Understanding the toxicity of carbon nanomaterials, such as carbon nanotubes and graphenes, is important for the development of nanotechnology. Studies have shown that surface redox capability is an important factor for toxicity of carbon nanomaterials. We have measured the surface reductive capacity for a number of carbon nanomaterials in previous studies, but the effects of various engineering processes on surface redox capability have not been investigated until this study. In this study, commercially available carbon black, carbon nanotubes, standard reference materials, fullerenes, graphenes and acetylene soot generated in the lab were used. The carbon nanomaterials were subjected to heating at various temperatures in various atmospheres up to 500 ˚C, and soaking in water at room temperature under various atmospheres, and weathering in the powder form at room temperature under various atmospheres. The redox capability of the carbon nanomaterials was quantified in terms of the reductive capacity towards Fe3+ ions (RCFI). The RCFI values of the asreceived nanomaterials and that of the nanomaterials after various treatments were compared. The carbon nanomaterials were also characterized using x-ray photoelectron spectroscopy (XPS), for understanding the surface chemistry mechanisms of RCFI and the effects of various treatments. In general, heating induced a significant increase in RCFI, regardless of the atmosphere under which the nanomaterials were heated. On the other hand, aging in O2- containing atmospheres brought about significant decrease in RCFI, either in water suspension or in the powder form. Water vapor enhanced the aging effect of O2. CO2 was found to affect the RCFI and the aging of carbon nanomaterials. The extent of RCFI change due to heating or aging was dependent on the type of material. According to the XPS results, the RCFI of some carbon nanomaterials such as carbon black may be correlated with the C-O surface functional groups. However, the definitive correlation between the oxygen-containing surface functional group and RCFI for all carbon nanomaterials couldn’t be determined by the XPS result. This indicates that the RCFI changes of carbon nanomaterials after treatments mainly derived from the factors such as the active sites of edges other than the oxygen-containing surface functional group changes as other studies show. This suggests that the RCFI measurement cannot be replaced by XPS analysis. The effects of heating and aging on RCFI, and more generally the surface redox capability of carbon nanomaterials, reveals that various engineering and environmental processes may significantly change the toxicity of carbon nanomaterials. The findings of this study suggest that it is important to take into account the effects of engineering and environmental processes when assessing the toxicity of carbon nanomaterials.

In this study, a reproducible spectrophotometric method was established and applied to measure reductive capacity of various nanomaterials. Reductive capacity had been implicated in the toxicity of nanomaterials, but a standardized measurement method had been lacking until this work. The reductive capacity of nanoparticles was defined as the mass of iron reduced from Fe3 to Fe2 by unit mass of nanoparticles, in an aqueous solution that initially contained ferric ions. To measure the reductive capacity, the nanomaterials were incubated in a ferric aqueous solution for 16 hours at 37 degrees C, and the reductive capacity of the nanoparticles was determined by measuring the amount of Fe3 reduced to Fe2 using a spectrophotometric method. The reagents 1,10-phenanthroline and hydroquinone were used as a Fe2 indicator and a reducing agent respectively for the assay. To standardize this method, various experiments were carried out. For the initial ferric solution, various Fe salts were tested, and Iron(III) sulfate was chosen as Fe salt for the standard method. The measured reductive capacity of nanoparticles was found to vary with the measurement conditions; the measured reductive capacity increased with increasing the Fe/nanoparticle ratio; the measured reductive capacity increased with incubation time and leveled off after 8 hours of incubation. For hydrophobic materials, the surfactant Tween-20 was added so that the particles could be wetted and suspended in the ferric aqueous solution. After incubation, the particles were removed from the solution by either filtration or centrifugation before applying the spectrophotometric method. In addition, optimal pH and minimum time to reach ultimate color intensity were also found. Carbon-based nanomaterials, standard reference material and metal oxides were measured for their reductive capacities with this method and characterized by transmission electron microscopy (TEM), energy dispersive x-ray spectroscopy (EDS), x-ray diffraction (XRD), BET measurement and Raman spectroscopy. For some nanoparticles, the reductive capacity was measured for both the pristine form and the form treated by oxidization or grinding. All carbon-based nanomaterials, except for pristine C60, have a significant reductive capacity while reductive capacity of metal oxides is very low. And it was found that reductive capacity can be increased by surface functional groups or structural defects and reduced by oxidization or heating (graphitization). The reductive capacity of a material can play an important role in its toxicology by synergistic toxic effects in the presence of transition metal ions through the Fenton reaction. Moreover, even without transition metal ions, the ability of a material to donate electrons can be involved in toxicity mechanisms via generation of reactive oxygen species.