The molecular beam epitaxy growth and characterization of zinc cadmium selenide/zinc cadmium magnesium selenide-indium phosphide quantum cascade structures for operation in the 3 - 5 um range

The quantum cascade (QC) laser has captured the interest of researchers for almost three decades. In the early stages, researchers were very interested in proving the QC concept1 proposed by Kazarinov and Suris in 1971. This new concept gave researchers hope that very bulky energy inefficient infra-red (IR) lasers would be replaced with ones that are very compact, tunable and portable. Since the proposal of the QC laser concept and its first demonstration by researchers at Bell Laboratories2 in 1994, this technology has progressed to the point where it is now finding commercial applications in a variety of areas such as military counter measures, free space telecommunications, infra-red imaging and chemical spectroscopy.3-5 The success of this technology can be attributed to the coming of age of the techniques of molecular beam epitaxy (MBE) semiconductor growth and bandgap engineering. 6,7Using MBE technology, the temperature of the source material can be stabilized by making use of a combination of proportional integral derivative (PID) controllers and thermocouple feedbacks. As a result, the material flux from the effusion cells can achieve stability better than (+/-) 1%. This flux stability together with a well-developed computer controlled shuttering mechanism make it possible to grow multi-quantum well (MQW) structures with excellent layer thickness precision (mono-layer scale) and interface quality. This stringent control of material flux is also a tool that is used by MBE growers to vary the material compositions for the growth of lattice matched and strain compensated QC structures. Today, MBE stands out as one of the premier methods for growing high performing QC lasers.The first successful demonstration of a QC laser2 was done using the InGaAs/InAlAs-InP material system. This demonstration was then repeated a few years later using GaAs/AlGaAs-InP.8 These III-V material systems were extensively studied to establish their material parameters. Given that material parameters are critically important in the process of modeling QC structures, it is not surprising that early success was achieved using these systems. Today, the best performing QC lasers operate in the 4--13 mum range and are produced using lattice matched InGaAs/InAlAs-InP. In order to produce short wavelength QC lasers, the well layer thicknesses in the active region of the device must be reduced in an effort to push the lasing energy states further apart. This reduction in well thicknesses results in the movement of the upper lasing state closer to the bandedge. This action increases the probability of the lost of lasing state electrons to the continuum. Therefore, in order to produce high performing short wavelength QC lasers, a large conduction band offset (CBO) is required. The CBO of lattice matched InGaAs/InAlAs-InP is 0.52 eV. In an attempt to produce high performing devices below 4 mum many researchers have resorted to the use of strain compensation9-11 . This approach has yielded very little improvement in performance due to electron scattering to the X and L intervalleys. This has lead to the exploration of wide bandgap material systems such as the antominides and nitrides.In this work the wide bandgap II-V Znx'Cd(1-x')Se/Zn xCdyMg(1-x-y)Se-InP will be explored for QC laser fabrication. To this end, QC lasers were designed for operation at 3--5 mum range. A Matlab-based program was written to calculate the energy level spacing within the active region of these devices. This simulation program was based on Schroindger's equation and the transfer matrix technique. Several calibration samples were grown to establish the doping levels and growth rate of the well and barrier materials. The growth rate was measured using scanning electron microscopy (SEM) and reflection high energy electron diffraction (RHEED) oscillations during MBE growth. X-ray diffraction measurements were performed to determine the lattice mismatch of the II-VI bulk layers, and therefore predict whether material composition adjustments were required to attain the lattice match condition. The samples that were grown were studied using photoluminescence (PL) to determine the bandgap of the well and barrier material. This information was then used to calculate the CBO of the II-VI MQW structure. In addition, PL studies were also carried out to look for material defects and assess the quality of the well/barrier interface. These II-VI QC samples were also subjected to Fourier transform infra-red (FTIR) absorption spectroscopy to determine the energy levels in the grown structures. After optimizing the active regions using simulation data and FTIR results, electroluminescence (EL) structures were grown and processed into QC emitters using a combination photolithography and electron beam contact deposition. The processed structures were then biased and investigated for IR emission at temperatures ranging from 80 K to room temperature.