Size controlled fullerene nanoparticles prepared by guest exchange of γ-cyclodextrin complexes in water

Abstract

Fullerene nanoparticles (nC_x; x = 60 or 70) with a monodisperse size and morphology are obtained through guest exchange of a γ-cyclodextrin (γ-CD) complex. The size of nC_x could be controlled by changing the initial concentration of the C_x–γ-CD complexes.

---

Fullerenes have received a considerable amount of attention owing to their unique physical and chemical properties. They are now recognized to have great potential in various research fields such as electronics, medicine, and biology. Although a fullerene itself is insoluble in water, it forms stable colloidal assemblies in water under certain conditions.^1–5 The relatively high solubility of fullerene nanoparticles, herein termed as nC_x (C_x; x = 60 or 70), in water has raised concerns regarding their potential impact on human and ecological systems. Many researchers have reported the preparation of nC_x using a solvent exchange method.² In this method, nC_x is produced by adding fullerene to an organic solvent such as toluene, tetrahydrofuran (THF), or some combination of organic solvents and then transferring the fullerene into water. nC_x prepared using a solvent exchange method usually forms a crystalline structure and shows monodispersity in size. However, it has been demonstrated that using organic solvents, including THF, in the preparation process of nC_x has toxic effects because the organic solvent is retained within a given nanoparticle.^1,6,7

Several studies have investigated the preparation of nC_x by the direct addition of fullerene to water with ultrasonication,⁸ long-term stirring,¹ laser irradiation,^9,10 and other methods.^7,11–13 Lyon et al. indicated that the nC₆₀ prepared directly in water with long-term stirring is about one order of magnitude less toxic than those prepared through a solvent exchange method using THF.¹⁴ However, as Vikesland et al. also reported, controlling the size or morphology of nC_x by direct preparation in water is difficult and challenging.¹ This might be because the solubility of fullerene is extremely low and the nanoparticles are formed in water through the weathering of large aggregates into smaller nanoparticles over time.

This paper reports the preparation of water-dispersible fullerene nanoparticles by guest exchange of an γ-cyclodextrin (γ-CD) complex (Fig. 1). Fullerenes bicapped with γ-CD (C_x–γ-CD; x = 60 or 70) are mixed with polyethylene glycol (PEG) in water. The fullerenes are expelled from the γ-CD cavity by PEG and immediately aggregate into nanoparticles (nC_x; x = 60 or 70) with monodispersity in size and morphology. nC_x is negatively charged and stably dispersed in water. Moreover, we demonstrate that the size of nC_x can be easily controlled by just changing the initial concentration of the C_x–γ-CD complex.

Fig. 1 Guest exchange of C_x–γ-CD from fullerene to PEG and the formation of nanosized fullerene nanoparticles (nC_x).

The C₆₀–γ-CD and C₇₀–γ-CD complexes were prepared according to a previously described procedure (see ESI† for details).^15,16 To investigate the effect of PEG, PEG (M_w = 2000) was added to a C₆₀–γ-CD aqueous solution ([C₆₀] = 0.2 mM, [PEG] = 50 g L⁻¹) and incubated at room temperature or 80 °C. We first obtained ¹H NMR spectra to examine the structure of the C₆₀–γ-CD complex in the presence of PEG. After the C₆₀–γ-CD mixture with 50 g L⁻¹ of PEG was incubated at room temperature for 1 h, peaks attributed to the C₆₀–γ-CD complex were still evident at 4.1 and 5.0 ppm (Fig. 2a, red line). This indicated that PEG does not affect the structure of the C₆₀–γ-CD complex at room temperature. In contrast, after the C₆₀–γ-CD mixture was heated at 80 °C for 1 h, these peaks completely disappeared (Fig. 2a, blue line). This suggests that the C₆₀–γ-CD complexes were decomposed by the interaction with PEG at 80 °C. C₆₀–γ-CD was mixed with various amounts of PEG ([PEG] = 0–50 g L⁻¹) and heated at 80 °C for 1 h; ¹H NMR spectra revealed that [PEG] = 5.0 g L⁻¹ is enough to decompose 0.2 mM of C₆₀–γ-CD complexes (Fig. S1a, ESI†).

Fig. 2 (a) ¹H NMR (○: C₆₀–γ-CD complex) and (b) absorption spectra of C₆₀–γ-CD aqueous solution ([C₆₀] = 0.2 mM) after incubating at room temperature (red line) or 80 °C for 1 h in the presence 50 g L⁻¹ of PEG.

