Khattak, Shaukat Ali and Hayne, Manus (2015) Exciton confinement in strain-engineered InAs quantum dots in metamorphic In_{x}Ga_{1-x}As. PhD thesis, Lancaster University.
2015khattakphd.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (9MB)
Abstract
In this work, magneto-photoluminescence at low temperature, 4.2 K, is used to probe the exciton confinement in strain-engineered InAs/In_{x}Ga_{1-x}As/GaAs metamorphic quantum dots (QDs), emitting at telecom wavelengths (1.3 µm - 1.6 µm). The emission wavelength can be tuned by changing two independent parameters, i.e.,indium content, x, in In_{x}Ga_{1-x}As upper and lower confining layers and thickness of lower confining layer (LCL), d. Varying x changes the band offset and QD-CLmismatch (strain inside the QD), while varying d changes only QD-CL mismatch.We investigate the dependence of confinement on the QD-CL mismatch and band offset. Zero-magnetic-field spectra showed that wavelength (PL energy) increases(decreases) with increasing x, for a constant d, and with increasing d, for a constant x, which was attributed to be due to relaxation of strain inside the QD that is, in turn,a function of x and d. No correlation between wavelength and intensity was observed. Magneto-photoluminescence results revealed that for a subset of samples, the exciton effective mass increases linearly, more or less, with increasing QD-CL mismatch (QD strain), while its Bohr radius has no correlation with mismatch. The diamagneticshift coefficient increases 12-fold with decreasing mismatch from ∼ 7.5 % to 4.5 %, which is attributed to low effective mass, which in turn, is due to low QD strain. For high mismatch (> 5.8 %), the Bohr radius is not determined, implying that it is less than10 nm, smaller than the dot radius. For indium composition x = 0.28 and 0.31, and for d > 1000˚A, the wave-function spills over out of the dot. For x = 0.35, the Bohr radii are, counter intuitively, found to be smaller than for samples with larger band offset (x = 0.31). Initially, it was explained as a spilling of the wave-function over vertically resulting in strong lateral confinement of exciton, but this explanation is not supported by our model calculations. Another explanation is, therefore, presented by carrying out temperature dependence and magnetic field dependence, at various temperatures, of PL energy: there are different dots, at x = 0.35, with different size where thermal escape of carriers from smaller dots to bigger ones occurs with increasing temperature, and the PL energy, in magnetic field, is contributed more by smaller dots than the bigger ones.