Search

A practical realization of a diode pumped actively Q-switched Er3+-doped fluoride fiber laser operating near 2.78 μm is reported. For the repetition rate of 100 Hz stable 26 ns pulses with an energy of 330 μJ and peak power of 12.7 kW are demonstrated. The laser performance is tested against variations in pump power, repetition rate and active fiber length.

The properties of highly nonlinear glasses for photonic devices and the advantages of processing these materials using femtosecond laser pulses are discussed in a brief review. A novel approach is proposed for the optimization of the modification process that takes into account the dispersion of the nonlinear coefficients of refraction and absorption. Numerical modeling of the pulse energy deposition into a sample of chalcogenide glass shows that the shapes and dimensions of the modified regions depend on the nonlinear coefficients.

An alternating direction implicit (ADI), scheme for the wide-angle finite-difference beam propagation method (FD-BPM) based on the wide angled Pade multistep method is presented. The scheme incorporates an iterative technique for correction of the operator splitting error. The resulting equations are efficiently solved by the Thomas algorithm for tri-diagonal band matrices. The dispersion characteristics, accuracy and stability of the scheme is verified analytically and numerically for the cases of a plane wave propagating at fixed angle to the assumed propagation direction of the algorithm and the three dimensional angled propagation of a Gaussian beam. The computational requirements of the method are assessed against a standard wide-angle Pade multistep method that uses direct and iterative sparse matrix solvers.

This paper presents a new full-vectorial higher order finite-difference scheme for rectangular dielectric waveguides. The method incorporates new difference equations for mesh points that are next to a dielectric interface and points that are near a dielectric corner. Its performance is assessed by means of a number of test problems and compared with that of an existing higher order scheme.

This paper presents an efficient and physically consistent method for modeling lumped electroabsorption (EA) RF modulators. The method extends the original spectral index (SI) waveguide simulation method to the time domain in order to model the time-varying phenomena occurring in the multiple-quantum wells of these modulators. The accuracy of the methodology is first verified on the simpler cases of time-varying homogeneous media and time-varying slab waveguides. The time domain SI method is then used to model pulse generation and chirp phenomena in lumped EA modulators operating at both 10 and 40 GHz producing predictions in good agreement with previously published experimental values

Two practically important effects in magnetooptic photonic bandgap (MO PBG) structures are investigated for the first time, namely 1) oblique incidence and 2) diffractive effects caused by incident beams of finite width. A rigorous solution of the full three-dimensional wave equation in a MO medium is derived and used to evaluate the effect of beam spot size on the transmission properties of MO PBG structures. It is shown that the properties of MO PBG structures are highly sensitive to the angle of incidence and, furthermore, that in the case of a small beam size, plane-wave analysis significantly overestimates the transmission and isolation properties of MO PBGs.

The computationally efficient DuFort-Frankel beam-propagation method (BPM) is ideally suited to parallel computing. Although the scheme is conditionally stable for structures with lossless materials, in the presence of material loss it can become unstable. It is shown that the use of perfectly matched layer (PML) boundary conditions can also cause instability, especially in the three-dimensional (3-D) case. These instabilities are characterized and a stabilized DuFort-Frankel scheme is presented that extends the scope of this powerful method to cover these practically important scenarios.

The use of optimization methods to automatically synthesize novel structures is attractive for optoelectronic design. However, the application of such methods to practical components is hindered by the long run times required to evaluate the performance of each particular trial design. This paper investigates, for the first time, how systematically adapting the accuracy of the simulations during the optimization process can be used to speed up the synthesis of effective designs. Adaptive accuracy variation is achieved in two ways: altering the acceptable error demanded of a particular simulation method and using a hierarchy of different methods. This novel method of synthesis is applied to quickly obtain optimal rib waveguide structures for two practical problems, namely modal degeneracy and low-loss fiber coupling.

The structure-related (SR) finite-difference beam-propagation method is applied to integrated optical couplers employing curved waveguide sections. A general nonorthogonal coordinate system is introduced that matches the local structure geometry, yielding a more accurate analysis with coarser meshes than standard schemes, thus providing significant computational resource savings. Furthermore, the algorithm is easily interfaced with conventional schemes. Examples of curved two-dimensional and three-dimensional directional waveguide couplers are examined to show the advantages and generality of SR schemes as well as to explore the role of the extra degrees of freedom offered by nonorthogonal meshing.

Results from detailed numerical analyses of the modal characteristics of large-cross-section silicon-on-insulator-based rib waveguides are presented. They highlight for the first time that satisfying widely used design criteria is not sufficient to ensure single-mode behavior. In particular, the geometries that the design formulas predict should be single-mode are shown to support higher order vertical modes that do not couple (leak) into the outer slab region and are thus low loss in nature. Fortunately, a wide range of practical rib geometries still remains for which the leakage loss of modes other than those of EH/sub 00/ and HE/sub 00/ is sufficiently high to make the waveguides effectively single mode for each polarization.

