Reliable, high power laser diodes at 980 nm are the essential devices for pumping Er or Er/Yb doped fibre amplifiers. Applications where a high fibre coupling efficiency is required, demand lasers with high spatial beam quality. Tapered lasers are today one of the most interesting solution in terms of high brightness (high power with high spatial beam quality). Furthermore, a reduced temperature shift of the wavelength is important for high power optical pump applications because this implies a less stringent temperature control. New GaInAs/(Al)GaAs quantum-dot materials exhibit a high wavelength stability vs. current and temperature. Using these quantum dots as active region, we have developed index guided tapered diode laser with a small aperture (width less than 30 μm) and standard AR/HR coatings. This device showed an optical power of 1 W CW at 1.6 A at 980 nm with a M² factor of 2.9 in the slow axis. Furthermore, we have measured a wavelength variation of 0.0075 Å/mA at 20°C with current and 1.7 Å/°C with temperature, which is weaker than typical variations of quantum well lasers. We also developed an array of seven quantum-dot lasers which emits 3 W CW at 5.7 A. Quantum well individual devices, using the same confinement layers and waveguide geometry were also developed for comparison.

The quantum dot laser is a complex nonlinear system in which light fields dynamically interact with the charge carriers in the dots and the embedding quantum well medium. In real laser systems, typical dot-to-dot variations in size, energy levels and material parameters exist. In addition, the dots are not equally positioned on a grid within the layers. The respective variance in quantum dot parameters and dot-to-dot distance depends on the material system and the epitaxial growth process of the particular quantum dot system. To elucidate the influence of spatial fluctuations, we calculate the temporal light field dynamics of quantum dot lasers with variable fluctuations in the characteristic dot parameters. The simulations on the coupled ultrafast spatio-temporal light-field and carrier dynamics in quantum dot lasers are based on a two-level Multi-Mode Maxwell-Bloch description. The constituent equations consist of coupled spatio-temporally resolved wave equations and Bloch equations for the carriers within each quantum dot of a dot ensemble constituting the active gain medium of a quantum dot laser. It is shown that the light field dynamics and the emission spectra are strongly determined by the nonlinear coupling between the light fields and the charge carrier plasma, spatially varying material properties of the quantum dot ensemble as well as device geometry and carrier injection.

The explosion of internet traffic, the increase in data or multimedia transmission are the main reasons for a huge rise in demand for transmission bandwidth especially in dense wavelength division multiplexing (DWDM) systems. Nowadays this technique must be developed in the 1.4 μm to 1.65 μm wavelength range to follow the progress of new low-loss fibres. A decade ago, a new class of gain material, based on quantum dot was intensively studied. For three years, researchers have succeeded in growing new elongated nano-structures based on InP, called quantum dashes, for applications beyond the wavelength limit of 1.3 μm using GaAs-based quantum dots. These great strides in the elaboration of these new gain materials could meet this gain bandwidth. In the framework of the European project, BIGBAND, we have developed 1.55 μm quantum dash Fabry-Perot lasers based on InP using a ridge waveguide operating in continuous wave at room temperature. These devices have reached the power of 40 and 50mW per facet in p side up and down configurations respectively and have shown a low relative intensity noise (RIN) of -162 dB/Hz ± 1.6 dB in 0.1-13 GHz range.

A model portraying the carrier dynamics for an inhomogeneous array of quantum dots (QDs) interacting with a number of photon modes is presented. The model treats an ensemble of QDs with one confined level coupled to a wetting layer or quantum well level and explicitly considers only the electrons. The model is derived by numerically solving a set of rate equations that includes the inhomogeneity of the dot size, multimode photon modes and temperature dependence. Explicitly the inhomogeneous size distribution is included within a inhomogeneous broadening parameter and the temperature dependence within the homogeneous broadening parameter as well as carrier thermal escape. This is similar to the well-known Sugawara model but in the Sugawara model the carriers are assumed to occupy the inhomogeneous quantum dots equally at all temperatures. Experimental and theoretical work in ref. (2) and (3) believes this is true only for a low temperature regime. Above the temperature where a global minima exists, Fermi-Dirac statistics have been used. This results in different gain and lasing behaviour for higher temperatures from those calculated using the Sugawara model.

We present experimental and numerical studies of the dynamics of two delay-coupled device-identical semiconductor lasers. We concentrate on the regime of short delay times where the coupling delay is comparable to the period of the relaxation oscillation frequency. We find characteristic scenarios in the intensity dynamics depending on the spectral detuning and the coupling phase. For small spectral detuning, the coupled lasers lock to the same frequency and exhibit stable emission. For larger detuning, we find pronounced oscillations of the output intensity. The frequency of oscillation shows a stair-like behavior which depends on the detuning frequency, the coupling time and the coupling phase. In addition, the coupling phase also affects the oscillation frequency and the phase relation between the two oscillating lasers.

We consider two single-mode semiconductor lasers, which are coupled face to face via the injection of the optical field. We describe the symmetries of the coupled rate-equations model, the associated stationary solutions (synchronous and antisynchronous), and the bifurcations between them for the case when the propagation delay of the injected field is small. For the coupled lasers with a detuning, we discuss the emergence of self-pulsations with specific properties (anti-phase, in-phase). We also identify parameter regions, where chaotic pulsations occur.

Distant lasers with mutual optical injection are subject to a delayed coupling. We consider the barely investigated case of delays shorter than the relaxation oscillation period. In order to illuminate the role of these short delays, the ultimate zero-delay limit is considered as a reference. We use a traveling wave equation model, which fully resolves the spatio-temporal distributions of optical fields and carriers in the lasers and treats the wave propagation between the lasers by delayed boundary conditions. Wavelength detuning between the otherwise identical single-mode DFB lasers is used as primary bifurcation parameter. The zero-delay reference exhibits a synchronisation scenario typical for coupled oscillators. The nonsynchronised regimes represents a self-pulsation of the nonlinear carrier-photon system. Additional effects appear when including a short delay. Resonances of the cavity formed by the laser pair cause a staircase dependence on detuning of the pulsation frequency. Irregular dynamics is observed at the borders of the locked regions as well as at the edges of the stairs.

The paper is concerned with a theoretical study of synchronization between two end-to-end coupled lasers. Lasers are treated within the framework of a special multimode theory, valid for arbitrary coupling between lasers, where the laser field is decomposed in terms of the composite-cavity modes of the entire coupled-laser system. Bifurcation continuation techniques are used to systematically investigate the resulting equations under class-A, and further under class-B approximation. We discovered that the mechanism leading to laser synchronization changes from strong composite-cavity mode competition in class-A regime to frequency locking of composite-cavity modes in class-B regime.

We present a theoretical study into the dynamics and bifurcations of a semiconductor laser subject to delayed optical feedback, as modelled by the Lang-Kobayashi equations. For the case of a short external cavity, of the order of a few centimeters, there is a limited number of external cavity modes (ECMs), which makes it possible to apply advanced techniques from dynamical systems, such as the continuation of ECMs and their bifurcations, and the computation of unstable manifolds. From the physical point of view, a short cavity is characterized by the fact that the delay time in the external cavity is of the same order of magnitude as the period of the relaxation oscillation of the laser. In this regime the optical feedback phase is known to play an important role. We provide a detailed overview of how the dynamics depends on the feedback phase, which is in good agreement with recent experimental measurements.

