IEE Proceedings - Optoelectronics
Volume 149, Issue 4, August 2002
Volumes & issues:
Volume 149, Issue 4
August 2002
-
- Author(s): Joachim Piprek
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, page: 121 –121
- DOI: 10.1049/ip-opt:20020665
- Type: Article
- + Show details - Hide details
-
p.
121
(1)
- Author(s): M. Wiedenhaus ; A. Ahland ; D. Schulz ; E. Voges
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 122 –130
- DOI: 10.1049/ip-opt:20020529
- Type: Article
- + Show details - Hide details
-
p.
122
–130
(9)
A synopsis of different methods of modelling electroabsorption modulators based on multiple quantum wells (MQW) is given. It contains two algorithms which are based on the density matrix formalism. A first estimation of the spectral absorption of the quantum wells is gained by expanding the excitonic states in terms of sub-band states. Another efficient and more appropriate model is the direct solution of the density matrix using a perfectly matched layer (PML) absorber to limit the calculation domain. A drift–diffusion model self-consistently corrected by the Bohm potential describes the quantum transport properties. The exciton equation and the transport model are iteratively coupled and, thus, account for nonlinear, carrier-dependent effects. The model includes bound and unbound states, avoids the Kramers–Kronig relation and considers important effects contributing to electroabsorption, such as the Franz–Keldysh effect, the Wannier–Stark effect and the quantum-confined Stark effect. The time-dependent problem is solved by implicit, rather than by explicit algorithms as the latter demand very restrictive stability conditions. - Author(s): K.J. Williams
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 131 –137
- DOI: 10.1049/ip-opt:20020504
- Type: Article
- + Show details - Hide details
-
p.
131
–137
(7)
Space-charge and thermal calculations are used to compare the performance of gigahertz-bandwidth, high-current p-i-n photodiodes utilising InGaAs absorbers and InGaAs/ InP drift layers. By varying the percentage of InGaAs and InP in the drift layer, comparisons can be made between traditional p-i-n photodiodes (100% InGaAs drift layer), dual-depletion region photodiodes (part InGaAs, part InP drift layer), and uni-travelling-carrier (100% InP drift layer) photodiodes. A local maximum in the maximum photocurrent occurs when a charge balance is present in the depletion layer. The results from space-charge and thermal calculations show that traditional p-i-n and dual-depletion region designs can potentially provide performance that exceeds that of the uni-travelling-carrier design. Charge compensation is proposed as a means of increasing the space-charge-limited photocurrent and for controlling the peak electric fields within the depletion region. - Author(s): H.S. Djie ; S.L. Ng ; O. Gunawan ; P. Dowd ; V. Aimez ; J. Beauvais ; J. Beerens
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 138 –144
- DOI: 10.1049/ip-opt:20020528
- Type: Article
- + Show details - Hide details
-
p.
138
–144
(7)
The interdiffusion effect on the strain build-up and refractive index profile of lattice-matched InGaAs/InGaAsP multiple quantum wells is reported. Interdiffusion is achieved experimentally using low energy (360 keV) arsenic or phosphorus ion-implantation-induced disordering, followed by an annealing step. A model of the interdiffusion process has been developed to analyse the effect of different interdiffusion ratios on the waveguide's polarisation behaviour through the strain build-up and the refractive index profiles for the transverse electric and transverse magnetic modes. Polarisation-resolved photocurrent absorption measurements of quantum-well waveguide structures have shown that sufficiently high ion implantation doses can lead to the realisation of polarisation-insensitive waveguides at 1.55 µm wavelength operation. Comparison with the modelling results shows that the polarisation-dependent behaviour of the waveguides is best described by a higher interdiffusion ratio for the group V than for the group III atoms. - Author(s): J. Piprek and S. Nakamura
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 145 –151
- DOI: 10.1049/ip-opt:20020441
- Type: Article
- + Show details - Hide details
-
p.
145
–151
(7)
The authors analyse the performance and device physics of nitride laser diodes that exhibit the highest room-temperature continuous-wave output power. The analysis is based on advanced laser simulation. The laser model self-consistently combines band structure and free-carrier gain calculations with two-dimensional simulations of wave guiding, carrier transport and heat flux. Material parameters used in the model are carefully evaluated. Excellent agreement between simulations and measurements is achieved. The maximum output power is limited by electron leakage into the p-doped ridge. Leakage escalation is caused by strong self-heating, gain reduction and elevated carrier density within the quantum wells. Built-in polarisation fields are found to be effectively screened at high-power operation. Improved heat-sinking is predicted to allow for a significant increase of the maximum output power. - Author(s): K.-H. Hasler ; H. Wenzel ; A. Klehr ; G. Erbert
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 152 –160
- DOI: 10.1049/ip-opt:20020505
- Type: Article
- + Show details - Hide details
-
p.
