A brief review is given of electronic and optical properties of carbon nanotubes mainly from a theoretical point of view. The topics cover an effective-mass description of electronic states, Aharonov-Bohm effects, and optical absorption including interaction effects on the band structure gap and excitonic effects.

Inorganic single crystal nanowires and nanowalls which exhibit rich growth morphogenesis are shown. More specifically, these were grown on lattice-matched substrates, which facilitate their specific growth directions with respect to the substrates' planes. Structural and optical characterizations suggest high single crystallinity of these nanostructures and possible applications in nano-optoelectronics are discussed.

Vapor-liquid-solid epitaxy process has been developed to synthesize high quality semiconductor nanowire arrays. The nanowires generally are single crystalline and have diameters of 10-200 nm and aspect ratios of 10-100. There is much current interest in the optical properties of these semiconductor nanowires, as the cylindrical geometry and strong two-dimensional confinement of electrons, holes, and photons makes them particularly attractive as potential building blocks for nanoscale electronics and optoelectronics devices. We recently reported the first study of laser action and nonlinear optical mixing in individual zinc oxide (ZnO) and GaN nanowires, demonstrating the potential of these structures as room temperature nanoscopic coherent light sources and frequency converters. These efforts further led to the demonstration of ZnO nanoribbon laser as well as GaN-based quantum wire lasers.

We review the basic electromagnetic properties of semiconductor nanowires which are required to evaluate their performance as lasers. These properties include the dispersions for guided modes, mode spacing, reflectivities from the nanowire facets, directionality and polarization of far fields, and confinement factors. We also discuss features that distinguish nanowire lasers from the usual heterostructure lasers.

The molecular beam epitaxy of self-assembled quantum dots (QDs) has reached a level such that the principal advantages of QD lasers can now be fully realized. We overview the most important recent results achieved to date including excellent device performance of 1.3 μm broad area and ridge waveguide lasers (J_th<150A/cm², I_th=1.4 mA, differential efficiency above 70%, CW 300 mW single lateral mode operation), suppression of non-linearity of QD lasers, which results to improved beam quality, reduced wavelength chirp and sensitivity to optical feedback. Effect of suppression of side wall recombination in QD lasers is also described. These effects give a possibility to further improve and simplify processing and fabrication of laser modules targeting their cost reduction. Recent realization of 2 mW single mode CW operation of QD VCSEL with all-semiconductor DBR is also presented. Long-wavelength QD lasers are promising candidate for mode-locking lasers for optical computer application. Very recently 1.7-ps-wide pulses at repetition rate of 20 GHz were obtained on mode-locked QD lasers with clear indication of possible shortening of pulse width upon processing optimization. First step of unification of laser technology for telecom range with QD-lasers grown on GaAs has been done. Lasing at 1.5 μm is achieved with threshold current density of 0.8 kA/cm² and pulsed output power 7W.

Model consideration is given to explain observed multi-shell emission spectra from InAs quantum dots embedded in GaAs or InGaAs. The shell model is based on the quantization of kinetic energy of lateral motion of carrier in the dot. 2-D oscillator is calculated on the basis of effective mass approximation. Profiles of inter-level separation are classified into categories that are connected with the lateral confining potential. Comparison is carried with experimental data on InAs/InGaAs quantum dot structures of the DWELL type (dot-in-a-well).

There has been great interests on semiconductor quantum dot (QD) due to its novel physical properties and potential applications such as semiconductor lasers with high gain and narrow linewidth. The collection of carriers by the QDs is a critical issue for efficient gain of QD lasers. A tunneling injection quantum-dot laser has been researched recently. Direct, photon-, phonon-, and Auger-assisted tunneling are all possible mechanisms for carrier transfer from QW to QD. In this talk, we present a theoretical model for the phonon-assisted tunneling from a quantum well (QW) state to the QD ground state in the conduction band. We assume a quantum-disk model and use its analytical wave functions to calculate the tunneling rate based on Fermi's Golden rule. The single-LO-phonon-emission and absorption processes are modeled by Froelich Hamiltonian. The dependence of the tunneling rate on the QW carrier density, temperature, barrier width between QW and QD, and energy difference between the QW state and the QD state are studied. The tunneling time ranging from several to a few tens of picoseconds are possible depending on the thickness of the barrier and the energy spacing between the QW and QD states.

