SEARCH

SEARCH BY CITATION

Keywords:

  • Buckminster fullerene;
  • Fullerenol;
  • Singlet oxygen;
  • Superoxide;
  • Hydroxyl radical

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Reactive oxygen species (ROS) are one of the most important intermediates in chemical, photochemical, and biological processes. To understand the environmental exposure and toxicity of fullerenes better, the production and consumption of ROS (singlet oxygen, superoxide, hydrogen peroxide, and hydroxyl radicals) by Buckminster fullerene (C60) and fullerenol were investigated in aqueous systems. Fullerenol exhibits higher photoproduction efficiency of singlet oxygen and superoxide than aqueous suspensions of C60 aggregates (aqu/nC60), and this higher efficiency results in higher steady-state concentrations of these two ROS. Transmission electron microscopy indicates that the C60 molecules in aqu/nC60 are much more closely packed than the C60 cages in fullerenol. These observations provide additional evidence that the lower ROS production efficiency of aqu/nC60 is attributable primarily to efficient self-quenching of C60 triplet states. Production of singlet oxygen by aqu/nC60 is accelerated by increasing oxygen concentration and in part is sensitized by fluorescent photoproducts that accumulate during irradiation. The fullerenes react slowly with singlet oxygen (second-order rate constant <4 × 105 M−1 s−1), but react rapidly with hydroxyl radicals (second-order rate constants of 5.4 × 109 and 4 × 108 M−1 s−1 for aqu/nC60 and fullerenol, respectively). These results show that environmental conditions, including light exposure and oxygen concentration, have the potential to impact the generation of toxic ROS by fullerenes. Environ. Toxicol. Chem. 2012;31:136–143. © 2011 SETAC


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

With the rapid development of nanotechnology and material science, nanomaterials are increasingly being or will be used in a broad variety of products and applications, such as magnetic recording tapes, sunscreens, cosmetics, biolabeling, electroconductive coatings, optical fibers, and medicine. The fact that these novel materials will inevitably enter the environment has raised concerns over their potential negative impact on the environment and human beings.

Fullerenes, including spherical fullerenes, carbon nanotubes (CNT), and their derivatives such as fullerenol, are potential chemical stressors to the environment and human beings. Increasing research efforts in exposure science related to nanomaterials have been focused on assessing exposure of organisms to fullerenes 1–6, and this research is providing not only data that can be used in environmental risk assessments but also data that are being used in the development of cancer therapeutics or bactericides. Many studies indicate that the toxicity of fullerenes is caused by exposure to photogenerated reactive oxygen species (ROS), such as singlet oxygen and superoxide 6–8. At the same time, Buckminster fullerene (C60) and its highly hydroxylated derivative fullerenol were found to protect cells from oxidative stress (free radical damage) by scavenging free radicals 9, 10. Because of the significant roles that ROS play in the toxicity of fullerenes 11, knowledge of ROS production and consumption by fullerenes can greatly enhance the understanding of their positive or negative impact on biological systems. Reactive oxygen species also play roles in the fate and transport of fullerenes in the environment 12, 13.

Fullerenes are well-known photosensitizers, with the ability to produce singlet oxygen (1O2) on irradiation. The formation of ROS by fullerenes typically follows two pathways: electron transfer (type I) resulting in the production of Omath image· and energy transfer (type II) leading to the generation of 1O2. Efficient photoinduced singlet oxygen production was observed on irradiation of molecular C60 in air-saturated organic solvents as a result of energy transfer from C60 triplet states to molecular oxygen 14. Fullerenes form aggregates in water, and these aggregates have photochemical properties different from those of molecular fullerenes. For example, although molecular C60 photoproduces singlet oxygen efficiently, no ROS production was detected from C60 water clusters (aqu/nC60) by measuring the concentration of 1O2 and Omath image· using electron paramagnetic resonance (EPR) 15. This result was attributed to the enhanced decay of excited triplet-state C60 (precursors of ROS) by the formation of aggregates in water. Another study also showed no photosensitized generation of 1O2 and Omath image· in bacterial media and no ROS-mediated damage in nC60-exposed bacteria 5. However, a later study found not only that singlet oxygen was produced in a C60 water suspension irradiated by sunlight but also that its production efficiency increased with irradiation time 16. The production of 1O2 and Omath image· was also detected for γ-CyD bicapped C60 and fullerenol irradiated in D2O using the EPR spin trapping technique with singlet oxygen production quantum yields of 0.76 and 0.08, respectively 6. Compared with the conflicting results for ROS production by C60 aqueous suspensions, results for ROS production by fullerenol are more consistent. Both Pickering and Wiesner 17 and Kong et al. 18 documented that the production of singlet oxygen by fullerenol is pH dependent. The quantum yield of singlet oxygen photoproduction by fullerenol was reported to be 0.10 in acidic solution and 0.05 at neutral and basic conditions 18, consistent with the value determined using an EPR spin trapping technique 6. Aqueous fullerenol more efficiently produced 1O2 and Omath image· than did C60 in aqueous suspension, with heterogeneities in fullerene nanoparticle aggregates believed to be responsible 19–21.

