Multifunctional Fe₂O₃/CeO₂ nanocomposites for free radical scavenging ultraviolet protection

Abstract

In this study we synthesized Fe₂O₃/CeO₂ composite nanoparticles for application as multifunctional ultraviolet (UV) filters. The precipitation of small ceria (CeO₂) nanoparticles onto the surface of larger hematite (α-Fe₂O₃) nanoparticles, results in stronger and more selective absorbance in the (UV) region. Through UV photocatalyst experiments, we show that the addition of these nanocomposite particles significantly reduces the degradation of crystal violet by P25 by scavenging the photogenerated OH˙ radicals.

---

1 Introduction

Since they were first introduced into cosmetic products, there has been concerns regarding the use of inorganic nanoparticles as ultraviolet (UV) blockers in sunscreen formulations. The possibility of negative health effects resulting from the use of cosmetic products containing such particles is a topic of debate in the scientific community, with more research needed.¹ Many sunscreen formulations contain a combination of organic or inorganic UV blocking agents, with the latter usually in the form of zinc oxide (ZnO) or titanium dioxide (TiO₂).² These inorganic nanoparticles are used due to their improved optical properties compared to their micro-sized counterparts, such as enhanced UV absorption and greater transparency in the visible spectrum.³ The use of such nanoparticles in cosmetic products also raises general safety concerns, as they have been shown to have different biological effects compared to their micro-sized counterparts, being able to more effectively penetrate into cells and come into close contact with sensitive biological targets. Nanoparticles of ZnO are well-known free radical generators and have shown to be highly cytotoxic⁴ and genotoxic⁵ to human cells. Similar concerns have also been raised for TiO₂ nanoparticles, as while they do not significantly effect cell death or cell cycle,⁶ they have a substantial effect on the mitochondrial activity of human keratinocytes.⁷ Titanium dioxide nanoparticles also can exhibit very strong photocatalytic activity under UV irradiation, producing highly reactive hydroxyl (OH˙) radicals.⁸ Despite that the photocatalytic properties of TiO₂ are well known, there is evidence to show that some companies use highly photoactive particles in their commercial cosmetic products.⁹ Direct interaction between these nanoparticles and human cells aside, there is also some concern with how these free radical generating particles interact with other organic compounds inside the sunscreen formulations, including other UV blocking agents. Despite many claims, organic UV filters are not very photostable, and their degradation is further enhanced when they are combined with other active organic or inorganic compounds, including TiO₂,¹⁰ more quickly reducing the sun protection factor (SPF) of the protective formulation after being applied on the skin. One reported example of this interaction is the photocatalytic transformation of 4-methylbenzylidene camphor (4-MBC) into unknown, more toxic intermediates when irradiated under UV light in the presence of TiO₂.¹¹ Such degradation products have unknown health effects, and as such, the free radical generating properties of these photocatalytically active nanoparticles are not ideal for use in cosmetic products.

