激光激发半导体使其降解的研究

Basics of semiconductor LEDs and lasers reliability in LEDs and lasers basic degradation mechanisms and enhancement factors reliability and degradation of AIGaAs/GaAs light sources reliability and degradation of InGaAs/InP surface emitting type LEDs reliability and degradation of InGaAsP/InP laser diodes degradation of MBE- and MOVPE-grown lasers degradation of bonds and heat sinks degradation modes and lifetime of semiconductor LEDs and lasers.

A new semiphenomenological statistical kinetic model for gradual degradation in semiconductor laser and light-emitting diodes is presented. In this model, the injection of a nonequilibrium electron-hole plasma increases the probability of structural changes and reduces their effective activation energy. Arrhenius-like expressions for the degradation rate with the pre-exponential factor and the effective activation energy as explicit functions of the material parameters are derived. Good agreement with experimental data is obtained.

It is experimentally demonstrated that the increase in residual spectral linewidth during device degradation is due to an increase in 1/f noise in a multiple quantum-well (MQW) distributed-feedback (DFB) laser. The origin of 1/f noise and its influence on device characteristics is discussed and clarified by observing the degradation behavior of the spectral linewidth.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Failure times of semiconductor lasers are usually lengthy under normal aging conditions. Determination of failure times typically involves extrapolation using a sublinear or linear model. It becomes increasingly difficult to experimentally determine activation energy and current exponent since data based on lower temperatures and lower stress currents are required. In this paper, the temperature and current dependences of 1310-nm buried heterostructure (BH) InP lasers were studied. We show that the activation energy of 1310-nm BH lasers based on life test data at 70/spl deg/C-100/spl deg/C is higher than the value of 0.4 eV suggested by Telcordia. The activation energies estimated by sublinear and linear models were 0.87 and 0.55 eV, respectively. We also show that the current exponents are 1.4 and 1.0, respectively, for sublinear and linear models. We discuss the implications of the reliability results in field reliability predictions.

After continuous-wave operation of ZnSe-based semiconductor laser diodes, the degradation of these devices was investigated using cathodoluminescence imaging and spectroscopy. Inside the stripe region, i.e., the carrier injection area, there were several dots in which the peak wavelength of emission from a ZnCdSe strained quantum well (QW) shifted with time to a shorter wavelength (blueshift). We consider that the blueshift is due to Cd/Zn interdiffusion. This interdiffusion is enhanced by the electron–hole recombination process (recombination enhanced interdiffusion). Furthermore, there were dark line defects (DLDs) in the 〈100〉 direction, outside the stripe region and running away from the dots, having emission with a blueshift. The peak wavelength of emission from the QW in the DLDs shifted to a longer wavelength (redshift). We consider that the redshift is due to the relaxation of strain in the QW by existing defects, which may originate in the blueshift dots and move outside the stripe region.

High-power semiconductor lasers have attracted widespread attention because of their small size, easy modulation, and high conversion efficiency. They play an important role in national economic construction and national defense construction, including free-space communication; industrial processing; and the medical, aerospace, and military fields, as well as other fields. The reliability of high-power semiconductor lasers is the key point of the application system. Higher reliability is sought in the military defense and aerospace fields in particular. Reliability testing and failure analysis help to improve the performance of high-power semiconductor lasers. This article provides a basis for understanding the reliability issues of semiconductor lasers across the whole supply chain. Firstly, it explains the failure modes and causes of failure in high-power semiconductor lasers; this article also summarizes the principles and application status of accelerated aging experiments and lifetime evaluation; it also introduces common techniques used for high-power semiconductor laser failure analysis, such as the electron beam-induced current (EBIC) technique and the optical beam-induced current (OBIC) technique, etc. Finally, methods used to improve the reliability of high-power semiconductor lasers are proposed in terms of the preparation process, reliability screening, and method application.

The degradation of AlGaAs based high power laser bars (808 nm) is modeled in terms of the thermal stress gradient induced by the overheating produced at a facet defect by self-absorption and nonradiative recombination. Using a thermomechanical model, the local heating at the defect is shown to induce local stress above the yield strength necessary for plastic deformation. Cathodoluminescence images of the facets show the formation of large facet defects. The role of the packaging stress is also elucidated. The power density dissipation and the local temperature necessary to achieve the plastic deformation are in good agreement with the experimental values reported for laser degradation.

Theory and experiment of the degradation of II-VI blue-green laser diodes using quantum-well (QW) structures are presented. We develop the fundamental equations for the operation current of laser diodes as a function of aging time under the constant optical power aging condition. We show experimental results for the increase of the operation current as a function of aging time and its dependence on the ambient temperature. We find that the increase of the threshold current and the operation current under constant optical output power aging condition is caused by the increase of the nonradiative recombination current due to the increase of the defect density. The generation of the defect density follows a kinetic mechanism caused by an electron-hole recombination enhancement. Thermal effects also accelerate the degradation of the laser diodes. Our theory agrees very well with the experiment.

A mechanism of degradation of semiconductor quantum-well lasers with a stripe contact more than 50 µm wide is proposed. This mechanism implies formation of several lasing channels at a carrier ambipolar diffusion length smaller than the contact width. The carrier diffusion length decreases with time due to the increase in the number of defects; as a result, the number of lasing channels increases and lasing spectrum is filled. The shape of the lasing spectrum can be used to predict the laser service life.

Various aspects of the failure mechanisms occurring with injection lasers are reviewed, with emphasis on the physical problems involved. Dislocation-related phenomena are known to dominate the short-term problems, and in general dislocations have to be avoided for longlife devices. Both glide processes, which are in general excitation-stimulated, and climb processes, are important, and are discussed in detail. Surface related problems also occur, and they may have a dominant influence on laser characteristics up to an operating time of say 103-104h, mainly due to the oxide formation of the surface. A special surface related problem is the so-called catastrophic degradation at high power levels of pulsed lasers, which is due to thermal runaway and melting at the mirror surface. The long term uniform degradation mode finally, involves recombination enhanced creation and transport of unidentified point defects within the active region of the device, resulting in a gradual reduction of the bulk radiative efficiency and consequently a raise in lasing threshold of a device.

Semiconductor lasers have been rapidly evolving to meet the demands of next-generation optical networks. This imposes much more stringent requirements on the laser reliability, which are dominated by degradation mechanisms (e.g., sudden degradation) limiting the semiconductor laser lifetime. Physics-based approaches are often used to characterize the degradation behavior analytically, yet explicit domain knowledge and accurate mathematical models are required. Building such models can be very challenging due to a lack of a full understanding of the complex physical processes inducing the degradation under various operating conditions. To overcome the aforementioned limitations, we propose a new data-driven approach, extracting useful insights from the operational monitored data to predict the degradation trend without requiring any specific knowledge or using any physical model. The proposed approach is based on an unsupervised technique, a conditional variational autoencoder, and validated using vertical-cavity surface-emitting laser (VCSEL) and tunable edge emitting laser reliability data. The experimental results confirm that our model (i) achieves a good degradation prediction and generalization performance by yielding an F1 score of 95.3%, (ii) outperforms several baseline ML based anomaly detection techniques, and (iii) helps to shorten the aging tests by early predicting the failed devices before the end of the test and thereby saving costs.

Abstract The origin of the increase in residual spectral linewidth during device degradation is experimentally and theoretically clarified in a multiple quantum well (MQW) distributed feedback (DFB) laser. Non‐radiative recombination current increases during device degradation and causes 1/ f noise to increase. This current 1/ f noise is the origin of the increase in the residual spectral linewidth. Through these degradation behaviours, a model showing a correlation between 1/ f noise and the semiconductor laser degradation is proposed.

No abstract

A statistical kinetic model for gradual degradation of semiconductor lasers and light-emitting diodes under the influence of pressure is presented. Within the framework of this model, the rate coefficient for disordering atom jumps, K, and the operating lifetime of the device, τ, are explicitly given in terms of temperature, pressure, material parameters, and free-carrier concentration. We find that a compressive pressure reduces the effective activation energy of the rate process and therefore accelerates degradation in GaAs- and InP-based devices.

The injection of a nonequilibrium electron-hole plasma increases the probability of structural changes in the semiconductor material and reduces the effective activation energy for defect formation and migration. The kinetic theory of short-lived large energy fluctuations of atoms is applied to link lasers and LED gradual degradation to material parameters. A comparison between GaAs and InP based devices is made based on diffusion experimental data.

Experimental results on LEDs show that a number of light-degrading mechanisms can operate which probably include migration of impurities or point defects. Dislocations when present in high densities also give rise to degradation in LEDs. Extrapolated lives to half intensity of between 105 and 107 h have been commonly projected for these devices, and this rather slow rate of degradation appears to have been achieved by minimizing impurity contamination, reducing strain and minimizing the dislocation density in the carrier recombination region. Relatively long lives have also been predicted for high-radiance, high-current-density lamps from thermally accelerated life-tests. Dark-spot and dark-line defects have been observed. In one case, the dark structure has been shown to result from the formation of tangles of dislocations generated during fabrication. High-current-density lamps which are free from dark structure also have projected lives of between 105 and 107 h. Generally, laser diodes can now be made which have lives in excess of 104 h. Details of the dark structure resulting from dislocations is reviewed and consideration is given to the remaining causes of degradation in lasers.

No abstract

&lt;p class="1Body"&gt;To enable high-performance fiber to the x (FTTx) and datacenter networks, it is important to achieve reliable and stable optical components over time. Laser diode is the essential building block of the optical components. Degradation analysis is critical for overall successful reliability design. In this paper, we study the modelling and experimental data of the InGaAs/InP laser degradation. We present a defect diffusion model that involves three propagation media (p-InGaAs contact, p-InP cladding and multi-quantum wells). We propose a simple constitutive equation based on the Gauss error function to describe the defect propagation. The physical model assumes that the p-InGaAs is the rate-limiting factor for the defect diffusion process.&lt;/p&gt;

Lead halide perovskite nanowires (NWs) are emerging as a class of inexpensive semiconductors with broad bandgap tunability for optoelectronics, such as tunable NW lasers. Despite exciting progress, the current organic-inorganic hybrid perovskite NW lasers suffer from limited tunable wavelength range and poor material stability. Herein, we report facile solution growth of single-crystal NWs of inorganic perovskite CsPbX3 (X = Br, Cl) and their alloys [CsPb(Br,Cl)3] and a low-temperature vapor-phase halide exchange method to convert CsPbBr3 NWs into perovskite phase CsPb(Br,I)3 alloys and metastable CsPbI3 with well-preserved perovskite crystal lattice and NW morphology. These single crystalline NWs with smooth end facets and subwavelength dimensions are ideal Fabry-Perot cavities for NW lasers. Optically pumped tunable lasing across the entire visible spectrum (420-710 nm) is demonstrated at room temperature from these NWs with low lasing thresholds and high-quality factors. Such highly efficient lasing similar to what can be achieved with organic-inorganic hybrid perovskites indicates that organic cation is not essential for light emission application from these lead halide perovskite materials. Furthermore, the CsPbBr3 NW lasers show stable lasing emission with no measurable degradation after at least 8 h or 7.2 × 10(9) laser shots under continuous illumination, which are substantially more robust than their organic-inorganic counterparts. The Cs-based perovskites offer a stable material platform for tunable NW lasers and other nanoscale optoelectronic devices.

No abstract

We studied the reliability of AlGaInP visible semiconductor lasers with three different coating films. The facet degradation was analyzed using an optical microscope and the micro-auger electron spectroscopy (µ-AES) sputter profiling method. We found that the degradation is associated with facet erosion. We discussed the causes of facet erosion.

The current status and understanding of various degradation phenomena in III-V optoelectronic devices are discussed, with special consideration given to semiconductor lasers and light emitting diodes fabricated from GaAlAs/GaAs, InGaAsP/InP, and InGaAsP/InGaP/GaAd double-heterostructure materials. Three major degradation phenomena are discussed: rapid degradation, gradual degradation, and catastrophic failure. For each type of these degradation phenomena, methods for eliminating degradation are proposed.

Cross-sectional scanning tunneling microscopy is performed on operating semiconductor quantum well laser devices to reveal real time changes in device structure. Low and nominally doped capping regions adjacent to the quantum well active region are found to heat under normal operating conditions. The increase in anion-vacancy defect formation and the generation of surface states pins the Fermi level at the surface and begins the process of catastrophic optical degradation which eventually destroys the device. The technique has implications for the study of defect generation and in-operation changes in all nanostructures.

. Catastrophic degradation takes place in case of reaching critical values of laser radiation density power in semiconductor lasers with electronically pumped energy made from single crystals of some compounds. It has been accompanied by mechanical destruction of the surface at resonator ends, an irreversible decrease in radiation power and an increase in generation threshold. Moreover, during the catastrophic degradation of semiconductor lasers under the action of intrinsic radiation, significant changes in the crystal structure occur within the single crystal: dislocation density reaches a value more 10 12 –10 15 cm –2 . It has been shown that initial density of dislocations and critical power density of the intrinsic radiation are inversely proportional. Thus, the degradation process of semiconductor lasers is directly related to generation and multiplication of dislocations during laser operation. Mechanical destruction of a crystal lattice occurs at critical values of laser radiation power and dislocation density. To clarify the proposed mechanism for the degradation of semiconductor lasers, it is necessary to take into account an effect of dislocations on optical properties of semiconductors. Typically, this effect is considered as follows: dislocations cause an appearance of a local deformation field and, in addition, form space-charge regions that surround a dislocation core in the form of a charged tube. The paper proposes a model of the phenomenon under study: large stresses arise in the dislocation core, leading to a displacement of individual atoms and deformation of the crystal lattice. Lattice deformation in the dislocation core leads to a local change in the width of a forbidden band. This change value is about 10 –2 eV for a screw dislocation and 10 –1 eV for a boundary dislocation. The mechanism of this change is that aforementioned deformation leads to a multiple rupture of electronic bonds and an increase in the electron concentration in the dislocation core to approximately value 10 18 cm –3 . The developed analytical model of the degradation mechanism allows to perform selection of a semiconductor and estimation of a laser operating mode under conditions of increased radiation power.

