Lanterna Rally

Author: YILING XIONG

Publisher:

Published: 2020

Total Pages: 0

ISBN-13:

The study of active region structure for semiconductor lasers began in the 1960s. Most recently, quantum dot (QD) based lasers have attracted increasing attention. Modeling is crucial for the design of semiconductor QD-based lasers. Many attempts have been made to the macroscopic and, particularly, the microscopic modeling of III-V semiconductor QD as well as its applications during these decades. However, these proposed approaches use a very similar but outdated way to calculate the elastic strain field, referred to as one-step model, without rigorous consideration of the influence of the growth interruption in double-capping procedure, as the latter is currently used in epitaxial self-assembly for the control over the size of QDs. This thesis aims to contribute to the design improvements of QD-based laser applications through more accurate modeling. In this thesis, we have focused on improving the modeling accuracy by elaborately analyzing the elastic strain and quantum confinement potential. By applying this accurate modeling methodology, not only the general semiconductor QD-based lasers but also the structures with an interlayer/sublayer or tightly coupled QD ensemble can be numerically modeled, giving rise to the possibility for predicting the behavior and even structural design of lasers, paving the way to potentially novel applications. The following work has been done in this thesis. Firstly, we propose an accurate method of modeling a single QD, including a thorough so-called two-step elastic strain analysis, by considering the influence of growth interruption. A series of settings in terms of the three-dimensional (3D) geometry of QD and surrounding matrix are considered. The 3D confinement potential profile is found significantly different compared with the counterpart using the conventional one-step model. The electronic band structure is then calculated by using the strain-dependent eight-band k ∙ p method. The simulation results by using the two-step model are found in better agreement than one-step model in comparison with measurements. Moreover, the impact of the quaternary compositions of barrier material is, for the first time, systematically studied. Secondly, the two-step model is further extended to three- and multi-step analysis to model the structures with additional GaP ultrathin layer above or beneath the QDs. It is found that, instead of preventing the As/P exchange, the main impact of GaP interlayer/sublayers is enhancing the quantum confinement and thereby blue-shifting the emission peak. Based on the ability to efficiently shifting the spectrum, a new vertically chirped multi-layer structure is proposed. By simultaneously optimizing the interlayer/sublayer thickness and double-capping settings, a total gain spectral bandwidth of 245.7 nm (i.e. 30% increase) is predicted, and peak wavelength is shortened to 1510 nm (i.e. 70 nm blueshift, in comparison to the case without interlayer/sublayer). Thirdly, laterally and vertically coupled QDs are modeled to investigate a variety of coupling effects in the active region of lasers. In particular, multi-step strain analysis is applied to the modeling of closely stacked QDs to reproduce a more realistic unidirectional compressive strain accumulation, evidenced by the morphological observation of cross-section images obtained from measurements. A "quasi continuum band" formed by the mixing of bonding and antibonding states is found, giving rise to the possibility of emission at excited state (ES) instead of the ground state (GS). Using this feature, a new laser structure allowing two-state lasing under continuous wave (CW) electrical pumping is proposed for the first time and characterized through the simulation of spectral linewidth and relative intensity noise (RIN). The new structure exhibits lower (i.e. −130 versus −110 dBc/Hz) integrated RIN compared with the conventional counterpart under relatively high CW current injection. Overall, this thesis sheds light on new device physics and provide guidelines to realize QD-based lasers with new features, and would be interesting to the scientific community.