The mixed solutions of C₆₀–γ-CD complexes with PEG ([C₆₀] = 0.2 mM, [PEG] = 50 g L⁻¹) incubated at room temperature maintained a purple color, which is characteristic of C₆₀ solubilized with γ-CD in water. In contrast, after heating at 80 °C for 1 h the mixed solutions showed a pale yellow color with slight turbidity, implying the formation of nC₆₀.^17–19 In the UV/Vis absorption spectra, the characteristic peak of solvated C₆₀ at 323 nm shifted to 347 nm in the presence of PEG (Fig. 2b). Furthermore, an additional broad absorption at 400–550 nm was also apparent. These are characteristic of solid-state, crystalline C₆₀ and arise from close electronic interactions between adjacent C₆₀ molecules.^17,20 These spectral changes were almost saturated at [PEG] = 5.0 g L⁻¹ (Fig. S1b, ESI†), which agreed with the saturation point observed in the ¹H NMR experiments shown in Fig. S1a (ESI†).

The interaction between CD of various sizes and PEG has been well studied.^21–25 Harada et al. first reported the double-stranded inclusion complexes between γ-CD and PEG.²¹ More recently, Huang et al. investigated the structure of double-stranded complexes more quantitatively and systematically.²⁶ These reports suggested to us that the interaction between γ-CD and PEG was the main force for the decomposition of the C₆₀–γ-CD complex and the formation of nC₆₀.

To confirm this hypothesis, the effect of the polymer structure on the decomposition of C₆₀–γ-CD complex was investigated. We first mixed polyvinylpyrrolidone (PVP) with a C₆₀–γ-CD complex and heated the mixture at 80 °C for 1 h because PVP interacts with C₆₀ and solubilizes it into water.^27,28 In UV/Vis absorption and ¹H NMR spectra, no change was observed in the examined concentration range ([PVP] = ∼20 g L⁻¹, Fig. S2, ESI†). Other water-soluble polymers such as polystyrene sulfonate (PSS) and polyethyleneimine (PEI) also did not induce the decomposition of the C₆₀–γ-CD complex. The concentration of γ-CD also strongly affects the decomposition of C₆₀–γ-CD complex by PEG. Fig. 3 shows the UV/Vis absorption spectra of a centrifuged reaction mixture of 0.2 mM C₆₀–γ-CD complex with 10 g L⁻¹ of PEG after 1 h incubation at 80 °C in the presence of various concentrations of γ-CD. In the presence of 1.3 mM γ-CD, less than 10% of C₆₀–γ-CD remained in the mixed solution after the reaction. Therefore, more than 90% of C₆₀ was expelled from the γ-CD cavity by PEG to form nC₆₀. The ratio of the guest exchange reaction decreased with an increase of the γ-CD concentration and was saturated at 10% at approximately 4 mM of γ-CD (Inset in Fig. 3). Notably, almost 10% of C₆₀–γ-CD decomposed after 1 h incubation at 80 °C even in the absence of PEG. This suggests that almost no C₆₀ was expelled from the γ-CD cavity in the presence of more than 4 mM of γ-CD. These results strongly indicate that PEG interacted with and removed γ-CD from C₆₀–γ-CD complex, and the expelled C₆₀ fullerenes immediately aggregated to form water dispersible nC₆₀.

Fig. 3 UV/Vis absorption spectra of a centrifuged mixed solution of 0.2 mM of C₆₀–γ-CD and 10 g L⁻¹ of PEG in the presence of various amounts of γ-CD. [γ-CD] = 1.3–8.0 mM. The inset is a plot of the rate of reaction with the concentration of γ-CD.

In the case of C₇₀–γ-CD, a color change from orange to dark red was observed after the incubation at room temperature for 1 h in the presence of PEG. Furthermore, the aggregation of C₇₀ and the decomposition of C₇₀–γ-CD were confirmed by UV/Vis absorption and ¹H NMR measurements, as shown in Fig. S3.† It should be noted that the difference in the reaction temperature between C₆₀–γ-CD and C₇₀–γ-CD might be derived from their different stability in water.^15,16,29

The dispersions of nC₆₀ and nC₇₀, prepared from 0.2 mM C_x–γ-CD and 10 g L⁻¹ PEG, exhibited a mean hydrodynamic diameter (intense mean values) of 164 nm and 169 nm, respectively. The diameter of nC₆₀ was in good agreement with the value of 158 nm calculated from UV/Vis absorption spectra (347 nm, Fig. 2b) using the reported formula.^13,20 The polydispersity indexes (PDIs) of nC₆₀ and nC₇₀ were 0.022 and 0.005, respectively. These PDIs were almost 1 order of magnitude smaller than the PDIs for the nC_x prepared in water with ultrasonication or after a few weeks of stirring, and slightly smaller than those prepared by the solvent exchange method.¹ Furthermore, UV/Vis absorption and dynamic light scattering (DLS) measurements revealed that the prepared C₆₀ and C₇₀ aggregates were stable for at least 1 month (Fig. S4, ESI†). It has been reported that water dispersible fullerene aggregates usually shows negative ζ-potential, in which the origin of the negative ζ-potential has yet to be ascertained.^2,30 In our system, C₆₀ and C₇₀ aggregate dispersions showed negative ζ-potential of −31.1 and −22.3 mV, respectively. These clearly indicate that monodisperse fullerene nanoparticles, nC₆₀ and nC₇₀, have been formed through guest exchange of γ-CD complexes.