Recombination-enhanced annealing of defects in semiconductors has been observed directly for the first time. The defects were produced in GaAs by 1-MeV electron irradiation and observed by transient-junction-capacitance techniques. The data clearly relate the enhanced defect annealing rate to electron-hole recombination processes at the defect.

We present a quantitative theory to describe enhancement of defect reaction rates upon electron-hole recombination. The theory is based on the following mechanism: energy liberated upon nonradiative electron or hole capture is converted largely into vibrational energy that is initially localized in the vicinity of the defect. This vibrational energy can be utilized to promote defect reactions such as diffusion. The process can be described using a formulation similar to the successful Rice-Ramsperger-Kassel theory of unimolecular reactions. The resulting expression for the enhanced reaction rate depends upon two physical properties of the defect: the number of effective oscillators of the "defect molecule" and the rate of dissipation of local vibrational energy to the lattice. The theory is consistent with recent experiments of Kimerling and Lang and may be useful for understanding several related processes occurring in semiconductors.

Abstract Significant evidence linking changes in the macroscopic properties of semiconductors with recombination processes associated with injection current components has accumulated during the past 15 years or so. These effects play a critical role in performance degradation for many semiconductor devices. In this paper we review the developments in understanding of this effect. We concentrate on some very recent detailed studies which provide convincing evidence that carrier capture or recombination processes at deep traps can induce or enhance dissociation and/or migration of the defect or impurity centres which produce them. Thus, the ‘phonon-kick’ mechanism invoked to explain the early macroscopic property changes has now received strong support. We show that the new findings provide a hierarchy of increasingly detailed information on the properties of traps which may show this effect, now termed a recombination enhanced (or induced) defect reaction (REDR). Carrier capture or recombination at these traps must be predominantly non-radiative, to provide the rapid, large evolution of localized vibrational energy necessary for the effect. Deep level transient capacitance spectroscopy has located such energy levels, particularly in irradiated GaAs and GaP, and has clearly demonstrated the sensitivity of the concentration of the relevant centres to electron-hole recombination processes. The resultant defect motility can improve the minority carrier lifetime, particularly adjacent to dislocations which act as sinks for these defects. The defect capture can also induce the type of dislocation climb which is a key feature in the degradation of semiconductor lamps. Direct evidence for these effects has been obtained very recently by electron microscopy, which has also shown the coalescence into small loops and precipitates of defects left behind climbing dislocations in degraded material without prior radiation damage. We show that luminescence spectroscopy has also recently provided evidence for the dissociation of complex radiative centres under REDR, for example the familiar Zn-O activator in red GaP LEDs and an H-related centre which is luminescent in several polytypes of H-implanted SiC. The dissociation and hence the luminescence can be largely restored by annealing in both examples. We discuss the properties of the H-related spectra which provide uniquely specific evidence for the vulnerability of this centre to dissociation under REDR, even at the lowest temperature. We also give recent evidence that a complex Zn-related luminescent centre can be formed by REDR in irradiated GaP. Finally, we review a recent theoretical description of REDR which assumes quasi-equilibrium for the energy liberated locally at the centre. We note that an adiabatic model may be more appropriate for those centres where the reactions are not enhanced thermal processes but rather are induced athermally.

The interaction between $〈100〉$ climb-induced dislocation networks and a naturally occurring deep level (the $\mathrm{DX}$ center) in ${\mathrm{Al}}_{x}{\mathrm{Ga}}_{1\ensuremath{-}x}\mathrm{As}$ has been analyzed by scanning deeplevel transient spectroscopy. The concentration of $\mathrm{DX}$ centers was observed to decrease markedly in the vicinity of climb-induced $〈100〉$ networks, but remained nearly unchanged near glide-induced $〈110〉$ dislocation networks.

No abstract

Nonradiative-recombination-related defects are significant for optoelectronic semiconductor devices. Here, we analyze nonradiative-recombination processes in ionic semiconductors using first-principles density-functional theory. In ionic group II-VI semiconductors, we find that large lattice relaxations of anion vacancies caused by strong Coulomb interactions between different charged defect states can significantly enhance recombination processes through a two-level recombination mechanism. Specifically, we show that the defect level of the 2+ charged anion vacancy $\mathrm{(}{V}_{\mathrm{Se}}^{2+})$ in group II-VI $\mathrm{Zn}\mathrm{Se}$ is close to the conduction-band minimum and easily captures an electron to form a metastable 1+ charged state $\mathrm{(}{V}_{\mathrm{Se}}^{+})$; then, the large lattice relaxation, on account of the change in Coulomb interactions locally in the different charged states, rapidly changes this metastable state to a stable one and simultaneously move the defect level of ${V}_{\mathrm{Se}}^{+}$ closer to that valence-band maximum, and thus, increases the hole-capture rate. Compared with the Shockley-Read-Hall nonradiative-recombination theory based on a single defect level, this two-level recombination mechanism involving anion vacancies can greatly increase the nonradiative-recombination rate in ionic group II-VI semiconductors. This understanding is expected to be useful for the study of the nonradiative-recombination process in ionic semiconductors for applications in the field of optoelectronic devices.

New techniques and apparatus are presented which overcome some of the limitations of previous capacitance transient techniques and extend the useful range of capacitance transient measurements to intermediate depth impurity and defect states in semiconductors. This development greatly enhances the usefulness of capacitance techniques as a tool to study nonradiative recombination. These techniques and apparatus are used here to measure the electron-capture cross sections and concentrations of isolated oxygen donors and ZnO complex luminescence centers in p-type Zn- and O-doped GaP. These results agree with the donor-acceptor pairing theory and with the conclusion of previous capacitance measurements that the ioslated O donor is of negligible importance as a recombination center in these samples. Data is also presented which shows the effect of the junction electric field in greatly enhancing the thermal emission rate of electrons trapped in ZnO centers.

No abstract

No abstract

No abstract

We present a kinetic model for the optical output degradation of light-emitting diodes based on the carrier-recombination enhanced defect motion. Our model leads to analytical solutions and universal curves for the optical output power and the defect density as a function of the normalized aging time with the initial quantum efficiency as the determining parameter. The theoretical results explain very well the time dependence of the II-VI light-emitting diodes under constant current aging condition. The faster aging rate with increasing bias current or temperature is also investigated both experimentally and theoretically, resulting in a very good agreement. Our model provides a quantitative description of the light-emitting diode aging characteristics for compound semiconductors in the presence of electron-hole recombination-enhanced defect generation.

Introduction. Materials and Structures of Optical Devices: Materials for Optical Devices. Structures of Optical Devices. Crystal Growth: Preparation of Materials by LPE. Preparation of Materials by MOVPE. Preparation of Materials by MBE. Fabrication Processes of Optical Devices: Fabrication Processes. Device Characteristics and Life Testing: Device Characteristics. Life Testing of Devices. Evaluation Techniques for III-V Compound Semiconductors and Degraded Optical Devices: Classification of Evaluation Techniques. Visual Inspection. Chemical Etching. Optical Measurement. Electrical Measurement. Structural Evaluation of Semiconductors by Transmission Electron Microscopy. Analytical Techniques. Flow Chart for Evaluation of Degraded Optical Devices. Materials Issues in III-V Compound Semiconductors I -- Defect Generation: Classification of Defects. Growth-induced Defects. Process-induced defects. Materials Issues in III-V Compound Semiconductors I -- Thermal Stability of III-V Alloy Semiconductors: Composition Modulated Structures. Ordered Structures. Influence of Modulated Structures on the Properties and Reliability of Optical Devices. Classification of Degradation Modes and Degradation Phenomena in Optical Devices: Classification of Degradation Modes in the Life Testing of Lasers and LEDs. Classification of Degradation Phenomena in Lasers. Degradation in LEDs. Influence of Stress on the Device Degradation. Degradation I -- Rapid Degradation: Rapid Degradation in GaAlAs/GaAs Optical Devices. Rapid Degradation in InGaAsP/InP Optical Devices. Rapid Degradation in InGaAsP/InGaP Optical Devices. Comparison of Recombination-enhanced Defect Reaction in Different III-V Materials. Elimination of the Rapid Degradation. Degradation II -- Gradual Degradation: Gradual Degradation in GaAlAs DH LEDs. Gradual Degradation in InGaAsP/InP DH LEDs. Comparison of Gradual Degradation in GaAlAs/GaAs and In GaAsP/InP Optical Devices. Enhancement of Gradual Degradation by Internal Stress in GaAlAs Visible Lasers. Degradation III -- Catastrophic Failure: Catastrophic Failure in Lasers. Catastrophic Failure in LEDs. Elimination of Catastrophic Failure.

No abstract

The early stages of catastrophic optical damage (COD) in 808 nm emitting diode lasers are mapped by simultaneously monitoring the optical emission with a 1 ns time resolution and deriving the device temperature from thermal images. COD occurs in highly localized damage regions on a 30 to 400 ns time scale which is determined by the accumulation of excess energy absorbed from the optical output. We identify regimes in which COD is avoided by the proper choice of operation parameters.

We experimentally and theoretically studied degradation phenomena and their mechanism in broad-area semiconductor lasers (BA-LDs) with optical feedback (OFB). We made two types of BA-LDs (one is consisted of AlGaAs emitting at 808nm in TE mode, and another one is consisted of AlGaInP emitting at 642nm in TM mode), and investigated conditions of the degradations caused by an optical feedback. The both types of BA-LDs showed degradations depending on feedback rate and output power. For example, both BA-LDs were damaged with about 20% of intensity feedback rate at half of an output power of a catastrophic optical mirror damage (COMD) levels. To describe a theoretical model for the degradation, the optical power at a front facet of the BA-LDs was calculated and compared with the COMD level of the solitary BA-LDs. In the theoretical model, we included a threshold reduction caused by the OFB. We found that the degradation was explained by a constructive interference between internal and the feedback optical fields. The BA-LDs are damaged when a coherent sum of those fields exceeds the solitary COMD level. We found that the threshold reduction decreases a critical value of the feedback rate corresponding to the damage at low output power regime, and also found that there is an optimum reflectivity of the front facet. The theoretical results show a good agreement with experimental results. According to this model, we can avoid the damages induced by the OFB in the various applications.

Among the limitations known from semiconductor lasers, catastrophic optical damage (COD) is perhaps the most spectacular power‐limiting mechanism. Here, absorption and temperature build up in a positive feedback loop that eventually leads to material destruction. Thus, this is truly an ultimate mechanism, and its continued suppression is a manifestation of progress in device design and manufacturing. After an overview of the current state of knowledge, new investigations of COD using artificially micrometer‐sized starting points created within the active zone in the cavity of 450 nm GaN semiconductor lasers are reported on. Defect growth mechanisms and characteristics are studied during 800 ns current pulses. The defect growth follows the highest light intensity. Secondary defect patterns are studied: complete destruction of the active zone and generation of a point defect cloud at least ≈10 μm into the remaining surrounding material. Extremely large angles (&gt;90°) of damage growth are traced back to the material properties and the aging scenario. The results are compared with former experiments with GaAs‐based lasers.

Mirror facet coating with a high-refractive-index film such as TiO 2 and Ta 2 O 5 is proposed to obtain high-power short-wavelength semiconductor lasers. A drastic increase in the output power attainable before catastrophic optical damage occurs is theoretically predicted for such a laser without the need to decrease facet reflectivity. This increase is shown to originate from destructive interference of laser light fields in the vicinity of the coated mirror facet when the film's refractive index is larger than the square root (≈1.8–1.9) of the laser's effective refractive index.

No abstract

Abstract Catastrophic Optical Damage (COD) occurs when the facet of high power semiconductor lasers melt due to extreme self‐absorption. COD tends to occur in short wavelength lasers but seems to be less of a problem in long wavelength devices. In this paper, we use hydrostatic pressure as a means of changing the wavelength in order to investigate the dependence of COD on emission wavelength. We find that the recorded output power threshold at the onset of COD decreases with increasing pressure (decreasing wavelength). Our measurements further reveal that this threshold is closely correlated with the increase in optical confinement factor with pressure which suggests that the wavelength dependence may largely be attributed to the change in mode size with operating wavelength. (© 2004 WILEY‐VCH Verlag GmbH &amp; Co. KGaA, Weinheim)

Mechanisms relevant for the COD in GaAs-based diode lasers are reviewed. Experiments, where COD is artificially provoked, represent a main topic. The sequence of events and the kinetics down to a nanosecond timescale are addressed.

