Elsevier

Journal of Crystal Growth

Volume 551, 1 December 2020, 125891
Journal of Crystal Growth

Intermittent growth for InAs quantum dot on GaAs(001)

https://doi.org/10.1016/j.jcrysgro.2020.125891Get rights and content

Highlights

  • We perform the intermittent growth of InAs quantum dots on GaAs(0 0 1) at 500 °C.

  • QDs nucleation was observed by STMBE where in situ STM equipped within MBE chamber.

  • The initial QDs formation occurred from 1.15 to 1.38ML with the intermittent method.

  • Detail RHEED observation indicates large (n×3) and (2×4) area formation before QDs.

Abstract

We performed an intermittent growth of InAs quantum dots (QD) on GaAs(001) at 500 °C. The transition of surface structures during the growth was investigated by using reflection high energy electron diffraction observation. We also performed in situ observation of QD nucleation by using STMBE system in which a scanning tunneling microscope equipped within a molecular beam epitaxy chamber. We have found that the initial QDs formation occurred from 1.15 to 1.38 ML with the intermittent InAs supply. This InAs supply amount was much smaller than that of ordinary continuous deposition (1.66 ML). Moreover, the QDs mainly appeared on the terrace while they mainly appear on step edges by continuous growth at 500 °C. This indicates that the annealing parts of the intermittent growth affected the surface atomic structures of InGaAs wetting layer, uniforming the In fluctuation and stabilized the surface morphology. By the preparation of large (n×3) and (2×4) area, the QD nucleation occurs with a smaller InAs supply on the surface.

Introduction

High-density arrays of quantum dots (QDs) are easily prepared by using Stranski–Krastanov (SK) growth mode of molecular beam epitaxy (MBE) [1], [2], [3]. Such QDs are strong candidates for advanced semiconductor quantum devices [4], [5], [6]. However, the method of precise controlling of QD size, density, and distribution depending on applications are important and necessary. So many investigations have been carried out on QDs forming process. Ramachandran et al. reported small 3D structures appearing just before QDs nucleation [7]. But, they observed the structures after stopping MBE growth and cooling down to the room temperature. This means it is difficult to determine when the small 3D structures form.

Tsukamoto et al. observed InAs QD formation processes at 430 °C by using the STMBE system [8] which is equipped with a scanning tunneling microscope (STM) inside a MBE growth chamber, performing true in situ imaging during the MBE growth. This can observe the characteristics of the surface reconstruction before the QD nucleation. Moreover, Honma et al. observed InAs QD formation processes at 500 °C during continuous STMBE growth [9]. In order to understand factors affecting QD nucleation sites, it is necessary to investigate more details of the surface structure before QD nucleation.

To understand the relationship between the QD formation and the surface reconstruction, it is necessary to focus on the thermal equilibrium state of the surface. The ordinary MBE growth keeps the surface non-equilibrium due to continuous In supply, and it is difficult to know the relationship between the surface and the QD formation. In order to reveal this relationship, it is important to keep the surface in thermal equilibrium state which directly leads to QD formation. Here, we have developed a new method of QD formation with continuously observing the same area by repeating to supply a small amount of InAs and to anneal under As4 flux (intermittent deposition method). In this method, In adatoms migration is enhanced during the annealing period compared to the continuous growth, which is expected to stabilize the surface in the thermal equilibrium state. This is similar to migration enhancement epitaxy reported by Horikoshi et al. [10] but they applied it to smooth the surface at low temprature growth.

In this paper, we report correlations between In adatoms migration and QD nucleation, applying the intermittent growth to regulate the surface reconstructions by the STMBE system. And we analyse the time evolution of reflection high energy electron diffraction (RHEED) pattern reflected to the surface morphology.

Section snippets

Experimental setup

All samples were prepared from the same type of epi-ready GaAs(001) just-cut substrate. After the samples were introduced into the load-lock chamber, the following procedures were performed under the same conditions: afrer pre-baking at 300 °C for 30 min in the load-lock chamber, transporting to the STMBE chamber, removing surface oxide layer at 580 °C, a GaAs buffer layer was grown at 560 °C using As4/Ga flux ratio of 89 measured by using a flux monitor. After growing the GaAs buffer layer,

Results and discussion

Figure 2(a-c) shows STM images of GaAs(001) surface after 1.15, 1.38, and 1.61 ML InAs supply, respectively. We can confirm surface morphologies before InAs QDs formation in the Fig. 2(a). There were 1.0 ML height islands but no QDs appeared yet. The InAs QDs were formed from 1.15 to 1.38 ML. The STM image shifted to the lower right, still showing the same particular area indicated by red broken rectangles. There were a few larger QDs as shown. This suggests that these QDs might be formed

Conclusion

We demonstrated intermittent growth of InAs QDs on GaAs(001) at 500 °C. We have found that the initial QDs formation occurred from 1.15 to 1.38 ML with intermittent In supply, which was much smaller than that of ordinary continuous deposition (1.66 ML). Moreover, the initial QDs mainly appeared on the terraces while they mainly appear on the step edges by ordinary InAs growth at 500 °C. This indicates that the intermittent In supply and annealing affected the surface atomic structures of

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (19)

There are more references available in the full text version of this article.

Cited by (1)

View full text