Temperature driven 3D ordering of quantum dots in (InGa)As/GaAs superlattices grown under As2 flux
|Peter Lytvyn 1,2, Yuriy Mazur 1, Euclydes Marega Jr 1, Leonardo de Souza 1,3, Mourad Benamara 1, Morgan Ware 1, Igor Prokopenko 2, Gilmar Marques 3, Gregory J. Salamo 1|
1. Institute for Nanoscience and Engineering, University of Arkansas,, 731 W. Dickson str., Fayetteville, AR 72701, United States
Strain-driven self-assembly has matured into a promising method for the fabrication of semiconductor quantum dot (QD) nanostructures. The basic mechanism of QD nucleation and growth is well described by the classical Stranski-Krastanov growth model. However, more complex multi-component systems as well as strongly anisotropic structures are not completely described by the simplified basic model. Elastic properties of the bulk crystal lattice and surface mass transport have a significant impact on the QD growth. So, in the case of multilayer structures of InGaAs QDs additional effects must be considered such as: the strain distribution through the spacer layers, In migration on the surface during the capping process, and possibly surface roughening throughout the deposition of the spacer layer. To control QD ordering the thickness and composition of corresponding layers are optimized as a rule. But, it is possible to achieve significant transformations in the three-dimensional ordering of the QDs by means of temperature control.
We systematically investigated the effect of deposition temperature of the QD layer on the three-dimensional ordering of InGaAs QDs within multilayer nanostructures grown using As2 flux as the group V source. The threshold nature for 3D ordering was established using direct observation (TEM and AFM) for the 510-520 °C and 540-555°C growth temperature transitions. At these temperatures, there is a significant decrease in strain propagation through the system as the distances between QDs in one of the characteristics directions <011> exhibits a sharp increase. Changes of inter-dot distances are minimal at 530 and 540°C. There is an increase in the anisotropy of the strain in the system which leads to the appearance of some anticorrelation (slopes) in the alignment of the QDs through the superlattice growth and a more pronounced planar ordering along the preferred direction (fig.1). It is shown that this ordering is accompanied by transition of the almost continuous wetting layer (low temperature) into large QDs (high temperature) where no In detected between them (fig. 2). Critical changes in the overall period of the superlattice and the reduction of the spacer thickness over buried QDs are observed at the 555°C where an intense desorption of indium atoms becomes possible. The above changes in 3D QDs alignment correlated well with the assessment of quality of the structures through photoluminescence measurements.
Fig.1. AFM height maps (left) and corresponding 2D autocorrelation functions (right) for the samples grown at 510, 530 and 555°C, (a,b), (c,d) and (e,f) correspondingly.
Fig.2. Bright field TEM images, taken under two-beam condition with (002) reflection and corresponding In distribution maps (electron energy loss spectroscopy) of the samples grown at 510°C (a,b) and 555°C (c,d).
These regularities can be explained by the competing processes of anisotropic mass transport on the growth surface and the influence of strain on the nucleation of 3D islands. Abrupt changes in the distances between QDs can be associated with changes in the surface atomic reconstruction which in turn affects the degree of anisotropy for the diffusion of adatoms. The indium surface segregation could play an important role in the behavior of the wetting layer. Less important are the effects of surface corrugation which caused fluctuations of superlattice period of ~0.5% in our case.
Presentation: Poster at 17th International Conference on Crystal Growth and Epitaxy - ICCGE-17, General Session 8, by Peter Lytvyn
See On-line Journal of 17th International Conference on Crystal Growth and Epitaxy - ICCGE-17
Submitted: 2013-03-30 14:09 Revised: 2013-04-06 12:09
|© 1998-2022 pielaszek research, all rights reserved||Powered by the Conference Engine|