Design considerations for fabricating highly efficient uncooled semiconductor lasers are discussed. The parameters investigated include the temperature characteristics of threshold current, quantum efficiency, and modulation speed. To prevent carrier overflow under high-temperature operation, the electron confinement energy is increased by using the Al/sub x/Ga/sub y/In/sub 1-x-y/As/InP material system instead of the conventional Ga/sub x/In/sub 1-x/As/sub y/P/sub 1-y//InP material system. To reduce the transparency current and the carrier-density-dependent loss due to the intervalence-band absorption, strained-layer quantum wells are chosen as the active layer. Experimentally, 1.3-/spl mu/m compressive-strained five-quantum-well lasers and tensile-strained three-quantum-well lasers were fabricated using a 3-/spl mu/m wide ridge-waveguide laser structure. For both types of lasers, the intrinsic material parameters are found to be similar in magnitude and in temperature dependence if they are normalized to each well. Specifically, the compressive-strained five-quantum-well lasers show excellent extrinsic temperature characteristics, such as small drop of 0.3 dB in differential quantum efficiency when the heat sink temperature changes from 25 to 100/spl deg/C, and a large small-signal modulation bandwidth of 8.6 GHz at 85/spl deg/C. The maximum 3 dB modulation bandwidth was measured to be 19.6 GHz for compressive-strained lasers and 17 GHz for tensile-strained-lasers by an optical modulation technique. The strong carrier confinement also results in a small k-factor (0.25 ns) which indicates the potential for high-speed modulation up to 35 GHz. In spite of the aluminum-containing active layer, no catastrophic optical damage was observed at room temperature up to 218 mW for compressive-strained five-quantum-well lasers and 103 mW for tensile-strained three-quantum-well lasers. For operating the compressive-strained five-quantum-well lasers at 85/spl deg/C with more than 5 mW output power, a mean-time-to-failure (MTTF) of 9.4 years is projected from a preliminary life test. These lasers are highly attractive for uncooled, potentially low-cost applications in the subscriber loop.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

High power semiconductor lasers have found increased applications. Indium solder is one of the most widely used solders in high power laser die bonding. Indium solder has some advantages in laser die bonding. It also has some concerns, however, especially in terms of reliability. In this paper, the reliability of indium solder die bonding of high power broad area semiconductor lasers was studied. It was found that indium solder bonded lasers have much shorter lifetime than AuSn solder bonded devices. Catastrophic degradation was observed in indium solder bonded lasers. Nondestructive optical and acoustic microscopy was conducted during the lifetime testing to monitor the failure process and destructive failure analysis was performed after the lasers failed. It was found that the sudden failure was caused by electromigration of indium solder at the high testing current of up to 7A. It was shown that voids were created and gradually enlarged by indium solder electromigration, which caused local heating near the facets of the laser. The local heating induced catastrophic optical mirror damage (COMD) of the lasers. It was discussed that current crowding, localized high temperature, and large temperature gradient contributed to the fast indium solder electromigration. It was observed that some bright pattern structures appeared on the front facet of the indium solder bonded lasers after the devices failed and the bright patterns grew and spread upon further testing. Failure analysis showed that the bright pattern structure apparent on the front facet was due to crystallization of the TiOx material of the front facet coating as a result of overheating during lifetime testing. It was concluded that indium solder is not suitable for high power laser applications due to electromigration at high current densities and high temperatures.

Semiconductor pump lasers are an important component in erbium-doped fiber amplifiers and Raman amplifiers. Thermal management has become one of the major obstacles of pump laser development. Understanding of the thermal behavior of high-power laser packages is crucial to the thermal design and optimization of pump lasers. In this paper, we report on the thermal characteristics of a high-power pump laser and discuss the issues associated with heat dissipation. The thermal management of high-power pump laser modules mainly consists of three aspects. One is the thermal resistance reduction which reduces bulk temperature rise in the laser diode chip. The second is facet temperature control, and the third is the thermoelectric cooler (TEC) coefficient of performance improvement. In this paper, the approaches to reduce thermal resistance and facet temperature at the chip level and package level will be reviewed, and the thermal design and optimization of the package assembly to improve the TEC coefficient of performance will be discussed. The thermal resistance of a pump laser could be reduced up to 40% by the proper design of the laser chip and epi-down bonding. An unpumped window design in the pump laser diode is proven to be very effective in reducing the facet temperature and increasing the catastrophic optical mirror damage level. Assembly and package optimization can provide more uniform temperature distribution on TEC cold plate which is critical in improving the TEC coefficient of performance.

We propose and demonstrate a novel approach to the coating of semiconductor laser facets. In this approach, processed semiconductor lasers are cleaved in a high-vacuum system immediately followed by coating of the vacuum-exposed facet with a very thin Si layer (≤100 Å) and a large band gap dielectric (Al2O3) layer. The Si layer is sufficiently thin to avoid the formation of quantized bound states in the Si. GaAs coated with thin Si and Al2O3 have a higher luminescence yield and a lower surface recombination velocity than bare GaAs surfaces as well as GaAs surfaces coated with Al2O3 only. A surface recombination velocity of 3×104 cm/s has been obtained using a modified dead layer model for the Si/Al2O3 sample. It is also shown that lasers which are cleaved in vacuum and subsequently coated with Si and Al2O3 have improved properties including an increased threshold for catastrophic optical damage.

No abstract

Improper back-coupling of optical feedback is a known cause for accelerated gradual degradation as well as catastrophic optical mirror damage of GaAs-based high-power semiconductor lasers. For those material systems emitting below the GaAs bandgap (870 nm at room temperature), such damage is induced by feedback-light absorption in the non-transparent cap layers and substrate, which intensifies the heating of the area close to the emission facet. For InGaAs/AlGaAs-based diode lasers in the spectral range from 880 to 1060 nm, however, all passive layers as well as the substrate are, with respect to interband absorption, transparent for the lasing wavelength so that, in this case, the degrading mechanism of misaligned optical feedback could not be explained entirely. By use of an experimental multi-device study of 950-nm broad-area lasers, we demonstrate that the threshold of catastrophic optical damage is decreased the most by positioning the feedback return spot such that it covers the p-side solder and the highly p-doped semiconductor layers. By means of thermoreflectance microscopy in conjunction with numerical simulations we find that such low damage thresholds are caused by a strong localization of the absorption of feedback light close to the laser front facet.

A numerical model is introduced that self-consistently calculates the time dependent axial variations of photon density, carrier density and temperature in semiconductor lasers. The most important approximations are outlined. In order to illustrate the capability of the model, some results are shown for an asymmetrically coated DQW GaAs/GaAlAs edge emitting laser diode. The temperature rise at the facets and the corresponding profiles of carrier and photon density are calculated. The asymmetric behaviour of the profiles is discussed. The heating is calculated as a function of surface recombination velocity at the mirrors and as a function of injection current. The calculations provide insight into the process of facet heating and catastrophic optical damage. The calculations are confirmed by experimental investigations.

We report a novel approach for increasing the output power in passively mode locked semiconductor lasers. Our approach uses epitaxial structures with an optical trap in the bottom cladding that enlarges the vertical mode size to scale the pulse saturation energy. With this approach we demonstrate a very high peak power of 9.8 W per facet, at a repetition rate of 6.8 GHz and with pulse duration of 0.71 ps. In particular, we compare two GaAs/AlGaAs epilayer designs, a double quantum well design operating at 830 nm and a single quantum well design operating at 795 nm, with vertical mode sizes of 0.5 and 0.75 μm, respectively. We show that a larger mode size not only shifts the mode locking regime of operation towards higher powers, but also produces other improvements in respect of two main failure mechanisms that limit the output power: the catastrophic optical mirror damage and the catastrophic optical saturable absorber damage. For the 830 nm material structure, we also investigate the effect of non-absorbing mirrors on output power and mode locked operation of colliding pulse mode locked lasers.

Near-field properties of the emission of broad area semiconductor diode lasers under extremely high pumping of up to ∼50 times the threshold are investigated. A transition from a gain to thermally-induced index guiding is shown under operation with single pulses of 300 ns duration. At highest output powers, catastrophic optical damage is observed which is studied in conjunction with the evolution of time-averaged filamentary near-field properties. Dynamics of the process is resolved on a picosecond time scale.

Recent advances in thermal management and improvements in fabrication and facet passivation enabled extracting unprecedented optical powers from laser diodes (LDs). However, even in the absence of thermal roll-over or catastrophic optical damage (COD), the maximum achievable power is limited by optical non-linear effects. Due to its non-linear nature, two-photon absorption (TPA) becomes one of the dominant factors that limit efficient extraction of laser power from LDs. In this paper, theoretical and experimental analysis of TPA in high-power broad area laser diodes (BALD) is presented. A phenomenological optical extraction model that incorporates TPA explains the reduction in optical extraction efficiency at high intensities in BALD bars with 100μm-wide emitters. The model includes two contributions associated with TPA: the straightforward absorption of laser photons and the subsequent single photon absorption by the holes and electrons generated by the TPA process. TPA is a fundamental limitation since it is inherent to the LD semiconductor material. Therefore scaling the LDs to high power requires designs that reduce the optical intensity by increasing the mode size.

No abstract

The temperature of single quantum well semiconductor laser facets increases during operation, eventually reaching a critical temperature, thermal runaway, and catastrophic optical damage. A study of changes in composition of the near-surface region of facets which accompany heating has been carried out for continuously operated, uncoated AlGaAs-GaAs-AlGaAs graded index separately confined heterostructure single quantum well lasers. High resolution depth profiles by scanning Auger microscopy show that the laser facets can be quite variable in initial composition, and undergo pronounced stoichiometry changes even during the first few minutes of operation. At longer times a continuing out-migration of the group III elements is observed. Unlike the double heterojunction lasers, facet oxidation is not pronounced and is not responsible for diffusion of Ga and Al. There are indications, however, that a slow leakage of oxygen into the crystal may occur. Spatially resolved analyses provide evidence that carrier-mediated elemental redistribution is an important factor in facet degradation. The progressive accumulation of defects which may act as non-radiative recombination centers provides a simple means of facet heating. Analyses of lasers which have suffered catastrophic damage indicate that the facets are not always melted, and that there is no typical chemical state which distinguishes them from facets of lasers which are fully operational. These results are compared to studies of facet degradation in double heterojunction lasers. Implications of the data for models of catastrophic optical damage are discussed.

An easily fabricated semiconductor laser, utilizing a mesa waveguide structure with a lenslike tapered active region for control of the optical laser mode, has been grown by metalorganic-chemical vapor deposition. This index-guided laser, which has thresholds of 40–60 mA and high differential quantum efficiency (∼40–50%), operates in the fundamental transverse mode up to its catastrophic damage limit. Single longitudinal mode is also obtained to greater than twice threshold.

The authors describe a straightforward experimental technique for measuring the facet temperature of a semiconductor laser under high-power operation by analyzing the laser emission itself. By applying this technique to 1-mm-long 980-nm lasers with 6- and 9-μm-wide tapers, they measure a large increase in facet temperature under both continuous wave (CW) and pulsed operation. Under CW operation, the facet temperature increases from ∼25°C at low currents to over 140°C at 500 mA. From pulsed measurements they observe a sharper rise in facet temperature as a function of current (∼400°C at 500mA) when compared with the CW measurements. This difference is caused by self-heating which limits the output power and hence facet temperature under CW operation. Under pulsed operation the maximum measured facet temperature was in excess of 1000°C for a current of 1000 mA. Above this current, both lasers underwent catastrophic optical damage (COD). These results show a striking increase in facet temperature under high-power operation consistent with the facet melting at COD. This is made possible by measuring the laser under pulsed operation.

Several factors limit the reliable output power of a semiconductor laser under CW operation, such as carrier leakage, thermal effects, and catastrophic optical mirror damage (COMD). Ever higher operating powers may be possible if the COMD can be avoided. Despite exotic facet engineering and progress in non-absorbing mirrors, the temperature rise at the facets puts a strain on the long-term reliability of these diodes. Although thermoelectrically isolating the heat source away from the facets with non-injected windows helps lower the facet temperature, data suggests the farther the heat source is from the facets, the lower the temperature. In this letter, we show that longer non-injected sections lead to cooler windows and biasing this section to transparency eliminates the optical loss. We report on the facet temperature reduction that reaches below the bulk temperature in high power InGaAs/AlGaAs lasers under QCW operation with electrically isolated and biased windows. Acting as transparent optical interconnects, biased sections connect the active cavity to the facets. This approach can be applied to a wide range of semiconductor lasers to improve device reliability as well as enabling the monolithic integration of lasers in photonic integrated circuits.

No abstract

Catastrophic optical mirror damage (COMD) and catastrophic optical bulk damage (COBD) are the main factors that affect the reliability of semiconductor lasers. In this paper, we characterize the COMD and COBD failure modes by examining the voltage changes at the current point where failure occurs, as well as by using the electroluminescent technique. Our study reveals that the voltage has an increase at the failure current point for COMD samples, in which failure occurs due to the damage to the facet mirrors; for COBD samples, in which failure occurs inside the laser cavity, the voltage exhibits a decrease and the amount of voltage decrease is roughly proportional to the size of the damaged area.