Using the segmented contact method, we have measured the passive modal absorption, modal gain and spontaneous emission spectra of an InAs “dot-in-well” (DWELL) system where the inhomogeneous broadening is sufficiently small that the ground and excited state transitions can be spectrally resolved. The modal optical gain from the ground state saturates with current at a maximum value of one third of the magnitude of the measured absorption. The population inversion factor spectrum, obtained from the measured gain and emission spectra, shows that the carrier distributions cannot be described by a single global Fermi distribution. However, the inversion factor spectrum can be described by a system where the ground state and excited state occupancies are each described by a Fermi distribution but with different quasi-Fermi energy separations.

We develop a general approach to including the internal optical loss in the description of semiconductor lasers with a quantum-confined active region. We assume that the internal absorption loss coefficient is linear in the free-carrier density in the optical confinement layer and is characterized by two parameters, the constant component and the net cross-section for all absorption loss processes. We show that the free-carrier-density dependence of internal loss gives rise, in general, to the existence of a second lasing threshold above the conventional threshold. Above the second threshold, the light-current characteristic is two-valued up to a maximum current at which the lasing is quenched. We show that the presence of internal loss narrows considerably the region of tolerable structure parameters in which the lasing is attainable; for example, the minimum cavity length is significantly increased. Our approach is quite general but the numerical examples presented are specific for quantum dot (QD) lasers. Our calculations suggest that the internal loss is likely to be a major limiting factor to lasing in short-cavity QD structures.

We have developed a theory of high-speed operation of quantum dot (QD) semiconductor optical amplifiers (SOAs), and showed that pattern-effect-free amplification of pulse trains, cross gain modulation (XGM) and cross phase modulation (XPM) can take place in QD SOA in the regime with maximum gain. Formulas, which relate the maximum bit-rate for the pattern-effect-free operation and the average SOA output power to the SOA pump current density, were derived. XGM without pattern effect can be realized in the regime with maximum gain due to spectral hole burning effects. Possibility of ultrafast frequency conversion and demultiplexing of data pulse streams through this nonlinearity is illustrated. Expression for the nonlinear refractive index η_nl due to spectral hole burning in QD structure was obtained. The value of η_nl in QD SOAs can be by 4-5 orders larger than η_nl in silica; and efficient ultrafast XPM without pattern effects can be carried out in QD SOA through this nonlinearity. In whole, usage of the regime with maximum gain in QD SOAs can lead to development of new generation of high-speed devices for ultrafast optical processing and communications.

Quantum-dot infrared photodetectors (QDIPs) have been researched intensively because normal incidence absorption of the quantum dot layer is possible, unlike quantum-well infrared photodetectors (QWIPs) which require a grating coupled structure or an off-normal incidence configuration to satisfy the polarization selection rules. In this paper, we present a theoretical model for the band structure of a strained quantum dot, modeled as a quantum disk, and the intersubband absorption. We first present analytical expressions for the polarization-dependent optical dipole moments and then calculate the absorption spectra for various carrier densities and temperatures. The effects of carrier density, temperature, and inhomogeneous broadening will be discussed.

We study the linear intersubband absorption spectra of a 15 nm InAs quantum well using the intersubband semiconductor Bloch equations with a three-subband model and a constant dephasing rate. We demonstrate the evolution of intersubband absorption spectral line shape as a function of temperature and electron density. Through a detailed examination of various contributions, such as the phase space filling effects, the Coulomb many-body effects and the nonparabolicity effect, we illuminate the underlying physics that shapes the spectra.

We designed and fabricated some components and devices by using Si photonic wire waveguides on SOI substrate. Because of the very high index contrast between the core and claddings, the waveguide allows a μ-bend. We applied this bend to form a μ-branch and μ-intersection, which exhibited a low of less than 0.3 dB in the experiment. The H-tree optical signal distribution circuit, the Mach-Zehnder interferometer, and the arrayed waveguide grating demultiplexer were demonstrated by this waveguide, for the first time. Although a more careful design and precise fabrication technologies are necessary for future high performance, this waveguide is expected to miniaturize any kind of conventional silica based devices by a factor over 10000 and realize more sophisticate functions by the dense integration of devices.

