Property Control of Single-Walled Carbon Nanotubes through Synthesis.

摘要

Single-walled carbon nanotubes (SWCNT) have enormous potential in electronic applications such as field effect transistors, but the mixture of metallic and semiconducting nanotubes in synthesized material limits the reproducible production of these devices.;The electronic property (such as bandgap) of pristine SWCNT is based on the diameter and chirality. Our main objectives are to control the properties, including diameter, chirality and electronic properties of SWCNT through synthesis, and to meet the need for certain electronic applications. The diameter and chirality control of SWCNT may be fulfilled by choosing appropriate catalysts and carbon sources. The bandgap changing with SWCNT diameter is not as sensitive in the large-diameter range (>1.5 nm) as in small-diameter range (<1 nm), which makes large-diameter tubes more suitable for some electronic device design. We will present a method of synthesizing large-diameter SWCNT (>1.5 nm) with small bandgap (<0.7 eV) through water-assisted ethanol pyrolysis using Co-MCM-41 catalysts in chapter 2. The data from X-ray absorption spectroscopy (XAS) collected at the cobalt K edge indicates that with the addition of water, the cobalt metal particles were oxidized. X-ray diffraction analysis of CoC X suggests that the formation of amorphous carbon on the surface of the Co particles was suppressed. Cobalt magnetization measurements were performed to study the size and anisotropy of cobalt particles. Thermogravimetric analysis (TGA) data demonstrates that with a water concentration of 7%, the yield increases by about 100% relative to pure ethanol synthesis. Raman and photoluminescence excitation spectroscopic (PLE) data demonstrate that the SWCNT diameter increases (within the diameter range detected) with the water/ethanol ratio. From the statistical SWCNT diameter distribution obtained from transmission electron microscopy (TEM), 65% of the SWCNT synthesized with 20% water in ethanol have diameters larger than 1.5 nm, but there is a severe decrease in yield and a modest decrease in selectivity of SWCNT. The mechanism for the water influence on the yield and diameter of SWCNT is also discussed in Chapter 2.;A more accurate control of diameter and chirality of SWCNT may be obtained from SWCNT templated growth -- `regrowth". The SWCNT were first cut into shorter seeds using different cutting techniques, which will be discussed in chapter 3. After solid state reaction cutting, the SWCNT are 150-500 nm long. And most of the shortened SWCNT after mechanical and liquid-phase oxidative cutting fall in the range of 50-200 nm. The SWCNT length was characterized by performing TEM statistical analysis. The material loss for both methods is about 10 wt%. The functional groups on the tube surface introduced by shortening were removed by refluxing in a soda lime/water suspension. Then, the carbon nanotubes were further annealed by sonicating in ethanol. After annealing, the defect level of shortened carbon nanotubes was reduced significantly, as determined by Raman spectroscopy, Fourier transform infrared spectroscopy and TGA.;The shortened SWCNT seeds were then impregnated with cobalt catalysts and uniformly deposited onto fumed silica, followed by ethanol pyrolysis to initiate regrowth. After regrowth, the regrown SWCNT were characterized by TGA, Raman, TEM and PLE. Some evidence of regrowth may be discovered but further experiments are necessary to prove it. The methodology of both synthesis and characterization will be proposed. Catalyst-free regrowth was also performed and the feasibility was discussed. The regrowth will be discussed in chapter 4. Though the bandgap of pristine SWCNT is determined on the diameter and chirality, doped SWCNT may have a different bandgap with the pristine ones. An appendix presents a demonstration of an in-situ nitrogen doping method for SWCNT through Ethanol Pyrolysis. The existence of nitrogen was characterized by XAS results collected at the carbon K edge. Raman spectroscopy shows that the electronic properties of N-doped SWCNT are different with those of pristine SWCNT. The future direction of the in-situ nitrogen doping method was also discussed in the Appendix.