Amorphous TiO2 nanoparticles: Synthesis and

Journal of Non-Crystalline Solids 459 (2017) 192–205

Contents lists available at ScienceDirect

Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/locate/jnoncrysol

Review

Amorphous TiO2 nanoparticles: Synthesis and antibacterial capacity Mónica Andrea Vargas, Jorge E. Rodríguez-Páez ⁎ CYTEMAC group, Deptarment of Physics, University of Cauca, Popayan, Colombia

a r t i c l e

i n f o

Article history: Received 11 November 2016 Received in revised form 7 January 2017 Accepted 9 January 2017 Available online xxxx Keywords: Amorphous TiO2 Nanoparticles Sol-gel Bacterial inactivation Escherichia coli

a b s t r a c t In this work, the sol-gel method was used to synthesize titanium dioxide (TiO2), a process that allowed elaboration of the different phases of TiO2, in a controlled manner through heat treatments, ensuring the purity of the oxide and the nanometric size of the particles. The results of X-ray diffraction (XRD) showed that the synthesized powders were amorphous up to a temperature T b 350 °C, with a particle size of ~100 nm, determined by electron microscopy (TEM and SEM). Considering the nature of the synthesis process used, a mechanism was put forward that would allow explanation of the formation of amorphous TiO2 nanoparticles (TiO2-ANPs). With the aim of studying the potential use of the synthesized TiO2-ANPs, their antibacterial capacity was studied. The inactivating effect on bacteria of synthesized amorphous TiO2 was analyzed by recording the effect of its presence on bacterial strains of Escherichia coli. As such, prior to the addition of amorphous TiO2 to the culture of E. coli, the oxide was activated by subjecting it to UV radiation for 1 h. The activated amorphous TiO2 was immediately then placed in contact with the culture, which was not irradiated with UV during the course of this test. The results indicate that in the first 30 min of exposure of the bacteria to the activated amorphous TiO2, the presence of E. coli colonies was significantly reduced, with no presence being detected in the culture. © 2017 Elsevier B.V. All rights reserved.

Contents 1. 2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Synthesis and characterization of amorphous TiO2 . . . . . . . . . . . . . . . . . . . . . 2.2. E. coli culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Inactivation of the E. coli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1. Preparation of the culture medium . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2. Preparation of the nutrient agar . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3. Preparation of the TiO2 suspensions . . . . . . . . . . . . . . . . . . . . . . . . 2.3.4. Colony count after incubation . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Material characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Thermal analysis of the solid obtained from the colloidal suspension. . . . . . . . . . 3.1.2. Crystalline structure of the synthesized solids. . . . . . . . . . . . . . . . . . . . 3.1.3. IR spectroscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.4. Morphology, particle size and agglomeration of the synthesized TiO2 particles. . . . . 3.1.5. UV-Visible spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.6. Suggested mechanism for the formation of amorphous TiO2 nanoparticles (TiO2-ANPs). 3.2. Assay of cell viability of Escherichia coli strain . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

⁎ Corresponding author. E-mail address: [email protected] (J.E. Rodríguez-Páez).

http://dx.doi.org/10.1016/j.jnoncrysol.2017.01.018 0022-3093/© 2017 Elsevier B.V. All rights reserved.

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

193 195 195 195 195 195 195 195 195 196 196 196 196 197 197 197 198 200 203 203 203

M.A. Vargas, J.E. Rodríguez-Páez / Journal of Non-Crystalline Solids 459 (2017) 192–205

