Carrier thermal escape in families of InAs/InP self-assembled quantum dots

Abstract

We investigate the thermal quenching of the multimodal photoluminescence from InAs/InP (001) self-assembled quantum dots. The temperature evolution of the photoluminescence spectra of two samples is followed from 10 K to 300 K. We develop a coupled rate-equation model that includes the effect of carrier thermal escape from a quantum dot to the wetting layer and to the InP matrix, followed by transport, recapture or non-radiative recombination. Our model reproduces the temperature dependence of the emission of each family of quantum dots with a single set of parameters. We find that the main escape mechanism of the carriers confined in the quantum dots is through thermal emission to the wetting layer. The activation energy for this process is found to be close to one-half the energy difference between that of a given family of quantum dots and that of the wetting layer as measured by photoluminescence excitation experiments. This indicates that electron and holes exit the InAs quantum dots as correlated pairs.

Spectral and Transient Luminescence Measurements on GaSb/AlGaSb Quantum Wells Grown on GaSb/GaAs Heterojunctions with and without Interfacial Misfit Arrays

“…When this ratio, ν is equal to 1, exciton escape to the barrier is dominant. When ν = ½, the carrier escape is in the form of correlated electron-hole pairs and this mechanism was observed with GaInP/AlGaInP QW, 35) InAs/GaAs QD, 36) InAs/InP QD, 37) and InAs/InGaAlAs QD. 38) This mechanism is probable when the band discontinuities both in the conduction and valence bands have relatively large values for carrier confinements.…”

Size dependent carrier thermal escape and transfer in bimodally distributed self assembled InAs/GaAs quantum dots

“…This parameter has been used to classify the carrier escape nature. 14 The value ¼ 1 is associated with correlated (excitonic) escape, while < 0.5 and ¼ 0.5 to unipolar and uncorrelated pair escape, respectively. In their model, Yang et al 4 obtain a single value, independent of the QD emission energy.…”

“…[23][24][25][26] Thermal escape can be also investigated attending to the nature of the particles being promoted to a higher energy state. 14 Depending on the model, the correlated (excitonic escape), 6,8,10,12,23 the uncorrelated electron-hole pair (ambipolar escape), 4,9,14 or just one of the carriers (unipolar escape) 15 can be considered. Whether one or the other is appropriate in a particular sample depends on the barrier height and therefore might vary for QDs of given composition but different size.…”

“…A variety of carrier redistribution effects caused by temperature has been investigated in QD ensembles. [4][5][6][7][8][9][10][11][12][13][14][15] It is commonly accepted that the sigmoidal evolution of the peak energy and full width at half maximum of the PL bands is due to carrier promotion from small QDs to larger ones. 6 Therefore, quantum dot size distributions, carrier capture, relaxation, and re-trapping among QDs of different sizes had to be considered to model correctly the QD recombination dynamics.…”

Femtosecond pulse generation in passively mode locked InAs quantum dot lasers

“…We observe that at temperatures below the minimum in threshold, the emission spectra broaden and analysis of the emission data at 20 K shows ground states in different size dots are populated with equal probability, increasing with drive current, indicative of random behaviour. 18 We have compared this data with rate equation calculations of gain spectra for inhomogeneous distributions of ground and excited states, interacting via a thermal phonon bath with each other and with the wetting layer, and in which population of dots by electrons and holes is correlated, 12,14,22 controlled by separation of the dot electron state from the wetting layer ($0.28 eV for the ground state). Measured modal absorption spectra provided input values for the gain coefficient and spontaneous recombination lifetime (via the Einstein relations), which were temperature independent, and for the wetting layer band edge energy; energies and distributions of dot transitions were determined by Gaussian fits.…”