Hybrid materials with an increased resistance to hard X-rays using fullerenes as radical sponges

The protection of organic and hybrid organic–inorganic materials from X-ray damage is a fundamental technological issue for broadening the range of applications of these materials. In the present article it is shown that doping hybrid films with fullerenes C₆₀ gives a significant reduction of damage upon exposure to hard X-rays generated by a synchrotron source. At low X-ray dose the fullerene molecules act as `radical scavengers', considerably reducing the degradation of organic species triggered by radical formation. At higher doses the gradual hydroxylation of the fullerenes converts C₆₀ into fullerol and a bleaching of the radical sinking properties is observed.

1. Introduction

Doping of hybrid organic–inorganic materials with fullerenes and fullerene derivatives (Prato, 1997) is a well exploited route for producing different types of functional materials (Brusatin & Innocenzi, 2001; Innocenzi & Brusatin, 2001). Sol–gel chemistry has been used for introducing fullerene molecules into a large variety of oxides, mesoporous (Innocenzi et al., 2004) and hybrid materials (Signorini et al., 2001) to exploit the functional properties of fullerenes. In particular, non-linear optical properties of fullerenes, such as optical limiting via reverse saturable absorption, have allowed the production of devices for photonics. In general, fullerene derivatives have shown to be a better solution for sol–gel processing of materials, because the synthesis design can allow fullerenes to be obtained with a higher solubility and the capability of being integrated into the hybrid network via sol-to-gel reactions. An example are the fulleropyrrolidines which have a silicon alkoxide end-group covalently linked to the fullerene; these molecules are characterized by the presence of a solubilizing appendage together with a silicon alkoxide end-group (Maggini et al., 1999). Among the different properties of fullerenes and its derivatives, two have attracted attention for developing applications (Innocenzi & Brusatin, 2001); the first one is optical limiting (Signorini et al., 2001) and the second one is the capability of fullerenes to work as a radical sponge (Krusic et al., 1991; McEwen et al., 1992). The latter, in particular, has nowadays allowed the development of fullerene products for the market; C₆₀-based skincare cosmetics with anti-radical properties, for instance, are currently commercialized (see, for instance, products of Vitamin C₆₀ BioResearch Corporation: https://www.vc60.com/ ). Fullerene and its derivatives have shown to be excellent antioxidant molecules and radical scavengers under UVA irradiation; this allows C₆₀ and its derivatives to be used as a preventive agent for skin degradation (Xiao et al., 2006). Fullerenes can sponge up and neutralize 20 or more free radicals per molecule with better performances than vitamin C and vitamin E. The radical scavenging property is due to the electronic structure of C₆₀ which has several conjugated double bonds and a low-lying lowest unoccupied molecular orbital (LUMO) which can easily take up an electron, capable of reacting with radicals. This property envisages that fullerene and its derivatives could also be successfully applied to prevent plastic degradation or more in general in developing new types of UV-resistant organic materials. The radical sponge property of fullerenes could also be exploited at high energy ranges, such as the X-ray and γ-ray ranges. While the application of fullerenes to biosystems has been largely studied, so far the possibility of developing materials with enhanced capability of resistance to UV light or even higher energy radiation has almost been ignored. We have, therefore, doped a hybrid material with fullerene molecules and studied the effect upon exposure to hard X-rays produced by synchrotron radiation, quite extreme conditions for organic or hybrid materials.

