Uncontrolled variability in the extinction spectra of C60 nanoparticle suspensions.

To properly investigate the environmental transport, fate, and impact of fullerene C60 nanoparticles (nC60), it is necessary to reproducibly obtain nC60 suspensions and to accurately determine their concentration ([C60]). The results in the present study, however, clearly illustrate that the production of nC60 via extended mixing and via sonication are highly stochastic top-down processes subject to widely divergent end points. nC60 suspensions exhibit variable characteristics (e.g., [C60], average particle size, size distribution, etc.) that make it challenging, if not impossible, to acquire reproducible UV-vis extinction spectra. The mass extinction coefficient, which is the absorptivity of a suspension with [C60] = 1 mM obtained by normalizing UV-vis spectra by the mass concentration of C60 in the suspension, decreases with a given suspension's hydrodynamic diameter, whereas the particle extinction coefficient, which is the absorptivity of a suspension containing one mole of nC60 nanoparticles with the same size distribution as the target suspension and calculated based upon the suspension nanoparticle size distribution, increases with its number weighted average diameter. Other spectroscopic properties of nC60 (e.g., absorbance bandwidth, position of absorption maximum, and relative extinction intensity) also change with average particle size. As a result of the extant variability between samples, when UV-vis spectra are employed to calculate or represent [C60] for fullerene nanoparticle suspensions, extreme care must be taken and other colloidal properties of this suspension must be measured to obtain an accurate result.