We study the influence of delayed optical feedback from a short external cavity on the emission dynamics of semiconductor lasers using the Lang and Kobayashi rate equation model. We present the bifurcation scenario leading to regular pulse packages (RPP) and give examples of bistability between RPP and time-periodic or steady state solutions. We investigate the change of the shape of the envelope of RPP in the transition to LFF. We analyze regions of feedback parameters for which RPP occurs. Detailed mapping shows that with increasing the delay time the windows of RPP broaden, merge and finally shrink when approaching the relaxation oscillation (RO) period. In such a way the largest region of RPP occurs for delays around half of the RO period of the solitary laser. Moreover, the period of RPP also possesses a minimum as a function of the delay time corresponding to approximately the half of the RO period. For smaller delays the RPP period shows an oscillatory behavior with the delay which we identify as being due to the destabilization of the RPP in the vicinity of newly born external cavity modes. Furthermore, we reveal continuous increase of the period of the RPP with the feedback rate. Finally, we study the scaling of the frequency of the pulse package envelope with the injection current. Our results contribute to better understanding of the origin and the peculiarities of the RPP dynamics.

We study analytically the bifurcations of phase-locked emission modes of a semiconductor laser subject to both optical feedback and CW optical injection. The theory is valid for a large external cavity and for weak injection and feedback rates. Full calculation details are given in a dedicated appendix.

Extensive mode-locking investigations are performed in InGaAs/InAs/GaAs quantum dot (QD) lasers. Monolithic mode-locked lasers are fabricated using QD material systems grown by MOCVD and MBE techniques and emitting at 1.1μm and 1.3μm, respectively. The mode-locking performance is evaluated using a variety of laser designs, with various ridge waveguide geometries, cavity and absorber lengths. Passive and hybrid mode-locking are studied and compared in 3.9mm long devices emitting at 1.1μm and operating at a repetition rate of 10GHz. Using 2.1mm long devices emitting at 1.3μm, 18GHz passive mode locking with 10ps Fourier transform limited pulses is demonstrated. This confirms the potential of quantum dot laser for low chirp, short optical pulse generation. Preliminary investigation of the timing jitter of QD passively mode-locked lasers and the behaviour of the QD absorber are also presented. Finally, we report 36GHz passive mode-locking with 6ps optical pulse obtained using 1.1mm long QD lasers emitting at 1.3μm.

We have investigated the characteristics of 1550 nm GaInAsP/InP multiple quantum well (MQW) structures to be used as the gain medium in monolithic mode-locked lasers. For this purpose, a series of laser structures with 3 QWs, 5 QWs and 8 QWs were grown and processed into ridge waveguide lasers. The impact of the quantum well number was studied by analyzing the changes in threshold current, external quantum efficiencies, gain-spectra and linewidth enhancement factors, which are valuable in design and modeling of the mode-locked lasers. Monolithic 20 GHz mode-locked lasers were fabricated. Pulse trains with a good extinction ratio of 14.8 dB and less than 14 ps in width were demonstrated, and an average power of 1 mW could be coupled into an optical fiber.

The use of a fully two-dimensional waveguide analysis of the optical modes in a self-pulsating laser is described. We show that there is a need for such an analysis as the effective index model often used for modeling semiconductor lasers is shown not to provide a correct optical mode when the lateral waveguide is only weakly guiding. An efficient technique for solving the large and sparse matrix calculation in the two-dimensional model, the implicitly restarted Arnoldi method, is used and shown to be useful for the repetitive calculation necessary to describe the laser dynamics. We find that the optical field and optical gain show a variation in the pulse cycle and that this variation that occurs due to the change in carrier density, is responsible for the development of self-pulsation in the diode device.

Semiconductor mode-locked lasers and colliding-pulse mode-locked lasers have proven to be effective sources of pulsed optical signals with repetition rate of several tens of GHz and above. Examples of application can be found in OTDM systems, radio-over-fiber networks, and millimeter-wave generation. This work reports on the characterisation of monolithic colliding-pulse passive mode-locked (CPM) lasers at 60 GHz in GaAs/AlGaAs double quantum well material, subjected to optical feedback. The characteristics of the optical-to-electrical converted signals are investigated both by means of an external fast commercial photodiode and by using the saturable absorber section of the device as an intra-cavity photodetector. The power of the electrical RF signal, linewidth, and central frequency stability are measured in unperturbed condition and under the effect of optical feedback. Measurements demonstrate the deterioration of the electrical properties of the signal as the optical feedback level is increased. The reduction of the stability region for mode-locking operation is also reported.

A semiconductor laser coupled to a gallium-made non linear mirror may exhibit pulse regime. In order to better understand this coupled cavity, stationary solutions and dynamics are described following the standard Lang and Kobayashi equations for a semiconductor laser submitted to nonlinear optical feedback. It is shown that the nonlinearity distorts the ellipse on which lied the stationary solutions, with a "higher" part corresponding to lower reflectivity and a "lower" part to higher reflectivity. Bifurcation diagrams and nonlinear analysis are presented while the conditions for pulsed operation are discussed.

Tunable laser diodes exploiting the free-carrier plasma-effect are renown for their large tuning range. To control the emission wavelength, carriers are injected into a tuning region inducing a refractive index change. Typically, the material system GaInAsP/InP is used for tuning regions. Since this material shows strong recombination, a large current has to be applied while tuning. Unfortunately, the tuning current causes a parasitic temperature increase of the device acting negatively in a twofold way. Firstly, the index change due to the temperature increase works directly against the index change that is brought about by the plasma-effect. Secondly, many parameters of the laser device depend critically on temperature. Therefore, a reduction of the current consumption would instantly improve all relevant device parameters. In this paper we propose a type-II superlattice consisting of Al_0.30Ga_0.17In_0.53As/Al_0.50Ga_0.50As_0.56Sb_0.44 as a tuning region. The staggered band alignment leads to a spatial separation of the electrons and holes. As a result, the recombination rate can be significantly reduced by over one order of magnitude, which in turn leads to an increase of the carrier density as a function of the current. An analysis shows that type-II superlattices can provide an equal tuning range with a reduced current consumption by a factor of six compared to conventional heterostructures.

Different design approaches and the corresponding fabrication technology of widely tunable lasers with vertically integrated Mach-Zehnder interferometer (VMZ) have been investigated with respect to the spectral selectivity and tuning performance. Calulations show for designs with InGaAsP bulk material as active region that a tuning range of 75nm and a side-mode suppression ratio (SSR) of more than 30dB are feasible. The tuning range can be further extended using multiple quantum well heterostructures as active regions. They enable tuning ranges covering the whole material gain spectrum, but with a reduced SSR. Due to the codirectional mode coupling of the laser, the use of an appropriate facet coating allows an enlargement of the SSR to more than 30dB even for the large tuning range designs. We present the different design concepts and discuss the theoretical data as well as the first experimental results of the corresponding VMZ-laser devices. The measured spectrum of the laser shows an SSR close to the theoretically predicted value.