2. Experimental

Phenyltriethoxysilane (PhTES), fullerene (C₆₀, 98%), hydrochloric acid (HCl) 1 M, 2-propanol and toluene were purchased from Aldrich and used as received; p-type boron-doped silicon wafers were employed as the substrates. According to previous experiments (Falcaro et al., 2011), the precursor sol was prepared adding 24 cm³ of PhTES into a mixture of 48 cm³ of toluene and 24 cm³ of 2-propanol; after stirring for 10 min at reflux temperature (∼343 K), 1.5 cm³ of HCl 1 M was added dropwise. The sol was kept under stirring for 4.5 h in a flask under reflux conditions and then doped with different amounts of fullerene C₆₀. To prepare the C₆₀-doped phenylsilane coating, increasing amounts of fullerene C₆₀ (1, 2.5 and 4 mg) were dispersed in 2 cm³ of toluene; after complete solubilization by sonication, the solution appeared violet and clear. The C₆₀ solutions were then added to the hydrolyzed sol and stirred for 10 min. The films were prepared by spin coating (400 rev min⁻¹) and top casting on a silicon wafer to obtain films with two different thicknesses, roughly 1.5 and 200 µm, as estimated by a digital micrometer.

Fourier-transform infrared (FTIR) measurements were performed using a Vertex 70 Bruker spectrometer in the 400–4000 cm⁻¹ range with a resolution of 4 cm⁻¹ on films deposited on a silicon wafer; silicon was used for background. To analyze the dataset, the spectra were cut in the 950–840 and 1420–1440 cm⁻¹ regions, after which the baseline was corrected by subtracting a concave rubber-band using OPUS 6.5 software with two iterations. The amount of residual silanol was estimated by integrating the spectra in the 950–840 cm⁻¹ range after subtraction of the concave rubber-band baseline.

A Raman microscope was used to observe the formation of aggregates in the films. A Bruker Senterra confocal Raman microscope working with a laser excitation wavelength of 785 nm at 100 mW of nominal power was used for optical microscopy (100× magnification) and Raman spectroscopy analysis. The spectra were at first cut in the 1500–1420 cm⁻¹ range, then a baseline was subtracted by using a concave rubber-band with ten iterations. The films were exposed to hard X-rays using the Deep X-ray Lithography beamline at the Trieste synchrotron facility (Italy). The films were irradiated with different doses by changing the exposure time (200, 400, 800 s). The films were exposed to energies per unit area incident to the sample surface corresponding to 1053.8, 2107.6 and 4215.2 J cm⁻². The energy per unit area is equal to the exposure time multiplied by a factor of 7.53, which is the power per unit area incident on the sample surface.

3. Results and discussion

The hybrid organic–inorganic material that we have selected is prepared using phenyltriethoxysilane as precursor, an organically modified alkoxide bearing a phenyl group which is covalently bonded to silicon; this precursor allows optically transparent and thick films to be obtained via spin coating or top casting. Layers several micrometres in thickness have been used to perform the experiments. Although the film is the best geometry for exposing the materials to X-rays, the detection of fullerenes in the hybrid matrix could be difficult to achieve if the coatings are not thick enough. The material has been doped with increasing amounts of fullerenes up to the solubility limit and the changes induced by X-ray exposure (Bhatta et al., 2009; Weon et al., 2010; Nygård et al., 2010) have been studied combining FTIR and Raman spectroscopy.

The effect of X-rays on fullerene has been observed, in particular, using Raman spectroscopy because C₆₀ has a quite well resolved and sharp band that can be used as a reference to check the changes induced by X-ray exposure. Fig. 1(a) shows the Raman spectra of samples doped with 4 mg of fullerene after exposure to hard X-rays. The samples have been exposed to increasing doses, 0 (not exposed; solid line), 1054 (dashed line), 2108 (dotted line) and 4215 J cm⁻² (dash-dotted line). The Raman spectra of an undoped sample and pure C₆₀ are also reported as a reference (Fig. 1b). At 1467 cm⁻¹ the typical signature of the fullerene pentagonal pinching mode, Ag(2) (Kuzmany et al., 2004), is clearly visible (dotted line) whereas the undoped sample does not exhibit any Raman active band in the same range (solid line). The fullerene Raman signal does not change in intensity in the samples exposed to 1054 and 2108 J cm⁻², whereas it is reduced to around half in the film exposed to the highest dose. Upon exposure to X-rays, a change in the Raman band has also been observed with an enlargement of the full wave half-maximum and a shift of the maximum towards lower wavenumbers. In accordance with the literature, these findings indicate a C₆₀ polymerization to form dimers or trimers (Cataldo, 1993); this effect is mainly induced by the secondary electrons produced by irradiation (Klesper et al., 2005; Zhao et al., 1994). Upon exposure to hard X-rays, in fact, the fullerenes can be excited by secondary electrons allowing addition reactions with other C₆₀ molecules.