A solution to this would be a nanoparticle system with similar UV blocking properties to TiO₂ and ZnO, which does not feature significant free radical generation or a free radical scavenging agent to act as an antioxidant to prevent accelerated decomposition of the organic components of the formulations. There has been a number of studies investigating other metal oxide nanoparticles for use in such products, including pure CeO₂,^12,13 rare earth-,¹⁴ alkali earth-^14,15 and transition metal¹⁶ doped CeO₂, α-Fe₂O₃,¹⁷ Ce-doped Fe₂O₃ (ref. 18) and most recently, Ti-doped SnO₂.¹⁹ Ceria (CeO₂) nanoparticles and compounds incorporating cerium are of particular interest for such application, as it features a optical bandgap close to that of TiO₂ (3.2 eV),²⁰ showing strong selective absorbance in the UV region, as well as the ability to scavenge radical species through the Ce³⁺ states that exist on their surface. These states are size-dependant property of CeO₂ nanoparticles, formed as a consequence of the oxygen deficiencies on their surface, and they are able to act as a catalyst and scavenge excess free radicals by moving between these two oxidation states.^21,22 In addition to this scavenging ability being highlighted through a number of in vitro^12,22,23 and in vivo²⁴ studies, they have been also shown to increase cell viability if applied before or after UV light exposure.¹² This would indicate that when exposed to UV light they do not form significant radicals in vitro, as well as that they are able to repair some of the damage to the cells upon ultraviolet light exposure by scavenging excess free radicals, acting as an antioxidant. One of the most effective UV-blockers of all the ceria-based nanoparticles has been shown to be Ca-doped CeO₂, shown to produce a higher SPF rating than ZnO when combined with TiO₂.²⁵ This doping however, decreases the concentration of Ce⁴⁺ states,¹⁵ which would in turn likely decrease any potential radical scavenging activity,²¹ as well as increases the band-gap of the nanoparticles, decreasing it's absorbance in the UVA (320–400 nm). There is at least one commercially available cosmetic product already containing ceria nanoparticles,²⁶ this material however, features a platinum coating thus potentially removing the positive free radical scavenging properties of the core particles. Furthermore, CeO₂ and Pt core–shell nanoparticles are well-known catalysts²⁷ and thus could have an unknown effect on both the other components of the sunscreen, and potentially the user.

A big drawback of cerium-based materials however, is their relatively high cost, especially when compared to the more abundant TiO₂ and ZnO. As such, in this study, we synthesised a nanocomposite system which incorporates hematite (α-Fe₂O₃) a low-cost inorganic compound broad band UV blocking properties, with nanoparticles of ceria (CeO₂) precipitated on the surface. Such nanocomposites have previously been studied as CO/CO₂ oxidation/reduction catalysis applications,^28,29 and more recently as visible light photocatalysts and supercapacitor electrodes.³⁰ The visible light photocatalytic properties was reported for a nanocomposite system which featured a higher amount of ceria relative to hematite and large particle size of the ceria phase. In this study we show that combining these two phases in the form nanocomposite particles, that there is a synergistic effect between the two materials. The final composite particles featuring greater UV absorption and free radical scavenging properties than the bare α-Fe₂O₃ or CeO₂ nanoparticles at particular ceria coating concentrations. In order to show these free radical scavenging properties, the synthesised nanoparticles were suspended in a solution of crystal violet (CV) and photocatalytically active commercial TiO₂ nanoparticles (P-25) before being exposed to broad spectrum ultraviolet light. The presence of the ceria particles as part of the nanocomposite system greatly reduces the degradation of the CV dye. We propose that the reason for this is the due to the scavenging or neutralization of the photogenerated OH˙ generated by the TiO₂ by the surface CeO₂ phase of the nanocomposite particles. These nanocomposite particles also showed improved UV absorbance and scavenging properties as compared to pure CeO₂ and α-Fe₂O₃ nanoparticles. Such multifunctional nanocomposite particles could have potential application as alternative UV filtering components in cosmetic sunscreen formulations, or to be used as an antioxidant stabilizer of these products by scavenging the photogenerated free radicals and thus preventing the accelerated decomposition of the organic filter components.

The hematite (α-Fe₂O₃) nanoparticles and Fe₂O₃/CeO₂ composite nanoparticles were synthesized following a procedure illustrated in Fig. 1. The core hematite particles were prepared using a simple thermal precipitation reaction, previously published by Matijević and Scheiner.³¹ A round bottom flask partially containing HCl (Sigma-Aldrich) in deionized water (DI) was heated to 100 °C in a silicon oil bath, under condenser reflux. Anhydrous FeCl₃ (Sigma-Aldrich) was dissolved in DI water separately and added to the flask, for a final concentration of 18 mM FeCl₃ and 1 mM HCl. This dark orange solution was then stirred magnetically at 100 °C for 24 hours, resulting in the gradual precipitation of an orange red solid hematite (α-Fe₂O₃). This precipitate was then separated via ultracentrifugation at 11 000 RPM and washed with DI water. The obtained solid was then dried at 100 °C overnight, and then 120 °C under vacuum for 1 hour. In order to prepare the α-Fe₂O₃/CeO₂ core–shell nanocomposite, 0.2 g of α-Fe₂O₃ nanoparticles were resuspended in DI water and added to a solution of Ce(NO₃)₃ (Sigma-Aldrich) for the desired stoichiometric Fe/Ce ratio of 2.5, 5 and 10 at%. The cerium oxide was then precipitated with the dropwise addition of 1 mL concentrated NH₄OH (27 w/w%) to the solution at 60 °C. The resulting particles were then separated and washed, after which, concentrated H₂O₂ was then added in order to crystallize the precipitated cerium oxide without the need for high temperature annealing. Following this, the particles were then dried following the same procedure for the uncoated α-Fe₂O₃ particles. Pure ceria nanoparticles were produced using spray pyrolysis in accordance to previous studies.¹⁵