The three-dimensional model of the packaging device is established based on ANSYS simulation platform. The thermal properties such as time response, axial and radial temperature distributions at different applied voltages are exhibited. With aids of Real Time Optical Spectrum Analyzing System and IR Camera System, time response of the device and axial temperature distribution along the coated fiber with intracore FBG are both demonstrated. Temperature responses to different applied voltages are achieved after measuring voltage induced wavelength shift and temperature dependent wavelength shift. Simulation shows results in agreement with those of experiment. Finally, regulations on length of the metal coating, size of the package, power consumption and tuning properties of the packaging device are discussed.

We demonstrate simultaneous stabilized operation of a mode locked ring fiber laser at two wavelengths. At one of the wavelengths the mode locked operation is at 10 GHz and it is at 40 GHz at the second wavelength. The laser has an intracavity LiNbO₃ modulator driven at 10 GHz. The 40 GHz pulses are obtained by rational harmonic mode locking. Pulses with widths in 5 to 8 ps range are obtained.

A new type of waveguide optical polarization splitter is proposed and investigated theoretically. The waveguide optical polarization splitter is composed of a Y branch waveguide and a microprism consisting of a dielectric periodic multilayer. As the dielectric periodic multilayer has large birefringence, the TM (x) - and TE (y) - polarized propagating waves are refracted with different angle each other at the microprism. This is the principle of the proposed waveguide polarization splitter. First we have designed the waveguide polarization splitters. An asymmetric Y branch, in which one output port is a straight waveguide for an input waveguide and the other is an abruptly bending waveguide, is used for the design. The refractive indices of the core and cladding (substrate) are 1.51 and 1.509, respectively. The dielectric periodic multilayer for the microprism has been designed so as the effective refractive index for the x-polarization become equal to the refractive index of the substrate. Therefore the x-polarized wave propagates for the output port consisting of the straight waveguide with low loss. The prism has been designed by using the method for the microorism-type of bending waveguide proposed by C.T. Lee and J.M. Hsu so as the y-polarized wave can propagate for the port consisting the abruptly bending waveguide with low loss. Finally we have calculated optical losses for the x- and y-polarizations by using a beam propagating method. The insertion losses of the typically designed waveguide optical polarization splitter for the x-and y-polarizations are 0.14 dB and 0.2 dB, respectively. It has also been confirmed that the crosstalks are <-35 dB for both polarizations.

A new method for solving the wave equation is presented, which, being non-paraxial, is applicable to wide-angle beam propagation. It shows very good stability characteristics in the sense that relatively larger step-sizes can be used. It is both faster and easier to implement. The method is based on symmetrized splitting of operators, one representing the propagation through a uniform medium and the other, the effect of the refractive index variation of the guiding structure. The method can be implemented in the FD-BPM, FFT-BPM and collocation schemes. The method is stable for a step size of 1 micron in a graded index waveguide with accuracy better than 0.001 in the field overlap integral for 1000-micron propagation. At a tilt angle of 50°, the method shows an error less than 0.001 with 0.25-micron step. In the benchmark test, the present method shows a relative power of ~0.96 in a 100 micron long waveguide with 1000 propagation steps and 800 sample points, while FD-BPM with Pade(2,2) approximation gives a relative power of 0.95 with 1000 sample points and 2048 propagation steps. Thus, the method requires fewer points, is easier to implement, faster, more accurate and highly stable.

Two schemes are developed to improve the computational accuracy of the full-vectorial imaginary-distance beam propagation method (FV-ID-BPM). In the first scheme, the cross-coupling terms (CCTs) demanded for the FV analysis are expressed in explicit forms, which are independent of specific types of waveguides, by using an improved finite-difference formula. In the second one, the generalized Douglas (GD) scheme is adopted for discretizing the second-order partial derivatives in the FV-ID-BPM equations. A detailed comparative study between the two schemes in improving the computational accuracy is performed by taking a strongly-guiding rib waveguide as a testing example. The highest accuracy is demonstrated in case of the combination of the two schemes. Nevertheless, the improved FD formula for the CCTs is proved to play a much more significant role than the GD scheme in improving the computational accuracy. Moreover, the effectiveness of the GD scheme diminishes as the FD grid is refined.

