Quantum Dot Epitaxy on Suspended Si Nanomembranes

We grow Ge and SiGe quantum dots (QDs) on suspended silicon nanomembranes (SiNMs). QDs form spontaneously when Ge is grown epitaxially on Si because the atomic spacing of Ge--its "lattice constant"--is 4% larger than Si's. A lot of strain builds up when Ge tries to fit onto the existing Si crystal, and after a few atomic layers Ge begins to form three-dimensional pyramids instead of smooth two-dimensional layers (Figure 1)[1].

 

STM of a Ge QD
Figure 1. A scanning tunneling microscope (STM) image of a Ge Quantum Dot (QD).

 

Nothing new, so far. Ge QDs were discovered almost 20 years ago. The new phenomena arise when the QDs are grown on SiNMs so thin (5-30 nm thick) that the strain that builds up in the QD actually deforms and stretches the SiNM underneath it. If we use chemical vapor deposition (CVD) to grow the QDs, we can grow on both sides at once (top and bottom). Growth has to start somewhere, and the first dot on the top actually bends the SiNM in such a way that it sets four sites on the bottom surface with lower "normalized strain energy"[2], sites on which the next QDs on the bottom surface will start to grow. The sites are shown in Figure 2.

 

Normalized strain energy contours on the surface
	  opposite a dot
Figure 2. Normalized strain energy contours on the surface opposite to a QD on a 6 nm thick membrane
(courtesy of Hyun-Joon Kim-Lee). The "+" shows the location of a minimum and the "X" shows a measured
distance from an experimental sample.

 

What you end up with is a self-organized array of quantum dots on the suspended region that you've made.

 

QDs on a suspended Ribbon
QDs on a suspended edge
Figure 3. A suspended Si nanoribbon and a partially undercut SiNM edge with QDs.

 

Author: Clark Ritz

References
[1]Mo et al. Phys. Rev. Lett. 65 1990
[2]H.-J. Kim-Lee, D.E. Savage, C.S. Ritz, M.G. Lagally, and K.T. Turner. Phys. Rev. Lett. 102, 226103 (2009)


  Copyright (c) 2010 Lagally Research Group. All rights reserved.