Fig. 1 Schematic representation of the synthesis procedure used in producing α-Fe₂O₃ and α-Fe₂O₃/CeO₂ nanoparticles.

The crystalline state of each of the synthesised nanomaterials was investigated using X-ray diffraction, and the diffraction patterns obtained for all synthesised nanoparticles (Fig. 2) exhibited sharp peaks corresponding to the hematite (α-Fe₂O₃) phase of iron oxide (PDF card 96-9000140). While the 2.5 at% nanocomposite particles showed no reflections from the CeO₂ phase (PDF card 96-900-9009), one broad peak was present in the samples with higher Ce content. This peak, located at ∼28° corresponds to the (111) CeO₂ reflection, and can be seen to increase in intensity as the amount of Ce was increased from 5 to 10 at%. This indicates that there is likely a higher amount of CeO₂ resulting from an increased Ce(NO₃)₃ concentration during synthesis. The low intensity and large FWHM for this ceria reflection indicates that the crystalline size of this cerium dioxide phase is quite small relative to the iron oxide phase and an increase in intensity between the 5 and 10 at% samples could be indicative of some crystalline growth of the surface phase.

Fig. 2 X-ray diffraction patters for α-Fe₂O₃ and α-Fe₂O₃/CeO₂ composite nanoparticles with 2.5, 5, and 10 at% Ce, with PDF lines for matching CeO₂ and α-Fe₂O₃.

High-resolution TEM images were obtained using a JEM-ARM200F (JEOL, Akishima, Tokyo, Japan) atomic resolution microscope in order to determine the morphology of the obtained nanocomposite particles (Fig. 3). These hematite particles can be seen to be roughly spherical and feature cracks or dislocations running from the center of the particle out to the surface. The size of individual particles were measured from TEM images, and the average size of the hematite nanoparticles was calculated to be 50 ± 10 nm. Images of the particles following the precipitation of ceria (Fig. 3C–E) shows the presence of much smaller particles to the surface of the hematite. These particles are comprised of ceria, and appear to have a much darker contrast than the previous formed hematite (Fig. 3C and D), which can be attributed to the significant difference in atomic number between iron (Z_Fe = 26) and cerium (Z_Ce = 58). This was also confirmed by measuring the d-spacings of these nanoparticles (Fig. 3E), which was found to be 3.17 Å, corresponding to the (111) plane of CeO₂. The size of these particles was also measured from the images, and it was found that the size of these particles were 2.5 ± 0.1, 3.5 ± 0.4, and 4.7 ± 0.4 nm, for the 2.5, 5 and 10 at% samples respectively, which shows the effect of the concentration of cerium nitrate upon precipitation of the ceria particles. Comparing the images of the 5 and 10 at% composite particles (Fig. 3C and D), it can be seen that increasing the cerium content did not result in a more evenly distributed surface covering of CeO₂ particles. The 10 at% composite certainly features a higher frequency of ceria particles on the surface of the hematite core, but they appear to be more aggregated with other ceria particles. It is likely that the observed defects in the hematite nanoparticles act as a precipitation center for the CeO₂ phase, and as the concentration is further increased, additional particles precipitate or aggregate on the surface of the other ceria particles, as opposed to the bare, smooth surface of the hematite core particles.

