Defining the Relationship Between Nanotube Length
and Optical Response
Lynn M. Savage
Single-walled carbon nanotubes have numerous optical properties of interest to physicists and electronics engineers. However, whereas these nanometer-scale constructs have been well-studied regarding the influence of diameter and chirality on their optical properties, little has been understood about the precise effect that nanotube length has on optical response.
Optical absorption spectra for various lengths of single-walled carbon nanotubes with a chirality of (6,5) are shown. AP = as produced (unsorted). Reprinted with permission of the Journal of the American Chemical Society.
Now, after exhausting nearly every photonics-based analytical technique available to them, Erik K. Hobbie and Angela R. Hight Walker of the National Institute of Standards and Technology in Gaithersburg, Md., and their colleagues have obtained batches of single-walled carbon nanotubes, have sorted the particles by length and have measured the optical properties of each set of particles.
They started with batches of single-walled nanotubes grown from several typical synthesis techniques, including laser ablation, electric arc, high-pressure carbon monoxide and, primarily, cobalt-molybdenum-catalyst methods. They purified the particles, wrapped them in DNA strands to encourage dispersion in water and used size-exclusion chromatography to separate them by length into 25 batches, or fractions.
During chromatographic separation, the researchers used a UV-VIS photodiode array made by Waters Corp. of Milford, Mass., and a multiple-angle light-scattering detector from Wyatt Technology Corp. of Santa Barbara, Calif., to record the absorbance and the light scattering of the particles as a function of time, thus obtaining measurements of length.
They further gleaned length data using a dynamic light-scattering device from Brookhaven Instruments Corp. of Holtsville, N.Y., an atomic force microscope made by Digital Instruments (now Veeco Instruments Inc. of Woodbury, N.Y.) and a transmission electron microscope from Philips of Eindhoven, the Netherlands. They determined that atomic force microscopy and dynamic light scattering provided the most robust data, so they used the average of the two sets of measurements to derive the lengths of the different fractions of nanotubes.
Once the particles were separated, the metrologists analyzed the optical responses of fractions number 5 to 15, which comprised nanotubes of approximately uniform lengths of about 50 to 800 nm. According to Hobbie, fractions 1 to 4 and 16 to 25 (batches of nanotubes larger than 800 nm and smaller than 50 nm, respectively) were problematic because particle concentrations fell sharply, or because the amounts of impurities and of free DNA became an issue.
The researchers used several techniques to measure the optical response of the nanoparticles. They used a UV-VIS-NIR absorption spectroscopy system from PerkinElmer Inc. of Fremont, Calif., to gauge the strength of the resonance associated with the particles’ interband transitions. They used a spectrofluorometer made by Horiba Jobin Yvon of Edison, N.J., to characterize the decay of the nanotubes’ fluorescence after the particles were excited with a 450-W xenon lamp.
In addition, they collected Raman spectra using a Ti:sapphire ring laser from Coherent Inc. of Santa Clara, Calif., that was pumped with an argon-ion laser, also from Coherent, coupled to a triple-grating spectrometer from Horiba Jobin Yvon. The scientists collected data at discrete wavelengths with the argon-ion laser by itself as well as at various wavelengths in the 690- to 850-nm range of the ring laser. According to Hight Walker, the Raman spectra provide a measure of the deformation of the lattice of carbon atoms that form the nanotubes, among other characteristics.
“Since all of these attributes vary with length [of the carbon nanotubes], our study was very systematic and exhaustive,” Hobbie said.
The researchers found that the overall optical response of the single-wall nanotubes increases nearly linearly with the length of the particles, with no evidence of saturation at the maximum length that they tested. They believe that this phenomenon results from the localization of a bound electron-electron hole pair, or exciton, along the spine of each nanotube, but, according to Hobbie, a definitive explanation remains to be found.
They now are exploring other methods for sorting nanotubes by length. They also are exploiting the intrinsic near-IR fluorescence of the nanotubes to study how the particles interact with biological systems using optical microscopy; searching for weaker Raman bands that have been predicted but are as yet unobserved; and measuring the decay rate of the nanotubes’ near-IR fluorescence as a function of their length to determine more about the role of excitons.
“[The results] represent one more piece of the puzzle that makes up these complicated and intriguing structures,” Hobbie said. “Our work might lead to a greater effort to manufacture and disperse longer tubes, and might … motivate a deeper understanding of the exact nature of optical excitations in extended one-dimensional systems.”
Journal of the American Chemical Society, Aug. 29, 2007, pp. 10607-10612.
- single-walled carbon nanotubes
- Referred to as SWCNTs, these cylindrical nanostructures composed of a folded sheet of graphene can be used as near-infrared markers for small animal multiplex imaging. Emitting at different wavelengths, which can be tuned in the visible or NIR region, SWNCTs provide photostable, narrow bandwidth fluorescence in a wide chromatic variety. That enables simultaneous imaging of a large number of spectrally separated markers, through the use of global hyperspectral imaging. In addition to being used...
MORE FROM PHOTONICS MEDIA