In numerical wave propagation methods, the perfectly matched layer (PML) boundary condition is employed to prevent spurious reflections. However, PML takes additional resources in number of computation points and time. In this study, the PML performance is examined with change in the distribution of sampling points and PML absorption profile with a view to optimizing its efficiency. We have used the collocation method in our examples. We have found that equally spaced field sampling points give better absorption of beams under both optimal as well as non-optimal conditions for lower PML widths. While at higher PML widths, unequally spaced basis points may be more advantageous. The behavior of different absorption profiles varies with point spacing. For numerical tests, Gaussian beam propagation in a homogeneous medium is considered. Comparing different profiles, we find that a new profile sin^p with p=4 and quartic profiles are best in equally spaced points, sin² and square profiles are best in unequally spaced points.

We present accurate closed form expressions for modal fields and propagation constant of the fundamental mode in diffused channel waveguides using a scalar variational approach. The results have been compared with those obtained by use of the Optiwave BPM_CAD software package. The closed form fields have been shown to be useful in optimizing fiber to waveguide coupling and evaluating gain in Erbium doped waveguides.

We have developed public domain numerical models of nonlinear three-wave mixing in birefringent crystals that include diffraction and dispersion. They are suitable for detailed and realistic modeling of mixing for both a single crystal pass and for multiple passes appropriate for a crystal in a resonant cavity. We routinely compare our models with laboratory devices, usually achieving excellent agreement.

In this work, we describe tunable wavelength converters based on a photodiode receiver integrated with a tunable laser transmitter. Devices are fabricated on a robust InP ridge/InGaAsP waveguide platform. The photodiode receiver consists of an integrated SOA pre-amplifier and a PIN diode to improve sensitivity. The laser transmitter consists of a 1550 nm widely tunable SGDBR laser modulated either directly or via an integrated modulator outside the laser cavity. An SOA post-amplifier provides high output power. The integrated device allows signal monitoring, transmits at 2.5 GB/s, and removes the requirements for filtering the input wavelength at the output. Integrating the SGDBR yields a compact wavelength agile source that requires only two fiber connections, and no off-chip high speed electrical connections. Analog and digital performance of directly and externally modulated wavelength converters is also described.

In this paper, we present three-dimensional (3D) simulation results for an integrated wavelength converter which monolithically combines a pre-amplifying receiver with a post-amplified sampled-grating distributed Bragg reflector tunable laser diode. The self-consistent physical model used in the simulation takes into account gain and absorption in the quantum wells, carrier drift and diffusion, and optical wave-guiding. In order to validate and calibrate the model, we compare the results to available experimental data. Microscopic physical processes inside the converter components are revealed and analyzed, such as receiver saturation effects.

We analyze the optoelectronic mixing characteristics of InAlAs, Schottky-enhanced, InGaAs-based, metal-semiconductor-metal photodetectors. For devices with Schottky-enhancement layers (SELs) of about 500 Å, the measured frequency bandwidth is less than that of a corresponding photodetector. The mixing efficiency decreases with decrease in optical power, decreases with increase in local oscillator frequency and decreases with decrease in mixed signal frequency. We attribute this behavior to the band-gap discontinuity associated with the SEL. For devices with thinner SELs (≈100 Å), the mixing characteristics are greatly improved: the bandwidth of the optoelectronic mixer (OEM) is similar to that of a corresponding photodetector and the mixing efficiency decreases only slightly with decrease in optical power. We attribute these results to the enhancement of the thermionic/tunneling current through the thinner SEL. We also present a circuit model of the Schottky-enhanced, InGaAs-based OEM to explain the experimental results.

We report results for broad area edge emitting lasers having AlInGaAs active regions that exhibit low thresholds, high T₀ and T₁ and high efficiencies. The lasers were grown on InP substrates using MOCVD. This paper analyzes the effects of doping, epilayer design, wavelength dependence and number of QWs on device performance. Our results show that the concentration and offset of zinc doping in the p-cladding layer plays a major role in carrier confinement and hence high temperature performance. The difference in surface mobility of Al adatoms (as compared to In or Ga) poses some challenges in the growth of AlInGaAs.

The emission spectrum of a two-color semiconductor laser is analyzed. We find four-wave mixing sidebands for difference frequencies up to 4 THz. The appearance of four-wave mixing signals is a clear sign for a modulation of the carrier plasma at the corresponding difference frequency. We prove experimentally that this difference frequency is also directly emitted out of the laser diode and suggest a new simple concept for the generation of tunable coherent THz-radiation.