1. Introduction Titanium dioxide (TiO2) can have four natural polymorphs (TiO2(B), brookite, anatase and rutile) and at least five polymorphs that are produced synthetically [1]. The main structural characteristics of the natural polymorphs are as follows: brookite – rhombohedrical, D15 2h – Pbca, a = 9.166 Å, b = 5.436 Å, c = 5.135 Å [1,2,3,4]; anatase – tetragonal, D19 4h – I41/amd, a = b = 3.782 Å, c = 9.502 Å [1,4]; rutile – tetragonal, D14 4h – P42/mnm, a = b = 4.584 Å, c = 2.953 Å [4,5] and TiO2(B) - low density monoclinic, C2/m, a = 12.1787 Å, b = 3.7412 Å, c = 6.5249 Å [6,7] having a monoclinic unit cell with a volume of 35.27 Å3 while anatase, rutile, and brookite are 34.02, 31.12 and 32.20 Å3, respectively. The TiO2 (titania) may also have other structures such as cotunnite TiO2 synthesized at high pressures and is one of the hardest known polycrystalline materials [8], TiO2(R) [9]; cubic TiO2 [10], TiO2-H (I4/ m) [11] and TiO2-m (P21/c) [12], among others. While the rutile structure is the most thermodynamically stable phase at all pressures and temperatures, the anatase and brookite phases are metastable. The base unit of the local order in each phase is described representatively by an octahedron, [TiO6]2. For the different structures of the titania, the octahedra are arranged spatially in different forms, sharing corners and edges in various ways, always ensuring that the overall stoichiometry is TiO2 [13,14]. In the specific case of amorphous TiO2, the titania structure of interest for this work, there is currently a growing interest in studying the structural characteristics of amorphous TiO2 nanoparticles - ANPs, given their potential technological applications [15–19]. To understand the structural properties of this amorphous oxide and investigate its local atomic structure, the researchers used x-ray absorption spectroscopy (XAS), both x-ray absorption near-edge structure (XANES) and extended x-ray absorption fine structure (EXAFS). Using TiO2 nanoparticles of 3 nm, Chen et al. [20] found a titanium coordination number of ZTi-O = 4.8 and concluded that the Ti sites were large octahedra. Meanwhile, Yeung et al. [21] found the average length of the Ti\\O bond to be RTi\\O = 1.93 Å and the coordination number ZTi\\O = 4.5. Despite the difference in the values reported for the TiO2-ANPs, obtained experimentally, it can be concluded that the local structure of the amorphous, nano-sized TiO2 presents a reduction in the coordination number and in the length of the Ti\\O bond. According to Zhang et al. [16], the structural model that best fits the data from the XAS and WAXS (synchrotron wide-angle X-ray scattering), obtained for the TiO2-ANP, considers a distorted surface layer and a slightly distorted, anatasetype crystalline core, with ZTi\\O = 5.3 and RTi\\O = 1.94 Å, while on the surface a reduction occurs in the coordination number of Ti and O atoms. Calculations obtained using computer simulation suggests that the core has a distorted octahedral structure with ZTi\\O ~ 6.0 and ZO\\Ti ~ 3.0, and the surface a more porous structure, with ZTi\\O ≠ 6.0 and ZO\\Ti ≠ 3.0 and greater amount of structural defects [15,22]; while the structure of the core is nearly independent of particle size, the surface structure is strongly dependent on the size and plays an important role in the dependence of the structure and properties of the TiO2-ANPs on particle size. The information reported by Zhang et al. in their work [16] indicates that the strained anatase-like crystalline core would have a dimension of approximately 2 unit cells with a highly distorted outer shell, approximately 2–4 atomic layers thick; the density of TiO2 nanoparticles would be 3.74 g/cm3, less than that of bulk anatase (3.90 g/cm3). There is some controversy surrounding the anatase-type nature of the core, a structural characteristic that ought to be carefully rechecked [19]. According to Hoang [18,19], the core exhibits an amorphous structure with an octahedral network. In turn, Petkov et al. [23] propose in their work that the atomic arrangement of the amorphous TiO2 materials resembles that occurring in brookite. They synthesized the amorphous TiO2 using controlled hydrolysis and condensation of tetraisopropoxytitanate (TPOT) in an ethanol solution and then characterized the powders using electron diffraction and XRD. The data obtained

