light-emitting diodes are studied in the temperature range of 80 K to 450 K. The forward-voltage-vs.-temperature curve has a “two-slope” characteristic with a slope of dVf/dT = −1.7 mV/K and −8.0 mV/K at room temperature and cryogenic temperatures, respectively. To investigate the two-slope characteristic, we perform transmission-line-model measurements on p-type GaN and show that both p-type contact and sheet resistance decrease drastically with increasing temperature. We conclude that dVf/dT in the high-slope region is limited by p-type sheet and contact resistance, and in the low-slope region by the GaN pn junction properties.

AlxGa1−xN/GaN superlattice electron blocking layers (EBLs) with gradually decreasing Al composition toward the p-type GaN layer are introduced to GaInN-based high-power light-emitting diodes (LEDs). GaInN/GaN multiple quantum well LEDs with 5- and 9-period Al-composition-graded AlxGa1−xN/GaN EBL show comparable operating voltage, higher efficiency as well as less efficiency droop than LEDs having conventional bulk AlGaN EBL, which is attributed to the superlattice doping effect, enhanced hole injection into the active region, and reduced potential drop in the EBL by grading Al compositions. Simulation results reveal a reduction in electron leakage for the superlattice EBL, in agreement with experimental results.

An unequivocal correlation between the onset of high injection and the onset of the efficiency droop is demonstrated in GaInN light-emitting diodes over a wide range of temperatures. The diode voltage at the onset of high injection and the voltage at the onset of the efficiency droop are correlated by the equation VHigh-injection onset + ΔV ≈ VDroop onset. The excess voltage, ΔV, determined to be 0.3 V, drops partially over the p-type neutral region. The resulting electric field sweeps electrons out of the active region and results in substantial electron leakage despite high barriers that confine the carriers to the active region.

The dependence of the polarization-induced electric field in GaInN/GaN multiple-quantum-well light-emitting diodes (LEDs) on the GaN quantum barrier (QB) thickness is investigated. Electrostatic arguments and simulations predict that a thin QB thickness reduces the electric field in the quantum wells (QWs) and also improves the LED efficiency. We experimentally demonstrate that the QW electric field decreases with decreasing QB thickness. The lower electric field results in a better overlap of electron and hole wave functions and better carrier confinement in the QWs. A reduced efficiency droop and enhanced internal quantum efficiency is demonstrated for GaInN/GaN LEDs when the QB thickness is reduced from 24.5 to 9.1 nm.

White light-emitting diodes (LEDs) based on the combination of a GaN-based blue LED chip and a yellow phosphor layer suffer from a low phosphor conversion efficiency (PCE) because a significant amount of the yellow fluorescence is emitted towards the blue LED chip where the fluorescence is partially absorbed. In this study, we greatly enhanced the PCE of a white LED by embedding a dichroic-filtering contact (DFC) which multi-functioned as a blue-transmitting but yellow-reflecting optical filter as well as a low-resistance ohmic contact to p-type GaN. Electrically conductive DFCs consisting of alternating layers of dense and nanoporous indium-tin-oxide were designed using a genetic algorithm optimization method, and experimentally realized by the oblique-angle deposition technique, which enables tuning of the refractive index of a single material. GaInN/GaN multiple quantum well LEDs with 3- and 5-layer DFCs show much enhanced PCEs, by 9.8% and 17.7%, respectively, while maintaining favorable electrical properties. In addition, the possibility of eliminating the trade-off between the color temperature and the luminous efficacy of typical warm-white LEDs by using a DFC is discussed.

The unrestricted control of the surface structure and refractive index would allow for new and improved functionalities in optoelectronic devices. Specifically, micro-patterned graded-refractive-index (GRIN) coatings can enable control of emission pattern and promote light extraction in GaInN light-emitting diodes (LEDs). We design and demonstrate coatings that are patterned into arrays of GRIN micro-pillars, each composed of five dielectric layers of (TiO2)x(SiO2)1−x with the bottom layer (adjacent to semiconductor) having the highest refractive index and the top layer (adjacent to air) having the lowest one. The GRIN micro-pillars, including their planar geometric shape and size, are structured for emission pattern control and maximum light-extraction efficiency. It is shown that the peak emission intensity of the GRIN LEDs is controllable from ±20° to ±50° off the surface-normal. In addition, LEDs patterned with an array of four-pointed-star-shaped GRIN micro-pillars with a pillar-size of 4 μm and spacing between neighboring pillars of 4 μm show a 155% enhancement in light-output power over an uncoated planar reference LED.