Fig. 3 Bright field TEM images of (A and B) α-Fe₂O₃, (C and E) 5 at%, and (D) 10 at% Ce composite nanoparticles.

High resolution EDS mapping images were also obtained (Fig. 4), using a JEM-ARM200F atomic resolution microscope fitted with a Centrino SDD 100 mm² detector (JEOL, Akishima, Tokyo, Japan) in order to further verify that the were CeO₂. The below EDS mapping images show a dark field image, followed by the iron highlighted in green and cerium in red. The pure hematite particles feature no cerium at all, while a significant contribution of cerium can be seen about the surface of the particles for the composite particles. These images confirms the presence of the secondary ceria phase that can be seen in the TEM images of the composite particles on the surface of the hematite particles. The iron/cerium ratio of each of these composites was calculated from EDS measurements, and it was found that the actual ratio was slightly lower than the target ratio for the 2.5 and 5 at% nanocomposites, which was 0.9 ± 0.9 and 4.5 ± 0.5% respectively. The actual ratio for the 10 at% sample was much closer to the target ratio, which was 10.1 ± 0.1%.

Fig. 4 High resolution EDS mapping of α-Fe₂O₃ (top) and α-Fe₂O₃/CeO₂ (5 at% Ce) nanoparticles. The images depict greyscale (A), iron (green) content (B), ceria (red) content (C) and an overlay of the two (D).

The ultraviolet-visible absorption spectra (Fig. 5) of the synthesised hematite and composite nanoparticles was measured using a UV-1800 spectrophotometer (Shimadzu, Kyoto, Japan), in order to determine the effects of the ceria precipitates on the optical bandgap and ultraviolet filtering properties of all the particles. Also measured was a commercial TiO₂ nanoparticle (Aeroxide® P25, Sigma-Aldrich) as a comparison of relative ultraviolet filtering compatibility, as TiO₂ is a common additive to commercial sunscreen formulations as an inorganic UVA filter. Conducting all experiments at the same concentration (25 μg mL⁻¹), it can be seen that P-25 had a higher and more selective absorbance within the ultraviolet region compared to the as prepared nanoparticles. The uncoated hematite nanoparticles showed significant absorption over the entire ultraviolet range, and the addition of the ceria phase to the surface of these particles further enhanced this. It can also be seen that the absorption of these composite particles had more selective absorbance in the ultraviolet region as compared to the uncoated hematite particles. This is most accurately reflected in the increase of the optical bandgap of composite particles compared to the hematite particles which was calculated to be 2.48 ± 0.03 eV, which is slightly larger than that for bulk hematite (2.1 eV) due to the effects of quantum confinement. The optical band gaps of the composite particles was shown to increase the optical band gap of the composite system up to 5 at%, with 2.74 ± 0.02 eV, 2.82 ± 0.03 eV and 2.77 ± 0.04 eV calculated for the 2.5, 5 and 10 at% composite particles. The calculation of the band gap was performed by creating a Tauc plot using the ultraviolet-visible absorption spectra,³² and details can be found in the ESI.† As ceria has a higher band gap relative to hematite, the additional of ceria content on to the surface of the core particles resulted in an increase in band gap for the nanocomposite particles. The significant aggregation and slight crystal growth of the surface ceria particles in the 10 at% nanocomposite is likely the reason that no further increase in optical band gap is seen relative to the 5 at% sample, despite the increase in ceria content. The UV blocking properties of these composites was also compared to pure ceria nanoparticles, synthesised using spray pyrolysis.²⁵ A band gap of 3.30 ± 0.02 eV was calculated for these pure CeO₂ nanoparticles from ultraviolet-visible absorption spectra. These particles featured a much lower absorbance than the other nanoparticles in this study, highlighting the synergistic effect of the hematite core on the UV absorbance of the nanocomposite particles.

Fig. 5 Ultraviolet-visible absorption spectra of commercial TiO₂ (P25), hematite and composite nanoparticles at 25 μg mL⁻¹ in DI water.