A diode laser based on an electrically pumped InGaP/GaAs/InGaAs two-quantum-well heterostructure with cavity completed by dynamic holographic gratings was created. The temporal dynamics, frequency spectra and spatial diagrams of lasing beams were examined.

Single-frequency grating outcoupled surface emitting (GSE) semiconductor lasers emitting at 1310 nm with output powers exceeding 5.25 mW into a multi-mode fiber, threshold currents below 13 mA and with > 30 dB side-mode suppression ratios are reported. These lasers consist of a 400 μm long horizontal cavity, and a 15 μm long second-order outcoupler grating sandwiched between 200 μm long first-order distributed Bragg reflector (DBRs) gratings. Higher output powers can be achieved with longer outcoupler gratings. These GSE lasers operate at 3.125 Gbps and have a full-width at half-maximum (FWHM) beam divergence of 5 x 12 degrees.

Short-wavelength GaInP/AlGaInP quantum-well (QW) laser diodes emitting in the 618-650 nm range at room temperature have been fabricated and characterized. Several variations in laser structures have been tested, including changes in QW composition, thickness, strain and number; changes in the barrier/waveguide composition and thickness; changes in cladding structure; use of multi-quantum-barriers and changes in the doping profile. The experiments showed that the threshold current characteristic temperature (T₀) increases with the number of QWs and is higher for compressive strain. The use of graded-index (GRIN) waveguides and higher p-cladding doping induced both a reduction in threshold current density and an increase in T₀, mostly at shorter wavelengths. Waveguide thickness optimization can be carried out, for both constant composition and GRIN waveguides, using the QW optical confinement as a first-approximation optimization criterion. Modified cladding structures reduced the vertical far-field full-width-at-half-maximum below 20° without significantly affecting the threshold current. Devices designed using some of the guidelines resulted from our study achieved, with different structures and under different operating conditions, performances like emitting more than 2W at 650 nm in continuous wave operation or lasing down to 618 nm at room temperature, which is among the shortest wavelengths from lasers grown by solid-source molecular-beam-epitaxy.

The linewidth enhancement factors of lattice-matched 1.5 μm wavelength InGaNAs/GaAs and InGaAs/InP single-quantum-well structures have been calculated using microscopic theory including many-body effects and a 10x10 effective-mass Hamiltonian. For applications which require high gain and carrier densities, InGaNAs/GaAs quantum wells have a much lower linewidth enhancement factor over a temperature range 300-400 K than InGaAs. The linewidth enhancement factor of InGaNAs is almost independent of both carrier density and temperature compared with InGaAs. The small-signal modulation characteristics of these 1.5μm lattice-matched structures and their temperature dependence have also been calculated. It is found that the maximum bandwidth of the InGaNAs/GaAs quantum well lasers is about 2.3 times larger than that of the InGaAs/InP quantum well lasers due to the high differential gain. The slope efficiency for the 3dB bandwidth as a function of optical density is twice as large for InGaNAs/GaAs as for InGaAs/InP quantum well lasers.

GaInNAs quantum well lasers have attracted significant interest in recent years. Their potential for operation at high temperatures without coolers and their application for low cost vertical-cavity surface-emitting lasers (VCSELs) are the main reasons for this interest. The main consequence of adding Nitrogen (N) to InGaAs materials is the band gap shrinkage. The reason for that is the interaction of N (acting as a localized defect) with the conduction band of the InGaAs. In previous studies, low temperature PL measurements of the impact of Nitrogen on the band structure of GaIn NAs have been examined. Pulsed measurements using a broad area GaInNAs QW laser were carried out and the results were analyzed in terms of the interaction of the N defect state with the GaInAs conduction band edge (band-anticrossing model). A detailed experimental temperature study of single quantum-well GaInNAs lasers at room temperature and above has been carried out. Experimental results of L-I, T₀, temperature dependence of lasing wavelength, optical gain and efficiencies are presented, discussed and compared with other materials. The temperature ranges studied is appropriate for most network applications. The gain spectra for moderate densities were experimentally measured using the method of Hakki and Paoli: the 600 μm long devise is biased below threshold and the gain is evaluated form the Fabry Perot modulation of the spontaneous emission spectra. A new concept will be introduced to study the bandwidth of the spectral gain and see its dependence with the temperature. The half-peak-BW will be the bandwidth where the gain decreases 50% from the peak gain. The temperature performance of the half-peak-BW has been studied obtaining a slope of 0.5871 nm/K. About the temperature dependence of the laser, a value of T_o (50 K) similar than the one found in InGaAsP has been found. This might disagree with the first results published of this new material system, giving extremely high values above 100 K. This is due to the high A parameter found in the previous materials. The improvement of the material is decreasing the A parameter and the characteristic temperature of the device. A small temperature dependence of the lasing wavelength was found (0.37 nm/K). This value was confirmed measuring the temperature dependence of the gain peak wavelength. This small temperature dependence can be understood by the interaction of the N state with the conduction band edge.

Optically-pumped semiconductor disk lasers offer high output power in combination with good beam quality. By optimizing epitaxial quality as well as thermal resistance, we have demonstrated more than 8W of continuous-wave, room-temperature emission at 1000nm. These high power-levels are tied to high optical-conversion efficiencies of more than 40%. Whereas available wavelengths for solid-state disk lasers are restricted to a set of atomic transitions, semiconductor disk lasers can be conveniently tailored to meet almost any wavelength. Building upon the high-power results at 1000nm, we have extended the emission range towards 900nm as well as 1100nm. Two prominent examples are devices realized at 920nm and 1040nm, in each case demonstrating several Watts of laser output.

Semiconductor lasers with high beam quality and high optical output power are very attractive for a variety of applications such as molecular spectroscopy, fiber optic communication and frequency conversion. In the used power regime, devices based on tapered gain sections are the most promising candidates to reach these demands. However, two disadvantages of the tapered laser concept are the reduced output power provoked by their additional resonator losses and the astigmatism of these diode lasers. In case of high brightness diode lasers it is important to discuss the methods needed for an advanced output power also from the point of view of beam quality. The knowledge about astigmatism is essential for designing micro-optics. For the experimental results low modal gain, single quantum well InGaAs/AlGaAs devices emitting at 980 nm were grown by molecular beam epitaxy. The influence of the thermal resistance and of the tapered section length on the output power as well as on the beam quality has been investigated. In addition the impact of these parameters on the astigmatism of tapered diode lasers has been analysed. The experimental results have been correlated with simulations of the current-power curves and BPM simulations of the nearfield behaviour.

Many industrial applications require high power semiconductor laser sources emitting beams of good quality. However, the emission of a free running high-power broad-area semiconductor laser contains many lateral modes that explains its poor beam quality and low brightness. One of the techniques to improve beam quality consists in placing a broad-area laser diode, used as a pure optical amplifier, in an extended cavity. Such a technique has proved its efficiency to produce a nearly diffraction limited beam at least for low pumping current level. For higher pumping currents, its spatial quality is deteriorated by the oscillation of higher order extended cavity modes. Using numerical simulations, we demonstrate that the insertion of a photorefractive crystal inside a broad-area laser diode extended cavity should extend the laser operating single mode range.