To understand the potential impact of fullerenes on environmental, and especially biological, systems better, we conducted a study that determined apparent quantum yields and compared factors that influence the production and consumption of ROS by aqu/nC60 and fullerenol. In the case of singlet oxygen, we focused particularly on the potential role that the nature of aggregates plays and also on the cause of increases in 1O2 production rates with increasing exposure of aqu/nC60 to light. A second objective of the present study was to examine effects of changing oxygen concentration on the production of singlet oxygen by aqu/nC60 and fullerenol. The third objective was to compare rate constants for reaction of aqu/nC60 and fullerenol with singlet oxygen. A fourth objective was to determine whether aqu/nC60, like fullerenol, photosensitizes the production of superoxide/hydroperoxyl radicals and hydrogen peroxide and, if so, to compare the quantum yields for this process. A final objective was determining rate constants for reaction of the potent oxidant hydroxyl radical (·OH) with aqu/nC60 and fullerenol.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Materials

High-purity C60 (99.9%) and fullerenol (C60(OH)24, acid-precipitated form) used in these experiments were purchased from Materials and Electrochemical Research, and the C60 powder was ground prior to use. Acetic acid, acetonitrile (HPLC grade), hydrochloric acid, sodium hydroxide, sodium nitrate, and sodium phosphate (mono- and dibasic) were purchased from Fisher Scientific. Potassium ferrioxalate was purchased from Alfa Aesar Products. Rose Bengal and methylene blue were obtained from Acros Organics. The source of ammonium acetate and ferrozine, furfuryl alcohol (FFA), hydrogen peroxide (30%, w/w), and superoxide dismutase (SOD; 2,680 units/mg) was Sigma-Aldrich. Ammonium chloride was from J.T. Baker. Horseradish peroxidase (HRP; 268 units/mg) was purchased from Thermo Scientific. Reagent-grade p-hydroxyphenylacetic acid (POHPAA) from Aldrich was purified by repeated recrystallization from ethanol. Sodium molybdate (99.5%) was purchased from Ventron, Alfa Division. Water purified with a Barnstead Nanopure D-7331 system (≥ 18.0 MΩ) was used in the experiments.

Preparation of aqueous suspensions of C60 and fullerenol

Aqueous suspensions of C60 were prepared by mixing ground C60 powder (∼400 mg) with Nanopure water (∼1 L) on a magnetic stirrer for at least four weeks. The stirring was stopped, and the suspension was allowed to sit for 24 h prior to subsequent filtration (0.75-µm glass fiber filters). The filters were prewashed by filtering 1.5 L Nanopure water prior to the filtration of C60 suspension. Filters were also changed after filtration of every 100 ml of C60 suspension to minimize the blockage by C60 crystals accumulated on the filter. A stock suspension of fullerenol was prepared according the methods described by Kong et al. 18 by adding fullerenol powder into Nanopure water and mixing for 24 h. Working suspensions of fullerenol were obtained by dilutions of aliquots from this stock suspension. The aqueous suspensions of C60 and fullerenol were adjusted to neutral pH using 0.05 M pH 7.0 phosphate buffer prior to use.