Photonic packet switching for all-optical networks is a rapidly developing technology since it circumvents many of the traditional bottlenecks created by the use of electronics. All-optical networking has application to both long-haul communications systems and high-performance computing systems. In each case, all-optical technologies are responsible for the routing, switching and logic decisions of the network. Characterizing the performance of a network includes calculating the latency and scalability of a given architecture assuming ideal behavior of its physical components. However, the physical layer ultimately determines the feasibility of data transmission. Thus accurately calculating the accumulated bit-error-rate (BER) is fundamental to evaluating the optical network as a whole, regardless of the network architecture. A new simulation technique, which is based upon experimental findings, is introduced which characterizes the physical layer performance of a given network architecture known as the Data Vortex. Experiments show that almost all the physical layer penalty is generated by the nodes which are used for switching and routing. Specifically, at each node data packets are amplified by a semiconductor optical amplifier so that coupling and routing losses are compensated. In this process, the data packets receive a noise penalty which results primarily from amplified spontaneous emission and in small part from spectral broadening. By using a phenomenological approach to modeling the noise penalties, the performance of the network nodes can be characterized. The modeling allows for a comprehensive understanding of the network and is a highly efficient computational tool for evaluating performance when compared to conventional time-domain techniques.

We report on wavelength-division-multiplexing (WDM) based on lithographically-fabricated slab-waveguide-contained planar holographic Bragg reflectors (HBRs). Partial HBR diffractive contour writing and contour displacement are successfully demonstrated to enable precise bandpass engineering of multiplexer transfer functions and make possible compact-footprint devices based on hologram overlay. Four and eight channel multiplexers with channel spacings of ~50 and ~100 GHz, improved sidelobe suppression and flat-top passbands are demonstrated. When a second-order apodization effect, comprising effective waveguide refractive index variation with written contour fraction, and the impact of hologram overlap on the hologram reflective amplitude are included in the simulation, excellent agreement between predicted and observed spectral passband profiles is obtained. With demonstrated simulation capability, the ability to fabricate general desired passband profiles becomes tractable.

Anti-competition of laser modes is observed in dual-wavelength semiconductor lasers with single gain medium. Under anti-competition, the increase of intensity of one lasing mode could enhance the intensity of another mode, which is opposite to the usual mode competition. In our experiment, anti-competition can be observed for wavelength separation larger than 111 nm, and gradually disappears for wavelength separation less than 100 nm. Besides, anti-competition can also be influenced by the intensity and the wavelength position of both modes. A simple theoretical analysis shows that anti-competition is due to the physics similar to optical pumping.

We present two methods for timing jitter measurement. The first method uses the spectral content of the noise, and, the second method uses autocorrelation and cross correlation of the pulsed output. For a distributed feedback (DFB) laser gain switched at 1 GHz, the timing jitter is about 3-4 ps. When an external CW laser injected into the DFB laser, the timing jitter can be reduced to 1-1.5 ps.

All-optical regeneration at 40 Gbit/s and beyond is a crucial element for future transparent networks. One solution to achieve the regeneration is an all-optical clock recovery element combined with a Mach-Zehnder interferometer. Among the different approaches investigated so far to accomplish the clock recovery function, a scheme based on a single self-pulsating distributed Bragg reflector laser is of particular interest from practical and cost viewpoints. In this structure at least two longitudinal modes beat together, generating power oscillation even though the laser is DC biased. The oscillation frequency is given by the free spectral range of the structure. In order to optimize the clock recovery performance of such a laser, a model based on four-wave-mixing has been developed. It takes into account the evolution of the amplitude and the phase of the complex electricfield of each longitudinal mode. From this model, a stability analysis is derived through the adiabatic approximation. The spectral density of the correlated phases of these modes is calculated and compared to the uncorrelated spectral density of each mode.

High-frequency spectra of free-running 5-mm-long triple-quantum-well graded-index separate-confinement heterostructure broad-area diode lasers emitting at ~1 μm are investigated in the range of 1-20 GHz using RF spectrum analyzer. The spectra reveal stable beat lines at ~8 and ~16 GHz, corresponding to single and double mode spacings between adjacent longitudinal modes. A current-dependent peak, varying from 0.6 to 2 GHz, is associated with the relaxation resonance. Measurements of mode beating spectra provide additional characterization of diode laser emission for coherent light applications.