Figure 1 (a) Raman spectra, in the 1480–1455 cm^−1 range, of 4 mg fullerene-doped samples after exposure to increasing X-rays doses: not-exposed (solid line), 1054 J cm^−2 (dashed line), 2108 J cm^−2 (dotted line) and 4215 J cm^−2 (dash-dotted line). (b) Raman spectra of an undoped sample (solid line) and pure C60 (dotted line) in the 1480–1455 cm^−1 range.

We have then studied the effect of exposure to hard X-rays of the C₆₀-doped hybrid. Previous studies on as-deposited sol–gel films have shown that, when the sample is only partially condensed, X-rays induce polycondensation of the inorganic oxide network and degradation of the organic part (Innocenzi et al., 2012). The samples are exposed to X-rays which are not monochromatic and whose energy, in accordance with the power density spectrum of the Elettra Deep X-ray Lithography synchrotron beamline, ranges from 2.5 keV (upper limit of soft X-rays) to 12 keV (hard X-rays) (Fig. 2). It has also to be stressed that with this energy and brilliance of the X-ray source the incident radiation is able to pass across all the sample. The penetration depth of X-rays is directly related to the absorption coefficient of the material, which is a function of wavelength (Cheng et al., 1997). High-energy photons have a high penetration depth and this means that the film is homogeneously exposed to the radiation and no changes are expected between the top and bottom of the layer.

Figure 2 Power density spectrum of the Elettra Deep X-ray Lithography synchrotron beamline, ranging from 2.5 keV (upper limit of soft X-rays) to 12 keV (hard X-rays).

The radiation-induced effects can be successfully employed for a lithographic process on sol–gel oxides or hybrid organic–inorganic materials (Falcaro et al., 2009; Costacurta et al., 2010; Faustini et al., 2010). Fig. 3 shows FTIR absorption spectra, in the range 950–840 cm⁻¹, of films containing different amounts of fullerenes (0 mg, dash-dotted line; 2.5 mg, dotted line; 4 mg, dashed line) and exposed to a dose of 1054 J cm⁻² (Fig. 3a) and 2108 J cm⁻² (Fig. 3b); the unexposed sample without fullerene (solid line) is also shown as a reference in Fig. 3(a). The spectra show the change of the Si–OH stretching band (Innocenzi, 2003) which is peaking at around 910 cm⁻¹, as a function of dose and fullerene concentration. The hybrid films immediately after deposition, prior to any thermal treatment, are largely uncondensed with a high amount of silanol groups; the exposure to X-rays promotes the silica condensation process through the formation of radicals. The samples exposed to the 1054 J cm⁻² dose show, however, a quite different response, because the presence of fullerenes within the materials affects the condensation process. With respect to the undoped and exposed sample, the hybrid containing fullerenes shows a reduced condensation which means a higher content of silanols. If we compare the undoped film with the one doped with the highest content of fullerene, the amount of residual silanols appears around 65% higher; on the other hand, in comparison with the unexposed sample, some decrease in area and intensity of the band is observed. These results clearly show the radical scavenging effect of fullerenes at low X-ray doses. Under these conditions, in fact, the C₆₀ molecules interact with the surrounding matter by absorbing the radical hydroxy species which are responsible for the most part of the densification and damaging of the materials (Innocenzi et al., 2011) (Fig. 4). The chemical addition of radicals produces hydroxylation of the fullerene and fullerol formation. The radical sponging, in fact, sensitively reduces the polycondensation degree of the silanols and, at the same time, the degradation of the aryl groups which are contained in the hybrid materials (see below). The samples exposed to 2108 J cm⁻² show, instead, a different response and only a small difference in silanol content is shown by the sample doped with 4 mg of fullerene. This measure reveals that the scavenging effect depends on the concentration of fullerenes, but also that there is a saturation effect of the scavenging capability of C₆₀. At higher doses, in fact, the degree of hydroxylation of fullerenes increases up to a threshold. Under high doping conditions, the increased hydrophilicity induces the aggregation of fullerol molecules into clusters, as clearly shown in Fig. 5, in agreement with previous findings (Vileno et al., 2006). The presence of aggregates has been detected only in the sample at high C₆₀ doping (4 mg in the sol) exposed at doses higher than 1054 J cm⁻².