Irradiation kinetic studies

An Atlas SunTest CPS solar simulator equipped with a 1-kW xenon arc lamp was used for simulated solar irradiations, and a rotating turntable reactor (a merry-go-round reactor [MGRR]) was used for monochromatic irradiations. The temperature was maintained at 22 ± 1°C using a Neslab recirculating water bath (Thermo Scientific). A combination of Corning 0–52 and 7–37 glass filters was used to isolate the 366-nm wavelength from a Hanovia mercury vapor lamp in the MGRR, and a set of Kopp 3480 cutoff filters was use to block wavelengths shorter than 547 nm. The 313-nm line was isolated with an aqueous solution of 1 mM potassium chromate in 2.3% potassium carbonate and a borosilicate glass sleeve as the filter 22. Detailed experimental protocols can be found in Kong et al. 18 and in the Supplemental Data. Aqueous suspensions of C60 scatter light significantly, and the scattering effect had to be taken into account to analyze the kinetic results reliably 23. Apparent quantum yields for production of singlet oxygen and superoxide/hydroperoxyl by aqu/nC60 were computed using procedures described in the Supplemental Data using molar absorption coefficients of aqu/nC60 that were measured with a Shimadzu UV-2450 spectrophotometer equipped with an ISR-2200 integrating sphere attachment (60 mm in inside diameter and with BaSO4 inside coating). To investigate the effect of oxygen on the production of singlet oxygen in the presence of aqu/C60 and fullerenol, irradiations of FFA in oxygen- and air-saturated suspensions were studied in the solar simulator (oxygen concentrations of 1.2 × 10−3 M and 2.5 × 10−4 M, respectively). For comparison, FFA photoreaction in nitrogen-saturated suspensions was studied (>5 × 10−6 M oxygen concentration). To prepare the pure oxygen- or nitrogen-saturated suspension, the suspensions were purged with pure oxygen for 30 min or purged with nitrogen for 1 h prior to simulated solar irradiation.

Thermal generation of singlet oxygen

Singlet oxygen was generated in the dark through the decomposition of aqueous H2O2 catalyzed by MoOmath image24, 25. The protocol was modified from Boreen and Arnold 26. As a brief description, MoOmath image solution (1 mM) containing FFA (100 µM) and fullerenol (3.4 µM) or aqu/nC60 (1.2 µM) was prepared in a pH 10 carbonate-buffered solution. Fifty microliters of 1 M H2O2 was added to 4.95 ml of this solution, and 1.7 ml of 0.5 M NaN3 was added to the mixture at a given time to quench the 1O2 produced from any remaining unreacted reagents. Furfuryl alcohol concentrations in the resulting solutions were analyzed by high-pressure liquid chromatography as described by Kong et al. 18.

Determination of Omath image·/HO2 concentration

The photochemical production of Omath image·/HO2 was estimated using a technique originally developed by Petasme and Zika 27. This procedure involves determining the photoproduction rate of hydrogen peroxide in the presence of SOD, which catalyzes the conversion of Omath image· to hydrogen peroxide.

Hydrogen peroxide that formed during the monochromatic irradiation (366 nm) of aqu/nC60 (4.1 µM, pH 7.0) and fullerenol (3.4 µM, pH 7.0) was quantified using a modified POHPAA method 28. After irradiation, fluorometric reagent was added to 5.00 ml of fullerene aqueous samples to achieve a final concentration of 10 mM Tris buffer, 107 µM POHPAA, and 39 mg/L HRP. The fluorescence of the solution was measured at 315 nm excitation and 400 nm emission using an ISA-APEX Jobin Yvon Fluorolog 3–12 scanning fluorometer. The solution was diluted 10 times, and then 100 µl NaOH (2 M) was added to adjust the pH to 11.5 prior to the fluorescence measurement.

Reaction of OH radicals with fullerenes

Known steady-state concentrations of hydroxyl radicals were photochemically produced using a technique described by Zepp et al. 22. This technique involved irradiation (313 nm) of aqueous suspensions of aqu/nC60 (8.3 µM) or fullerenol (8.3 µM) in an aqueous solution of nitrate (0.53 mM) and 1-octanol (0.11 mM) that was phosphate-buffered at pH 7.1.