We report a method to determine the radiative efficiency (RE) of a semiconductor by using room-temperature excitation-dependent photoluminescence measurements. Using the ABC model for describing the recombination of carriers, we show that the theoretical width of the RE-versus-carrier-concentration (n) curve is related to the peak RE. Since the normalized external quantum efficiency, EQEnormalized, is proportional to the RE, and the square root of the light-output power, LOP, is proportional to n, the experimentally determined width of the EQEnormalized-versus-n curve can be used to determine the RE. We demonstrate a peak RE of 91% for a Ga0.85In0.15N quantum well.

An effective method to enhance the light extraction for GaInN light-emitting diodes (LEDs) is reported. The method employs TiO2 micro-pillars with tapered sidewalls, which are refractive-index-matched to the underlying GaN. The tapered micro-pillars are fabricated by using reflowed photoresist as mask during CHF3-based dry etch, with O2 added in order to precisely control the taper angle. LEDs patterned with TiO2 micro-pillars with tapered sidewalls show a 100% enhancement in light-output power over planar reference LEDs. The measured results are in good agreement with ray-tracing simulations, showing strong potential of optical surfaces that are controlled in terms of refractive index and lateral structure.

GaInN blue light-emitting diodes (LEDs) emitting at 445 nm exhibit a spatially uniform cyan emission (480 nm) that dominates the emission spectrum at low injection current. Photoluminescence using resonant optical excitation shows that the cyan emission originates from the active region. The blue-to-cyan intensity ratio, which depends on the electrical and optical excitation density, reveals that the cyan emission is due to a transition from the conduction band to a Mg acceptor having diffused into the last-grown quantum well of the active region. The Mg in the active region provides an additional carrier-transport path, and therefore can explain the high subthreshold forward leakage current that is measured in these LEDs.

We demonstrate a back-emitting (sapphire-substrate emitting) linearly polarized GaInN light-emitting diode (LED) embedded with a subwavelength-sized aluminum wire-grid polarizer (WGP). Rigorous coupled wave analysis is implemented to study the polarization characteristics of such a WGP LED. The aluminum nanowire grating with a period of 150 nm is located on the sapphire backside of a GaInN LED structure and is fabricated by electron-beam lithography and inductively coupled plasma reactive-ion etching. A polarization ratio of 0.96 is demonstrated for a WGP GaInN LED in good agreement with simulation results.

Patterned graded-refractive-index (GRIN) coatings that offer the controllability of far-field emission pattern and polarization properties of GaInN light-emitting diodes (LEDs) are investigated. Compared with a planar reference LED, the light-output power of an LED with patterned GRIN coatings (GRIN LED) is enhanced by about 69%. Furthermore, the GRIN LED has bidirectional emission peaks at about 45° off-surface-normal and polarized light emission with the maximum polarization ratio occurring at the same angle, i.e. the intensity maximum and the polarization-ratio maximum coincide. The large off-surface-normal emission of the GRIN LED results from the strong light extraction through the sidewalls of the GRIN pillars, which is in good agreement with results predicted from ray-tracing simulations.

not Available.

An analytic model is developed for the droop in the efficiency-versus-current curve for light-emitting diodes (LEDs) made from semiconductors having strong asymmetry in carrier concentration and mobility. For pn-junction diodes made of such semiconductors, the high-injection condition is generalized to include mobilities. Under high-injection conditions, electron drift in the p-type layer causes a reduction in injection efficiency. The drift-induced leakage term is shown to have a 3rd and 4th power dependence on the carrier concentration in the active region; the values of the 3rd- and 4th-order coefficients are derived. The model is suited to explain experimental efficiency-versus-current curves of LEDs. V C 2012 American Institute of Physics .[ http://dx.doi.org/10.1063/1.4704366]

At room temperature, AlGaInP pn-junction light-emitting diodes (LEDs) emitting at 630 nm do not exhibit an efficiency droop. However, upon cooling the AlGaInP LEDs to cryogenic temperatures, they show a pronounced efficiency droop. We attribute the efficiency droop in AlGaInP LEDs to electron-drift-induced reduction in injection efficiency (i.e., carrier leakage out of the active region) mediated by the asymmetry of the pn junction, specifically the disparity between electron and hole concentrations and mobilities, with the concentration disparity exacerbated at low temperatures.