Figure 3 (a) FTIR absorption spectra, in the 950−840 cm^−1 range, of phenylsilane-doped films exposed to a 1054 J cm^−2 dose and containing decreasing amounts of C60 in the sol used for films deposition: 4 mg (dashed line), 2.5 (dotted line) and 0 mg (dash-dotted line). The unexposed sample without fullerene (solid line) is also shown as a reference. (b) FTIR absorption spectra, in the 950−840 cm^−1 range, of phenylsilane-doped films exposed to 2108 J cm^−2 dose and containing decreasing amounts of C60 in the sol used for films deposition: 4 mg (dashed line), 2.5 (dotted line) and 0 mg (dash-dotted line).

Figure 4 Hybrid materials with an increased resistance to hard X-rays using fullerenes as radical sponges.

Figure 5 Optical image of 4 mg fullerene-doped films after exposure to a 2108 J cm^−2 X-ray dose.

The effect of radiation on the organic part of the fullerene-doped hybrid material has been studied as a function of dose and C₆₀ concentration by taking as reference the FTIR absorption band at 1430 cm⁻¹ which is assigned to the Si-aryl band (Olejniczak et al., 2005). In particular, Fig. 6(a) shows FTIR absorption spectra, in the 1440–1420 cm⁻¹ range, of three different samples after exposure to the same dose of 1054 J cm⁻². The sample without fullerenes (dotted line) has the lowest absorbance which in turn increases with the content of fullerenes within the film; the absorbance in the 4 mg C₆₀-doped hybrid is double (solid line) that of the sample without fullerenes. This indicates that the presence of fullerenes reduces the organic bond cleavage owing to X-ray exposure; when a higher dose of 2108 J cm⁻² is used, however, the radical sponge effect appears less effective (Fig. 6b).

Figure 6 (a) FTIR absorption spectra, in the 1440–1420 cm^−1 range, of phenylsilane-doped films exposed to a 1054 J cm^−2 dose and containing decreasing amounts of C60 in the sol used for films deposition: 4 mg (solid line), 2.5 (dashed line) and 0 mg (dotted line). (b) FTIR absorption spectra, in the 1440–1420 cm^−1 range, of phenylsilane-doped films exposed to a 2108 J cm^−2 dose and containing decreasing amounts of C60 in the sol used for films deposition: 4 mg (solid line), 2.5 (dashed line) and 0 mg (dotted line).

The data show that fullerenes within the hybrid material behave as a radical sponge, which effectively change the response of the material upon exposure to an intense source of X-rays. By decreasing the available amount of radicals, in fact, both the inorganic polycondensation and the organic bonds cleavage are reduced. On the other hand, a careful design of the material in terms of fullerene concentration and dispersion is necessary to make the resistance of the material effective to exposure to hard X-rays.

4. Conclusions

The experiments, performed on C₆₀ hybrid materials exposed to high-energy doses of radiation, give two main indications: (i) owing to its radical scavenging effect, fullerene enhances the resistance threshold of organic bond damage by X-rays; (ii) a radical scavenging saturation threshold exists and the concentration, dispersion and solubility of C₆₀ largely affect the capability of fullerene to act as a radical sponge. Using specifically designed fullerene derivatives could allow the X-ray resistance of hybrid materials to be optimized. These experiments open the route to new classes of plastics and hybrids with the enhanced capability of resisting high-energy radiation.