Excitation–emission matrix spectra of fullerenol

Excitation–emission matrix (EEM) scans were performed by using an ISA-APEX Jobin Yvon Fluorolog 3–12 scanning fluorometer equipped with a double-grating spectrometer in the emission position. Both excitation and emission slits were set to produce 5-nm bandwidths. High-resolution scans were performed from 300 to 600 nm excitation (5-nm intervals) and 350 to 700 emission (5-nm intervals). Scans were corrected for instrument configuration and converted to quinine sulfate equivalents. Rayleigh and Raman scattering peaks were eliminated during postprocessing of data in Matlab (MathWorks) using the Fluorescence Toolbox (version 2.0) software 29.

Transmission electron microscopic analysis

Transmission electron microscopic (TEM) images were obtained using a Hitachi 7600 transmission electron microscope operated at 120 kV. A small amount of aqu/nC60 or fullerenol aqueous suspension was dropped onto a carbon-coated 200 mesh copper TEM grid, followed by drying in air on a 70°C hot pan for 30 min prior to TEM analysis.

Chemical analysis

Analysis of C60 concentrations in the aqueous suspensions was accomplished by first extracting the C60 into toluene with a water suspension to toluene ratio of 1:1. Aqueous 1 M manganese perchlorate (100 µl) was added to the water suspension to enhance the C60 partitioning into the toluene phase. The mixture was then vortex mixed for 2 min, and the toluene layer (upper layer) was transferred to an amber glass vial (1.5 ml) for determination of the C60 concentration. The C60 was analyzed using a Dionex Ultimate 3000 HPLC system with a mobile phase of acetonitrile:toluene = 20:80 at a wavelength of 333 nm 30. The fullerenol concentration was determined using a PerkinElmer Lambda 35 UV/vis Spectrometer. The FFA was measured using the HPLC system with a mobile phase of acetonitrile:water (30:70) at a wavelength of 209 nm 18.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Singlet oxygen photoproduction from aqu/nC60 and fullerenol

Singlet oxygen is the lowest excited state of molecular oxygen and is one of the most active intermediates in chemical and biochemical reactions. To quantify the production of 1O2 by fullerenes, aqu/nC60 suspensions or fullerenol in aqueous media (∼pH 7) were irradiated by monochromatic light (366 nm) or simulated solar radiation in the presence of the singlet oxygen scavenger FFA (100 µM) in air-saturated systems. The loss of dilute FFA in the presence of singlet oxygen photosensitizers typically follows pseudo-first-order kinetics, with the rate constant proportional to the rate of singlet oxygen production 31. Prior research by Kong et al. 18 on fullerenol showed the highest 1O2 production quantum yield of 0.10 under acidic conditions (pH 4.6) and a lower 1O2 production quantum yield of 0.05 under neutral to basic conditions. Our previous results also indicated that the chemical reactions involving diffusive encounters between 1O2 or Omath image· and fullerenol are too slow to contribute significantly to the fast component of fullerenol photoreactions in sunlight. In the current study, production of 1O2 or Omath image·/HO2 by aqu/nC60 was evaluated and compared with the previous fullerenol results.

Unlike the FFA first-order photooxidation kinetics observed in fullerenol aqueous suspension (inset in Fig. 1a), the FFA photooxidation kinetics in the presence of aqu/nC60 (Fig. 1a) did not strictly obey first-order kinetics. The slope of first-order plots of the data (ln[C0/C] vs time where C0 and C are the concentrations at time zero and time t, respectively) increased with irradiation time, which is consistent with previous observations 16. This phenomenon was attributed to the formation of hydroxylated C60 photoproducts, which may possess a stronger capability to produce singlet oxygen. To verify this finding further, FFA photooxidation kinetics was also compared in the presence of nonirradiated aqu/nC60 and of aqu/nC60, which had been exposed to simulated solar irradiation for 3 d prior to the addition of FFA. Furfuryl alcohol was photooxidized more rapidly in the preirradiated aqu/nC60 than in nonirradiated aqu/nC60 (Fig. 1b), indicating that preirradiation enhanced photoproduction of singlet oxygen by aqu/nC60.

thumbnail image

Figure 1. (a) Furfuryl alcohol (100 µM) oxidation kinetics under monochromatic irradiation (366 nm) in aqueous suspension of Buckminster fullerene aggregates (aqu/nC60) (4.1 µM; circles) or fullerenol (3.4 µM; triangles); inset shows FFA oxidation data in the presence of fullerenol. Lines represent the first-order kinetics fitting to experimental data. (b) Simulated solar irradiation in the presence of nonirradiated aqu/nC60 (solid circles) and irradiated aqu/nC60 (open circles) after 3 d of irradiation.