Abstract Output power and reliability are the most important characteristic parameters of semiconductor lasers. However, catastrophic optical damage (COD), which usually occurs on the cavity surface, will seriously damage the further improvement of the output power and affect the reliability. To improve the anti-optical disaster ability of the cavity surface, a non-absorption window (NAW) is adopted for the 915 nm InGaAsP/GaAsP single-quantum well semiconductor laser using quantum well mixing (QWI) induced by impurity-free vacancy. Both the principle and the process of point defect diffusion are described in detail in this paper. We also studied the effects of annealing temperature, annealing time, and the thickness of SiO 2 film on the quantum well mixing in a semiconductor laser with a primary epitaxial structure, which is distinct from the previous structures. We found that when compared with the complete epitaxial structure, the blue shift of the semiconductor laser with the primary epitaxial structure is larger under the same conditions. To obtain the appropriate blue shift window, the primary epitaxial structure can use a lower annealing temperature and shorter annealing time. In addition, the process is less expensive. We also provide references for upcoming device fabrication.

High power, 0.81-μm-emitting, semiconductor diode lasers are used as pump sources for Nd:YAG solid-state lasers. Devices (1-mm-long) consisting of a InGaAsP/In0.5(Ga0.9Al0.1)0.5P/In0.5(Ga0.5Al0.5)0.5P laser structure provide a threshold-current density, Jth, of 290 A/cm2 and a relatively high threshold-current characteristic temperature, T0 (140 K). Uncoated diode lasers (1.2-mm-long) have a maximum continuous wave output power of 5 W (both facets) at 20 °C. The internal power density at catastrophic optical mirror damage (COMD), P̄COMD, is determined to be 9.1 MW/cm2; that is, 1.8 times that for GaAs-active layer, Al-free, uncoated devices. Coated, InGaAsP-active devices are expected to have P̄COMD=18 MW/cm2, more than twice the P̄COMD of AlGaAs-active, 0.81-μm-emitting devices with the same emitting aperture. Therefore, 0.81-μm-emitting, InGaAsP-active diode lasers should operate reliably at powers at least twice those of AlGaAs-based devices with the same contact-stripe geometry.

To reduce the surface states of GaAs and related semiconductors which originate from native oxides on a surface, we developed a simple surface treatment method in which the surface oxides could be physically sputtered by Ar plasma irradiation in an electron cyclotron resonance chemical vapor deposition (ECR-CVD) apparatus. In the experiment of Ar irradiation of a GaAs surface, we were able to determine the optimum irradiation time at which the native-oxide-related surface states were almost removed without damaging the irradiated surface. Then, we applied this method to an actual 0.98-µm semiconductor laser in which catastrophic optical damage (COD) failure occurred due to the surface states of the output facet. By irradiating Ar plasma to the output facet of the laser and removing the oxide-related surface states there, tolerance to COD was remarkably improved compared with that of a conventional non-Ar-irradiated one.

An investigation was made of catastrophic degradation of electron-beam-pumped gallium arsenide lasers operating at temperatures of 80 and 300°K. The critical radiation density in the semiconductor resonator resulting in its damage ranged from 2 to 15 MW/cm2, depending on the temperature, optical homogeneity, and quality of the surface treatment of the crystal. An analysis was made of the mechanisms of the fracture of gallium arsenide crystals under the action of their own laser radiation.

No abstract

The microscopic processes accompanying the catastrophic optical damage process in semiconductor lasers are discussed. For 808 and 650 nm edge-emitting broad-area devices relevant parameters such as surface recombination velocities, bulk and front facet temperatures are determined and discussed. Facet temperatures vs. laser output and temperature profiles across laser stripes reveal a strong correlation to near-field intensity and degradation signatures. Furthermore, the dynamics of the fast catastrophic optical damage process is analyzed by simultaneous high-speed infrared thermal and optical imaging of the emitter stripe. The process is revealed as fast and spatially confined. It is connected with a pronounced impulsive temperature flash detected by a thermocamera.

Current high performance optical communication systems demand high power erbium doped fiber amplifiers. These amplifiers are powered by edge emitting 980 nm pump lasers, which are limited in optical power by catastrophic optical damage and require external gratings for wavelength stabilization. Conventional surface emitting lasers (Miller et al, 2000) have demonstrated very high powers from large apertures (320 /spl mu/m), but only in a highly multimode beam (/spl sim/3500 modes). In this paper, we report electrically pumped GaInAs surface emitting semiconductor lasers that have both the optical mode and the wavelength controlled by an extended cavity. These lasers have produced record power and brightness levels in both multi-mode and single-mode operation. Output power levels of more than 1 W CW from a multi-mode (beam quality parameter M/sup 2//spl sim/10-20) device and more than 500 mW CW in a TEM/sub 00/ mode (M/sup 2//spl sim/1.2) have been achieved.

A single frequency laser structure is obtained by coupling a high order mode of a semiconductor waveguide to a low index polymer waveguide. The device does not require a grating or regrowth, emits in a mode compatible with optical fibers, and may be immune to catastrophic mirror damage. The epilayers of the semiconductor waveguide use quarterwave reflectors to support a mode with a low enough effective index to phase match to the polymer waveguide. The coupling between the two waveguides is highly frequency selective and therefore stabilizes the wavelength. Preliminary structures emit in a single longitudinal and spatial mode, have 30 dB of sidemode suppression, and emit about 6 mW into a fiber compatible mode.

In optical systems employing high-power diode lasers, back-irradiance of emission onto the laser facet has been found to contribute to catastrophic optical damage. In this paper, thermoreflectance imaging has been used to measure quantum well temperature rise at the facet for diode lasers emitting at several wavelengths under a wide range of back-irradiance beam positions. For TM-polarized diode lasers operating near 800 nm, the quantum well temperature at the facet is found to reach a maximum when back-irradiance is positioned at the cladding-substrate interface. For TE-polarized lasers operating near 900 nm, a similar effect is observed, albeit with lower magnitude. Interestingly, a second maximum of similar magnitude is observed when back-irradiance is centered on the metal-semiconductor ohmic contact interface. For TE-polarized lasers operating near 1000 nm, the cladding-substrate critical point disappears and the critical point at the metal-semiconductor interface intensifies. The dependence of the critical back-irradiance spot location on operating wavelength originates in the spectral absorptivities of the device's constituent materials. In addition, the incident light's polarization can affect its absorption in the metallic solder. These measurements provide key insights into the potential thermal contribution of back-irradiance to catastrophic optical damage in diode lasers at various operating wavelengths.

In this paper catastrophic optical mirror damage(COMD) mechanism of the semiconductor laser is analyzed.COMD is one of major device damage mechanisms, which is drastically limited laser lifetime and output optical power.The theoretical model that based on heat source with injection current and optical power to describe the temperature distribution of laser is builded.Through analyzing the edge-emitting semiconductor lasers by using ANSYS, we can describe change of facet temperature distribution at COMD events.the results of simulation show that the main reason of COMD is oxidation of the semiconductor laser in facet which caused by optical absorption.

No abstract

Electronic-nose devices have received considerable attention in the field of sensor technology during the past twenty years, largely due to the discovery of numerous applications derived from research in diverse fields of applied sciences. Recent applications of electronic nose technologies have come through advances in sensor design, material improvements, software innovations and progress in microcircuitry design and systems integration. The invention of many new e-nose sensor types and arrays, based on different detection principles and mechanisms, is closely correlated with the expansion of new applications. Electronic noses have provided a plethora of benefits to a variety of commercial industries, including the agricultural, biomedical, cosmetics, environmental, food, manufacturing, military, pharmaceutical, regulatory, and various scientific research fields. Advances have improved product attributes, uniformity, and consistency as a result of increases in quality control capabilities afforded by electronic-nose monitoring of all phases of industrial manufacturing processes. This paper is a review of the major electronic-nose technologies, developed since this specialized field was born and became prominent in the mid 1980s, and a summarization of some of the more important and useful applications that have been of greatest benefit to man.

Power electronics has progressively gained an important status in power generation, distribution, and consumption. With more than 70% of electricity processed through power electronics, recent research endeavors to improve the reliability of power electronic systems to comply with more stringent constraints on cost, safety, and availability in various applications. This paper serves to give an overview of the major aspects of reliability in power electronics and to address the future trends in this multidisciplinary research direction. The ongoing paradigm shift in reliability research is presented first. Then, the three major aspects of power electronics reliability are discussed, respectively, which cover physics-of-failure analysis of critical power electronic components, state-of-the-art design for reliability process and robustness validation, and intelligent control and condition monitoring to achieve improved reliability under operation. Finally, the challenges and opportunities for achieving more reliable power electronic systems in the future are discussed.

A new field limit of silicon devices stressed at pulsed high fields is proposed, namely, the threshold field for the onset of the phase-shifted current response of the device. While the classical limit by the breakdown of the semiconductor-dielectric system, generally by surface flashover, causes a total damage of the semiconductor, the new limit prevents any damage of the device, even at high applied fields. Depending on the overall quality of the device under test, catastrophic failure of the device can be avoided even at large fields of ∼70 kV/cm by not exceeding the threshold for the onset of the phase-shifted current response, the proposed new field limit. Beyond this limit the device response is unstable and a surface filament, distinctly different from the surface flashover tracks, may appear on the device surface, permanently degrading its quality. Experimental results are presented supporting the new predamage high-field limitation. Scanning electron microscopy and optical micrographs are presented for a clear description of the nature of the damage. The present results are discussed using a recent physical model of prebreakdown and breakdown phenomena in high-field semiconductor-dielectric systems [G. Gradinaru and T. S. Sudarshan, J. Appl. Phys. 73, 7643 (1993)].

The package structure critically influences the major characteristics of semiconductor lasers, such as thermal behavior, output power, wavelength, and far-field distribution. In this paper, a new single emitter package structure called F-mount is designed and compared with the conventional package structure C-mount. The influence of package structure on their performances is characterized and analyzed. The thermal resistances of lasers with different package structures are calculated through simulation, and are contrasted with experimental results. Some devices are also tested for the maximum output power level. Under the continuous wave (CW) condition, the maximum power of F-mount reaches 12.6 W at 808 nm while the output power only reaches 10.9 W for C-mount. Under the condition of 0.5% duty cycle (100 μs, 50 Hz), the catastrophic optical mirror damage level reaches 58.7 W at 74 A for F-mount, and 54.8 W at 57 A for C-mount are reported for the first time. It is experimentally found that there is an obvious wavelength difference between the two type structure lasers: about 1.37 nm in CW mode and 2.89 nm in quasi CW mode. Theoretical analysis shows that red-shift and blue-shift is a result of external strain in the package process of F-mount and C-mount, respectively. It is also found that the package structure has an effect on the divergence angle of slow axis far fields, but little impact on that of fast axis far fields. The analysis shows that package structure has a strong influence on the performance of the laser; therefore, the package should be optimized to achieve better performance for some special applications.

No abstract

The influence of threading dislocations on the properties of GaN-based metal-semiconductor-metal (MSM) ultraviolet photodetectors was investigated. It was found that screw dislocations had a strong influence on the dark current of the photodetectors, while edge dislocations had the predominant effect on their responsivity. The dark current increased as the screw dislocation density increased due to their lowering of the Schottky barrier height. However, the responsivity of the photodetectors decreased with increasing edge dislocation density because of the dangling bonds along those edge dislocation lines which enhance the recombination of photogenerated electron-hole pairs. The results suggest that reducing both the screw and edge dislocation densities is an effective way to improve the photoelectric property of GaN-based MSM ultraviolet photodetectors.

No abstract

The dream of epitaxially integrating III-nitride semiconductors on large diameter silicon is being fulfilled through the joint R&D efforts of academia and industry, which is driven by the great potential of GaN-on-silicon technology in improving the efficiency yet at a much reduced manufacturing cost for solid state lighting and power electronics. It is very challenging to grow high quality GaN on Si substrates because of the huge mismatch in the coefficient of thermal expansion (CTE) and the large mismatch in lattice constant between GaN and silicon, often causing a micro-crack network and a high density of threading dislocations (TDs) in the GaN film. Al-composition graded AlGaN/AlN buffer layers have been utilized to not only build up a compressive strain during the high temperature growth for compensating the tensile stress generated during the cool down, but also filter out the TDs to achieve crack-free high-quality n-GaN film on Si substrates, with an X-ray rocking curve linewidth below 300 arcsec for both (0002) and (102) diffractions. Upon the GaN-on-Si templates, prior to the deposition of p-AlGaN and p-GaN layers, high quality InGaN/GaN multiple quantum wells (MQWs) are overgrown with well-engineered V-defects intentionally incorporated to shield the TDs as non-radiative recombination centers and to enhance the hole injection into the MQWs through the via-like structures. The as-grown GaN-on-Si LED wafers are processed into vertical structure thin film LED chips with a reflective p-electrode and the N-face surface roughened after the removal of the epitaxial Si(111) substrates, to enhance the light extraction efficiency. We have commercialized GaN-on-Si LEDs with an average efficacy of 150–160 lm/W for 1mm2 LED chips at an injection current of 350 mA, which have passed the 10000-h LM80 reliability test. The as-produced GaN-on-Si LEDs featured with a single-side uniform emission and a nearly Lambertian distribution can adopt the wafer-level phosphor coating procedure, and are suitable for directional lighting, camera flash, streetlighting, automotive headlamps, and otherlighting applications.