TEM micrographs of the nC₆₀ and nC₇₀ dispersions prepared from 0.2 mM C_x–γ-CD and 10 g L⁻¹ PEG, are shown in Fig. 4a and b, respectively. Clusters, fairly monodisperse in size, are observed. The average diameter of the individual nanoparticles determined from the TEM images, 152 nm for nC₆₀ and 155 nm for nC₇₀, agrees well with the average hydrodynamic diameter obtained by DLS. The majority of nC₆₀ exhibited lattice fringes, which indicate that the particles are crystalline (Fig. 4c). In addition, diffraction patterns were also observed using selected-area electron diffraction and the pattern is shown in Fig. 4d. From these results, it was found that nC₆₀ maintain their face-centered cubic (fcc) crystalline structure with a lattice constants of 0.82 nm. The shapes of the nC₆₀ and nC₇₀ were spherical or faceted and almost no irregular shapes were detected. In contrast, the majority of the fullerene nanoparticles prepared in water with ultrasonication or long stirring time were irregular in shape.^1,4 The differences in the morphology might be relics of their formation. In the case of the reported nanoparticles prepared in water, it is likely that the nanoparticles formed through the weathering of large aggregates into smaller nanoparticles over time. Conversely, considering the regularity in size and shape in our system, the expulsion of fullerenes from γ-CD by PEG causes nucleation of the fullerenes in water, which results in the formation of crystalline nanoparticles of regular size and shape.

Fig. 4 TEM images of (a) nC₆₀ and (b) nC₇₀. Scale bars are 100 nm. (c) High resolution TEM image and (d) selected-area electron diffraction pattern of nC₆₀. Scale bar is 2 nm.

Solutions with initial C₆₀–γ-CD concentrations of 0.1, 0.2, 0.4, and 0.8 mM with 50 g L⁻¹ of PEG were produced to study the effect of the initial concentration on nC₆₀ formation. After 1 h incubation at 80 °C, the guest exchange reaction completely proceeded in all the solutions, as confirmed by UV/Vis absorption measurements (Fig. S5, ESI†). TEM observations and DLS measurements revealed that the size of nC₆₀ increased when the initial concentration of C₆₀–γ-CD was increased from 0.1 to 0.8 mM (Fig. 5). Furthermore, the PDIs for each dispersion were maintained below 0.02. These results suggest that the size of nC₆₀ can be controlled with a narrow size distribution by using the guest exchange reaction.

Fig. 5 TEM images of nC₆₀ prepared at various initial concentration of C₆₀–γ-CD with 50 g L⁻¹ of PEG. [C₆₀–γ-CD] = (a) 0.1, (b) 0.2, (c) 0.4, and (d) 0.8 mM. Scale bars are 200 nm. (e) Changes of hydrodynamic diameter (blue square, left axis) and number (red square, right axis) of nC₆₀ prepared by guest exchange reaction in the presence of 50 g L⁻¹ of PEG with initial concentration of C₆₀–γ-CD.

The number of nanoparticles in each dispersion was calculated from the average diameter and initial concentration of C₆₀–γ-CD. As shown in Fig. 5e, the calculated number of nanoparticles showed an approximately constant value of 5.0 × 10¹², which did not depend on the initial concentration of C₆₀–γ-CD. It is generally known that the size of a crystal depends on the number of nuclei and the concentration of the supersaturated molecules in solution. In our system, both the nucleation and growth processes are very fast because of the very low solubility of fullerene in water. Therefore, the fullerenes dispelled from γ-CD by PEG form nuclei and grow to nanosized particles immediately so that the number of nanoparticles is constant regardless of the initial concentration of C₆₀–γ-CD. Therefore, the size of the nanoparticle increases linearly with the initial concentration of C₆₀–γ-CD and the size distribution remains narrow even at high concentration conditions.

In conclusion, we have demonstrated the preparation of crystalline fullerene nanoparticles by guest exchange of a γ-CD complex. Fullerenes, bicapped by γ-CD, were expelled from the γ-CD cavity by PEG and aggregated in a crystalline manner with a diameter of 110–340 nm depending on the initial concentration of the C₆₀–γ-CD complex. In this method, no organic solvent was used and the size is controllable, which will be useful for actual medical applications involving photodynamic therapy because the accumulation of colloidal particles in tumor tissue is highly dependent on the residual organic solvent and particle size.¹⁴

Acknowledgements

This work was supported by the Japanese Society for the Promotion of Science (JSPS) KAKENHI Grant-in-Aid for Scientific Research (B) (Grant No. 16H04133), a Grant-in-Aid for Challenging Exploratory Research (Grant No. 16K13982) and the Electric Technology Research Foundation of Chugoku.