The effect of chip area on the temperature-dependent light-output power (LOP) in GaInN-based light-emitting diodes (LEDs) is investigated. The larger the chip size, the faster the reduction in LOP with increasing temperature becomes, indicating that increasing the size of LED chips, a technology trend for reducing the efficiency droop at high currents, is detrimental for high temperature-tolerant LEDs. In addition, it is found that regardless of chip size, the temperature-dependent LOP is identical for the LEDs operating at the same current density.

A low-cost, wideband, flexible antenna for radio frequency identification tags mountable on curved metal surfaces is presented. The antenna consists of a label-type patch adhering to a flexible polyvinyl chloride substrate. A novel balanced feed method using two orthogonal slits on the patch is proposed to perform simple and wideband impedance matching between the antenna and a tag chip. A prototype antenna was fabricated and tested with a commercial tag chip. It has a 3 dB return loss bandwidth of 70 MHz and a uniform read range of over 9.7 m in the 902 - 928 MHz range when it is mounted on a cylindrical metal plate with a radius of curvature of 150 mm.

Light-emitting diodes are becoming the next-generation light source because of their prominent benefits in energy efficiency, versatility, and benign environmental impact. However, because of the unique polarization effects in III–V nitrides and the high complexity of light-emitting diodes, further breakthroughs towards truly optimized devices are required. Here we introduce the concept of artificial evolution into the device optimization process. Reproduction and selection are accomplished by means of an advanced genetic algorithm and device simulator, respectively. We demonstrate that this approach can lead to new device structures that go beyond conventional approaches. The innovative designs originating from the genetic algorithm and the demonstration of the predicted results by implementing structures suggested by the algorithm establish a new avenue for complex semiconductor device design and optimization.

The effect of the asymmetry in carrier concentration and mobility is studied in GaInN pn-junction light-emitting diodes (LEDs). We propose and present experimental evidence that the asymmetry in carrier concentration and mobility, and associated high-level injection phenomena, cause efficiency droop in GaInN LEDs. Low temperatures exacerbate the degree of asymmetry of the junction by reducing acceptor ionization, and shift high-injection-phenomena to lower currents. Accordingly, at temperatures near 80 K, we measure a greater droop compared to room temperature. The analysis of temperature-dependent I–V curves shows an excellent correlation between the onset of high-level injection and the onset of droop.

GaInN/GaN light-emitting triodes having two anodes for promoting the injection of holes into the active region were fabricated and characterized. It was found that the anode-to-anode bias modulates not only the hole-injection efficiency but also the effective light-emitting area and hence the current density through the active region. As the anode-to-anode bias increases, the efficiency at the same current density increases, whereas the efficiency droop decreases substantially, indicating that the limited hole-injection efficiency is one of the dominant mechanisms responsible for the efficiency droop in GaN-based light-emitting diodes.

Transparent dielectric layers with varying compositions of TiO2 and SiO2, and ITO are deposited on sapphire and Si substrates by using an RF sputter system. Inductively coupled plasma (ICP) reactive ion etching (RIE) of the ITO hard mask is examined under H2, CH4, and Cl2 chemical environments. The slope of the sidewall and the etch residue on the sidewall of the ITO hard mask are controlled by the flow rates of H2, CH4, and Cl2. ICP-RIE dry etch of TiO2 and SiO2 is investigated under fluorinated environments. Comparable etch rates of TiO2 and SiO2 (ratio ≈ 2:1) and high selectivity ≫ 1 over ITO are found. Graded-refractive-index (GRIN) layers, made up of multiple dielectric layers of TiO2 and SiO2, are patterned to form cylindrical pillars by ICP etching using the ITO hard mask. Fluorine containing residues are identified on the TiO2 and SiO2 surfaces. Various etch chemistries are investigated to obtain smooth, vertical, and residue-free sidewalls of the GRIN pillars.