Time-domain Volterra integral equations are obtained for the electromagnetic fields in longitudinal uniform dielectric waveguides with time-varying media in their core. Resolvent operators are constructed for the case when the core permittivity changes abruptly in time and new features are observed in the consequent field transients in comparison with those occurring in a simple unbounded space.

High-Q semiconductor microdisk resonators laterally coupled to optical waveguides serve as compact and robust components in wavelength-division-multiplexing systems. We present an efficient algorithm for the two-dimensional design of wavelength-selective filters based on waveguide-coupled whispering-gallery-mode (WGM) microresonators of arbitrary cross-sectional shape. A full-wave integral equation formulation enables us to accurately account for the coupling between the resonator and bus waveguide as well as for radiation and material losses. Furthermore, we apply this approach to the analysis of elliptic-resonator filters. Filter tuning by changing the resonator shape from circular to elliptic reduces the dependence of the coupling efficiency on the width of the air gap. Two alternative coupling designs are suggested for monolithic and hybrid integration technologies. One is based on increasing the coupling length, while the other exploits the field concentration at the increased-curvature portion of the deformed WGM resonator.

The application of the Volterra integral equation method for investigations of the interaction of electromagnetic fields with a one-dimensional time-varying dielectric layer is considered. The medium permittivity inside the layer may change in time owing to a nonlinearity as well as to extrinsic sources. Original software for the computer modeling of such an interaction is developed. Numerical results are presented and a comparison made between the accuracy and stability of integral equation method and the finite-difference time-domain (FDTD) method.

An efficient and accurate algorithm based on the combination of the Green's function method and an analytical regularization technique is applied to study the complex natural frequencies and Q factors of two-dimensional (2-D) dielectric resonators (DRs) in a layered environment. Frequency shift and degeneracy splitting of the whispering-gallery modes (WGMs) of circular-cylindrical and tubular DRs immersed into layered dielectric media are computed and analyzed.

The highly efficient free space radiation mode method is extended, enabling for the first time its use in determining the lossy modes of curved buried semiconductor waveguides. Both the real and imaginary parts of the propagation constant along with the corresponding field profiles, are found with an accuracy comparable to that of more computationally intensive methods.

Efficient spot-size convertors are an essential component of modern integrated optoelectronics, providing a simple and reliable interface with fibers. One recently proposed design functions by forcing the field from a small spot rib waveguide into a large spot one by using a taper. Accurate models of these structures, and a clear understanding of the coupling mechanisms they rely upon, are essential if optimized designs are to be produced. Furthermore, it is necessary to develop analysis techniques which are accurate yet not computationally intensive, suitable for use within an iterative design environment. Here, the well known and proven spectral index method is extended to efficiently analyze this class of vertically coupled rib waveguide spot-size transformer, providing both the guided modes and for the first time, the substrate radiation modes of the structure. These modes can then subsequently be used to design and determine the performance of the full tapered structure.

The use of an oblique coordinate system with the finite difference beam propagation method has previously been demonstrated to offer significant computational advantages over using rectangular coordinates for a wide range of practical optical structures. The effects of finite mesh resolution, step size, and order of the algorithm in terms of numerical dispersion and dissipation are here investigated and quantified for the first time.

Novel beam propagation algorithms for tapers have been developed that eliminate errors caused by staircase approximations. Nonorthogonal coordinate systems are introduced that model the effects of beam spreading and obliqueness while retaining the standard coordinate z to advance the solution forward uniformly. Both two- and three-dimensional (2-D) and (3-D) examples are considered to illustrate both the accuracy and utility of the scheme. The new method maintains its accuracy with fewer sample points allowing significant reduction in calculation time.

A nonlinear iterative (NLI) method, originally developed by Hewson-Browne in geomagnetism, is applied to the vectorial analysis of optical waveguides. The method explicitly shows the interrelations between the scalar, polarized and vectorial operators and can be conveniently implemented using finite difference methods. Excellent accuracy in the normalized propagation constant is claimed along with agreement with earlier work and field distributions. The method presented enables vector results to be obtained simply by using computer programs, often available, for either polarized or scalar modes. This procedure proves more efficient than a standard vectorial finite difference technique.

A novel finite difference beam propagation method (BPM) analysis based upon a bi-oblique coordinate representation is presented. The analysis yields a computationally efficient algorithm that permits accurate simulation of a wide variety of structures to be made without staircase approximations. Further, the independent paraxial and wide angle approximations, centered upon the direction of optical field propagation in each branching waveguide, may be incorporated into the algorithm. This feature allows a low-order approximation to be used in the algorithm even if one or both of the branching angles is large. The approach has been applied to both symmetric and asymmetric optical waveguide Y-junctions and produces results in excellent agreement with those in the literature.

Bent-guide structures will undoubtedly play an important role in increasing packing density in future integrated optical circuits. A method for forming bends in waveguides in III-V semiconductors by reflection off an etched vertical wall is proposed for providing wave-guiding through 90° with negligible loss.