Each of these materials were then tested for their photocatalytic activity under a broad UVA-UVB ultraviolet spectrum, using a RPR-200 photochemical reactor (Rayonet, Branford, CT, USA) fitted with 3000 and 3500 Å lamps. In this experiment, crystal violet (tris(4-(dimethylamino)phenyl)methylium chloride) (99% anhydrous, Sigma-Aldrich) was used as a indicator of the amount of generated hydroxyl radicals, as the P-25 generates OH˙ radicals upon exposure to ultraviolet light, which subsequently react with the central double bond of crystal violet, rendering it colourless.³³ Exposing the solution of crystal and P-25, a well known and highly active photocatalyst, to the ultraviolet light of the photochemical reaction caused the dye to be almost completely degraded over a 30 minute period (Fig. S3† left). The addition of 5 mg L⁻¹ of any of the as-synthesised nanoparticles resulted in a decrease in this degradation of the crystal violet dye (Fig. S3† right), and the relative absorbance at 590 nm after different exposure times, for each material, can be seen in Fig. 6. The linear rate constants k for the effect of each of the materials on the degradation of crystal violet was also calculated using the Langmuir–Hinshelwood model³⁴ and the plots used to find this constant can be seen in Fig. 7.

Fig. 6 Relative decrease in absorbance of crystal violet solutions containing P25 and nanocomposite particles both at 5 mg L⁻¹ concentration under ultraviolet light exposure.

Fig. 7 Linear rate constant curves for ultraviolet dye degradation of crystal violet shown in Fig. 6.

The addition of the uncoated hematite particles caused a slight decrease in the degradation and rate constant ((6.3 ± 0.3) × 10⁻² min⁻¹) as compared to the control ((8.65 ± 0.06) × 10⁻² min⁻¹). This is likely due to the effects of absorption and scattering by the additional particles. While this decrease in degradation was more pronounced when the 2.5 at% composite particles were added ((5.5 ± 0.2) × 10⁻² min⁻¹), a very large decrease in the dye degradation was seen upon the addition of the 5 at% nanocomposite ((3.1 ± 0.1) × 10⁻² min⁻¹) and 10 at% nanoparticles ((3.3 ± 0.5) × 10⁻² min⁻¹). While this effect could be attributed to the increase absorption in the ultraviolet region compared to the uncoated α-Fe₂O₃ nanoparticles, further experiments were conducted in which 10 and 25 mg L⁻¹ of hematite nanoparticles were added to the system (Fig. S4†). Even at a substantially higher concentration of particles, the uncoated hematite particles and did not have as significant effect on the dye degradation, decreasing the rate constant to (4.8 ± 0.3) × 10⁻² min⁻¹ at the highest concentration of hematite particles (25 mg L⁻¹).

Thus, this substantial decrease in the degradation of the crystal violet upon addition of the 5 at% coated hematite nanoparticles can be attributed to the scavenging of OH˙ generated from the photocatalytic activity of the P-25 by this nanocomposite. This is a well-known property of ceria nanoparticles, and the very small diameter of the ceria particles in our nanocomposite system maximises this effect. The lack of any increase in this scavenging effect after increasing the cerium content to 10 at% is likely due to the significant aggregation of the ceria particles at this concentration. This aggregation reduces the available surface area of the ceria content to engage in the OH˙ radical scavenging reactions, compared to if the particles were evenly distributed about the surface of the hematite particles. The pure ceria particles only slightly decreased the dye degradation rate when added to the P25 and dye mixture ((7.2 ± 0.1) × 10⁻² min⁻¹), which could be attributed to a very weak scavenging or absorbance effect. The small size of the ceria particles in the nanocomposite material resulting in a large concentration of Ce³⁺ surface states, is the likely the reason for their substantial free radical scavenging properties as it has been shown in other studies that ceria nanomaterials,³⁵ and nanocomposites containing ceria can be photocatalytically active.^30,36 Furthermore, when added to the reaction mixture in the absence of P-25, the 5 at% composite did not degrade the dye over the 30 minute exposure period, meaning that these nanocomposite particles do not generate OH˙ radicals upon exposure to UV light. The linear fitting used to determine the kinetic dye degradation coefficient k all returned correlation coefficients (R²) of 0.99 except for the 25 mg L⁻¹ hematite particles reacted along with P25, as well as the crystal violet alone, and the 5 at% composite nanoparticles with crystal violet and without P25. The R² value obtained for 25 mg L⁻¹ α-Fe₂O₃ particles was 0.97, and was likely lowered due to the high concentration of particles resulting in some small amounts of aggregation and enhanced interparticle scattering of incident ultraviolet light. Both experiments with crystal violet only, and 25 mg L⁻¹ of the 5 at% nanocomposite sample showed a negligible overall degradation to the crystal violet, and as such did not return good R² values or meaningful values for the linear rate constant k.