Graded-refractive-index (GRIN) coatings, composed of multiple dielectric layers of TiO2 and SiO2 are sputter-deposited on the nitrogen-face of thin-film GaInN/GaN light emitting diodes (LEDs). The thickness and refractive index of each layer in the GRIN stack is designed to minimize light trapping inside the LED caused by total internal reflection at the semiconductor–air interface. Patterning the GRIN stack forms an optically functional surface, which converts trapped modes of light into desirable extracted modes that have preferential directions. Inductively coupled-plasma reactive-ion-etching is used to fabricate various patterns, including arrays of cylindrical pillars and diamond-shaped pillars on the GRIN coatings. In comparison to an uncoated planar reference device, the light-output power is enhanced by 131% and 104% for an array of GRIN diamond-shaped pillars and an array of GRIN cylindrical pillars, respectively. This enhancement in light-output power is comparable to N-face roughened LEDs, which show an enhancement of 124%. In addition, the peak emission intensity of the GRIN LEDs with an array of GRIN pillars is between 25° and 55° off-surface-normal.

III-V nitrides form the backbone of light-emitting diode (LED) technology. However, the relevance of the very strong polarization fields in III-V nitride LEDs remains unclear. Here, we demonstrate the tuning of polarization fields by mechanical force. For compressive strain in a GaInN LED epitaxial layer, we find: (i) redistribution of intensity within the electroluminescence spectrum; (ii) a decrease in the peak efficiency at low current densities; and (iii) an increase in light-output power at high current densities. These findings show the relevance of transport effects in the efficiency droop.

Reduction in the light-output power in GaN-based light-emitting diodes (LEDs) with increasing temperature is a well-known phenomenon. In this work, temperature dependent external-quantum-efficiency versus current curves are measured, and the mechanisms of recombination are discussed. Shockley-Read-Hall recombination increases with temperature and is found to greatly reduce the light output at low current densities. However, this fails to explain the drop in light-output power at high current densities. At typical current density (35 A/cm2), as temperature increases, our results are consistent with increased Shockley-Read-Hall recombination and increased electron leakage from the active region. Both of these effects contribute to the reduction in light-output power in GaInN/GaN LEDs at high temperatures.

The design, fabrication, and characterization of a broadband, omnidirectional, graded-index anti-reflection (AR) coating on a glass substrate, fabricated by using nanostructured low-refractive-index (n = 1.05–1.40) silica, is reported. The AR coating is designed by using a genetic algorithm and fabricated by using oblique angle deposition. The AR coating is designed for the wavelength range of 400 to 2500 nm and 0 to 40° angle of incidence. The measured average optical transmittance between 1000 and 2000 nm is improved from 92.6 to 99.3% at normal incidence by using a two-layer AR coating deposited on both surfaces of the glass substrate.

We report on a self-organized nanoscale patterning method by using oblique angle deposition to enhance the light extraction in a GaInN light-emitting diode (LED). The method offers one-step processing with good controllability of the feature size and density of the nanopatterns by varying the deposition angle during oblique angle deposition, eliminating the need for photolithography and annealing. A 5-nm-thick silver (Ag) film, when deposited by using oblique angle deposition, spontaneously forms a nanoscale island-like morphology on the substrate. This method is used to texture p-type GaN with nanoscale features, which results in increased light extraction from a GaInN LED. At 100 mA, the nanotextured LED shows a 46% higher light output than a standard LED with unpatterned (planar) p-type GaN.

The internal quantum efficiency (IQE)-versus-carrier-concentration (n) curves of GaN-based light-emitting diodes have been frequently described by the ABC model: IQE=Bn2/(An+Bn2+Cn3). We show that this model predicts IQE-versus-n curves that have even symmetry. Phase-space filling makes the B and C coefficients concentration-dependent. We also show that IQE-versus-n curves that take into account phase-space filling possess even symmetry. In contrast, experimental IQE-versus-n curves exhibit asymmetry. The asymmetry requires a fourth-power or higher-power contribution to the recombination rate and provides insight into the mathematical form of the droop-causing mechanisms.