A Novel structure of high power dual-wavelength semiconductor laser diode is proposed and fabricated. Two laser structures are cascaded by a high doping tunnel junction during the epitaxial growth. The lasers can emit at wavelength of 951nm and 987nm at the same time. Without facet coating, the output power of the dual-wavelength laser is as high as 3.1W at 3A. And the slope efficiency of these devices is about 1.21A/W. Much higher output power can be reached for those dual-wavelength lasers when modifying the structure. The external differential quantum efficiency of different cavity length devices is analyzed.

We study the behaviour of a semiconductor laser subject to phase-conjugate feedback when the interaction time of the phase-conjugating mirror changes. With continuation techniques we present two-parameter bifurcation diagrams in the plane of feedback strength versus pump current, which change qualitatively as the interaction time of the mirror is increased. This reveals that for small interaction times the assumption of instantaneous feedback is justified. On the other hand, increasingly larger interaction times lead to considerable changes in the locking region. By investigating how curves of Hopf bifurcations change with the interaction time, we show how more complicated, chaotic dynamics become suppressed. One-parameter bifurcation diagrams as a function of the pump current, obtained by simulation, complement the continuation analysis.

The rate equations describing a laser with phase conjugate feedback are analyzed in the case of non-zero detuning. For low feedback rates and detuning, the stability diagram of the steady state is similar to the laser subject to injection. A stable steady state may loose its stability through a Hopf bifurcation exhibiting a frequency close to the relaxation oscillation frequency of the solitary laser. We also construct time-periodic pulsating intensity solutions exhibiting frequencies close to an integer multiple of the external cavity frequency. These solutions have been found numerically for the zero detuning case and play an important role in the bifurcation diagram.

Semiconductor lasers subject to filtered feedback are examined in the limit of a narrow filter. Under particular conditions, the laser dynamical equations reduce to the equations of a laser subject to an externally injected signal. The conditions for steady state locking and Hopf bifurcation can be determined analytically allowing a deeper understanding of the role of the laser parameters on the perturbed semiconductor laser dynamics. The success of our analysis motivates the study of other feedback cases such as the laser subject to slow phase-conjugated feedback.

The stability properties of an external cavity laser with strong grating-filtered optical feedback to an anti-reflection coated facet are studied with a general frequency domain model. The model takes into account non-linear effects like four wave mixing and gain compression. A small-signal analysis in the frequency domain allows a calculation of the range of operation without mode hopping around the grating reflectivity peak. This region should be as large as possible for proper operation of the tunable laser source. The analysis shows the stabilizing effect of mode coupling and of gain compression on the lasing mode. An integral equation for the electrical field is derived from the frequency domain model and used for time domain simulations of large-signal behavior.

Semiconductor external cavity lasers (ECL) have a wide range of applications in the field of DWDM and measurement systems. One of their most important features is the continuous tuning without mode hopping in a wide wavelength range. In this paper we present a modelling approach for an ECL in Littman-Metcalf configuration carried out for optimising: 1) the laser diode position inside the cavity in order to maximize the range of continuous wavelength tuning without mode hopping and without cavity-length adjustment and 2) the choice of the detuning of the operating wavelength respect to the Bragg condition in order to minimize the four-wave mixing (FWM) effects and the effect of a non-perfect antireflection coating (ARC). A realistic example has been analyzed and therefore we considered: the wavelength dependence of the modal gain, linewidth enhancement factor and grating selectivity, as well as the modal refractive index change with carrier injection, operating wavelength and temperature. The implemented numerical tools allow also to obtain some specifications on the grating selectivity and the ARC design.

It is shown that the nonlinear dynamics of semiconductor lasers with electro-optical feedback can be reconstructed using a new type of neural network with two modules: one for non-feedback part with input data delayed by the embedding time, and a second one for the feedback part with input data delayed by the feedback time. The delay time is obtained by using a simple method based on the forecast error of a linear model. Two values of the feedback strength have been considered that yield nonlinear feedback functions with 1 and 3 extremes. We have found that the complexity of the neural network model required to reconstruct nonlinear dynamics increases with the feedback strength, but not with the delay time. Therefore, the confidentiality level of systems with electro-optical feedback increases with the feedback strength. Chaos synchronization of the simulation data with data obtained from the approximated neural system is obtained when both systems are coupled in a diffusive way. Synchronization with the neural network model is used to extract by substraction a message added to the chaotic carrier.

MBE growth of high quality diluted Nitride materials have been investigated. Photoluminescence intensity of high nitrogen content InGaAsN/GaAs SQW can be improved significantly by decreasing the growth temperature due to suppressd phase separation of InGaAsN alloy. The longest room temperature PL peak wavelength obtained in this study is 1.59 μm by increasing the nitrogen composition up to 5.3%. High performance ridge-waveguide InGaAsN/GaAs single quantum well lasers at wavelength 1.3 μm have been demonstrated. Threshold current density of 0.57 KA/cm² was achieved for the lasers with a 3-μm ridge width and a 2-mm cavity length. Slope efficiencies of 0.67 W/A was obtained with 1 mm cavity length. The cw kink-free output power of wavelength 1.3 μm single lateral mode laser is more than 200 mW, and the maximum total wallplug efficiency of 29% was obtained. Furthermore, monolithic MBE-grown vertical cavity surface emitting lasers (VCSELs) on GaAs substrate with an active region based on InGaAsN/GaAs double quantum wells emitting at 1304 nm with record threshold current density below 2 KA/cm² also have been demonstrated. The CW output power exceeds 1 mW with an initial slope efficiency of 0.15 W/A. Such low threshold current density indicates the high quality of InGaAsN/GaAs QW active region.

Cavity solitons are stationary self-organized bright intensity peaks which form over a homogeneous background in the section of broad area radiation beams. They are generated by shining a writing/erasing laser pulse into a nonlinear optical cavity, driven by a holding beam. The ability to control their location and their motion by introducing phase or amplitude gradients in the holding beam makes them interesting as mobile pixels for all-optical processing units. We show the generation of a number of cavity solitons in broad area vertical cavity semiconductor microresonators electrically pumped above transparency but slightly below threshold. The observed spots can be written, erased and manipulated as independent objects. We analyze experimentally the cavity solitons domain of existence in the parameter space and how their characteristics are affected by inhomogeneities and impurities of the vertical cavity devices. A theoretical model, keeping into account the devices characteristics, reproduces numerically the experimental observations with good agreement.

CSs have been theoretically predicted and recently experimentally demonstrated in broad area, vertical cavity, driven semiconductor lasers (VCSELs) slightly below the lasing threshold. Above threshold, the simple adiabatic elimination of the polarization variable is not correct, leading to oscillatory instabilities with a spuriously high critical wave-number. To achieve real insight on the complete dynamical problem, we study here the complete system of equations and find regimes where a Hopf instability, typical of lasers above threshold, affects the lower intensity branch of the homogeneous steady state, while the higher intensity branch is unstable due to a Turing instability. Numerical results obtained by direct integration of the dynamical equations show that writable/erasable CSs are possible in this regime, sitting on unstable background.

