University of Houston • University of Houston-Clear Lake • ISSO Annual Report Y2002—pp. 33-36

Characterization of Carbon Nanotubes Using Raman Spectroscopy: A Parametric Study of Carbon Nanotube Production

Milko N. Iliev (UH), Alexander Litvinchuk (UH), Carl D. Scott (NASA-JSC), Sivaram Arepalli (GB Tech/NASA-JSC), Pavel Nikolaev (GB Tech/NASA-JSC), and Victor G. Hadjiev (UH)

Abstract
Raman scattering capabilities to characterize the variation of single wall carbon nanotube (SWCNT) material are demonstrated in a parametric study of carbon nanotube production by a double-pulse laser oven process. The effect of various operating parameters on the production of carbon single wall nanotubes is estimated by characterizing the nanotube material using analytical techniques, including SEM, TEM, TGA, and Raman. The study included changing the sequence of the laser pulses, laser energy, pulse separation, type of buffer gas used, operating pressure flow rate, inner tube, and lower flow rates.

DURING THE LAST FEW YEARS, INTEREST HAS GROWN IN single-wall carbon nanotubes (SWCNTs) that can be produced by different processes, including the arc process, laser ablation process, chemical vapor deposition (CVD), and gas phase processes, such as carbon monoxide disproportionation. The laser ablation process developed by Rice University researchers is known to produce the highest percentage of SWCNTs and uses a double-pulse laser oven method. Normal operating conditions for this method utilize a green laser pulse (532 nm) followed by an infrared laser pulse (1064 nm) within a few nanoseconds to ablate a metal-containing graphite target. The target is located in a flow tube maintained in an oven at 1473K with argon flow of 100 sccm (standard cc per minute) at a pressure of 500 Torr. These conditions are important in optimizing the production of single-wall carbon nanotubes and can be used to scale up this process for large-scale commercial use. Previous work, mostly accomplished at Rice University, reported on the effect of oven temperature and briefly mentioned the possible effect of flow conditions. No systematic study had been done earlier.

We present results of a parametric study of this process where several production runs are carried out changing one operating condition at a time for each run. The study of the effects of nine parameters involved (1) changing the sequence of the laser pulses, (2) pulse separation times, (3) laser energy densities, (4) the type of buffer gas used, (5) oven temperature, (6) the operating pressure, (7) flow speeds, (8) inner flow tube diameters, and (9) inter tube materials. To avoid the possible influence of target variation, all the runs are achieved by using the same graphite target. Collected nanotube materials are characterized by a variety of analytical techniques, including scanning electron microscopy (SEM), transmission electron microscopy (TEM), and thermo gravimetric analysis (TGA). We demonstrate that Raman spectroscopy provides an effective and express means for characterization of large varieties of carbon allotropes.

Experimental and Technical Approach
Eighteen production runs were completed for this parametric study for nine variables and run conditions. The same laser target had been used for all these measurements, and the target was polished flat after each day’s run. Each run lasted about 30 minutes, including 20 minutes of spectral data collection (transient, spectral and images with fixed laser beams), followed by ten minutes of production when the laser beams were scanned across the target face. Dispersed spectral data from the plume was collected using quartz fiber optic assemblies, a spectrometer, and an ICCD. The same ICCD is also used to capture plume images during production runs, as described in previous work. After each run, the material produced was collected by scraping from the inside of the flow tube. No attempt was made to remove material from the 25 mm inner tube.

Raman Spectra of the raw samples are measured using a Raman spectrometer equipped with a microscope. A 50X objective is used to focus the laser beam to a spot size of about 2 microns on the sample surface. Raman spectra covering the radial breathing mode (RBM) and tangential G-mode (TM) regions are recorded in static grating scans. This protocol helped us accurately measure the variation of the RBM and TM frequencies in different samples. The Raman spectra are excited by 514.5 nm or 780.6 nm lasers. The 514.5 nm laser line excites resonantly SWNTs with diameter distributions around 1.3+/-0.1, whereas that at 780.6 nm brings to light the tube diameters around 1+/0.1 nm. Throughout the study, the correspondence between the RMB frequency w_RMB for a tube in a bundle and the SWNT diameter d was given according to w_RMB(cm^-1) = 6.5 + 232/d(nm).

Results
The normal oven temperature at 1473 K yielded larger diameter tubes (1.3 nm), while a lower temperature of 1073 K produced smaller (1.0 nm) diameter tubes, as can be seen in Fig. 1.