Download figure to PowerPoint

Previous studies have demonstrated the formation of water-soluble photoproducts with oxygen-containing functional groups during the phototransformation of aqu/nC6023, 32. The present study found new evidence to support this mechanism by first extracting the irradiated aqu/nC60 with toluene, then measuring the fluorescence EEM of the water layer after the extraction. Prior to irradiation, no fluorescence signal was observed in the water layer of aqu/nC60 after toluene extraction. The water layer of the irradiated aqu/nC60 after extraction displayed an EEM spectrum very similar to that of irradiated fullerenol aqueous suspension, as shown in Figure 2a,b. This result implies that the formation of oxygen-containing or hydroxylated C60 intermediates may occur during the aqu/nC60 photolysis. Results of the present study showed increased efficiency of singlet oxygen production by aqu/nC60 with increasing conversion, which likely is caused by the formation of hydrophilic intermediates such as fullerenol that are more capable of producing singlet oxygen than aqu/nC60.

thumbnail image

Figure 2. Comparison of excitation–emission matrix (EEM) spectra (contour view) of aqueous suspension of Buckminster fullerene aggregates aqu/nC60 and fullerenol following simulated solar irradiation. Irradiation times are shown on the graphs; t0 denotes the EEM at time zero. (a) EEM spectra of the water layer of aqu/nC60 extracted with toluene. (b) EEM spectra of fullerenol suspension.

Download figure to PowerPoint

For the comparison with other fullerenes, first-order kinetics were assumed for the analysis of FFA photooxidation kinetics in aqu/nC60. Under the assumption that the illumination intensity is constant and the concentration of FFA is low enough (<200 µM), the steady-state concentration of 1O2 ([1O2]ss) can be estimated using the equation

  • equation image(1)

where kobsd, FFA is the pseudo-first-order rate constant of FFA photooxidation, and kr, FFA is the rate constant of the reaction between FFA and 1O2 (1.2 × 108 M−1  s−1) 33. Under simulated solar irradiation, the steady-state concentration of 1O2 was computed to be 5.4 × 10−14 M for fullerenol (3.4 µM) and 1.1 × 10−14 M for aqu/nC60 (4.1 µM) in neutral aqueous media. Quantum yields for singlet oxygen production by the fullerenes were determined using monochromatic irradiation with 366-nm light. Estimation of apparent quantum yields in irradiated aqu/nC60 is complicated by the fact that light absorbance is attributable to a combination of absorption and scattering by the aggregates, but absorption coefficients are needed for the computation of apparent quantum yields 23. The molar absorption coefficient at 366 nm (1.74 × 104 M−1 cm−1) was determined using a Shimadzu UV-2450 spectrophotometer equipped with an ISR-2200 integrating sphere attachment. The scattering albedo for the aqu/nC60 suspension (4.1 µM), which is defined as the ratio of the scattering coefficient to the total attenuation coefficient, was determined to be 0.3 at 366 nm. That is, only 70% of the incident light attenuated by the aqu/nC60 was absorbed, and the other 30% was scattered. Light scattering had a negligible effect on absorbance in the fullerenol aqueous suspension. By using the molar absorption coefficients measured with the integrating sphere, the apparent quantum yield of 1O2 production by aqu/nC60 was estimated to be 0.015 (see Supplemental Data for detailed calculation) with 0.05 for fullerenol in neutral aqueous media 18. The difference in singlet oxygen production between aqu/nC60 and fullerenol can be explained in part by the fact that they form different structured aggregates in water. As shown in the TEM images (Fig. 3), aqu/nC60 aggregates are tighter assemblies than those formed by fullerenol. This result is in agreement with the experimental observations and theoretical prediction of Hotze et al. 19, 34 in which the nearest-neighbor triplet C60 quenching was believed to be particularly efficient in more closely packed aqu/nC60 aggregates, and this efficient quenching resulted in a lower production of singlet oxygen. In addition, this estimated apparent quantum yield in the case of aqu/nC60 was enhanced by the production of 1O2 by photoproducts. This phenomenon partially offset the decrease related to aggregate structure.

thumbnail image

Figure 3. Transmission electron microscopic (TEM) images of Buckminster fullerene aggregates (left) and fullerenol aggregates (right). A small amount of the aqueous suspensions of the aggregates was dropped onto a carbon-coated 200 mesh copper TEM grid, followed by drying in air on a 70°C hot pan for 30 min prior to TEM analysis.