No abstract

No abstract

Abstract The evolution of wide bandgap semiconductor materials has led to dramatic improvements for electronic applications at high powers and temperatures. However, the propensity of extended defects provides significant challenges for implementing these materials in commercial electronic and optical applications. While a range of spectroscopic and microscopic tools have been developed for identifying and characterizing these defects, such techniques typically offer either technique exclusively, and/or may be destructive. Scattering‐type scanning near‐field optical microscopy (s‐SNOM) is a nondestructive method capable of simultaneously collecting topographic and spectroscopic information with frequency‐independent nanoscale spatial precision (≈20 nm). Here, how extended defects within 4H‐SiC manifest in the infrared phonon response using s‐SNOM is investigated and the response with UV‐photoluminescence, secondary electron and electron channeling contrast imaging, and transmission electron microscopy is correlated. The s‐SNOM technique identifies evidence of step‐bunching, recombination‐induced stacking faults, and threading screw dislocations, and demonstrates interaction of surface phonon polaritons with extended defects. The results demonstrate that phonon‐enhanced infrared nanospectroscopy and spatial mapping via s‐SNOM provide a complementary, nondestructive technique offering significant insights into extended defects within emerging semiconductor materials and devices and thus serves as an important diagnostic tool to help advance material growth efforts for electronic, photonic, phononic, and quantum optical applications.

Preface to the First Edition Preface to the Second Edition INTRODUCTION Arrhenius Plot The Relationship between Kinetics and Thermodynamics The Boltzmann Distribution Kinetic Theory of Gases Collisions DIFFUSION IN FLUIDS Diffusion in a Gas Diffusion in Liquids DIFFUSION IN AMORPHOUS MATERIALS Amorphous Materials Network Glass Formers The Glass Transition The Free Volume Model Fictive Temperature Diffusion in Polymers The Stokes-Einstein Relationship DIFFUSION IN CRYSTALS Diffusion in a Crystal Diffusion Mechanisms in Crystals Equilibrium Concentration of Vacancies Simmons and Balluffi Experiment Ionic and Covalent Crystals Stoichiometry Measurement of Diffuion Coefficients Surface Diffusion Diffusion in Grain Boundaries Kirkendall Effect Whisker Growth Electromigration DIFFUSION IN SEMICONDUCTORS Introduction Diffusion in Silicon Diffusion of Zinc in GaAs Recombination Enhanced Diffusion Doping of Semiconductors Point Defect Generation in Silicon during Crystal Growth Migration of Interstitials (and Liquid Droplets) in a Temperature Gradient Oxygen in Silicon Gettering Solid-State Doping ION IMPLANTATION Introduction Ion Interactions Implantation Damage Rutherford Backscattering Channeling Silicon-on-Insulator MATHEMATICS OF DIFFUSION Random Walk The Diffusion Equation Solutions to the Diffusion Equation Numerical Methods Boltzmann-Matano Analysis Diffusion During Phase Separation STEFAN PROBLEMS Steady State Solutions to the Diffusion Equation Deal-Grove Analysis Diffusion Controlled Growth of a Spherical Precipitate Diffusion Limited Growth in Cylindrical Coordinates Diffuion Controlled Growth of a Precipitate PHASE TRANSFORMATIONS Transformation Rate Limited Growth Diffuion Limited Growth Thermally Limited Growth Casting of Metals Operating Point CRYSTAL GROWTH METHODS Melt Growth Solution Growth Vapor Phase Growth Stoichiometry SEGREGATION Segregation During a Phase Change Lever Rule Scheil Equation Zone Refining Diffusion at a Moving Interface Segregation in Three Dimensions Burton, Primm and Schlicter Analysis INTERFACE INSTABILITIES Constitutional Supercooling Mullins and Sekerka Linear Instability Analysis Anisotropic INterface Kinetics CHEMICAL REACTION RATE THEORY The Equilibrium Constant Reaction Rate Theory Reaction Rate Constant Transition State Theory Experimental Determination of the Order of a Reaction Net Rate of Reaction Catalysis Quasi-Equilibrium Model for the Rate of a First Order Phase Change PHASE EQUILIBRIA First Order Phase Changes Second Order Phase Changes Critical Point Between Liquid and Vapor NUCLEATION Homogenous Nucleation Heterogeneous Nucleation Johnson-Mehl-Avrami Equation SURFACE LAYERS Langmuir Adsorption CVD Growth by a Surface Decomposition Reaction Langmuir-Hinschelwood Reaction Surface Nucleation Thin Films Surface Reconstruction Amorphous Deposits Surface Modification Fractal Deposits Strain Energy and Misfit Dislocations Strained Layer Growth THIN FILM DEPOSITION Liquid Phase Epitaxy Growth Configuration for LPE Chemical Vapor Deposition Metal-Organic Chemical Vapor Deposition Physical Vapor Deposition Sputter Deposition Metallization Laser Ablation Molecular Beam Epitaxy Atomic Layer Epitaxy PLASMAS Direct Current (DC) Plasmas Radio Frequency Plasmas Plasma Etching Plasma Reactors Magnetron Sputtering Electron Cyclotron Resonance Ion Milling RAPID THERMAL PROCESSING Introduction Rapid Thermal Processing Equipment Radiative Heating Temperature Measurement Thermal Stress Laser Heating KINETICS OF FIRST ORDER PHASE TRANSFORMATIONS General Considerations The Macroscopic Shape of Crystals General Equation for the Growth Rate of Crystals Kinetic Driving Force Vapor Phase Growth Melt Growth Molecular Dynamics Studies of Melt Crystallization Kinetics The Kossel-Stranski Model Nucleation of Layers Growth on Screw Dislocations The Fluctuation Dissipation Theorem THE SURFACE ROUGHENING TRANSITION Surface Roughness The Ising Model Cooperative Processes Monte Carlo Simulations of Crystallization Equilibrium Surface Structure Computer Simulations Growth Morphologies Kinetic Roughening Polymer Crystallization ALLOYS: THERMODYNAMICS AND KINETICS Crystallization of Alloys Phase Equilibria Regular Solution Model Near Equilibrium Conditions Phase Diagrams The DLP Model PHASE SEPARATION AND ORDERING Phase Separation versus Ordering Phase Separation The Spinodal in a Regular Solution Analytical Model for Diffusion during Spinodal Decomposition Microstructure Development Modeling of Phase Separation and Ordering NON-EQUILIBRIUM CRYSTALLIZATION OF ALLOYS Non Equilibrium Crystallization Experiment Computer Modeling Analytical Model Comparison with Experiment Crystallization of Glasses COARSENING, RIPENING Coarsening Free Energy of a Small Particle Coarsening in a Solution Coarsening of Dendritic Structures Sintering Bubbles Grain Boundaries Scrath Smoothing DENDRITES Dendritic Growth Conditions for Dendritic Growth Simple Dendrite Model Phase Field Modeling Faceted Growth Distribution Coefficient EUTECTICS Eutectic Phase Diagram Classes of Eutectic Microstructures Analysis of Lamellar Eutectics Off-Composition Eutectics Coupled Growth Third Component Elements CASTINGS Grain Structure of Castings Dendrite Re-Melting

The studies of wide-gap semiconductor materials using electron-beam induced current (EBIC) method are reviewed. The main methods for measuring the diffusion length of nonequilibrium carriers in semiconductor structures using the EBIC method in a scanning electron microscope are analyzed. The experimental results of measuring the diffusion lengths in GaN, Ga2O3, 4H-SiC, and ZnO are considered. The reliability of the obtained values is discussed. The EBIC possibilities for detecting dislocations and studying their recombination activity are demonstrated. The strategy for achieving high lateral resolution at EBIC measurements in crystals with submicron diffusion length is discussed. Examples of the influence of irradiation of wide-gap semiconductor materials by a low-energy electron beam on their electrical and optical properties are shown. The results of studying the recombination-enhanced dislocation glide in GaN and 4H-SiC under their electron-beam irradiation in a scanning electron microscope are presented.

Abstract Degradation of 4H‐SiC PiN diodes – a rapid increase in the forward voltage drop with operation time – has been a focus of research since it was first reported over a decade ago. It was soon discovered that associated with this degradation is a rapid increase in the number and extent of intrinsic stacking faults (SFs) in SiC. Further research showed that the expansion of SFs is due to the injection of electron–hole pairs (ehps) in the active region of the device under forward biasing. This effect indicated that recombination‐induced dislocation glide – a phenomenon that was known to occur in many other semiconductors and in SiC – may play an important role in enhancing the motion of the leading partial dislocations and thus in expanding SFs in PiN diodes operated under forward bias conditions. This was later supported by enhanced nucleation and expansion of SFs by bandgap optical excitation in virginal SiC crystals itself – whether the material was of the 4H or 6H polytype – rather than in forward‐biased SiC diodes. At the same time, we used the concept of quasi‐Fermi energy – as the transient value of the electron population during excited states of 4H‐ or 6H‐SiC – to explain the driving force for expansion of SFs. Recently, Caldwell et al. have used this concept to develop various experimental observations that have been made on these diodes, including SF expansion under forward biasing, saturation of the maximum forward voltage drift, annealing‐induced contraction of SFs so generated and consequent drift recovery of the forward voltage drop. In this paper, we expand the concept of quasi‐Fermi level to get a better understanding of faulted loop expansion in 4H‐SiC bipolar devices and in virgin 4H‐ and 6H‐SiC crystals.

The contrast formation of localized non-radiative semiconductor defects in cathodoluminescence (CL) micrographs is analyzed using a linear relationship between the CL intensity and the excess carrier density δp. The latter is evaluated on the basis of Donolato's Born approximation and for generation by a point source. Numerical calculations for a threading dislocation at right angle to a surface of infinite recombination velocity yield an exponential decay of the CL contrast at sufficient distance from the dislocation with a decay constant of 0.63 minority carrier diffusion lengths. The contrast mechanism is seen to be based primarily on the reduction of the total number of excess carriers due to enhanced recombination at the dislocation rather than on the associated reduction of the luminescence quantum efficiency. Die Kontrastbildung lokalisierter nicht-strahlender Halbleiterdefekte in Kathodolumineszenz (KL)-Mikrobildern wird unter Zugrundelegung einer linearen Beziehung zwischen der KL-Intensität und der Überschußladungsträgerdichte δp analysiert. δp wird auf der Grundlage von Donolatos Bornscher Näherung und unter der Annahme punktförmiger Generation ermittelt. Numerische Berechnungen für eine Versetzung senkrecht zur Oberfläche, für die eine unendliche Oberflächenrekombinationsgeschwindigkeit zugrunde gelegt wird, ergeben für genügend große Abstände von der Versetzung einen exponentiellen Abfall des KL-Kontrasts mit einer Abfallkonstante von 0,63 Minoritätsträgerdiffusionslängen. Es zeigt sich, daß der Kontrastmecbanismus primär auf der Reduktion der gesamten Überschußträgerzehl durch verstärkte Rekombination an der Versetzung beruht und nicht so sehr auf der damit verbundenen Reduktion der Lumineszenzquantenausbeute.

Significant advancements have been made in the characterization and understanding of the degradation behavior of the III-V semiconductor materials employed in Vertical Cavity Surface Emitting Laser (VCSEL) diodes. Briefly, for the first time a technique has been developed whereby it is possible to view the entire active region of a solid state laser in a Transmission Electron Microscope (TEM) using a novel Focussed Ion Beam (FIB) prepared plan-view sample geometry. This technique, in conjunction with TEM cross-section imaging has enabled a three-dimensional characterization of several of the degradation mechanisms that lead to laser failure. It is found that there may occur an initial drop in laser power output due to the development of cracks in the upper mirror layers. In later stages of degradation, dislocations are punched out at stress-concentrating sites (e.g. oxide aperture tips) and these dislocations can then extend over the active region in a manner consistent with recombination enhanced dislocation motion. Alternatively, complex three-dimensional dislocation arrays which exhibited dendritic-like growth and which cover the entire active region can nucleate on a single defect.

No abstract

The development of metal oxide semiconductor field effect transistors (MOSFETs) utilizing epitaxially grown 4H-SiC has accelerated in recent years due to their favorable properties, including a high breakdown field, high saturated electron drift velocity, and good thermal conductivity. However, extended defects in epitaxial 4H-SiC can affect both device yields and operational lifetime. In this work, we demonstrate the importance of a multiscale luminescence characterization approach to studying nondestructively extended defects in epitaxial 4H-SiC semiconducting materials. Multiscale luminescence analysis reveals different aspects of excess charge carrier recombination behavior based on the scale of a particular measurement. Combining measurements of the same extended defect area at different scales tells us more about the essential nature of that defect and its microstructure. Here, we use photoluminescence imaging and cathodoluminescence spectrum imaging to investigate the recombination behavior of several different types of extended defects, including stacking faults, inclusions, and basal plane dislocations. A detailed understanding of the optoelectronic properties of extended defects in epitaxial SiC helps elucidate the microstructure of extended defects and can provide pathways to mitigate detrimental changes during device operation related to their evolution, such as the recombination enhanced dislocation glide effect that affects SiC-based MOSFETs.

The dependence of the high-temperature internal friction of germanium and silicon, both intrinsic and highly n type, was measured as a function of temperature, frequency, dislocation density, and dopant concentration. An acoustoelectric peak in both germanium and silicon was detected and found to agree well with the theory of Weinreich. The high-temperature dislocation-dependent damping in intrinsic germanium and silicon was studied and seen to be consistent with most previous studies. If deformed at high temperature and allowed to anneal, highly doped n-type material behaved intrinsically due to preferential precipitation at dislocations; however, if deformed at moderate temperatures and not allowed to anneal, such crystals exhibited a greatly enhanced dislocation-dependent internal friction which depended on the extrinsic carrier concentration. A theory was developed for dislocation damping in semiconductors and was found to agree well with experimental results. The model is based upon electronic viscous damping of dislocations by excess current carriers whose lifetimes are controlled by Auger recombination processes.