The high-speed modulation dynamics of semiconductor lasers is determined by a complex interplay of ultrafast carrier and light field dynamics. The characteristic time scales of the underlying physical processes determine the relaxation oscillations and set an upper limit to the modulation of a single-mode semiconductor laser. In spatially extended semiconductor lasers the longitudinal and transverse dimensions enable the coexistence of numerous longitudinal and transverse modes. A suitable design of the laser cavity and electronic contacts should consequently allow one to directly influence the lateral coupling and transverse mode dynamics. The twin-stripe semiconductor laser (realized with two parallel contacts on top of the active area) represents one of the simplest semiconductor lasers with coexisting transverse modes. Modulation of the current in the stripes with a beat frequency corresponding to the frequency separation of transverse modes may then lead to a significant increase of the high-frequency modulation response of the laser. In this paper, we present results of simulations on the modulation characteristics of twin-stripe semiconductor lasers on the basis of multi-mode Maxwell Bloch equations that include propagation effects and spatio-temporally varying mode competition. In particular, we analyze the dependence of light field dynamics and spectral properties on laser dimensions, carrier injection and modulation frequency. Our simulations reveal that it is both, the transverse and the longitudinal degree of freedom that influence the transverse mode dynamics as well as the laser response to high-frequency modulation.

We theoretically investigate the dynamical properties of a system of two semiconductor lasers that are mutually coupled via their optical fields. An intrinsic feature of the coupling is its time delay which generically arises from the finite propagation time of the light form one laser to the other. In our system the coupling time is in the sub-ns range, which is of the same order of magnitude as the period of laser's internal relaxation oscillations. We model this system with Lang-Kobayashi-type rate equations where we account for the mutual coupling of the two lasers by a delay term. The resulting set of nonlinear delay differential equations is analyzed by using recently developed numerical continuation. We consider the case of two nearly identical lasers with symmetrical coupling conditions but different frequencies, and present an analysis of the coupled laser modes (CLMs) of the system.

A method for the investigation of the dynamics of two semiconductor lasers, grown side-by-side on the same wafer to enhance the lateral optical coupling, is presented. Using steady state analysis, parameter regimes of relevant dynamics are identified. This is completed by a spectral analysis, were two routes to chaos are implicated. Finally, we confirm the calculations by showing an avoided crossing type of behavior for the coupling strength.

We investigate correlations of the intensity fluctuations of two-dimensional arrays of non-identical, locally-coupled lasers, numerically and experimentally. We find evidence of a power-law dependence of spatial correlations as a function of laser pair distance (or coupling strength) near the phase-locking threshold.

Nonlinear delayed dynamics was first proposed in Optics by Kensuke Ikeda in 1979. Since then, many different setups based on similar dynamical principles were carried out experimentally, first to explore the numerous and various behaviours, and then to use the high complexity chaotic regimes for optical data encryption. After a brief review of the different setups and principles, we will report on 4 different optoelectronics realizations developed in our group, emphasizing on the characteric properties of each setup, and their implementation in chaos-based secure communication systems.

A detailed investigation of the decoding properties of different receiver configurations in an all-optical chaotic transmission system is presented for two data-encoding techniques and for various dispersion compensation maps. A semiconductor laser subjected to optical feedback generates the chaotic carrier while data is encoded either by Chaotic Modulation (CM) or Chaotic Shift Keying (CSK) methods. The complete transmission module consists of various dispersion management maps, in-line amplifiers and Gaussian optical filters. The receiver, employing a high facet reflectivity laser, is either forming a closed-loop configuration operating at the non-amplification regime or a strongly injected open-loop one. For the latter configuration the possibility of utilizing an anti-reflection (AR) coated laser is also investigated. System's performance is numerically tested by calculating the Q-factor of the eye diagram of the 1 Gb/s received data. The influence of the optical power launched into fibre or the transmission distance to the quality of the decoded message has been investigated. The closed-loop scheme had better performance relative to the open-loop, while CSK method and maps utilizing Dispersion Shifted Fibres are superior to CM and that employing Dispersion Compensating Fibres respectively. When an AR-coated laser is used in the open-loop receiver setup, improved decoding performance occurs.

The performance of an all-optical closed-loop chaotic communication system in a transmission link consisting of single mode fibers (SMF) applying two different dispersion management techniques is numerically studied. The first technique is implemented by the usage of dispersion compensating fibers (DCFs), while the second utilizes optical phase conjugators (OPCs). The latter is implemented by means of four wave mixing (FWM) in a dispersion shifted fiber (DSF), where the chaotic carrier corresponds to the signal wave and a high power continuous wave corresponds to the pump wave. Calculation of the recovered message Q-factor values obtained from the corresponding eye diagrams has been carried out applying chaotic modulation (CM) and chaos shift keying (CSK) encryption techniques for two repetition rates (2.4Gbps, 5Gbps). It is shown that the optical phase conjugation is an effective dispersion and non-linear effects compensation technique even if high-bit rate message encoding is applied. The superiority of a transmission system including OPCs to that utilizing (DCFs) is presented. The influence of key system parameters such as optical power, OPC spacing, pump power level, etc. to the transmission performance has been investigated. Acceptable system performance is presented for approximately 600Km at 2.4Gbps and 400Km at 5Gbps.

In this manuscript we analyze the modal dynamics of multimode semiconductor quantum-well lasers. Modal switching is the dominant feature of the devices analyzed and it obeys a highly organized antiphase dynamics which leads to an almost constant total intensity output. For each active mode a regular switching at frequencies of few MHz is observed. The activation order of the modes follows a well defined sequence starting from the lowest wavelength (bluest) mode to the highest wavelength (reddest) mode, then the sequence starts again from the bluest mode. Using a multimode theoretical model and a simpler phenomenological model we identify that four wave mixing is the dominant mechanism at the origin of the observed dynamics. The asymmetry of the susceptibility function of semiconductor materials allows to explain the optical frequency sequence.

Low Frequency Fluctuations (LFF) are defined by an abrupt (1 ns) drop-out of the emitted power followed by a gradual (50 ns) build-up of the power until the next drop-out event, when the laser with feedback is biased close to threshold. In this paper experimental and theoretical results on a vertical-cavity surface-emitting laser (VCSEL) with polarized optical feedback are presented. Experimentally, we observe single-mode low frequency dynamics when the VCSEL is biased below the solitary laser threshold. We can choose one of the two typical polarization modes (PM) of the VCSEL to be lasing, by an adequate choice of the polarization direction in the external cavity. Our theoretical analysis is based on a model developed by Loiko et al. which is an extension of the Spin-Flip model. We confirm the appearance of single-mode LFF and also reproduce the response of the orthogonal polarization mode above the solitary laser threshold, both deterministically and in presence of noise. This analysis shows that aiming the feedback at the passive mode (in absence of feedback) forces the active mode to react with short pulses, due to parasitic carrier theft, while targeting the feedback at the active mode induces a smaller response from the orthogonal polarization mode. This difference in response allows us to conclude that the secondary polarization does not play an essential role in the LFF dynamics.