Chaos synchronization in polarization selected mutually coupled vertical-cavity surface-emitting semiconductor lasers (VCSELs) is experimentally investigated in a low-frequency fluctuation regime. Two lasers synchronize in one of the orthogonal polarization modes selected for synchronization. The counterpart polarization components also show synchronized outputs due to anti-phase oscillations of VCSELs.

We have developed a fiberoptics setup which can be easily specialized with minor changes to implement different schemes of optical chaos generation and synchronization using semiconductor lasers. Long and short cavity, open and closed loop configurations have been compared, as well as various encoding/decoding methods for secure transmission based on chaotic carriers, such as CSK (Chaotic Shift Keying), ACM (Additive Chaotic Masking), CM (Chaos Modulation). Different transmission media, possibly including optical amplifiers, have been also tested.

A direct experimental observation of chaotic synchronous scenarios, namely chaotic optical modulation, is demonstrated in a unidirectional chaotic-coupling semiconductor laser system. In this fully optical system, the channel signal is different from the output field of the transmitter laser by an additional monochromatic optical field. Different from the chaos synchronization explainable by theory of chaos synchronization, the output field of the receiver laser is not synchronized to that of the transmitter laser. Instead, it is synchronized to the channel signal. However, the optical frequency of the receiver is not locked to that of the transmitter. It is observed that not only is the intensity of the receiver output is synchronized to that of the channel signal, but also the chaotic slowly-varying phase of the receiver. The synchronization of the slowly-varying phase is verified by optical interference between the output of the receiver and the channel signal, and the interference result is recorded through a photodetector.

We investigate the synchronization properties of two mutually-coupled semiconductor lasers (SL) in a face to face configuration, when a non-negligible injection delay time is taken into account. Under the appropriate conditions, we derive a thermodynamic potential analog to the one studied by Mork et al. and by Lenstra for a semiconductor laser subject to an optical feedback. In this context, the role that noise and detuning play in the dynamics of the system is clearly identified. When operating in the Low Frequency Fluctuations (LFF) regime, the effect of the detuning on the leader-laggard operation is also analyzed. Finally, we focus on the short intercavity regime and we study the influence of the detuning and the propagation phase on the dynamics of each laser.

Mutually-coupled semiconductor lasers are of great current interest because of the important insight they provide into coupled physical, chemical, and biological systems. Two semiconductor lasers either with or without optoelectronic feedback are mutually coupled together through optoelectronic paths. It is found that mutual coupling can significantly affect the dynamics of the semiconductor lasers, depending on the coupling delay time and the coupling strength. Interesting phenomena such as generation of chaos, quasiperiodic and period-doubling bifurcation to chaos, and death by delay are observed. Synchronization of the chaotic outputs from mutually coupled semiconductor lasers is also observed.

We study the dynamics of two Vertical Cavity Surface Emitting Lasers (VCSELs), when they are bidirectionally coupled through the mutual injection of their coherent optical fields. In the long distance limit between the lasers, we focus on the Low Frequency Fluctuations (LFF) regime and we investigate the polarization-resolved dynamics of each laser under the effect of detuning. In the short distance limit, the influence of the propagation phase parameter is also evaluated. For large spin-flip rates, it is found that a change in the propagation phase may induce a sudden switch in the polarization mode that becomes dominant. Extensive simulations scanning the Coupling-Detuning space are performed for both long and short injection delay times.

The characteristics of period-one oscillations in semiconductor lasers subject to optical injection is experimentally and quantitatively investigated. The changes in the frequency separation and in the magnitude difference between the principal oscillation and the sideband of the injected laser are studied as a function of experimentally accessible parameters, the detuning frequency and the injection strength of the injection signal. The frequency separation decreases as the injection strength and the detuning frequency decrease. The magnitude of the principal oscillation decreases with the decreasing injection strength and the increasing detuning frequency, while that of the sideband grows at the same time. At some operating conditions, these characteristics leads to a situation that the magnitude of the sideband becomes larger than that of the original principal oscillation, resulting in a frequency shift of the principal oscillation from the injection frequency to the sideband.

The dynamical behavior of a single-mode laser subject to optical feedback is investigated in the limit, when the delay time is much shorter than the period of the relaxation oscillations. Use of an integrated DFB device allows us to control the feedback phase. The system shows a very rich manifold of nonlinear phenomena. Among them are two kinds of Hopf bifurcations associated with regular self-pulsations of different frequencies as well as a fold and period doubling bifurcation.