Download figure to PowerPoint

Dependence of 1O2 photoproduction on oxygen concentration

To understand the mechanism of singlet oxygen production in the presence of aqu/nC60 and fullerenol better, we conducted FFA oxidation experiments under simulated solar irradiation using oxygen- and air-saturated suspensions of aqu/nC60 (1.2 × 10−3 and 2.5 × 10−4 M oxygen, respectively). For comparison, we also examined the photodegradation of FFA in a nitrogen-saturated suspension (estimated to be <5 × 10−6 M oxygen). Furfuryl alcohol photooxidation was oxygen dependent and increased with increasing oxygen concentration (Fig. 4). The comparatively slow photochemical loss of FFA in nitrogen-saturated aqu/nC60 could be attributable to oxygen-independent photoproduction of reactive transients by aqu/nC60 or its photoproducts or, alternatively, to photooxidation mediated by trace amounts of oxygen in the nitrogen-outgassed suspension. The rate of FFA loss increased slightly with irradiation time in all these systems (Fig. 4a), whereas FFA photooxidation displayed first-order kinetics in the presence of fullerenol (Fig. 4b). This observation further confirms the involvement of aqu/nC60 photoproducts in the photosensitized reactions of FFA. In the case of air- or oxygen-saturated systems, the FFA photooxidations likely were mediated by production of singlet oxygen by energy transfer from C60 triplets.

thumbnail image

Figure 4. Effect of oxygen concentration on photooxidation kinetics of the singlet oxygen probe, furfuryl alcohol (FFA), in an aqueous suspension of Buckminster fullerene aggregates aqu/nC60 (4.1 µM; a) and fullerenol under simulated solar irradiation (b). Symbols are experimental data for oxygen-saturated conditions (triangles), for air-saturated conditions (solid circles), and for nitrogen-saturated conditions (open circles). Lines in (a) are connections between experimental data, and lines in (b) are corresponding linear fitting curves.

Download figure to PowerPoint

Reaction of 1O2 with aqu/nC60 and fullerenol

To examine the reaction of singlet oxygen with C60 and fullerenol, it was necessary not only to provide an independent source of 1O2, but also to prevent or minimize the production of 1O2 by these fullerenes. Under dilute conditions, the rate constant (kr,FUL) for reaction of the fullerenes with singlet oxygen can be computed using the equation

  • equation image(2)

where kobsd, FUL is the first-order rate constant for oxidation of the fullerene.

In the present study, two techniques were used to determine the reaction between fullerenes and singlet oxygen. Aqueous suspensions of C60 aggregates, fullerenol, and FFA were present in the system at low concentrations. In the first technique, aqu/nC60 or fullerenol was amended with Rose Bengal, a widely used photosensitizer, and subsequently irradiated under the visible light produced by filtration (cutoff at 546 nm) of radiation from a mercury lamp in the MGRR. The average absorption coefficient of Rose Bengal (3.4 µM) over the wavelength range of 567 to 578 nm was about 0.032 cm−1, nearly tenfold higher than that of fullerenol (3.4 µM). Given that Rose Bengal has a higher quantum yield for 1O2 production than fullerenol (0.75 for RB vs. 0.05 for fullerenol at pH 7) 18 and that it had a much higher light absorption in the evaluated system, Rose Bengal clearly was the predominant source of 1O2. Therefore, the possible self-photosensitization of fullerenol can be neglected. No degradation of fullerenol was observed in the absence of Rose Bengal, but a slight loss of fullerenol (<10%) was measured in the presence of Rose Bengal over a 5-d period with a rate constant (kobsd, FUL) of ∼2 × 10−7 s−1. In the presence of Rose Bengal (3.4 µM), kobsd, FFA at 578 nm was 9.2 × 10−5 s−1 (Supplemental Data, Fig. S1). Therefore, the rate constant for the fullerenol reaction with singlet oxygen (kr, FUL) was 2.6 × 105 M−1 s−1, nearly three orders of magnitude lower than the rate constant between FFA and water (1.2 × 108 M−1 s−1). As for the aqu/nC60 aggregates, which had higher light absorption in the visible spectral region, with an average absorption coefficient of 0.14 cm−1 from 567 to 578 nm, no difference in the degradation kinetics in the absence and presence of Rose Bengal was observed.