We present a study of the time-scale at which current induced polarization switching (PS) in VCSELs takes place. To this end, we measure the step and frequency response in three different types of PS VCSELs, showing that the dominating time-scales differ strongly from one VCSEL structure to another. We characterize the current-driven polarization modulation frequency response by measuring the critical modulation amplitude necessary to steadily force PS back and forth across the PS point as a function of the modulation frequency. The polarization step response is obtained by measuring the stochastic properties of the delay between the applied current step and the resulting change in the polarization, for various values of the initial and final current. For the studied proton-implanted VCSEL the polarization response is characterized by the thermal relaxation time. The measured polarization response of the air-post VCSEL also shows a clear signature of thermal effects, however PS is not at all inhibited at higher frequencies. In the oxide-confined device studied, there seems to be no thermal influence on the PS at all. Comparing the frequency response and the step response measurements done on the same device leads to similar conclusions and allows us to crosscheck our results. In all cases, we are able to reproduce our experimental findings using a rate-equation model, where PS is supposed to be induced by changes in the gain balance between the two polarization modes.

We present results of comprehensive investigations of the intensity noise and the angular-resolved spectral emission characteristics of resonant-cavity light-emitting diodes (RCLEDs), demonstrating an interesting interplay between these two properties. First, we find that the intensity noise of the investigated RCLEDs, detected within a full solid angle of detection, is up to -0.15 dB below the shot noise in a quite large pumping regime, i.e., we demonstrate the successful generation of squeezed states of light with these optoelectronic devices. Second, we investigate the spectral and angular emission characteristics and find that the cavity-like character of the Bragg mirrors and the quantum well active medium manifests itself in a blue shift of the central emission wavelength from 847 nm at zero degree to 825 nm at an emission angle of sixty degree. By varying the temperature we are able to detune the quantum-well emission wavelength and the cavity resonance wavelength and observe a broader angular intensity profile. Third, we measure the angular resolved intensity noise. Its super-shot noise behavior can be explained by anticorrelations between radial components of the output intensity emitted at different angles. Finally, the possible origin of the observed anticorrelations in the angular-resolved intensity noise, as well as possibilities for future trends, applications and the limitations of these non-classical states of light with respect to sensing and spectroscopic applications are discussed.

We present 2D-spatially and temporally resolved measurements of the emission of selectively oxidized state-of-the-art VCSELs of intermediate aperture sizes, showing their rich modal, polarization, spectral, and spatial dynamics. For the characterization of the repetitive dynamics we performed experiments with a technique we developed called TRIDA (Temporally Resolved Imaging by Differential Analysis) providing a temporal resolution down to 10 ps. From the temporal evolution of the near-field emission we provide a comprehensive overview of the phenomena in the turn-on dynamics. The evolution of the spatial intensity distribution of the transverse modes over a range of some nanoseconds is presented and discussed. Complementary, we have taken single-shot images of the near-field emission of 3ns short pulses in order to characterize the nonrepetitive part of the dynamics and to get a deeper understanding of the dynamical processes arising from the interaction of transverse modes in the turn-on process. We find strong variations in the intensity distribution among the transverse modes and the polarization directions, indicating the onset of spatio-temporal chaos. A correlation analysis for the modal intensities is performed showing the influence of the competition for the available spatial and spectral gain.

We have developed a theory of the nonlinear refractive index in Quantum Dot (QD) Semiconductor Optical Amplifiers (SOAs) due to Spectral Hole Burning (SHB). Estimates show that this SHB nonlinear refractive index can be of order of 4x10^-16 m²/W that is by four orders higher than the nonlinear refractive index in silica, and offers the possibility of an efficient ultrafast Cross-Phase-Modulation (XPM) in QD SOAs. The opportunity of XPM without patterning effects via this refractive index nonlinearity is discussed. The Pattern-Effect-Free (PEF) XPM is possible in QD SOAs at high pumps, when maximal (constant) gain is achieved in SOAs, and the linear and nonlinear refractive indices also become independent of the total carrier density in the QD structure. In whole, use of the ultrafast refractive index nonlinearity in the regime of maximum gain in QD SOAs can lead to the development of a new generation of nonlinear interferometers for ultrafast optical switching.

The dynamics of ultrashort pulses propagating in a quantum dot amplifier is determined by a complex nonlinear coupling and dynamic interplay of light fields and carriers in the spatially inhomogeneous quantum dot ensemble. Computational modeling shows that in spite of the large complexity the strong localization of the carrier inversion and the low amplitude phase coupling may allow the amplification and transmission of ultrahort light pulses with minimum deterioration of the pulse properties (e.g. pulse shape, duration). The theoretical description is based on spatially resolved Quantum Dot Maxwell-Bloch equations that describe the spatio-temporal light field and inter-/intra-level carrier dynamics in each quantum dot of a typical quantum dot ensemble. In particular, this includes spontaneous luminescence, counterpropagation of amplified spontaneous emission and induced recombination as well as carrier diffusion in the wetting layer of the laser. Intradot scattering via emission and absorption of phonons, as well as the scattering with the carriers and phonons of the surrounding wetting layer are dynamically included on a mesoscopic level. Spatial fluctuations in size and energy levels of the quantum dots and irregularities in the spatial distribution of the quantum dots in the active layer are simulated via statistical methods. Simulation results of the nonlinear pulse propagation in quantum dot optical amplifiers allow visualization and interpretation of fundamental nonlinear processes such as selective depletion and re-filling of quantum dot energy levels leading to a complex gain and index dynamics that affect the amplitude and phase of a propagating light pulse. Computational modelling thus may lay the foundation for an optimization and tayloring of pulse properties.

Semiconductor Optical Amplifiers (SOAs) are of central interest as multifunctional, easy-to-integrate components for the development of future optoelectronic systems. Their dependence upon the incoming light polarization is a well-known, but still debated, issue in the context of emerging optical telecommunication networks, fueling the need for a detailed polarimetric characterization of such structures. In this paper, we present what we believe to be the first systematic polarimetric analysis within the frame of the Mueller-Stokes formalism of an integrated InP/InGaAsP SOA around 1550 nm. The challenge stems from the amplifying, active, spectrally broadband and nonlinear nature of the component. For the sake of our study, we have developed a highly sensitive, free-space, polarimetric set-up, with the additional experimental challenge induced by the spatial constraints of a guided-wave device, most notably in terms of light injection. Physical phenomena (intrinsic noise contribution of internal sources, carrier saturation due in particular to Amplified Spontaneous Emission, modal birefringence for index and gain...) responsible for the polarization dependence of the amplification process are identified, and discussion of the data highlights the need for an extended matrix formalism taking explicitly internal sources into account.

A detailed numerical and experimental study of the additional RIN induced to the input waves by the FWM process, is presented. The nonlinear medium used in the experiments is a bulk Semiconductor Optical Amplifier (SOA). The measurements are carried out for different operating conditions (pumping level of the SOA, input power, input signals with different intensity noise characteristics, etc.). A complete numerical model is employed to simulate the FWM process, taking into account interaction of four waves in the SOA (two input waves and two product waves). The latter is used in order to obtain realistic behavior of the model, when operation at a wide range of input power is considered. The theoretical interpretation of the above results is based on the static transfer function of the FWM process where all waves interacting in the SOA are continuous waves (CW).