Self-pulsating semiconductor lasers are deviced to reduce optical feedback noises as light sources in optical data storage systems. However, they themselves include instabilities in their solitary oscillations without any optical feedback and they also show unstable behaviors induced by optical feedback. We experimentally investigate instabilities and dynamics of self-pulsating semiconductor lasers without and with optical feedback from a distant reflector on the order of several tens of centimeters to one meter.

We demonstrate novel electrically driven 1330 and 1550 nm VCSELs using conventional InGaAsP active regions. The VCSELs employ two TiO2/SiO2 DBR mirrors and an InAlAs tunnel junction that converts electrons to holes, minimizing free carrier losses in the p-type material. The active layers are transferred onto Si wafers using wafer-scale Pd silicide bonding. We have obtained single-mode room-temperature output powers as high as 2.4mW at 1330nm and 2.7mW at 1550nm. At 80C we have obtained 0.6mW of single-mode power at 1330nm and over 1 mW at 1550nm. These are the highest power single-mode InP-based VCSELs reported in these wavelength ranges.

This paper discusses the design and the internal device physics of novel high-performance vertical-cavity surface-emitting lasers (VCSELs) emitting at 1.32 µm wavelength. Our VCSEL design features intra-cavity ring contacts, strain-compensated AlGaInAs quantum wells, and an AlInAs/InP tunnel junction. The tunnel junction is laterally confined forming an aperture for current injection and wave guiding. Undoped AlGaAs/GaAs mirrors are bonded on both sides to the InP-based active region. These devices have recently demonstrated continuous-wave (CW) lasing at stage temperatures up to 134°C, the highest temperature reported thus far for any long-wavelength VCSEL. In order to increase the single mode output power at high temperatures, we simulate, analyze, and optimize our VCSEL using advanced numerical software tools. The two-dimensional model self-consistently combines electrical, optical, thermal and gain calculations. It gives good agreement with measurements after careful calibration of material parameters. Design optimization promises single mode output power of 2mW in CW operation at 80°C ambient temperature.

The role of metal apertures in the mechanism of lateral mode confinement in vertical-cavity surface-emitting lasers (VCSELs) is clarified by means of a detailed effective-frequency-method analysis of an oxide-confined VCSEL structure with the radius of the oxide window exceeding that of the metal aperture. Ring metal contact layer on top of the VCSEL structure is shown to be able to change the conditions for the lateral waveguiding in VCSELs by significantly modifying the local resonant properties of the VCSEL cavity. The resonant effects are demonstrated in the longitudinal coupled-cavity system consisting of the designed laser cavity, determined by the lower and top DBRs, and a very short cavity formed by the top DBR and semiconductor-metal interface. The conditions for higher-order lateral mode suppression using metal apertures are established.

Electrically conductive zirconium diboride (ZrB₂) is a promising lattice-matched substrate for GaN-based nitride semiconductors. In this paper, important properties of ZrB₂ as a substrate for nitrides, such as, thermal expansion coefficient, thermal conductivity, optical reflectivity and cleavage, are reviewed. Then, heteroepitaxial growth of GaN and AlN on the substrate by molecular-beam epitaxy (MBE) are discussed. Direct growth and two-step growth using low-temperature GaN nucleation layers as well as characterization of the surface condition of ZrB₂ substrates by X-ray photoelectron spectroscopy (XPS) and the effect of surface treatment on grown layers are presented.

Properties of InGaN/GaN multiple-quantum-well (MQW) light-emitting diodes (LEDs) grown by MOCVD on sapphire substrates are investigated over a temperature range from 290 to 340 K. Two types of wafers are used to fabricate the devices: one with Mg dopants in p-type epilayers pre-activated in N₂ ambient for 4 min at 800 °C, and the other as-grown, without any pre-activation of Mg acceptors. Measured specific resistances of p-side contacts are 1.49x10^-4 Ωcm² for contacts on pre-activated samples annealed at 650°C for 4 min, and 1.55x10^-5 Ωcm² for contacts on as-grown samples annealed at 600 °C for 30 min. Based on the specific contact resistance experiments, interdigitated LEDs are fabricated using either the standard annealing procedures (separate annealings for p-type conduction activation and for ohmic contact formation) or a single-step annealing process (simultaneous annealing for activation of p-type conduction and for ohmic contact formation). In devices fabricated using the standard annealing procedures, the electroluminescence (EL) peak position at 300 K is at 2.379 eV (~521.3 nm) and the full width at half maximum (FWHM) is ~132 meV, while in devices fabricated using a single-step annealing, the EL peak position shows a red shift by ~10 meV without affecting the FWHM. Over the entire voltage range up to 4 V, tunneling is the dominant carrier transport mechanism. The operating voltage is comparable in both types of LEDs, and the output power of LEDs fabricated using the single-step annealing process is somewhat improved.

