SiNWs

Ni thin films were used as catalysts by Jin et al. and Yan etal. to produce amorphous SiNW at 1200°C. No external supplies of Si were needed, but hydrogen gas was used. It was unclear whether the nanowires were amorphous silicon nanowires covered with silica layers, or silica nanowires. 

On the other hand, the work by Yan et al. and Jin et al. using silicon wafers and Ni as catalysts has suggested that bulk silicon would diffuse through the nanoparticles to produce SiNW. In this case, solid silicon in the wafer reacts with Ni catalysts to directly make SiNW. If this is true, it falls into the category of root growth. However, as we will illustrate below, the use of hydrogen in the presence of metal catalysts may activate a new reaction pathway that converts Si in the substrate into silane. As a result, the suggested solid-liquid-solid model may actually be the VLS model at work. 

Other mechanisms have also been proposed. For example, SiO has been used to produce SiNW. In this case, disproportionation reactions between SiO nanoparticles produce Si and Si02 nanoparticles. These nanoparticles then coalesce to form SiNW. 

Morphology of Si-Based Nanowires. The overall diameters of SiNW produced so far range from 3 nm to 100 nm, and many methods have yielded a frequently occurring diameter of 15 nm when they are made between the temperature of 1100 and 1200°C. This size of nanowires, as speculated at the end of this chapter, may be determined by the selected growth temperature. 

SiNW (made of silicon, silicon-covered with silica, or silica) can be either crystalline or amorphous. The interaction of silicon and metal catalysts, the evolution of silicon precipitating out of the catalysts, and the reactions followed the precipitation determine the morphology and crystallinity of SiNW. In many cases, SiNW have been observed with a silicon core covered with an amorphous silicon dioxide sheath. 

The crystalline core can make SiNW either fairly straight, or coiled or curly. It was discovered that CH4 may affect the helicity. This may be caused by a more complicated phase diagram that involves carbon, silicon, and metal. 

Other morphologies have also been made and observed for SiNW. Although straight nanowires dominate, other forms such as string beans (strings of pearls), shells. 

Milligram quantity SiNW have been obtained, although no stand-alone materials have been generated for easy handling. In general, nanowires so produced are attached to the substrate, and removal has been difficult or nearly impossible due to the small quantities available. 

Correlation between Sizes of Catalysts and SiNW. A loose correlation exists between the nanowire diameter and the size of the catalysts. The size of... 

Improved morphology and compositions of SiNW can further enable many new applications. For example, SiO c nanowires usually possess the most intense blue PL 47,48 ]-ecent study, long wavelength PL from oxygen vacancies in Si02 has... 

SiNW can also be used as sensors. In this case, SiNW of a silicon core without or with a very thin Si02 sheath are preferred. Since oxygen in the air normally reacts with Si, thin Si nanowires will eventually become Si02 nanowires if no protective layers are in place. 

Nanoparticles were prepared on the basis of our previous report. In brief, Co nanoparticles were synthesized via a high-temperature thermo-decomposition method.Three-, five-, and twelve-nanometer Co nanoparticles in dichlorobenzene (DCB) solutions were made and used to make self-aligned nanostructures (SAN) and SiNW. 

Gold nanoparticles were synthesized according to the literature procedure. Fifteen-nanometer Au nanoparticles were made and used for SiNW production. 

Synthesis of SiNW. SiNW were made by reacting Co nanoparticles with Si wafers in H2 and Ar between 1000 and 1100°C. " After nanoparticle deposition, the... 

Si wafers were either cleaned in Piranha solution, washed with deionized water and dried with Ar gas, or used without cleaning. The samples were then inserted into the high temperature tube furnace. Since SAN can form extremely fast, we believe that SAN existed in all samples. When the furnace temperature was set between 1035 and 1100°C, SiNW were formed. 

The apparatus was identical to that shown in Fig. 10.1. The gases were Ar (99.997%, Praxair), H2 (99.95%, Praxair), and CH4 (99.99%, Quadren Cryogenic Processing LTD). CH4 was not necessary to produce either SAN or SiNW. The flow rates were the same as for the growth of SAN. Only small amounts of CH4 were used. 