An unwanted green satellite emission in blue GaInN light emitting diodes (LEDs) has been investigated under various electrical bias conditions and temperatures. The dot-shaped green satellite emission appears only under electrically biased conditions of the LED (but not under photoluminescence excitation) and contributes directly to a high subthreshold leakage current of the LED. A weak temperature dependency of the green satellite emission intensity is observed indicating that tunneling-assisted radiative recombination involving Mg acceptors is the origin of the green emission.

Light-emitting diodes (LEDs) using phosphor conversion inherently have a wide variation of multiple parameters, including correlated color temperature (CCT), light output power (LOP), and forward voltage VF. A method, based on the formation of LED pairs with complementary charac- teristics, is presented to produce LED-based light sources with narrow CCT, LOP, and VF distributions. A weighted matching algorithm was developed to select LED pairs under a multiple selection criteria. Based on the weighted matching algorithm, 96 LEDs were combined to form 48 LED pairs, and it was experimentally demonstrated that the variance of LED CCT and LOP distributions decreased by 93% and 71%, respectively.

Optical interference filters utilizing tunable refractive index layers are shown to have higher spectral fidelity as compared to conventional filters consisting of non-tunable refractive index layers. To demonstrate this increase in spectral fidelity, we design and compare a variety of optical interference filters employing both tunable and non-tunable refractive index layers. Additionally, a five-layer optical interference filter utilizing tunable refractive index layers is designed and fabricated for use with a Xenon lamp to replicate the Air Mass 0 solar irradiance spectrum and is shown to have excellent spectral fidelity.

The thermal properties, including thermal time constants, of GaInN light-emitting diodes LEDs and laser diodes LDs are analyzed. The thermal properties of unpackaged LED chips are described by a single time constant, that is, the thermal time constant associated with the substrate. For unpackaged LD chips, we introduce a heat-spreading volume. The thermal properties of unpackaged LD chips are described by a single time constant, that is, the thermal time constant associated with the heat spreading volume. Furthermore, we develop a multistage RthCth thermal model for packaged LEDs. The model shows that the transient response of the junction temperature of LEDs can be described by a multiexponential function. Each time constant of this function is approximately the product of a thermal resistance, Rth, and a thermal capacitance, Cth. The transient response of the junction temperature is measured for a high-power flip-chip LED, emitting at 395 nm, by the forward-voltage method. A two stage RthCth model is used to analyze the thermal properties of the packaged LED. Two time constants, 2.72 ms and 18.8 ms are extracted from the junction temperature decay measurement and attributed to the thermal time constant of the LED GaInN/sapphire chip and LED Si submount, respectively. © 2010 American Institute of Physics.

We model the carrier recombination mechanisms in GaInN/GaN light-emitting diodes as R=An+Bn2+Cn3+f(n), where f(n) represents carrier leakage out of the active region. The term f(n) is expanded into a power series and shown to have higher-than-third-order contributions to the recombination. The total third-order nonradiative coefficient (which may include an f(n) leakage contribution and an Auger contribution) is found to be 8×10−29 cm6 s−1. Comparison of the theoretical ABC+f(n) model with experimental data shows that a good fit requires the inclusion of the f(n) term.

The confinement of electrons to the active region of GaInN light-emitting diodes(LEDs) is limited by the (i) inefficient electron capture into polar quantum wells, (ii) electron-attracting properties of electron-blocking layers (EBL), (iii) asymmetry in electron and hole transport, and (iv) unfavorable p-doping in the EBL for high Al content. To counteract these mechanisms, we employ tailored Si-doping in the quantum barriers (QBs). Experiments show a 37.5% enhancement in light-output power at high currents of one-QB-doped LEDs over all-QB-doped LEDs. These results are consistent with simulations showing that QB doping can be used to symmetrize the electron and hole distribution.

The characteristics of the reverse leakage current of GaInN/GaN multiple quantum well light-emitting diodes (LEDs) are examined with various n -type GaN doping concentrations and interpreted by using a tunnelling current model. Changing the doping concentration of the n -type GaN influences the tunnelling probability of electrons into the conduction band and thus the reverse leakage current. Reducing the doping concentration of the top 150 nm portion of the n -type GaN layer by half decreases the tunnelling probability, resulting in decrease of the reverse leakage current by 80% at %10%V without deterioration of any forward electrical properties of LEDs.