Operation-induced degradation of internal quantum efficiency of high-brightness (Al_xGa_1-x)_0.5In_0.5P light-emitting devices (LEDs) is analysed experimentally and theoretically. A test series of LEDs was grown by MOCVD with identical layer sequence but different Aluminum content x in the active AlGaInP layer resulting in devices emitting light between 644 nm and 560 nm. The analysis yields the wavelength dependence of both the nonradiative recombination constant A as well as the carrier leakage parameter C of devices before and after aging. While test devices with λ>615 nm are very stable, LEDs with shorter emission wavelengths exhibit both an increase of A and a slight decrease of C upon aging. Possible degradation mechanisms are discussed.

We have studied the effects of composition and hydrostatic pressure on the direct optical transitions at the Γ point of the Brillouin zone in MBE-grown ZnO_xSe_1-x and ion-implantation-synthesized Zn_1-yMn_yO_xTe_1-x alloys. We observe a large O-induced band-gap reduction and a change in the pressure dependence of the fundamental band gap of the II-O-VI alloys. The effects are similar to those previously observed and extensively studied in highly mismatched III-N-V alloys. Our results are well explained in terms of the band anticrossing model that considers an anticrossing interaction between the highly localized oxygen states and the extended states of the conduction band of II-VI compounds. The O-induced modification of the conduction band structure offers an interesting possibility of using small amounts of O to engineer the optoelectronics properties of group II-O-VI alloys.

We present a new distributed time domain model (DTDM) using Maxwell's wave equations with a time dependent polarization in the form of classical electron oscillators (CEO)s with randomly excited spontaneous emission using a virtual field. The model is based upon the neoclassical rate equations of A.E. Siegman and includes effects such as chromatic dispersion, line-width enhancement, gain suppression, optically induced gratings, and excess noise. Although our equations were independently derived we have found that they do resemble the Maxwell-Bloch equations. However, most authors appear to favor the Ginzburg-Landau equations for their DTDM models. We demonstrate that the model can reproduce results comparable with those of others, as well as new results.

The laser diodes and laser bars with InGaAlAs/GaAs active region are attractive as high power devices operating at around 808 nm. The quaternary InGaAlAs active region seems to have distinctive advantages over the standard GaAs quantum well construction. The most important of them is that quantum wells, required to achieve desired wavelength can be wider, which provides better carrier confinement. Another advantage is better thermal conductivity of InGaAlAs as comparing to GaAs. We have modeled single and double quantum well separate confinement heterostructure lasers with various cavity lengths. The well thickness and indium content in the active region were optimized to obtain 808 nm wavelength with acceptable threshold current density. Numerical simulation based on the selfconsistent solution of drift diffusion equations, Schrödinger equation and photon rate equation has been used to optimize the high power lasers design. In this work we have used commercial simulation package PICS3D developed by Crosslight Soft. Inc.

Crystalline silicon being ubiquitous throughout the microelectronics industry has an indirect bandgap, and therefore is incapable of light emission. However, strong room temperature visible and near-IR luminescence from non-crystalline silicon, e.g., amorphous silicon, porous silicon, and black silicon, has been observed. These silicon based materials are morphologically similar to each other, and have similar luminescence properties. We have studied the temperature dependence of the photoluminescence from these non-crystalline silicons to fully characterize and optimize these materials in the pursuit of obtaining novel optoelectronic devices.

Theoretical work of our group is placed in the general frame of efforts to improve numerical performance and efficiency of rigorous coupled-wave analysis of grating diffraction. Mathematical transformation of Maxwell equations for a multi-layered structure to evolution equations in functional space is presented. By-construction numerically stable symbolic algorithm to solve these equations using the notion of in-layer scattering operator is proposed. On the base of this algorithm a toolbox for simulation of diffraction from multi-layered grating structures, implemented by a graphical user interface is developed. An example of simulation using this in-house software is exposed.