We analyse the sensitivity of quantum dot semiconductor lasers to optical feedback. While bulk and quantum well semiconductor lasers are usually extremely unstable when submitted to back reflection, quantum dot semiconductor lasers exhibit a reduced sensitivity. Using a rate equation approach, we show that this behaviour is the result of a relatively low but nonzero line-width enhancement factor and strongly damped relaxation oscillations.

An analysis of the transverse and longitudinal mode structure of broad area quantum dot lasers emitting at 1060 nm is presented. In particular, temperature is shown to play an important role in the stabilisation of the transverse mode structure of the devices. In addition, the investigation of the interaction between these transverse modes, through the measurement of the spatial intensity correlation, shows that the laser retains some modal properties in the unstable regime. Finally, measurements of spectral correlations between longitudinal mode groups display a strong dependency on their respective transverse mode structures indicating the importance of spatial overlap.

InP based quantum dash active materials have been recently presented in the literature for their promising emission characteristics in the 1.55 μm wavelength range and for their broadband emission spectrum, that makes them interesting candidates as active material of semiconductor optical amplifiers and widely tunable lasers. The effect of p-type modulation doping in quantum well and quantum dot semiconductor active materials has been extensively studied in the literature as a technological possibility to increase the differential gain and the modulation bandwidth of semiconductor lasers, but, to our knowledge, it has never been considered for quantum dash lasers. In this work the effect of p-type modulation doping in a quantum-dash active material is theoretically analyzed, using a quasi-equilibrium model that accounts for the carrier distribution in the dashes of the ensemble, in the wetting-layer and in the barriers. We will show that the carrier distribution in these states significantly influences the efficacy of p-type modulation doping.

Feeding light inside a semiconductor laser gives rise to a wealthy set of dynamics that have been largely described. The two fundamental parameters are the detuning and the injected power. We present here theoretical and experimental evidence that other parameters may play a role. For instance, injecting with an orthogonal polarization is not equivalent to nullify the effect of optical injection as demonstrated by our experimental map. We will focus the main part of the talk on the influence of coherence. Some part of this study has been made possible by the use of a single-frequency fibre laser which enables us to establish an enlightening comparison between these two different types of lasers and also to surround the influence of coherence.

In this paper we theoretically investigate the influence of the injection of a so-called holding beam - a CW beam at the transparency wavelength of the gain medium on the direct modulation properties of semiconductor lasers. Both the small-signal and the large-signal behaviour are investigated. The influence of the holding beam is studied using a simple rate equation model. From the solution of the rate equations it follows that the injection of the holding beam doesn't increase the modulation bandwidth or resonance frequency itself, but that it results in a significant increase of the damping of the relaxation oscillations. Furthermore, it is shown that the chirp (both the adiabatic and dynamic chirp) will be significantly reduced. A few numerical examples - calculated with our CLADISS software - are given to show that the modulation response at not too high bias levels can be significantly improved when a holding beam is applied. For direct, large signal modulation, the injection of a holding beam leads to significantly reduced transient effects and reduced chirping, even at low bias.

Picosecond-range single optical pulses with peak power in the range 10-100 W are fairly attractive for various practical applications. A laser diode structure has lately been suggested which produces powerful (~ 50 W) picosecond (~20 ps) optical pulses near the trailing edge of the current pulse by means of field-assisted gain control. Lasing onset is delayed in this diode by a few nanoseconds due to intendance-reduced pumping efficiency caused by the implementation of internal optical pumping. The ps operating mode is based on a compromise between the dynamics of carrier accumulation and of the transverse electric field, controlled by the efficiency of the internal optical pumping. The pumping efficiency is determined to a large extent by competition between stimulated and spontaneous radiative recombination at the source of optical pumping. An effect of the laser diode switching from the picosecond to the quasi-steady-state (ns) mode was observed when the length of the laser cavity was reduced from 400 μm to 200 μm. This phenomenon is studied and attributed to an increase in the fraction of spontaneous photons due to reduction in the density of the stimulated emission at the source of the optical pumping.

There is currently interest in using novel modulation formats for high bit-rate datacoms systems. 4-level modulation is an attractive method of halving the line-rate required for 40Gb/s systems. This 20GBaud line rate enables reduced bandwidth direct modulation of semiconductor lasers, thus reducing laser chirp, increasing transmission distances and also enabling simplified drive electronics to be used. In this experiment the 4-level signal is generated by electrically combining 2 de-correlated 20Gb/s data streams of differing amplitude from a pattern generator and then used to modulate a DFB laser. The directly modulated source is a DFB laser, emitting at 1310nm with a 3dB frequency response of 20GHz. This laser also has a very linear modulation response, with a spurious free dynamic range of over 100dBHz^2/3 at 25°C and over 90 dBHz^2/3 at 85°C. This highly linear behaviour is necessary to allow direct 4-level modulation source even at high temperature. The 40Gb/s 4-level signal is then transmitted along standard fibre and detected with an electrical receiver. In order to overcome the attenuation limited transmission distance of 20km a semiconductor optical amplifier, with a saturation power of 11dBm and fibre to fibre gain of 20dB, is used. The addition of an SOA enables transmission distances of 40km to be achieved with transmission penalties of as low as 2.6dB, even with the laser operating at 70°C. The robustness of the 4-level modulation is compared to NRZ and the impairments to both signals upon optical amplification are examined.

The performance of an external-cavity mode-locked semiconductor laser is analyzed theoretically and numerically. Passive mode-locking is described using a fully-distributed time-domain model including fast effects, spectral hole burning and carrier heating. We provide optimization rules in order to improve the mode-locking performance, such as reducing the pulsewidth and time-bandwidth product as much as possible. Timing jitter is determined by means of extensive numerical simulations of the model, demonstrating that an external modulation is required in order to maintain moderate timing-jitter and phase-noise levels at low frequencies. The effect of the driving conditions is investigated in order to achieve short pulses and low timing jitter. Our results are in qualitative agreement with reported experiments and predictions obtained from the master equation for mode-locking.

In this paper, we describe a low cost pulse generator based on bipolar transistors designed to emit very short light pulses with a simple laser diode. Its main characteristics are laser pulses as short as 60 ps Full Width at Half Maximum (FWHM) and a repetition rate from DC up to 100 MHz. The circuit is based on three standard RF bipolar transistors which emit a short electrical pulse of 1.2 ns FWHM with 5 Volt amplitude peak-peak into 50 Ω. This unit can directly drive a classical laser diode to emit short laser pulses. A DC current generator is added in order to polarize the laser diode near its emission point and reduce the power delivered by the pulse generator and enhance its life time. This generator allows the adjustment of the emitted power of the laser diode. If the power emitted is too high, secondary pulses are observed. There is an optimal power where there is a single short pulse with duration down to 60 ps FWHM. The system is simply triggered by a TTL signal and its structure allows the generation of a single shoot, or a complex pattern generation and repetitive generation. The unique limitation is the delay between pulse emissions which must be greater than 10 ns. The typical optical power with a standard laser diode at 660 nm (45 mW continuous) is 1.7 mW with a repetition rate of 80 MHz and a pulse width down to 60 ps measured with a synchroscan streak camera. The peak power is about 350 mW. The pulse to pulse jitter is less than 10 ps rms. This system can be a competitive alternative to expensive commercial products.