193

was used to obtain three-dimensional structure models, using Reverse Monte Carlo (RMC) Simulations, which were consistent with the experimental structure functions. The model structured was based on the atomic arrangement of brookite, which was able to reproduce all the important features of the experimental atomic distribution functions for synthesized amorphous titania. They concluded further that the synthesis method used and the nature of the precursor used, affect details of the atomic arrangement, such as the degree of atomic order, the relative abundance of defective [TiO6], and the proportion of sides and vertices that the octahedral-like units share. Since many of the physical and chemical processes take place on the surface of the particle, or initiate there, knowledge of the atomic structure of the surface of the solid is important. While for the structural characteristics of the surface of crystalline polymorphs of TiO2 excellent reviews can be found [24], this is not the case for the surface of TiO2ANPs. Previous studies [15,18,19] indicate that the structure is strongly dependent on particle size, such that surface defects, due to bond breaking, mean that the concentration of undercoordinated structural units ([TiO3], [TiO4] and [TiO5]) in the TiO2-ANPs is greater than that of the amorphous bulk, generating a lot of structural defects on the surface; these undercoordinated Ti sites can be considered as defects with a deficiency in oxygen as occurs in crystalline TiO2 [24]. Additionally, due to the lack of periodicity in the structure of the nanoparticles of amorphous TiO2, other point defects can be expected, for example vacancies, but given the small dimensions of these and their amorphous structure, undercoordinated defects would predominate over other types of structural defects normally observed on the surface of crystalline TiO2: step edges, oxygen vacancies, line defects, crystallographic shear planes, etc. [24]. These predominant point defects would strongly influence the different properties of the materials, including their antibacterial capacity, since they can cause changes in the electronic structure of the amorphous TiO2, as in the crystalline polymorphs of TiO2 [24], which would lead to a strong surface activity in the amorphous nanoparticles compared to crystals of the same size [17]. The properties that the structural characteristics mentioned above lend to titania (TiO2), especially the anatase and rutile crystalline polymorphs, which have been the most widely studied, make them very attractive for applications such as: solar cells [25,26], photocatalysis [27, 28], electrode in semiconductor devices [29], adsorption of proteins [30], catalyst [31], or catalyst support [32,33], white pigment in paints, cosmetics and toothpastes [34], the material for electrodes in lithium ion batteries [35,36], waste water purification [37,38], disinfection based on the antibacterial properties of TiO2 [39,40], self-cleaning coatings [41], gas sensor [42a,b], among others. These polymorphs of TiO2 were synthesized using different techniques as indicated by the important and careful reviews that are found on this topic in the scientific literature [14,43,44,45,46]. Each crystalline polymorph of titania has been considered for a wide range of applications, as indicated above, in which the structural differences they present mean that their physicochemical and optoelectronic properties are different, controlling the electronic structure and diffusion in the bulk of the charge carriers [14]. Although the crystalline nanoparticles have a well defined structure, a large fraction of the atoms located on the surface show structural disorder, leading to them having unique properties different to their bulk crystalline counterparts. This explains why the amorphous nanoparticles, with their disordered structure can have different properties [17–19], for advanced applications, relative to the crystal structures that have well-defined properties [14,43,44]. This is why the amorphous nanoparticles - ANPs - have aroused great interest recently, because of their great importance in science and nanotechnology [19]. The common structural characteristics of the ANPs, which determine the properties they present and define their potential applications could be listed as follows [17,18,19,22]: (1) the ANPs would have two areas: the surface shell and core (as indicated by Zhang et al. [16] in their work on amorphous TiO2), where the first would have a more porous