In order to understand the irreversible failure mechanisms of planar InGaAs p-i-n photodiodes, 32 devices from 19 different wafers that shorted during aging were first examined in the scanning electron microscope. Included were devices that failed during long term aging (&amp;gt;103 h) as well as those that failed during short term aging (&amp;lt;102 h) at higher reverse bias. With a few exceptions, the diodes failed as a result of a single localized leakage source located at the perimeter of the p-n junction. Three types of leakage sources were found: (a) a microplasma, (b) a microplasma associated with a region of high recombination rate, and (c) a microplasma associated with a thermally damaged region. Analysis of ∼40 devices before and after aging shows that leakage paths found after aging result from microplasmas initially present in the device. Defect analysis shows that neither threading dislocations nor misfit dislocations are generally responsible for these microplasmas. Analysis of the processing shows that the p-contact/semiconductor interface is stable during device operation. Thus, the leakage source, attributed to contact migration in other studies, is not present in our devices. However, pinholes in the SiNx diffusion mask close (&amp;lt;5 μm) to the perimeter of the p-n junction were found to be the major source of microplasmas. The high-electric field at the shallow (≲0.5 μm) p-n junctions formed by unintentional diffusion through the dielectric pinholes is believed to cause the microplasmas. Electron-hole pairs, recombining at the microplasma site, are believed to create defects. These defects were observed as a localized region of enhanced recombination in electron-beam induced current (EBIC) images of the p-n junction without applied bias. The enhanced leakage current as a result of the defects leads to thermal runaway.

No abstract

This paper reviews the low temperature dislocation climb process in III-V compounds semiconductors and points out areas in which more experimental information is needed to understand this complex problem. A dislocation climb model requiring the supersaturation of point defects of only one element of the coumpound is found to account for the main climb features. Rapid dislocation climb is attributed to recombination enhanced defect motion. Finally evidence of an interaction between the climb dislocations and a deep level donor center suggest that it might possibly be associated with the source of point defects needed for dislocation climb in Ga1-xAlxAs structures.

Solid state UV emitters have many advantages over conventional UV sources. The (Al,In,Ga)N material system is best suited to produce LEDs and laser diodes from 400 nm down to 210 nm-due to its large and tuneable direct band gap, n-and p-doping capability up to the largest bandgap material AlN and a growth and fabrication technology compatible with the current visible InGaN-based LED production. However AlGaN based UV-emitters still suffer from numerous challenges compared to their visible counterparts that become most obvious by consideration of their light output power, operation voltage and long term stability. Most of these challenges are related to the large bandgap of the materials. However, the development since the first realization of UV electroluminescence in the 1970s shows that an improvement in understanding and technology allows the performance of UV emitters to be pushed far beyond the current state. One example is the very recent realization of edge emitting laser diodes emitting in the UVC at 271.8 nm and in the UVB spectral range at 298 nm. This roadmap summarizes the current state of the art for the most important aspects of UV emitters, their challenges and provides an outlook for future developments.

Direct epitaxial integration of III-V optoelectronic devices on Si offers a substantial manufacturing cost and scalability advantage over heterogeneous integration via wafer bonding. The challenge in utilizing direct epitaxy of III-Vs on Si is that epitaxial growth introduces high densities of crystalline defects that limit device performance and lifetime. As an optical gain medium, quantum dots exhibit a unique tolerance to crystalline defects due to their three-dimensional quantum confined structure.Quantum dot lasers epitaxially grown on Si are showing promise for achieving low-cost, scalable integration with silicon photonics. Their atom-like, inhomogeneously broadened, discrete density of states yields unique gain properties that show promise for improved performance and new functionalities relative to their quantum well counterparts (even on native substrates). By reducing the dislocation density in III-V/Si material and improving quantum dot size homogeneity, several world record results have been achieved for epitaxial laser performance on silicon.A subset of the results achieved include continuous-wave threshold currents below 1 mA in micro-scale ring laser cavities, single-facet output powers of 175 mW at 20 °C, continuous wave lasing up to 105°C, near zero linewidth enhancement factor, isolator-free stability at optical feedback levels of up to 90%, and record long device lifetimes on silicon of more than 100 years at 35°C based on extrapolated 8,000-hour aging studies, and &amp;gt;100,000 h lifetimes at 60°C from extrapolated 4,000-hour aging studies. These results show potential to revolutionize integrated photonics through economic advantages and performance capabilities not achievable in quantum well lasers.

We have extensively studied the frequency noise and relative intensity noise spectra in a tunable external-cavity InGaN diode laser at blue (420 nm) wavelengths. We report flicker (1/f) frequency-noise behavior at low Fourier frequencies measured using offset frequency-absorption spectroscopy on 85Rb vapor cells, which yields an estimated lasing linewidth of 870 kHz. From considerations of high-dislocation density in III nitride epitaxy, 1/f noise and linewidth were expected to be larger than in conventional III-V lasers. Surprisingly, the measured noise characteristics are comparable to or better than those of near-infrared distributed feedback lasers and external-cavity diode lasers. The noise-reduction mechanism is attributed to the wavelength dependence of 1/f noise. We discuss challenges in atomic spectroscopy applications caused by defects and mode-clustering effect in GaN lasers. Using the Hakki-Paoli analysis in an aged laser diode, we provide possible explanation about the origin of observed mode clustering.

No abstract

Abstract The direct bonding process of InP chips on a silicon-on-insulator (SOI) wafer is investigated using surface hydrophilization by UV-ozone treatment. The influence of the treatment on surface roughness is observed by atomic force microscopy, and found to be negligibly small. A high-quality III–V/Si bonding interface without crystal defects is verified in a scanning transmission electron microscope observation. The III–V layers remaining on the SOI wafer after the removal process of the InP substrate show uniform photoluminescence intensity over the whole region of the chip, indicating a bonding interface without the influence of lateral etching and peeling-off. From this bonding process, Fabry–Perot lasers with III–V gain and Si waveguide regions are fabricated, and cw operation is successfully achieved at a stage temperature from 20 °C to 85 °C. Stable operation is also confirmed from the changes of threshold current in the aging test (200 mA at 85 °C) after 2000 h.

No abstract

No abstract

Triple-negative breast cancer (TNBC) is a heterogeneous disease that can be classified into distinct molecular subtypes by gene expression profiling. Considered a difficult-to-treat cancer, a fraction of TNBC patients benefit significantly from neoadjuvant chemotherapy and have far better overall survival. Outside of BRCA1/2 mutation status, biomarkers do not exist to identify patients most likely to respond to current chemotherapy; and, to date, no FDA-approved targeted therapies are available for TNBC patients. Previously, we developed an approach to identify six molecular subtypes TNBC (TNBCtype), with each subtype displaying unique ontologies and differential response to standard-of-care chemotherapy. Given the complexity of the varying histological landscape of tumor specimens, we used histopathological quantification and laser-capture microdissection to determine that transcripts in the previously described immunomodulatory (IM) and mesenchymal stem-like (MSL) subtypes were contributed from infiltrating lymphocytes and tumor-associated stromal cells, respectively. Therefore, we refined TNBC molecular subtypes from six (TNBCtype) into four (TNBCtype-4) tumor-specific subtypes (BL1, BL2, M and LAR) and demonstrate differences in diagnosis age, grade, local and distant disease progression and histopathology. Using five publicly available, neoadjuvant chemotherapy breast cancer gene expression datasets, we retrospectively evaluated chemotherapy response of over 300 TNBC patients from pretreatment biopsies subtyped using either the intrinsic (PAM50) or TNBCtype approaches. Combined analysis of TNBC patients demonstrated that TNBC subtypes significantly differ in response to similar neoadjuvant chemotherapy with 41% of BL1 patients achieving a pathological complete response compared to 18% for BL2 and 29% for LAR with 95% confidence intervals (CIs; [33, 51], [9, 28], [17, 41], respectively). Collectively, we provide pre-clinical data that could inform clinical trials designed to test the hypothesis that improved outcomes can be achieved for TNBC patients, if selection and combination of existing chemotherapies is directed by knowledge of molecular TNBC subtypes.

A technique, namely bonding by atomic rearrangement has been invented to realize high quality heteroepitaxy for lasers and optoelectronics. High performance lasers of 1.5 μm wavelength have been fabricated on GaAs substrates using this method. The laser has the same threshold current and quantum efficiency as lasers on InP substrates. No performance degradation has been observed. The transmission electron microscopic results show that the heteroepitaxy is excellent, without a single threading dislocation or stacking fault.

We demonstrate electrically injected InGaAsP (1.3 μm) vertical cavity lasers (VCLs) fabricated on GaAs substrates and employing GaAs/AlAs mirrors. The technique of wafer fusion allows for integration of GaAs/AlAs mirrors with InP double heterostructures without degradation of device performance, despite a 3.7% lattice mismatch between the wafers. The wafer fused VCLs have the lowest threshold current (9 mA) and lowest threshold current density (9.5 kA/cm2) and the highest characteristic temperature (T0=67 K) reported to date of any room-temperature long wavelength VCL.

Compound-semiconductor-based lasers grown directly on silicon substrates would constitute an important technology for the realization of on-chip optical interconnects. The characteristics of GaAs-or InP-based devices on silicon can be degraded by the large density of propagating dislocations resulting from the large lattice mismatch (> 4%). The use of multiple layers of self-organized In(Ga, Al)As/GaAs quantum dots (QDs) as a 3D dislocation filter to impede the propagation of dislocations and to reduce dislocation density in GaAs/Si lattice-mismatched heterostructures has been investigated. The effectiveness of this technique, depending on QD composition, size, areal density, and number of dot layers, is analyzed by a quasi-3D model of strain-dislocation interaction. It is found that ten layers of InAs QDs of size ~20-30 nm constitute the most effective dislocation filter. This is experimentally verified by cross-sectional transmission electron microscopy, photoluminescence, and performance characteristics of In <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.5</sub> Ga <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.5</sub> As/GaAs QD separate confinement heterostructure lasers on Si. The lasers exhibit J <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sub> ~900 A/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> at 273 K, the large characteristic temperature (T <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> =278 K) is in the temperature range of 5degC-85degC, and the output slope efficiency (~0.4 W/A) is independent of temperature in the range of 5degC-50degC.

Recent progress in semiconductor quantum-dot (QD) lasers approaches qualitatively new levels, when dramatic progress in the development of the active medium already motivates search for new concepts in device and system designs. QDs, which represent coherent inclusions of narrower bandgap semiconductor in a wider gap semiconductor matrix, offer a possibility to extend the wavelength range of heterostructure lasers on GaAs substrates to 1.3 /spl mu/m and beyond and create devices with dramatically improved performance, as compared to commercial lasers on InP substrates. Low-threshold current density (100 A/cm/sup 2/), very high characteristic temperature (170 K up to 65/spl deg/C), and high differential efficiency (85%) are realized in the same device. The possibility to stack QDs (e.g., tenfold) without an increase in the threshold current density and any degradation of the other device parameters allow realization of high modal gain devices suitable for applications in 1.3-/spl mu/m short-cavity transmitters and vertical-cavity surface-emitting lasers (VCSELs). The 1.3-/spl mu/m QD GaAs VCSELs operating at 1.2-mW continuous-wave output power at 25/spl deg/C are realized, and long operation lifetime is manifested. Evolution of GaAs-based 1.3-/spl mu/m lasers offers a unique opportunity for telecom devices and systems. Single-epitaxy VCSEL vertical integration with intracavity electrooptic modulators for lasing wavelength adjustment and/or ultrahigh-frequency wavelength modulation is possible. Arrays of wavelength-tunable VCSELs and wavelength-tunable resonant-cavity photodetectors may result in a new generation of "intelligent" cost-efficient systems for ultrafast data links in telecom.

This paper reports on catastrophic degradation, called sudden failure (SF), that is observed in both AlGaAs/GaAs and InGaAsP/InP double-heterostructure lasers. The SF observed here is not associated with electrical surge effects and appears unexpectedly in the middle of a long-term, stable operation. It was found that this type of SF can be caused by aging-induced metallurgical deterioration at the interfacial bonding solder layer. Among the metallurgical deteriorations observed were (1) solder migration into the laser crystal due to current-induced local heating near the end mirror of the laser, (2) In whisker growth due to electromigration in In solder, and (3) Sn whisker growth, when using an Au-Sn alloy as solder, due to strain relaxation. All of these effects cause SF. Countermeasures against these deteriorations are described and some successful results are presented.

Aging tests were carried out on as-cleaved InGaAs/GaAs strained quantum-well ridge waveguide lasers. Although the lasers have immunity to sudden failure and have degradation rate as low as 2×10−5 h−1, after over 6000 h of operation, they readily suffered facet oxidation. The measured oxidation rate was comparable to that of GaAs quantum-well lasers and one order of magnitude higher than that of lattice-matched InGaAs/InP lasers. This high oxidation rate is considered to be caused by light absorption in the vicinity of the facet where the band gap is reduced because of the stress variation from biaxial to uniaxial.