152
–160
(9)
The dynamical behaviour of three-section distributed Bragg reflected (DBR) lasers containing an active gain section and passive phase-shift and Bragg-grating sections was studied and modelled by numerically solving the travelling wave equations and a rate equation for the carrier density. By changing the parameters determining the optical gain or absorption, respectively, in the phase section a switching behaviour, hysteresis phenomena and self-sustained pulsations were obtained. By injecting small current pulses through the phase section, high-power optical pulses with small spectral width and repetition rates in the GHz range were generated. Experimentally, this was achieved by a selective heating of the phase section and by modulating the current injected into the phase section. In contrast to DBR lasers emitting at 1.3 µm and 1.55 µm, in the DBR lasers under investigation with an emission wavelength of 1.06 µm the active layer extends over all sections. Thus by the heating the interband absorption in the active layer of the phase section is strongly increased, while owing to the current pulses the absorption is rapidly reduced. Thus, giant light pulses are emitted which have a small spectral width owing to the wavelength selectivity of the Bragg grating. - Author(s): P. Bienstman and R. Baets
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 161 –165
- DOI: 10.1049/ip-opt:20020547
- Type: Article
- + Show details - Hide details
-
p.
161
–165
(5)
A novel optical VCSEL model is presented, based on vectorial eigenmode expansion combined with perfectly matched layer (PML) boundary conditions. It is fully rigorous and computationally efficient, as the PML boundaries eliminate parasitic reflections and allow the metal discretisation wall to be placed much closer to the device under study. The model is illustrated with a number of simulation results on proton-implanted, airpost and oxide-confined VCSELs. The trade-off between tight transverse optical confinement and scattering loss is clearly illustrated. - Author(s): M. Streiff ; A. Witzig ; W. Fichtner
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 166 –173
- DOI: 10.1049/ip-opt:20020527
- Type: Article
- + Show details - Hide details
-
p.
166
–173
(8)
The authors discuss the comprehensive computation of optical modes in the context of physics-based semiconductor laser device simulation. In this approach a generic frequency domain finite element (FE) formulation of the dielectric optical resonator problem is used. The FE expansion of a realistic vertical-cavity surface-emitting laser (VCSEL) device yields very large sparse generalised complex non-Hermitian eigenproblems. In the course of solving the device equations self-consistently over the entire operation range of the laser device, a sequence of many such large eigenproblems has to be solved. A distinct property of this sequence is that each eigenproblem can be seen as the perturbed version of the former in the sequence. This fact is exploited by using an iterative numerical method that uses the solution of one eigenproblem to solve the next. - Author(s): J.-F.P. Seurin and S.L. Chuang
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 174 –181
- DOI: 10.1049/ip-opt:20020526
- Type: Article
- + Show details - Hide details
-
p.
174
–181
(8)
Recent vertical-cavity surface-emitting lasers (VCSELs) include complex features, such as oxide apertures or intracavity index-guides, which complicate the electromagnetic modelling problem. Determining the optical characteristics of the cavity is essential to correctly interpret the physical behaviour of these devices. The authors have developed an efficient vector optical solver combining a mode-matching method and Krylov subspace techniques for a general VCSEL structure with cylindrical geometry. The solver generates accurate results for the cavity-mode profiles, resonant wavelengths, losses and threshold gain coefficients. Comparison with experimental data is shown. Scalar and vector modelling results are directly compared for the first time and the limits of the scalar approximation are investigated. It is found that the scalar approximation leads to large errors for small, strongly index-guiding apertures. The model is applied to both oxide-apertured and implant-apertured index-guided VCSEL structures. - Author(s): Y. Liu ; W.-C. Ng ; F. Oyafuso ; B. Klein ; K. Hess
- Source: IEE Proceedings - Optoelectronics, Volume 149, Issue 4, p. 182 –188
- DOI: 10.1049/ip-opt:20020544
- Type: Article
- + Show details - Hide details
-
p.
182
–188
(7)
The comprehensive semiconductor laser simulator, Minilase, has been extended to simulate the dynamic response of vertical cavity surface emitting lasers (VCSELs). Unlike conventional rate equation approaches, Minilase is capable of correctly modelling nonlinear gain effects in the modulation response without the introduction of an empirical gain suppression factor. Various oxide-confined single-mode VCSEL structures are simulated with Minilase to examine the effects of real-space carrier transport on the modulation response in both vertical and lateral directions. It is demonstrated that a roll-off in the dynamic response is closely associated with the diffusion capacitance caused by vertical carrier leakage. By either grading the separate confinement regions or reducing the thickness of the SCH cavity, the vertical carrier leakage is shown to be greatly suppressed, and the modulation response is significantly improved. Simulations also reveal that the nonuniform HE11-like transverse optical intensity in the quantum well results in an overdamping of the relaxation oscillation, and this effect is greatly reduced by making the electrical aperture smaller than the optical aperture. Finally, a comparison with experimental results is presented and discussed.
Editorial: Simulation of semiconductor optoelectronic devices
Modelling and simulation of electroabsorption modulators
Comparisons between dual-depletion-region and uni-travelling-carrier p-i-n photodetectors
Analysis of strain-induced polarisation-insensitive integrated waveguides fabricated using ion-implantation-induced intermixing
Physics of high-power InGaN/GaN lasers
Simulation of the generation of high-power pulses in the GHz range with three-section DBR lasers
Rigorous and efficient optical VCSEL model based on vectorial eigenmode expansion and perfectly matched layers
Computing optical modes for VCSEL device simulation
Efficient optical modelling of VCSELs using a vector mode-matching method
Simulating the modulation response of VCSELs: the effects of diffusion capacitance and spatial hole-burning
Most viewed content for this Journal
Article
content/journals/ip-opt
Journal
5
Most cited content for this Journal
We currently have no most cited data available for this content.