Buffer Gas
The dependence of the SWCNT yield on buffer gas in the flow tube is shown in Fig. 2. Using helium definitely gave the worst material seen, as confirmed by SEM images and Raman data. In effect, the oxidation temperature was lower and the metal content higher (per TGA). This characterization may be attributed to the reduced plume temperature resulting from better thermal conductivity of helium. The helium run produced smaller diameter tubes (TEM) and also produced Raman spectra indicating more amorphous carbon and defective material. Nitrogen seemed to produce slightly smaller diameter tubes with similar quality as argon.

Gas Flow Rate
It was not easy to change the flow rates without simultaneously slightly affecting the pressure. Flow rate dependence is shown in Fig. 3, where four different flow rates are shown (0, 100, 300, and 500 sccm). The no-flow condition (0 sccm) produced very little material with almost double the metal. The tubes have similar oxidation temperatures in normal run. Runs with faster flows of 300 and 500 sccm seemed to enhance the quality of the tubes (SEM); i.e., their oxidation temperature and decreased metal content. Also, the diameter slightly increased with increasing flows.

Laser Energy Density
The second set of experiments focused on varying laser energy densities of both the laser beams (0.5, 1.5, and 4.5 J/cm²) by changing the size of the laser beams using telescopes in their path; the results are shown in Fig. 4. There is a drastic fall in material yield with lower energy density (0.5 J/cm²) that probably signifies inadequate or decreased ablation. This fall also suggests an ablation threshold for these laser targets, which may vary if the process of making the target is changed. The lower energy case also seemed to yield worse material without much change in diameter distribution. Ahigh energy density run (4.5 J/cm²) yielded almost twice the material, but large (micron size) chunks are seen in the product. This higher energy run, it has been noted, yielded more fullerenes (Raman) with a slightly reduced metal content (TGA).

Laser Pulse Separation
The laser pulse separation study (Fig. 5) did not yield any substantial changes when the pulse separation was changed from the nominal value of 50 nscec to 500 nsec. There was a slight increase in diameter for the 500 nsec case. We have also noted that a pulse separation of the order of several microseconds (earlier work, using 10 Hz system) reduced the yield considerably.

Conclusions
Characterization of the nanotube material procured during the parametric study helped to identify possible effects of the run conditions on the quality and quantity of the SWCNTs. Investigators realized that tube diameter can be changed by either the oven temperature and flow conditions, or both. Impurity levels in the product also seemed to decrease with increased flow rates, reduced pressure, and slightly elevated laser energy conditions. This study also indicated areas of concern that need to be avoided for some of the parameters. Further study is warranted to establish interdependencies of these parameters. Further study is needed particularly on the parameters that control flow conditions.

Publications
Arepalli, S., P. Nikolaev, V. G. Hadjiev, W. Holmes, and C. D. Scott. "A Parametric Study of Single-Wall Carbon Nanotube Growth by Laser Ablation," Appl. Phys. A (2002). (Submitted.)
Yowell, L., B. Files, B. Mayeaux, E. Sullivan, S. Arepalli, P. Nikolae v, O. Gorelik, W. Holmes, and V. G. Hadjiev. "Nanotube Composites and Applications for Human Spaceflight," 53rd International Astronautical Congress, Houston, TX, Oct. 10-19, 2002. 1-6.

Presentations
Gorelick, O., P. Nikolaev, W. Holmes, S. Arepalli, V. G. Hadjiev, R. Devivar, and L. Yowell. "Comparison of Purification Procedures for Carbon Nanotube Materials," NANOSPACE 2002, 5th Annual International Conference on Integrated Nano/Micro/Biotechnology for Space Medical and Commercial Applications, Galveston, TX, June 24-28, 2002.
Hadjiev, V. G., S. Arepalli, P. Nikolaev, and S. Jandl, "Raman Imaging of Single-Wall Carbon Nanotubes: Improving Resolution by Mapping Elementary Excitations," NANOSPACE 2002, 5th Annual International Conference on Integrated Nano/Micro/Biotechnology for Space Medical and Commercial Applications, Galveston, TX, June 24-28, 2002.
Holmes, W., S. Arepalli, P. Nikolaev, V. G. Hadjiev, and C. Scott. "Carbon Nanotube Production by Laser Ablation Process at NASA-JSC," NANOSPACE 2002, Fifth Annual International Conference on Integrated Nano/Micro/Biotechnology for Space Medical and Commercial Applications, Galveston, TX, June 24-28, 2002.

PDF (K)
Table of Contents

Institute for Space Systems Operations - Y2002 Annual Report
Copyright © 2003