Search

Experimental results showing room temperature normal incidence mid-infrared detection by AlGaN/GaN quantum well infrared photodetectors are presented. Designed structures have intersubband transitions corresponding to wavelengths in the region of 3 and 4 µm, where strong absorption in a sapphire substrate dominates. The intersubband spectra, therefore, were characterized by electronic Raman scattering and infrared photocurrent spectroscopy. The absorption spectra agree well with theoretical predictions. Details of device fabrication are presented with sensitivity estimates for the devices.

We propose an optically pumped laser based on intersublevel transitions in InAs-GaAs pyramidal self-assembled quantum dots. A theoretical rate equations model of the laser is given in order to predict the dependence of the gain on pumping flux and temperature. The energy levels and wave functions were calculated using the 8-band k/spl middot/p method where the symmetry of the pyramid was exploited to reduce the computational complexity. Carrier dynamics in the laser were modeled by taking both electron-longitudinal optical phonon and electron-longitudinal acoustic phonon interactions into account. The proposed laser emits at 14.6 /spl mu/m with a gain of g/spl ap/ 570 cm/sup -1/ at the pumping flux /spl Phi/=10/sup 24/ cm/sup -2/ s/sup -1/ and a temperature of T=77 K. By varying the size of the investigated dots, laser emission in the spectral range 13-21 /spl mu/m is predicted. In comparison to optically pumped lasers based on quantum wells, an advantage of the proposed type of laser is a lower pumping flux, due to the longer carrier lifetime in quantum dots, and also that both surface and edge emission are possible. The appropriate waveguide and cavity designs are presented, and by comparing the calculated values of the gain with the estimated losses, lasing is predicted even at room temperature for all the quantum dots investigated.

A systematic design of a GaN/AlGaN active region for a three-level intersubband laser emitting in the near-infrared is presented. The proposed system is based on a double coupled quantum well structure with the laser level energy separation corresponding to 1.55 µm (ΔE32~800 meV), a wavelength important for optical communications. The lower laser level depopulates mostly via longitudinal optical phonon resonance, requiring ΔE21~90 meV. The model for scattering rate calculations, including many-body carrier screening effects, was first tested by comparison with recent experimental measurements in GaN/AlGaN quantum wells, and a very good agreement was obtained. It was then employed to calculate the gain coefficient, a commonly used figure of merit for three level laser systems, and this in turn allowed the optimal structural parameters of GaN/AlGaN quantum well to be found.

The design of ZnCdSe-ZnMnSe-based quantum wells is considered, in order to obtain a large shift of the peak absorption wavelength in the far infrared range, due to a giant Zeeman splitting with magnetic field, while maintaining a reasonably large value of peak absorption. A triple quantum-well structure with a suitable choice of parameters has been found to satisfy such requirements. A maximal tuning range between 14.6 and 34.7 meV is obtained, when the magnetic field varies from zero to 5 T, so the wavelength of the absorbed radiation decreases from 85.2 to 35.7 /spl mu/m with absorption up to 1.25% at low temperatures. These structures might form the basis for magnetic field tunable photodetectors and quantum cascade lasers in the terahertz range.

We present a Raman scattering study of two GaN/AlxGa1−xN multiple quantum well (MQW) structures with x = 0.3 and x = 0.4 and well widths of 6 nm and 4 nm respectively, they were nominally undoped but are expected to contain a low-density, n-type, carrier population due to the residual donors in the barriers. Polarization dependent Raman scattering was performed at room temperature, strong scattering due to intersubband transitions (ISBT's) in the GaN quantum wells was observed from both the e1–e2 and e1–e3 transitions for both samples. The first sample has an e1–e2 transition at 300 meV and e1–e3 at 480 meV whilst the second sample has an e1–e2 transition at 360 meV and e1–e3 at 580 meV. The full width half maximum of the Raman peaks is lower than that reported in infra-red absorption for similar, heavily doped, MQW structures. (© 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

A method is proposed for the design and optimization of structural parameters of GaN-AlGaN Bragg-confined structures with respect to peak intersubband absorption from the ground to the first excited state,1 /spl rarr/ 2 electronic transition, in the near infrared spectral range. An above-the-barrier bound state was used to extend the range of transition energies above the values available in conventional quantum wells. Intrinsic polarization fields and nonparabolicity effects were taken into account. The selection of optimal parameters, maximizing the absorption at wavelengths of 1.55 and 1.3 /spl mu/m, was performed by using a simulated annealing algorithm, and optimal structures with infinite superlattices as confinement regions were thus designed. These optimal parameters were then used to set realistic, finite structures with a small number of layers, the performance of which was re-evaluated by solving the Schrodinger-Poisson equation self-consistently for a few different levels and profiles of doping.