The free radical generating properties of ZnO and TiO₂ nanoparticles currently used in some commercial sunscreen products could have the potential to cause adverse health effects, either through the direct interaction of these radicals with human tissue, or through causing the accelerated decomposition of other organic compounds within the formulation. These reactions not only have the potential to decrease the SPF of the formulation more rapidly, but also yield unknown products. In this study we synthesised a number of hematite–ceria nanocomposite particles through a simple two-step precipitation approach. The result of combining these two materials is a composite system with greater, more selective absorbance in the UV range, and greater free radical scavenging properties than either pure ceria or hematite nanoparticles. This effect was maximized at 5 at% loading of Ce, as a further increase in ceria content did not increase the distribution of ceria particles about the surface of the hematite particles, instead resulting in larger regions of aggregated ceria nanoparticles. Nanoparticles such as these could provide a novel and cost-effective approach to both improve the UV blocking properties of commercial sunscreen formulations, as well as act as an antioxidant stabilizer for the organic components of the product, reducing potential harm of these products on the health of the consumer.

References

1. M. Crosera, M. Bovenzi, G. Maina, G. Adami, C. Zanette, C. Florio and F. Filon Larese, Int. Arch. Occup. Environ. Health, 2009, 82, 1043–1055.
2. N. Serpone, D. Dondi and A. Albini, Inorg. Chim. Acta, 2007, 360, 794–802.
3. A. P. Popov, J. Lademann, A. V. Priezzhev and R. Myllyä, J. Biomed. Opt., 2005, 10, 064037–064039.
4. V. Sharma, D. Anderson and A. Dhawan, Apoptosis, 2012, 17, 852–870.
5. K. Soto, K. Garza and L. Murr, Acta Biomater., 2007, 3, 351–358.
6. M. Hamzeh and G. I. Sunahara, Toxicol. In Vitro, 2013, 27, 864–873.
7. P. Tucci, G. Porta, M. Agostini, D. Dinsdale, I. Iavicoli, K. Cain, A. Finazzi-Agró, G. Melino and A. Willis, Cell Death Dis., 2013, 4, e549.
8. J. Zhang and Y. Nosaka, J. Phys. Chem. C, 2014, 118, 10824–10832.
9. P. J. Barker and A. Branch, Prog. Org. Coat., 2008, 62, 313–320.
10. N. Serpone, A. Salinaro, A. V. Emeline, S. Horikoshi, H. Hidaka and J. Zhao, Photochem. Photobiol. Sci., 2002, 1, 970–981.
11. V. Sakkas, P. Calza, M. A. Islam, C. Medana, C. Baiocchi, K. Panagiotou and T. Albanis, Appl. Catal., B, 2009, 90, 526–534.
12. N. Zholobak, V. Ivanov, A. Shcherbakov, A. Shaporev, O. Polezhaeva, A. Baranchikov, N. Spivak and Y. Tretyakov, J. Photochem. Photobiol., B, 2011, 102, 32–38.
13. T. Boutard, B. Rousseau, C. Couteau, C. Tomasoni, C. Simonnard, C. Jacquot, L. J. Coiffard, K. Konstantinov, T. Devers and C. Roussakis, Mater. Lett., 2013, 108, 13–16.
