Spatial Fluctuation Fluorescence Spectroscopy: A New Method for Studying Single-Walled Carbon Nanotube Dispersions

We describe a novel spectroscopic method that has been applied to characterize liquid suspensions of single-walled carbon nanotubes (SWCNTs). When fluorescence spectra are sequentially measured from small volumes of very dilute samples, the intensities of emission features can fluctuate significantly because relatively few nanotubes of each type are present within the observation volume, and these concentrations vary statistically from observation to observation. By restricting acquisition times to ~200 ms and slowly flowing the sample so that a different region is sampled by each spectrum, we can observe such spatial concentration variations as spectral intensity fluctuations. Using observation volumes of only 10^-7 mL and sample mass concentrations down to ~60 ng/mL, we acquire sequences of 3000 spectral snapshots and then analyze the set of data. At each emission wavelength, we compute the mean intensity and the variance (squared standard deviation) of intensity. The ratio of squared mean intensity to variance is then plotted as a function of wavelength to give an abundance spectrum, reflecting the number of nanoparticles contributing to emission at that wavelength.

Spatial fluctuation spectroscopy thus gives relatively direct access to the absolute (n,m) composition of SWCNT samples. Another benefit of this technique is that emission features from different (n,m) SWCNTs are better resolved in the variance spectrum than in the squared mean emission spectrum, allowing clearer detection of species with overlapping fluorescence spectra. In addition, the mean intensity spectrum can be divided by the abundance spectrum to obtain the relative fluorescence efficiency per nanotube for all of the (n,m) species represented in the data. These values can be used to validate the calibration factors that are central to conventional quantitative fluorimetric analysis of mixed SWCNT samples.

Furthermore, we have analyzed the spectral data sets to find correlations between signal fluctuations at different wavelengths. These are interpreted as reflecting the spatial correlations of nanotubes of different types, which implies the presence of aggregates. Studies will be reported showing how these cross-correlations depend on sample condition and processing. Spatial fluctuation spectroscopy appears to be a powerful new tool for analyzing and studying SWCNT suspensions.