Small- and large-signal analyses of transistor lasers (TLs) are demonstrated for 0.98-μm wavelength GaInAs/GaAs and 1.3-μm AlGaInAs/InP systems. The modulation bandwidth of the TL was larger than that of a laser diode due to the lower damping effect in the former. Comparisons between TLs with different numbers of quantum wells indicated that a large signal response and high modulation bandwidth could be realized simultaneously. However, in the case of large-signal analysis, the calculated eye diagrams were degraded by a resonance oscillation peak. By changing structural parameters such as the facet reflectivity and by controlling the damping effect, the resonance frequency peak was suppressed and clear eye diagrams of >;40 Gb/s were obtained.

No abstract

Ageing characteristics of conventional double heterostructure lasers have been explained by extending the defect model proposed in connection with remote junction heterostructure lasers. Two major phenomena are responsible for laser degradation: One is that mobile defects remaining in the cladding layer before ageing move toward the pn junction and accumulate there during ageing. Another component is due to the mobile defects created in the active region by the non-radiative recombination of injected carriers. Using this model, the ultimate lifetime of InGaAsP–InP lasers was, for the first time, concluded to be far longer than that of AlGaAs–GaAs lasers.

We have grown ZnCdSe/ZnCdMgSe quantum well (QW) structures nearly lattice matched to InP substrates. Emission energies from 2.307 to 2.960 eV were measured by low-temperature photoluminescence at 10 K for samples with QW thicknesses between 5 and 80 Å. By using exactly lattice-matched QWs, the lower limit of the energy range can be lowered to about 2.2 eV (at 10 K). We propose that these structures could be used in entirely lattice-matched semiconductor lasers operating at room temperature in the blue, green, and yellow regions. Because of the absence of strain, these materials are expected to be less prone to degradation than the current blue-green lasers grown on GaAs.

We have investigated the radiation enhanced diffusion of ion defects during reactive ion beam etching of GaAs and InP, using the multiple quantum well (MQW) probe technique. During low energy (sub-keV) Ar+ ion exposure, illumination with light of energy above the band gap can substantially reduce the photoluminescence efficiency of MQW samples, relative to those which were not laser illuminated; the degradation of luminescence efficiency increases with the intensity of the light. Illumination with light of energy below the band gap produces a slight increase in the damage profiles. The observation of enhanced defect diffusion due to optical radiation in our studies suggests that in ion-assisted etching of semiconductors, the generation of excess electron-hole pairs and their subsequent recombination can play an important role in the propagation of defects into the substrate.

We present experimental results on thin-film wafer fusion of InP/GaAs to fabricate InP-based lasers on a GaAs substrate. We have studied the load pressure dependence of the photoluminescence intensity (PL) of the InP-based layers and electrical properties at the fused interface. Although a higher load pressure results in better electrical contact, it can degrade the PL intensity of InP-based quantum-wells structure fused to a GaAs substrate due to the generation of recombination centers. Buried-heterostructure InP-based lasers are consequently fabricated on a GaAs substrate by thin film wafer fusion, and these lasers are demonstrated to have good performance under continuous-wave operation.

Material transformations occurring at the facets of optically ‘‘stressed’’ planar InGaAsP/InP diode lasers have been investigated by transmission electron microscopy and energy dispersive x-ray spectroscopy. Catastrophic degradation lines (CDLs) which are characteristic of catastrophic optical damage are observed for optical power densities ∼107 W/cm2. Analysis of the microstructure reveals a series of 150 nm wide GaAs-rich tracks and the formation of unique void/InGa-rich precipitate pairs within the InGaAsP active layer. These observations suggest that the formation of local group III-rich regions is the first stage in the formation of CDLs. Subsequently, the strong absorption of the impinging laser beam leads to propagation of an InGa-rich melt, thereby producing the GaAs-rich tracks through a process similar to liquid phase epitaxy. These results are discussed in the context of standard physical models for CDLs.

Catastrophic degradation of V-grooved substrate buried-heterostructure InGaAsP/InP lasers (λ=1.3 μm), by large pulsed currents, has been investigated using scanning electron microscopy, etching technique, energy dispersive x-ray spectroscopy, and spatially resolved photoluminescence topography. After the degradation, the diode stops lasing and becomes ohmic. These are associated with the following phenomena: (i) facet erosion inside or outside the stripe region; (ii) penetration of the electrode-metals into the epitaxial layer. These phenomena are presumably caused by the abrupt passage of large current along the facet or by local heating at the contact region outside the stripe region. Catastrophic optical damage, which frequently occurs in GaAlAs/GaAs double-heterostructure lasers, is not observed in any part of the degraded diodes.

Quantum dots (QDs) epitaxially grown on Si are promising for monolithic integration of light sources on a Si photonics platform. Unlike quantum well (QW) lasers on Si, 1.3 μm InAs QD lasers on Si show similar threshold current to those grown on GaAs owing to their better dislocation tolerance. To date, research on dislocation-tolerant QDs has exclusively focused on materials emitting at telecom wavelengths. In this work, we report visible InP QDs on Si with photoluminescence (PL) intensity similar to their counterparts grown on GaAs despite high threading dislocation density (TDD). In contrast, visible InGaP QWs grown on Si with the same TDD value show 9× degradation in PL intensity compared to QWs grown on GaAs. The dislocation tolerance of InP QDs arises from their high density relative to TDD and the lateral carrier confinement that they provide. InP QDs on Si with bright PL are promising for low-cost light emitters and integrated photonics applications requiring monolithic red-light sources.

III-V lasers based on self-assembled quantum dots (QDs) have attracted widespread interest due to their unique characteristics, including low threshold current density (Jth), low sensitivity to backreflections, and resistance to threading dislocations. While most work to date has focused on 1.3 μm InAs/GaAs QDs, InP QDs have also aroused interest in lasers emitting at visible wavelengths. Molecular beam epitaxy (MBE) enables the growth of high-density InP/AlGaInP QDs on exact (001)-oriented GaAs substrates but requires a relatively low substrate temperature of &amp;lt;500 °C. The low substrate temperature used for phosphide growth in MBE leads to degraded optical properties and makes post-growth annealing a crucial step to improve the optical quality. Here, we report the exceptional thermal stability of InP/AlGaInP QDs grown using MBE, with up to 50× improvement in room temperature photoluminescence intensity with the optimization of annealing temperature and time. We also demonstrate the room temperature pulsed operation of InP multiple quantum dot (MQD) lasers on GaAs (001) emitting close to 735 nm with Jth values of 499 A/cm2 after annealing, a factor of 6 lower than their as-grown counterparts and comparable to such devices grown by MOCVD. In0.6Ga0.4P single quantum well (SQW) lasers on GaAs (001) also exhibit a substantial reduction in Jth from 340 A/cm2 as-grown to 200 A/cm2 after annealing, emitting at 680 nm under pulsed operation conditions. This work shows that post-growth annealing is essential for realizing record-performance phosphide lasers on GaAs grown by MBE for applications in visible photonics.

Facet degradation of InGaAsP/InP double-heterostructure (DH) lasers in accelerating environment of water was studied from the standpoint of quantitative comparison with GaAlAs DH lasers. Facet degradation of InGaAsP/InP DH lasers proceeds by oxidation mechanism and its degradation rate is 2–3 orders below that of GaAlAs DH lasers. The difference of oxidation behavior between InGaAsP/InP and GaAlAs DH lasers was examined by using large wafers having the same composition as lasers. The following results were obtained: the oxidation proceeds with the inward diffusion of molecule water through the oxide film, the oxidation rate of n-InP is more than two orders of magnitude lower than that of n-GaAs, In oxide and P oxide were uniformly distributed in the depth profile of the oxide for InP and InGaAsP, and As oxide has the abnormally depleted distribution in the front layer of the oxide for GaAs and GaAlAs. The facet degradation rate is influenced by some experimental factors, i.e., the optical output power, the injection current in dc-biased experiment, and the water temperature in unbiased experiment. To explain the degradation behavior caused in dc-biased and unbiased experiment under identical mechanism, we attempted to evaluate the degradation rate as a function of the facet temperature; the optical output power and the injection current in dc-biased experiment were assumed to contribute to the supplying source of carriers, which rapidly recombine nonradiatively, raise the facet temperature, and finally promote the diffusion of molecule water through oxides and the reaction rate at oxide-facet interface. Combining these results with the fact that the oxidation rate varies in proportion to the partial pressure of a molecule water in the operation environment, we estimated the degradation rate of InGaAsP/InP DH lasers operated under 6.7 mW/μm2 at 70 °C in a nearly dry nitrogen environment (relative humidity of 1–10%).

Experimental data for an InP-based 40-stage quantum cascade laser structure grown on a 6-in. GaAs substrate with a metamorphic buffer are reported. The laser structure had an Al0.78In0.22As/In0.73Ga0.27As strain-balanced active region composition and an 8 μm-thick, all-InP waveguide. High reflection coated 3 mm × 30 μm devices processed from the wafer into a ridge-waveguide configuration with a lateral current injection scheme delivered over 200 mW of total peak power at 78 K with lasing observed up to 170 K. No signs of performance degradation were observed during a preliminary 200-min reliability testing. Temperature dependence for threshold current and slope efficiency in the range from 78 K to 230 K can be described with characteristic temperatures of T0 ≈ 460 K and T1 ≈ 210 K, respectively. Lasing was extended to 303 K by applying a partial high reflection coating to the front facet of the laser.

The stability of InGaAsP/InP DH lasers has been investigated in water and mirror facet oxidation has been confirmed by Auger sputtering analysis. The pulse threshold current increasing rate for InGaAsP/InP lasers is about two orders of magnitude smaller than that for AlGaAs/GaAs lasers under both storage and pulsc operation tests in water. The facet degradation has not been enhanced by light output power, but rather is enhanced by the active layer temperature rise due to carrier injection.

InGaAsP epitaxial layers were excited by a Nd: YAG laser (λ=1.064 µm), directly, or through InP cladding layers. The photoluminescence intensity pattern of the InGaAsP active layer in InP–InGaAsP double heterostructure (DH) wafers remained unchanged even of excitations exceeding 100 kW/cm2, when excited through InP cladding layers. This result should be compared with an A1GaAs–GaAs system, where photoluminescence degradation is observed for about 5 kW/cm2 excitation. However, photoluminescent dark regions were formed when the InGaAsP surface was excited directly. Moreover, once these dark regions were formed, they grew continuously without further excitation. The observed dark regions are not due to InGaAsP bulk degradation, because the degraded photoluminescence intensity could be recovered to the initial intensity by rinsing the InGaAsP surface with dilute HC1, which does not attack a bulk InGaAsP crystal.

Abstract Degradation behaviours of 980 nm InGaAs/GaAs strained quantum well (QW) lasers are clarified and compared with normal AlGaAdGaAs, InGaAsP/InP and GaAs/GaAs QW lasers. Through various ageing tests, it is confirmed that 980 nm InGaAs/GaAs strained QW lasers are applicable to optical fibre transmission systems where the components are required to be highly reliable.

Self-sustained pulsation (SSP) has been observed in InGaAsP/InP double heterostructure (DH) lasers during accelerated operation. The development of SSP depends on the number and absorption coefficient magnitude of dark defects associated with degradation. The absorption coefficient of a dark defect that induces SSP is experimentally estimated to be from 300 to 600 cm -1 . The frequency of the SSP is a linear function of the square root of the current density. The fact that dark defects in the active region are one origin of SSP in InGaAsP/InP lasers, as well as AlGaAs/GaAs lasers, is made clear.

Effective and economical burn-in screening is important for technology development and manufacture of semiconductor lasers. We study the burn-in degradation behavior of wavelength-division multiplexing semiconductor lasers to determine the feasibility of short burn-in. The burn-in is characterized by the sublinear model and correlated with long-term reliability.

We are concerned with assuring the reliability of semiconductor lasers intended for an application in which the design lifetime is long, replacement or redundancy is impossible or impractical, and the failure of even a few lasers could be disastrous. In the search for a reliability assurance strategy that will meet our objectives, we have carefully examined the well-known and widely used bathtub and lognormal approaches. Based upon our understanding of the expected aging behavior of lasers, we propose an alternative reliability assurance strategy that we believe to be an improvement over the traditional approaches. The object is not how to make reliable lasers, but rather how to confidently predict, in a timely fashion, which lasers in a given population will endure beyond the intended system lifetime. Particular emphasis is placed upon initially imposed overstress regimes that address the anticipated presence of transient modes of depadation and infant failure modes with low thermal activation energies that may be invulnerable to detection during accelerated thermal aging. Since lasers degrade gradually rather than fail suddenly, comparable emphasis is placed upon monitoring the stabilized long-term degradation rates of the survivors of the overstress regimes so as to permit lifetime predictions of individual lasers.