In the second technique, singlet oxygen was generated thermally in the absence of light by the MoOmath image-catalyzed decomposition of aqueous H2O226. The first-order rate constant of FFA degradation (kobsd, FFA) under these conditions was 6.3 × 10−4 s−1 (Supplemental Data, Fig. 2S). No observable loss of fullerenes was found in either aqu/nC60 or fullerenol aqueous suspensions over a 6-h period (data not shown). Given that we could detect 5% loss of the fullerenes, this result indicates that the maximum oxidation rate constant for the fullerenes was 2.2 × 10−6 s−1. Using Equation 2 and the known values of kr, FFA, we compute that the rate constants for reaction of aqu/nC60 or fullerenol with singlet oxygen are less than 4.2 × 105 M−1 s−1. This result is consistent with the low reactivity estimated from the Rose Bengal studies. However, these findings do not necessarily rule out the possibility that aqu/nC60 or fullerenol could react with photochemically generated 1O2 within the clusters of these fullerene aggregates where steady-state concentrations of 1O2 may be substantially higher than the concentrations produced in bulk water by these external sources of 1O2.

Photoproduction of hydrogen peroxide and superoxide/hydroperoxyl

The reduction of molecular oxygen can result in the generation of superoxide anion radicals, a highly toxic oxygen species, especially in the presence of transition metal ions. Superoxide production by fullerenol has been described elsewhere 18. In the present study, the aqu/nC60 was irradiated under the same conditions (366 nm); the production of superoxide and its conjugate acid hydroperoxyl radicals (Omath image·/HO2) was also monitored by measuring H2O2 produced by reactions of Omath image·/HO2. As previously observed in fullerenol aqueous suspensions, the production rate of the hydrogen peroxide in aqu/nC60 increased on addition of SOD as a result of the catalysis of dismutation of superoxide into oxygen and hydrogen peroxide (Fig. 5). Based on the production rate of hydrogen peroxide in the presence of the SOD, the apparent quantum yield of superoxide production was estimated to be 9.3 × 10−5 for aqu/nC60 (the quantum yield was 6.2 × 10−4 for fullerenol as previously reported 18). The efficient quenching of the excited state precursors of superoxide/hydroperoxyl within aqu/nC60 aggregates can also help to account for the much lower superoxide quantum yield for aqu/nC60. Alternatively, the much higher production rate observed for fullerenol has also been attributed to photoreactions mediated by hemiketal moieties in the fullerenol cage. In environmental and biological systems, the production of superoxide could be facilitated by fullerenes via radical anions formed by photoreduction of the fullerene cage that could shuttle electrons to oxygen. Electron donors such as NADH and humic substances are present in these systems.

thumbnail image

Figure 5. Hydrogen peroxide production during the monochromatic irradiation (366 nm) of an aqueous suspension of Buckminster fullerene aggregates aqu/nC60 (circles) and fullerenol (triangles) at 366 nm. Solid and open symbols represent the data in the presence and absence of superoxide dismutase, respectively. Lines represent the first-order kinetics fitting to experimental data. Bars represent ± one standard deviation.

Download figure to PowerPoint

Reaction of aqu/nC60 and fullerenol with hydroxyl radicals

The hydroxyl radical is highly reactive and short-lived, and this reactive oxygen species is believed to initiate lipid peroxidation, which can result in cell damage or death via the impairment of the cell membrane. Pristine Buckminster fullerene is reported to be a good free radical scavenger or is referred to as a “free radical sponge” 9, 35, 36 and therefore protects the biological system from radical attack. A previous toxicity study indicated that the aqu/nC60 was not capable of quenching free radicals resulting from the formation of aggregates, because C60 must be in solution in order to scavenge radicals 9. However, a recent study 37 observed the reaction between aqu/nC60 and hydroxyl radicals generated by γ-radiation using pulse radiolysis in oxygen-free systems, with an overall slow conversion that was postulated to be due to buildup of unreactive intermediates. In the present study in air-saturated water, production of ·OH from photolysis of nitrate (Fig. 6) was estimated by using a previously described method in which the steady-state concentration of ·OH was manipulated by addition of the scavenger 1-octanol 22. The second-order rate constant k (L mol−1 s−1) and steady-state ·OH concentration ([·OH]ss) in this system could be described as