This paper compares and contrasts different theoretical approaches based on incoherent electron scattering transport with experimental measurements of optoelectronic devices formed from semiconductor heterostructures. The Monte Carlo method which makes no a priori assumptions about the carrier distribution in momentum or phase space is compared with less computationally demanding energy-balance rate equation models which assume thermalised carrier distributions. It is shown that the two approaches produce qualitatively similar results for hole transport in p-type Si1-xGex/Si superlattices designed for terahertz emission. The good agreement of the predictions of rate equation calculations with experimental measurements of mid- and far-infrared quantum cascade lasers, quantum well infrared photodetectors and quantum dot infrared photodetectors substantiate the assumption of incoherent scattering dominating the transport in these quantum well based devices. However, the paper goes on to consider the possibility of coherent transport through the density matrix method and suggests an experiment that could allow coherent and incoherent transport to be distinguished from each other.

This paper compares and contrasts different theoretical approaches with experimental measurements of transport in optoelectronic devices based on semiconductor heterostructures. The Monte Carlo method which makes no a priori assumptions about the carrier distribution in momentum or phase space is compared with less computationally demanding energy-balance rate equation models which assume thermalised carrier distributions. It is shown that the two approaches produce qualitatively similar results for the specific case of hole transport in p-type Si1-xGex/Si superlattices designed for terahertz emission though there are significant differences which originate in the absolute values of the carrier kinetic energy in the plane of the quantum wells. In contrast to this, simulations of electron transport in n-type GaAs/Ga1-xAlxAs terahertz quantum cascade lasers show a more similar behaviour between both theoretical approaches. In addition, the very good comparison with experiment substantiates the incoherent scattering approach which under-pins both methods. Further evidence to support the applicability of carrier transport in semiconductor heterostructures at finite temperature being dominated by incoherent scattering, which can be modelled by self-consistent energy-balance rate equations, is given by further comparisons with experimental measurements of two other categories of devices namely quantum well infrared photodetectors and quantum dot infrared photodetectors

A theoretical model of electron transport in quantum cascade lasers subjected to a magnetic field is developed. The Landau level electronic structure was calculated from the envelope-function Schr\"odinger equation within the effective-mass approximation. The electron transport in a magnetic field was modeled using the self-consistent rate-equation description for the full period of the cascade and its interaction with adjacent periods. The scattering processes included in the model are electron\char21{}longitudinal-optical-phonon, electron\char21{}longitudinal-acoustic-phonon, and electron-electron scattering. All these processes show oscillatory behavior with magnetic field, and their interplay determines the electron transport and the output characteristics of quantum cascade lasers in magnetic field. The model was applied to investigate the influence of magnetic field on the performance of a $\mathrm{GaAs}∕\mathrm{AlGaAs}$ quantum cascade laser emitting at $\ensuremath{\lambda}\ensuremath{\approx}11.4\phantom{\rule{0.3em}{0ex}}\mathrm{\ensuremath{\mu}}\mathrm{m}$ [P. Kruck et al., Appl. Phys. Lett. 76, 3340 (2000)]. The calculated results show good overall agreement with the available experimental data.

In this paper we show that the idea of a mid‐infrared quantum‐cascade laser with gain regions based on a double‐phonon resonance can be implemented in the GaAs/AlGaAs system. In contrast to the usual GaAs/AlGaAs laser active region design, which involves a triple quantum well active region, here we identify an optimal heterostructure design by using a simulated annealing algorithm which is programmed to maximize the laser gain for a given wavelength and for subband spacings prescribed to satisfy the double‐phonon resonance condition. The output characteristics are calculated using a full self‐consistent rate equation model of the intersubband electron transport. Initial devices grown according to this design show laser action up to 240K in pulsed mode with good agreement between the calculated and measured characteristics.

We report the growth and characterization of AlGaN/GaN multiple quantum well structures designed to have intersubband transitions in the mid-infrared region of the spectrum. The samples were nominally undoped but were found to contain a high electron population in the wells induced by the local polarization fields. The sample was characterized by the use of the Raman spectroscopy and photocurrent spectroscopy. The Raman spectroscopy shows electronic Raman scattering from intersubband transitions in the AlGaN/GaN quantum wells. The e1-e2 and e1-e3 transitions of the confined 2d electron population in the wells can clearly be observed. A sample designed to absorb at 4μm was fabricated into mesa structures and the vertical photocurrent measured under normal incidence illumination from the free-electron laser FELIX. A wavelength and bias dependent photocurrent was observed in the mid-IR region of spectrum. The peak responsivity was of the order of 50μA/W at 4 K, the photocurrent still being measurable at room temperature.

A fully quantum-mechanical model for carrier scattering transport in semiconductor intersubband devices was applied to modelling of carrier dynamics in quantum cascade lasers. The standard model uses the envelope function and effective mass approximations to solve electron band structure under an applied bias. The k(.)p model has been employed in p-type systems where the more complex band structure requires it. The resulting wave functions are then used to evaluate all relevant carrier-phonon, carrier-carrier and alloy scattering rates from each quantised state to all others within the same and the neighbouring period. This piece of information is then used to construct a rate equation for the equilibrium carrier density in each sub-band and this set of coupled rate equations are solved self-consistently to obtain the carrier density in each eigenstate. The latter is a fundamental description of the device and can be used to calculate the current density and gain as a function of the applied bias and temperature, which in turn yields the threshold current and expected temperature dependence of the device characteristics