High-power semiconductor lasers have found increasing applications in industrial, military, commercial, and consumer products. The thermal management of high-power lasers is critical since the junction temperature rise resulting from large heat fluxes strongly affects the device characteristics, such as wavelength, kink power, threshold current and efficiency, and reliability. The epitaxial-side metallization structure has significant impact on the thermal performance of a junction-down bonded high-power semiconductor laser. In this paper, the influence of the epitaxial-side metal (p-metal) on the thermal behavior of a junction-down mounted GaAs-based high-power single-mode laser is studied using finite-element analysis. It is shown that a metallization structure with thick Au layer can significantly reduce the thermal resistance by distributing the heat flow to wider area laterally, and the thermal resistance of a junction-down bonded laser with thick Au metallization is much less sensitive to the voiding in the die attachment solder interface than a laser with thin Au metallization. A metallization structure of Ti-Pt-thick Au-Ti-Cr-Au is designed and implemented, and the metallurgical stability of this metallization scheme is reported. It was found that, without a diffusion barrier, the thick Au layer in the epi-side metallization would be mostly consumed and form intermetallics with the Sn from the AuSn solder during soldering and thermal aging. The Ti-Pt-thick Au-Ti-Cr-Au metallization scheme prevents the diffusion of Sn into the thick Au layer and preserves the integrity of the metallization system. It is a promising candidate for junction-down bonding of high-power semiconductor lasers for improved thermal management and reliability

Semiconductor lasers, one of the key components for optical communication systems, have been rapidly evolving to meet the requirements of next generation optical networks with respect to high speed, low power consumption, small form factor etc. However, these demands have brought severe challenges to the semiconductor laser reliability. Therefore, a great deal of attention has been devoted to improving it and thereby ensuring reliable transmission. In this paper, a predictive maintenance framework using machine learning techniques is proposed for real-time heath monitoring and prognosis of semiconductor laser and thus enhancing its reliability. The proposed approach is composed of three stages: i) real-time performance degradation prediction, ii) degradation detection, and iii) remaining useful life (RUL) prediction. First of all, an attention based gated recurrent unit (GRU) model is adopted for real-time prediction of performance degradation. Then, a convolutional autoencoder is used to detect the degradation or abnormal behavior of a laser, given the predicted degradation performance values. Once an abnormal state is detected, a RUL prediction model based on attention-based deep learning is utilized. Afterwards, the estimated RUL is input for decision making and maintenance planning. The proposed framework is validated using experimental data derived from accelerated aging tests conducted for semiconductor tunable lasers. The proposed approach achieves a very good degradation performance prediction capability with a small root mean square error (RMSE) of 0.01, a good anomaly detection accuracy of 94.24% and a better RUL estimation capability compared to the existing ML-based laser RUL prediction models.

In recent year, the combination of III-V semiconductor devices with Si manufacturing techniques to develop optoelectronic integrated circuits (OEICs) has been widely studied. Flip-chip bonding has been used widely because superior electrical performance, proper reliability, efficient heat conduction and self-alignment are the advantages of this technology. Because optoelectronic devices are quite sensitive to temperature and stress-induced degradation, the bonding medium should be chosen to have high thermal conductivity and stress-relief. The indium (In) based alloy solders are generally recognized to provide lower melting point, longer fatigue life, and higher thermal conductivity. In this study, we have successfully developed a fluxless bonding process to manufacture In-Au microjoint between laser diode and silicon substrate. During the soldering, the solder reacts with the bonding pad metal to form the intermetallic compound at the interface. Such an intermetallic compound is crucial to the quality of solder joint. We utilized SEM, EDX, and XRD to observe and identify the intermetallic compounds. These results indicate that AuIn/sub 2/ is the main intermetallic phase and plays an important role on the quality of joints. Moreover the reliability of solder joint is strongly depended on the initial microstructure. The optimum bonding temperature is found to be about 200/spl deg/C by the microstructure of the solder joint by SEM and optoelectronic characteristics (I-V and L-I) of the laser diodes. Shear force test has also been performed according to MIL-STD-883C. The results reveal the fact that all well-boned devices meet the shear force requirement. To verify the thermal stability, the bonded samples were tested by thermal shock test. The bonded specimens endure 500 cycles of thermal shock between liquid nitrogen temperature and a hot plate (80/spl deg/C). To evaluate the long-term reliability, the bonded laser diodes were subjected to an accelerated aging test at 90/spl deg/C for 500 h. These devices show no abrupt degradation from I-V and L-I plots and their mechanical strength is nearly unchanged as before. This shows that indium could achieve the requirements of thermal stability. The flip-chip bonding technique by using indium solder shows good feasibility for the integration of laser diodes on silicon substrates.

We report the first uncooled nonhermetic 1.3-/spl mu/m InP-based communication lasers that have reliability comparable to their hermetically packaged counterparts for possible applications in fiber in the loop and cable TV. The development of reliable nonhermetic semiconductor lasers would not only lead to the elimination of the costs specifically associated with hermetic packaging but also lead the way for possible revolutionary low-cost optoelectronic packaging technologies. We have used Fabry-Perot capped mesa buried-heterostructure (CMBH) uncooled lasers with both bulk and MQW active regions grown on n-type InP substrates by VPE and MOCVD. We find that the proper dielectric facet passivation is the key to obtain high reliability in a nonhermetic environment. The passivation protects the laser from the ambient and maintains the proper facet reflectivity to achieve desired laser characteristics. The SiO facet passivation formed by molecular beam deposition (MBD) has resulted in lasers with lifetimes well in excess of the reliability goal of 3,000 hours of operation at 85/spl deg/C/90% RH/30 mA aging condition. Based on extrapolations derived experimentally, we calculate a 15-year-average device hazard rate of <300 FITs (as against the desired 1,500 FITs) for the combination of thermal-and humidity-induced degradation at an ambient condition of 45/spl deg/C/50% RH. For comparison, the average hazard rate at 45/spl deg/C and 15 years of service is approximately 250 FITs for hermetic lasers of similar construction. A comparison of the thermal-only degradation (hermetic) to the thermal plus humidity-induced degradation (nonhermetic) indicates that the reliability of these nonhermetic lasers is controlled by thermal degradation only and not by moisture-induced degradation. In addition to device passivation for a nonhermetic environment, MBD-SiO maintains the optical, electrical, and mechanical properties needed for high-performance laser systems.

Room‐temperature continuous wave (RT‐CW) electrically pumped 1550 nm indium phosphide (InP)‐based laser diodes are realized on complementary metal‐oxide‐semiconductor (CMOS) compatible silicon (Si) substrates by direct heteroepitaxy. Dynamic properties are investigated by gain switching and small signal modulation measurements. A maximum 3 dB bandwidth of 5.3 GHz is demonstrated, along with a narrow optical pulse with a width of 1.5 ns. The dark current density of 490 mA cm −2 at −1 V bias is an order of magnitude higher than identical devices grown and fabricated on native InP substrates. Also, reliability measurements and failure analysis are carried out for the lasers on Si. The lasers operate stably over 200 hours (h) at 10 °C under CW operation without apparent change in threshold or output power. In sharp contrast, a rapid failure occurs at 60 °C under pulsed operation following 5.6 h of aging. To further improve device characteristics for lasers on Si, the dislocation density of the InP template is reduced by introducing a 2 μm‐thick compositionally graded In 0.4 Ga 0.6 As buffer. The resulting surface defect density is as low as 4.5 × 10 7 cm −2 , which is expected to improve the performance and reliability of long wavelength lasers grown directly on Si.

This paper investigated the reliability of semiconductor 1.3-/spl mu/m multiquantum-well (MQW) Fabry-Perot laser diodes (LDs) in a quarter 2-in wafer level that are measured to have uniform threshold currents, slope efficiencies, and wavelengths within 4% of the maximum deviation. By performing the accelerated aging test under a constant optical power of 3 mW at 85/spl deg/C for 2100 h, the lifetime of the fabricated optoelectronic devices was estimated, where the failure rate was matched on the fitted line of the lognormal distribution model resulting in the mean-time-to-failure (MTTF) of 2/spl times/10/sup 6/ h operating at room temperature.

No abstract

No abstract

Direct diode laser systems gain importance in the fields of material processing and solid-state laser pumping. With increased output power, also the influence of strong optical feedback has to be considered. Uncontrolled optical feedback is known for its spectral and power fluctuation effects, as well as potential emitter damage. We found that even intended feedback by use of volume Bragg gratings (VBG) for spectral stabilization may result in emitter lifetime reduction. To provide stable and reliable laser systems design, guidelines and maximum feedback ratings have to be found. We present a model to estimate the optical feedback power coupled back into the laser diode waveguide. It includes several origins of optical feedback and wide range of optical elements. The failure thresholds of InGaAs and AlGaAs bars have been determined not only at standard operation mode but at various working points. The influence of several feedback levels to laser diode lifetime is investigated up to 4000h. The analysis of the semiconductor itself leads to a better understanding of the degradation process by defect spread. Facet microscopy, LBIC- and electroluminescence measurements deliver detailed information about semiconductor defects before and after aging tests. Laser diode protection systems can monitor optical feedback. With this improved understanding, the emergency shutdown threshold can be set low enough to ensure laser diode reliability but also high enough to provide better machine usability avoiding false alarms.

No abstract

No abstract

High-power semiconductor lasers at the 980-nm wavelength are becoming increasingly important as the pump-light sources for Er3+-doped-fiber amplifiers. High reliability and high-power characteristics are required as we move from the development phase into deployment in optical communication systems. Lasers with InGaP and InGaAsP cladding layers are expected to be more reliable than those having AlGaAs cladding layers.1,2 To assure their applicability, aging tests at high temperatures and high bias levels will be required. This paper reports 85°C 6000-h aging tests at 150 mA (equivalent to 10X threshold current) and without facet protection for Al-free 980-nm strained-layer quantum-well lasers.

No abstract

No abstract

No abstract

III-V semiconductor nanowires (NWs) such as GaAs NWs form an interesting artificial materials system promising for applications in advanced optoelectronic and photonic devices, thanks to the advantages offered by the 1D architecture and the possibility to combine it with the main-stream silicon technology. Alloying of GaAs with nitrogen can further enhance performance and extend device functionality via band-structure and lattice engineering. However, due to a large surface-to-volume ratio, III-V NWs suffer from severe non-radiative carrier recombination at/near NWs surfaces that significantly degrades optical quality. Here we show that increasing nitrogen composition in novel GaAs/GaNAs core/shell NWs can strongly suppress the detrimental surface recombination. This conclusion is based on our experimental finding that lifetimes of photo-generated free excitons and free carriers increase with increasing N composition, as revealed from our time-resolved photoluminescence (PL) studies. This is accompanied by a sizable enhancement in the PL intensity of the GaAs/GaNAs core/shell NWs at room temperature. The observed N-induced suppression of the surface recombination is concluded to be a result of an N-induced modification of the surface states that are responsible for the nonradiative recombination. Our results, therefore, demonstrate the great potential of incorporating GaNAs in III-V NWs to achieve efficient nano-scale light emitters.

No abstract

The works on degradation of light emitting diodes (LEDs) with quantum wells (QW) were analysed. The calculation model of the relation between LED luminous flux and duration of LED current flow and current density was proposed. It allows us to forecast service life of such radiators with the pre-set electric modes and temperature. It is demonstrated that: – Reduction of quantum yield of LED with QW based on high-bandgap semiconductors with longterm flow of forward current occurs due to generation of point defects in the QW areas; – The dot defects occur as a result of interaction between hot electrons and semiconductor atoms caused by subthreshold displacement out of the lattice dots; – The dot defects create non-radiative recombination centres with particular concentration in the energy gap of a semiconductor, as a result of which intensities of recombination flows in QWs and in barriers between QWs redistribute towards the non-radiative component of the ABC model.

No abstract

No abstract

The lifetime of deep-ultraviolet light-emitting diodes (LEDs) is still limited by a number of factors, which are mainly related to semiconductor defects, and still need to be clarified. This paper improves the understanding of UV LED degradation, by presenting an analysis based on combined deep-level transient spectroscopy (C-DLTS), electro-optical characterization, and simulations, carried out before and during a constant current stress test. The original results of this paper are (i) C-DLTS measurements allowed us to identify three traps, two associated with Mg-related defects, also detected in the unaged device, and one related to point defects that were generated by the ageing procedure. (ii) Based on these results and on TCAD simulations, we explain the variation in the forward I–V by the degradation of the p-contact, due to Mg passivation. (iii) On the other hand, optical degradation is ascribed to an increase in defectiveness of the active region and surrounding areas, which led to a decrease in injection efficiency, to an increase in non-radiative recombination, and to an increase in trap-assisted tunneling processes.

No abstract

No abstract

InGaN micro-LEDs (μLEDs) with their potential high-volume applications have attracted substantial research interest in the past years. In comparison to other III–V semiconductors, InGaN exhibits a reduced susceptibility toward non-radiative surface recombination. However, efficiency degradation becomes more prominent as dimensions shrink to a few μm or less. Due to the high surface-to-volume ratio of the miniaturized devices, the non-radiative recombination increases and reduces the internal quantum efficiency. While many groups focus on surface passivation to mitigate surface defects, the influence of crystallographic orientation of the μLED sidewall on the efficiency remains unexplored. This study addresses this gap by investigating the impact of crystallographic orientation of the sidewalls on the emission properties of the μLEDs. Hexagonal and elongated μLEDs with dimensions as small as 3.5 μm and sidewalls with crystallographically well-defined m- and a-planes were fabricated. Electrical and optical properties were investigated using photo- and electroluminescence. External quantum efficiency (EQE) is assessed based on well-known carrier recombination models. It can be shown that μLED performance intrinsically depends on the crystallographic orientation of the sidewalls. Comparing hexagonal μLED structures with a-plane and m-plane sidewalls, an increase in the EQE by 33% was observed for structures with a-plane sidewalls, accompanied by reduction in the current density of the peak EQE by a nearly two orders of magnitude compared to structures with m-plane sidewalls. By analyzing the EQE characteristics at the μLED center and near the sidewalls, the improvements can be directly attributed to the increased radiative recombination from sidewalls with a-plane orientation.