We present an experimental and theoretical study of optical feedback in a semiconductor laser for the case of an extremely short external cavity (EC) configuration. When the length of the EC is changed both the output power and the voltage drop onto the laser are modulated with a period of half of the solitary laser wavelength. We also perform modulation experiments in which the EC length is modulated with amplitude corresponding to the half of the solitary laser wavelength and with different signal shapes. In this way we prove that by using optical feedback we are able to detect very small features. Such detection is of general interest from an application point of view, e.g. for optical data readouts, resulting in a reduced number of optical components. Optical feedback also affects the frequency of the laser light and results in a longitudinal mode hopping. With increasing the EC length we observe mode hops between neighboring solitary laser modes followed by large jumps at the EC frequency splitting. These large EC mode hops can be exploited for broad band frequency tuning of the emitted light. We also study the dependence of the amplitude and the period of the EC mode hops on the EC length. We reveal the existence of a cut-off EC length of a few micrometers for which the amplitude of the EC mode hops reaches a maximum and then strongly decreases. We give a theoretical explanation of our experimental findings based on Fabry-Perot resonant condition for coupled cavities.

Single-mode high-power emission from a single VCSEL can be achieved by means of mode control in an external resonator. A problem arises how to analyse interference between wave field coming from external resonator to VCSEL structure. Analytical approach is formulated allowing for simplified description of the system. Numerical approach is based on 3-D bi-directional beam propagation method. Modal content and discrimination of higher-order modes in passive resonator are examined. Besides, above threshold operation of optical modes is simulated using multiple iterations. A method based on functions of Krylov's subspace, is developed to find a number of optical modes in a VCSEL with gain and index distributions established by the oscillating mode. In calculations, both Fourier and space variable descriptions of beam propagation are combined. This approach allows us to evaluate critical conditions for single-mode operation.

Laterally coupled laser-diode pairs, also called "twin stripe lasers", are coupled by means of the evanescent em-field in between two dielectric laser cavities. The coupling strength between the two lasers is determined by the decay length of this evanescent em-field. This length depends on the indices of refraction of the material which separates the two cavities compared to the cavity material. However the index of refraction of the cavity material also depends on the inversion density in the lasing material. Therefore the coupling coefficient in the rate equation for the twin stripe device can be written as κ₀[1 + ½(α_c(1) ρ₁ + α_c(2) ρ₂)], where ρ₁ and ρ₂ are the inversion densities in the two stripes. We give an expression for α_c>(i) in terms of the properties of the device.

Self-seeded, gain-switched operation of an InGaN multi-quantum-well laser diode has been demonstrated for the first time. An external cavity comprising Littrow geometry was implemented for spectral control of pulsed operation. The feedback was optimized by adjusting the external cavity length and the driving frequency of the laser. The generated pulses had a peak power in excess of 400mW, a pulse duration of 60ps, a spectral linewidth of 0.14nm and maximum side band suppression ratio of 20dB. It was tunable within the range of 3.6nm centered at a wavelength of 403nm.

We analyse the dynamics of a self-pulsating semiconductor laser with optical feedback. Without re-injection of light the laser displays periodic oscillations. At very weak feedback levels we observe an amplitude instability whose frequency increases with the feedback level until the laser enters the low frequency fluctuation regime commonly observed in cw lasers with optical feedback. We show that such behaviour can be observed within the framework of the Lang Kobayashi equations for self-pulsating semiconductor lasers.

The polarisation dynamics of vertical cavity surface emitting lasers (VCSELs) in the bistable regime is well described by Kramers theory for noise induced transitions. By employing feedback, a memory mechanism can be introduced, which make the dynamics of the system non-Markovian. Here we analyse theoretically and experimentally the residence time distribution of the bistable systems in the presence of noise and time-delayed feedback, using an opto-electronic feedback cycle for a VCSEL. We demonstrate and explain various non-exponential features of the residence time distribution using a continuous as well as a two-state model. Additionally we compare the results to an electronic Schmitt trigger, which represents an experimental realization of the two-state model.

The combination of high power, small linewidth and rapid tuneability is essential for many fields in high resolution spectroscopy. Furthermore these optical features are essential for laser-cooling techniques. Enhancement of high power lasers with excellent spectral and spatial quality is currently an important research subject. The requirements for a laser system applied in both fields of application are demanding: a mode-hop free tuning range of a few GHz, with a linewidth in the order of 1MHz and an output power of a few 100mW. We report a very compact external cavity diode laser system (ECDL) with an output power of up to 800mW with an almost Gaussian shaped beam quality (M²<1.2). The coupling efficiency for a single mode fibre exceeds 60%. The centre wavelength can be preadjusted within the tuning range of 20 nm. This laser operates single mode with a mode-hop free tuning range of up to 15GHz without current compensation and a side-mode-suppression better than 50dB at different wavelength between 730 and 1060nm. To demonstrate the suitability for neutral atom cooling we used this laser as light source in the production of a BEC of over a million ⁸⁷Rb atoms. In addition we approved this light source for high resolution spectroscopy, more precisely for the Cavity-Ring-Down-Spectroscopy (CRDS). Our ECDL was part of a MIR-light source which utilizes difference-frequency-generation in PPLN. At the wavelength of 3.3μm we were able to perform a high resolution absorption measurement of 50ppb Ethane. Both applications clearly demonstrate the suitability of this laser for high-precision measurements.

We investigate the threshold-crossing of a single-mode semiconductor laser, with special emphasis on its transfer function. Spontaneous emission, looked upon as the driving source of the radiation, is described in a semi-classical way in the spectral domain. The internal and emitted fields are filtered into the resonance modes of the whole structure: their spectral density are described by the generalized Airy-like transfer function, well approximated by a Lorentzian, which contains all essential mechanisms at work in a laser oscillator: Gain, losses and sources. The active zone is saturated through Amplified Spontaneous Emission, integrated over its whole spectral range. Continuously valid across threshold, the method enables one to derive in a simple way the main steady-state properties of the laser oscillation. Most specifically, we obtain analytical expressions, in normalized units, for emitted power, linewidth sharpening, carrier clamping and frequency shift, with the pumping rate as the only external parameter. In this approach, the optical properties of the active medium (the gain, the source and the refractive index) are supposed to be uniquely determined by the steady-state values of the carrier and photon density, obtained within the framework of the rate equation formalism and assumed uniform along the active zone.

In this work we present experimental results on the auto-correlation and cross-correlation properties of the two counterpropagating modes of a monolithic semiconductor ring laser. The ring laser can operate in a bidirectional regime where the two modes have equal power, and also in a unidirectional regime where one of the modes is almost suppressed. Auto-correlation measurements, that are carried out using an unbalanced Mach-Zehnder fiber interferometer, allows to determine the coherence length and linewidth of the ring laser. Cross-correlation measurements are carried out using a modifed interferometric set-up, and they reveal that the two counterpropagating modes are phase-locked.