  • equation image(3)

where kw (1.44 × 104 s−1) and koct (6 × 109 L mol−1 s−1) are the rate constant for decay of ·OH in water and the second-order rate constant for reaction of 1-octanol with ·OH, respectively 22. [OCT] is the concentration of 1-octanol (0.1063 mM). The rate of ·OH photoproduction from nitrate, v, equates the specific light absorption rate (ka,λ) times the quantum yield for production of ·OH by nitrate (ΦOH,313 nm = 0.015) and nitrate concentration ([NOmath image]) as follows

  • equation image(4)
thumbnail image

Figure 6. Phototransformation kinetics (313 nm) of aqueous suspensions of Buckminster fullerene aggregates (aqu/nC60) and fullerenol in an aqueous nitrate (0.53 mM) and 1-octanol (0.11 mM) solution that was phosphate buffered at pH 7.1. Solid circles and open circles represent the data for aqu/nC60 and fullerenol, respectively. Lines represent the first-order kinetics fitting to experimental data.

Download figure to PowerPoint

The calculated values of second-order rate constants are 5.4 × 109 and 4 × 108 L mol−1 s−1 for aqu/nC60 and fullerenol, respectively. Hydroxyl radicals reacted readily with aqu/nC60, most likely to form hydrophilic products 23, but reaction with fullerenol was much slower. A recent study using pulse radiolysis to generate ·OH, concluded that the rate constant for reaction of C60 aggregates with ·OH was 7.3 × 109 L mol−1 s−1, approximately 20% higher than the rate constant estimated here 37. Given that different methods were used to produce the C60 aggregates and to estimate the rate constants in these studies, the results are in satisfactory agreement.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Both Buckminster fullerene and fullerenol can photosensitize the production of singlet oxygen, superoxide ions, and hydrogen peroxide under environmentally relevant conditions. Aqu/nC60 generally exhibited lower photoproduction efficiencies of singlet oxygen and superoxide than fullerenol, most likely as a result of the different structures of the aqu/nC60 aggregates and consequently an efficient self-quenching of C60 triplet states within the aggregates. This reduction in photochemical efficiency caused by the structure of C60 aggregates in water is partially offset by the production of hydroxylated, hydrophilic aqu/nC60 photoproducts that sensitize the production of singlet oxygen. This latter process results in increases in 1O2 production efficiencies with increasing light exposure. Both aqu/nC60 and fullerenol react very slowly with singlet oxygen. However, aqu/nC60 scavenges ·OH radicals at close to a diffusion-controlled rate and more than 12-fold more rapidly than fullerenol. Taken together, these results indicate that sensitivity to light exposure, oxygen concentration, and other environmental conditions are likely to affect greatly the photosensitization of ROS and ROS scavenging properties of fullerenes and ultimately affect their roles in other physicochemical and biological processes.

SUPPLEMENTAL DATA

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Detailed protocols of irradiation kinetics and calculation of apparent quantum yields of 1O2 production, FFA photooxidation in the presence of Rose Bengal at 578 nm, and oxidation of FFA by thermally generated singlet oxygen (147 KB DOC).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

This article has been reviewed in accordance with the U.S. Environmental Protection Agency's (U.S. EPA) peer and administrative review policies and approved for publication. Mention of trade names or commercial products does not constitute an endorsement or recommendation for use by the U.S. EPA. We thank Dr. Haijun Qian at the Electron Microscope Facility at Clemson University for conducting the transmission electron microscope analysis.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSIONS
  7. SUPPLEMENTAL DATA
  8. Acknowledgements
  9. REFERENCES
  10. Supporting Information

All Supplemental Data may be found in the online version of this article.

FilenameFormatSizeDescription
etc_711_sm_SupplData.doc147KSupplementary Data

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.