High resolution TEM (CEM300) was used to study the morphology of SiNW, and in situ TEM (JEOL 3010) experiments were performed at the National Center for Electron Microscopy (NCEM) to study migration of nanoparticles on Si wafers as a function of temperature. 

Raman samples were prepared by peeling SiNW thin films off the silicon wafers to avoid Raman signals from the silicon substrate. A razor blade was used to shave the thin films on top of the silicon wafers. The thin films were mounted on a carbon tape for SEM and Raman inspection, or on a transparent Mylar film for Raman measurements. 

The optimal densities of nanoparticle deposition for the formation of SAN and SiNW were different. For SAN formation, low-density deposition of Co nanoparticles was preferred. In contrast, high-density deposition was needed to produce high yield SiNW. 

FIGURE 10.3. Nanoparticle deposition for SiNW production. High-density deposition of nanoparticles was used. 

Figure 10.3 shows a Co nanoparticle deposition prior to SiNW growth. The average size of the nanoparticles used in this deposition was 12 nm. The standard deviation was 5 nm. As shown in Fig. 10.3, no individual nanoparticles were detected by SEM. The nanoparticles were embedded in large amounts of organic solvents or hydrocarbons such as oleic acid (OA) and tri-octylphosphine oxide (TOPO). 

Under the optimum condition for SAN formation, where the samples prior to high temperature synthesis were washed with Dl-water and DCB, the nanoparticle density was much lower. In this case, fewer SiNW were produced at 1100°C. An SEM image of the sample before growth is shown in Fig. 10.4. The cleanliness of the surface was due to the DI-water wash and the DCB rinse. Individual and strings of nanoparticles are clearly seen. 

At 1100°C, SiNW dominated. Figure 10.8 shows SiNW grown at 1100°C for 5 s. Reaction time was also a factor for SAN formation. Reaction times of 30 min and a few seconds produced similar results, although longer times generally produced more SiNW. Long SiNW fully covered the Si wafers after a few hours. The nanowires were between 5 and 50 nm in diameter, and microns to millimeter long. 

FIGURE 10.8. Samples of SiNW grown at 1 lOO C. The SiNW are microns to millimeters long, and have an average diameter of 16 nm. 

FIGURE 10.9. A SiNW sample produced with the finger printing technique (see text). The scale bar on the left is 1mm, and 10 fim on the right. The darker regions are bare Si surfaces. [Chem Comm 2005]-Reproduced by permission of The Royal Society of Chemistry, (ref 54)... 

FIGURE 10.10. SiNW grown on a Si wafer coated with Co nanoparticles at 1100°C, with H2. Several short SiNW are visible under SEM. 

We also found that addition of methane did not yield any SiNW at up to 1100°C. This may be caused by an increase in the melting point of cobalt carbide nanoparticles when methane was added. 

The substrate also played an important role in controlling the movement of nanoparticles at moderate temperatures before the nanoparticles melt. This can be clearly seen in Fig. 10.5 where nanoparticles are aligned with the atomic lines on Si (001). In addition, H2 and SAN together may play an even more important role in the catalysis of SiNW, as shown below. 

The nanowires in Fig. 10.5, straight and coiled, are believed to be SiNW. Experiments were performed to verify that they were not carbon nanotubes or metal nanorods. One method to prove this was by using a substrate other than Si. In this case, AI2O3 wafers replaced Si as the substrate to avoid interference from the Si substrate for characterization. The SiNW grown on Si are shown in Fig. 10.11, and those grown on the AI2O3 wafer are shown in Fig. 10.12. Characterization of these nanowires yielded that they were SiNW. 

This experiment also proved that the Si source became airborne and migrated from the Si wafer to the AI2O3 substrate because initially there was no Si on the AI2O3 substrate. Therefore, silicon in the silicon wafer had to move across the space between the two wafers to enable the growth of SiNW on the AI2O3 wafer. 

---