Notes and references

1. L. K. Duncan, J. R. Jinschek and P. J. Vikesland, Environ. Sci. Technol., 2008, 42, 173–178.
2. R. G. Alargova, S. Deguchi and K. Tsujii, J. Am. Chem. Soc., 2001, 123, 10460–10467.
3. H. L. Li, Y. W. Zhang, Y. L. Luo and X. P. Sun, Small, 2011, 7, 1562–1568.
4. N. O. McHedlov-Petrossyan, Chem. Rev., 2013, 113, 5149–5193.
5. J. A. Brant, J. Labille, J. Y. Bottero and M. R. Wiesner, Langmuir, 2006, 22, 3878–3885.
6. X. K. Cheng, A. T. Kan and M. B. Tomson, J. Chem. Eng. Data, 2004, 49, 675–683.
7. S. Deguchi, S. Mukai, M. Tsudome and K. Horikoshi, Adv. Mater., 2006, 18, 729–732.
8. W. B. Ko, Y. H. Park and M. K. Jeong, Ultrasonics, 2006, 44, E367–E369.
9. H. Tabata, M. Akamatsu, M. Fujii and S. Hayashi, Jpn. J. Appl. Phys., 2007, 46, 4338–4343.
10. T. Sugiyama, S. Ryo, I. Oh, T. Asahi and H. Masuhara, J. Photochem. Photobiol., A, 2009, 207, 7–12.
11. B. S. Murdianti, J. T. Damron, M. E. Hilburn, R. D. Maples, R. S. H. Koralege, S. I. Kuriyavar and K. D. Ausman, Environ. Sci. Technol., 2012, 46, 7446–7453.
12. D. Iohara, F. Hirayama, K. Higashi, K. Yamamoto and K. Uekama, Mol. Pharm., 2011, 8, 1276–1284.
13. S. Deguchi, S.-a. Mukai, T. Yamazaki, M. Tsudome and K. Horikoshi, J. Phys. Chem. C, 2010, 114, 849–856.
14. D. Y. Lyon, L. K. Adams, J. C. Falkner and P. J. J. Alvarez, Environ. Sci. Technol., 2006, 40, 4360–4366.
15. K. Sugikawa, A. Kubo and A. Ikeda, Chem. Lett., 2016, 45, 60–62.
16. Y. Doi, A. Ikeda, M. Akiyama, M. Nagano, T. Shigematsu, T. Ogawa, T. Takeya and T. Nagasaki, Chem.–Eur. J., 2008, 14, 8892–8897.
17. Y. Ishibashi, M. Arinishi, T. Katayama, H. Miyasaka and T. Asahi, Chem. Lett., 2012, 41, 1104–1106.
18. M. E. Hilburn, B. S. Murdianti, R. D. Maples, J. S. Williams, J. T. Damron, S. I. Kuriyavar and K. D. Ausman, Colloids Surf., A, 2012, 401, 48–53.
19. H. Kato, A. Nakamura, K. Takahashi and S. Kinugasa, Phys. Chem. Chem. Phys., 2009, 11, 4946–4948.
20. X. Chang and P. J. Vikesland, Environ. Sci. Technol., 2011, 45, 9967–9974.
21. A. Harada, J. Li and M. Kamachi, Nature, 1994, 370, 126–128.
22. J. Li, A. Harada and M. Kamachi, Polym. J., 1994, 26, 1019–1026.
23. M.-M. Fan, Z.-J. Yu, H.-Y. Luo, Z. Sheng and B.-J. Li, Macromol. Rapid Commun., 2009, 30, 897–903.
24. J. Xue, L. Zhou, P. He, X. Zhu, D. Yan and X. Jiang, J. Inclusion Phenom. Macrocyclic Chem., 2008, 61, 83–88.
25. C. C. Rusa and A. E. Tonelli, Macromolecules, 2000, 33, 1813–1818.
26. K. Huang, L. Li, J. Wang, Z. Zhou and X. Guo, Colloid Polym. Sci., 2016, 294, 311–319.
27. T. M. Benn, P. Westerhoff and P. Herckes, Environ. Pollut., 2011, 159, 1334–1342.
28. C. Ungurenasu and A. Airinei, J. Med. Chem., 2000, 43, 3186–3188.
29. A. Ikeda, J. Inclusion Phenom. Macrocyclic Chem., 2013, 77, 49–65.
30. J. Brant, H. Lecoanet, M. Hotze and M. Wiesner, Environ. Sci. Technol., 2005, 39, 6343–6351.

---

Footnote

† Electronic supplementary information (ESI) available: Experimetal details and Fig. S1–S5. See DOI: 10.1039/c6ra16513c

---