New Self-Organized Nanostructure: Chain of Crystalline-Silicon Nanospheres
- PDF / 2,042,189 Bytes
- 6 Pages / 417.6 x 639 pts Page_size
- 70 Downloads / 151 Views
293 Mat. Res. Soc. Symp. Proc. Vol. 571 ©2000 Materials Research Society
Fig. 1: TEM image of the chains of crystalline-Si nanospheres.
Fig. 2: High-resolution TEM image of a chain of crystalline-Si nanospheres. Lattice planes of Si crystal are clearly seen.
294
Fig. 3: Energy-filtered TEM images of a chain of crystalline-Si nanospheres with an energy window of (a) 16 ± 2.5 eV and (b) 22 ± 2.5 eV.
etry (XEDS) analysis in the TEM showed that the necks consist of silicon, oxygen and carbon. Accordingly, we conclude that the amorphous material is a mixture of silica and carbon. GROWTH MECHANISMS We discuss how Si was supplied and formed into chains of nanospheres. As mentioned above, Si whiskers/wires are grown via the VLS mechanism using various kinds of liquid-forming catalysts such as Au. In this mechanism, Si is supplied as a gas phase such as Sill4 or SiCI 4 and captured by the liquid catalysts, and the supersaturated Si in the molten catalyst is deposited in succession on the top of the growing whisker/wire. We confirmed that there is a Au-silicide
particle at the end of the chain of nanospheres. Also, as the by-product of the chains, many wires of polycrystalline Si were grown, which show similar morphology as that of the wires grown by the ordinary VLS technique. We also observed the transition from a nanosphere chain to a Si wire with an uniform diameter in a single wire. Accordingly, we think that the chains as well as wires were grown via the VLS mechanism. We propose that Si is supplied via a carbothermal reduction process: SiO2 (s) + C(s) -> SiO(g) + CO(g), SiO(g) + C(s) -> Si(s) + CO(g), where (s) and (g) represent solid and gas phases, respectively. Considering that the SiO(g) may be supplied, we propose the Au/SiO VLS mechanism for the formation of chains of nanospheres. Next, we discuss how the Si nanospheres are arrayed in a chain-like manner (Fig. 4): (i) The diameter of wires may change periodically during their growth. Givargizov has observed a similar periodic structure in much thicker whiskers by means of scanning electron microscope (SEM) 9. The diameter of his whiskers were about ten times larger than ours and grown epitaxially on a Si substrate. Single crystallinity will extend over his thicker whiskers, since the necks are so thick that the oxidization will be limited to only the surface. In order to reveal the mechanism of the periodic instability, we discuss the feedback mechanisms which are based on Givargizov's idea and modified by us. We consider both positive and negative feedback mechanisms of the
295
Growth direction
10Au-Si •-"
(a)
droplet
Crystalline -silicon nanosphere
Silica
Fig. 4: Growth model of a chain of crystalline-Si nanospheres. (a) At a neck, the large contact angle of the droplet and the small diameter. Oxidization reaches the core. (b) At a knot, the small contact angle and the large diameter. Only the surface is oxidized. instability of the molten catalyst (droplet). First, we assume the following mechanism for positive feedback: if the diameter of the
Data Loading...