14. S. Yabe and T. Sato, J. Solid State Chem., 2003, 171, 7–11.
15. L. Truffault, M.-T. Ta, T. Devers, K. Konstantinov, V. Harel, C. Simmonard, C. Andreazza, I. P. Nevirkovets, A. Pineau, O. Veron and J.-P. Blondeau, Mater. Res. Bull., 2010, 45, 527–535.
16. L. Truffault, Q. W. Yao, D. Wexler, I. P. Nevirkovets, K. Konstantinov, T. Devers and S. Nightingale, J. Nanosci. Nanotechnol., 2011, 11, 1–10.
17. L. Truffault, B. Choquenet, K. Konstantinov, T. Devers, C. Couteau and L. J. Coiffard, J. Nanosci. Nanotechnol., 2011, 11, 2413–2420.
18. D. Cardillo, K. Konstantinov and T. Devers, Mater. Res. Bull., 2013, 48, 4521–4525.
19. A. Morlando, D. Cardillo, T. Devers and K. Konstantinov, Mater. Lett., 2016, 171, 289–292.
20. Q. Xiang, J. Yu, W. Wang and M. Jaroniec, Chem. Commun., 2011, 47, 6906–6908.
21. E. G. Heckert, A. S. Karakoti, S. Seal and W. T. Self, Biomaterials, 2008, 29, 2705–2709.
22. I. Celardo, M. De Nicola, C. Mandoli, J. Z. Pedersen, E. Traversa and L. Ghibelli, ACS Nano, 2011, 5, 4537–4549.
23. J. Colon, N. Hsieh, A. Ferguson, P. Kupelian, S. Seal, D. W. Jenkins and C. H. Baker, J. Nanomed. Nanotechnol., 2010, 6, 698–705.
24. J. Colon, L. Herrera, J. Smith, S. Patil, C. Komanski, P. Kupelian, S. Seal, D. W. Jenkins and C. H. Baker, J. Nanomed. Nanotechnol., 2009, 5, 225–231.
25. L. Truffault, B. Winton, B. Choquenet, C. Andreazza, C. Simmonard, T. Devers, K. Konstantinov, C. Couteau and L. J. Coiffard, Mater. Lett., 2012, 68, 357–360.
26. Y. Tokyo, Key Ingredient: Aqua Ceria, 2015.
27. N. Singhania, E. A. Anumol, N. Ravishankar and G. Madras, Dalton Trans., 2013, 42, 15343–15354.
28. H. Bao, X. Chen, J. Fang, Z. Jiang and W. Huang, Catal. Lett., 2008, 125, 160–167.
29. V. V. Galvita, H. Poelman, V. Bliznuk, C. Detavernier and G. B. Marin, Ind. Eng. Chem. Res., 2013, 52, 8416–8426.
30. N. S. Arul, D. Mangalaraj, R. Ramachandran, A. N. Grace and J. I. Han, J. Mater. Chem. A, 2015, 3, 15248–15258.
31. E. Matijević and P. Scheiner, J. Colloid Interface Sci., 1978, 63, 509–524.
32. J. Tauc, Amorphous and Liquid Semiconductors, Springer, 1974.
33. Y. Xue, Q. Luan, D. Yang, X. Yao and K. Zhou, J. Phys. Chem. C, 2011, 115, 4433–4438.
34. R. Djellabi and M. Ghorab, Desalin. Water Treat., 2015, 53, 3649–3655.
35. Y. Zhai, S. Zhang and H. Pang, Mater. Lett., 2007, 61, 1863–1866.
36. H. Eskandarloo, A. Badiei and M. A. Behnajady, Ind. Eng. Chem. Res., 2014, 53, 7847–7855.

---

Footnotes

† Electronic supplementary information (ESI) available: Further details on nanocomposite synthesis and characterisation, calculation of optical band gaps, raw photodegradation curves, additional photodegradation experiments and FT-IR characterization. See DOI: 10.1039/c6ra10951a

‡ These authors contributed equally to this work.

---