194


structure compared to latter, while the core has a more compact packaging; while the core is independent of size, the structure of the surface shell depends strongly on size and plays an important role in the properties the ANPs have; (2) the surface shell contains a great many structural defects, including undercoordinated sites and dangling bonds that can act as surface sites; and (3) the ANPs exhibit a layer structure commonly found in crystals with free surfaces. Specifically, TiO2-ANPs have aroused great interest from the experimental point of view [47–52] and in computer simulation [15,53,54]. Several of the physico-chemical properties of the TiO2-ANPs have been considered, including their electronic properties [55] and the effect of the content of amorphous structure on their photocatalytic properties [56]. Based on the results of these studies and the lack of long range order as well as through the metastability in thermodynamics, amorphous TiO2 exhibits novel properties. This material has shown potential applications as an electrode in solar batteries [57] and semiconductors [58] and as thin film in capacitors, given its high dielectric constant [59] as well as self-cleaning techniques due to its super-hydrophilicity [60,61,62]. It has more recently generated particular interest in the use of nanostructures of amorphous TiO2 as anodes for sodium ion rechargeable batteries [63], oxygen sensors at low temperature [64], photocatalysts active under visible light [65,66] and biomedical applications [67]. The synthesis of amorphous TiO2 has been carried out using different techniques, as described in the literature [17,18,19]. Wang et al. [68] synthesized amorphous TiO2 using as organic template a neutral amine surfactant (dodecylamine) and titanate tetrabutyl as a precursor, so that on adjusting the surfactant-Ti alkoxide ratio it was possible to obtain wormhole like framework mesostructure, with a high value of surface area (221 m2/g for the sample calcined at 450 °C for 2 h). In their work, Li et al. [69] synthesized TiO2 nanoparticles using titanium alkoxides (tetrabutyl titanate - TBOT and titanium tetraisopropoxide TTIP) as precursors, using the sol-gel method with subsequent calcination. To control the size of the TiO2-ANPs, acetic acid was used, allowing regulation of the system pH, while polyethylene glycol was used to improve dispersion. More precisely, when acetic acid was added to the TBOT solution, equiaxed TiO2-ANPs were obtained, between 38 and 52 nm. Using TTIP as precursor resulted in nanoparticles with a size of 21 nm. Among the novel synthesis techniques employed to synthesize TiO2-ANPs, the Pressure-Induced Amorphization [70,71] and impulse plasma in dielectric liquid [66] methods stand out. In this work, the amorphous TiO2 was synthesized using the solgel method. In this process, an alkoxide metal M(OR)n (M = metal ion, R = alkyl group) is hydrolyzed by the addiction of water, through the next reaction [72,73]: MðORÞn þ XH2 O→MðOHÞx ðORÞn−x þ xROH

ð1Þ

Later, propitiating a condensation reaction can be formed bonds \\M\\O\\M\\ through the dehydration reaction: M\\OH þ HO\\M→\\M\\O\\M\\ þ H2 O

ð2Þ

or the dealcoholation: M\\OH þ RO\\M\\→\\M\\O\\M\\ þ ROH

ð3Þ

reactions that lead to the progressive formation of inorganic polymeric oxide networks. The hydrolysis, condensation and polynucleation reactions depend of several factors, which included: the molar ratio of water to alkoxides, nature of the solvent, temperature, and system pH. Making a proper adjustment of this factors, can be favored or a linear polymeric gel or more crosslinked polymeric gel [73]. The gel formation can be propitiated stirring the system solvent, increasing the probability of crosslinking between the polymeric chains, or aging the solution with the objective of favored the hydrolysis and condensation reactions and promotes the polymerization reaction. When sufficient crosslinking

occurs, a sol-gel transition is observed, what causes that the viscosity increases abruptly. An adequate programmed thermic treatment let the obtainment of the oxide of interest, the amorphous TiO2 in the present case. Specifically, considering the application of TiO2 in environmental remediation, where the process of photodegradation shown by the material is important, few papers have been published that show the photocatalytic activity of amorphous TiO2 [74–76]. In these few, samples of commercially available amorphous non-hydrated TiO2 were used. The results reported by these studies indicate a negligible activity in comparison with that of the anatase structure, a behavior that researchers have justified by taking into account that in the amorphous metal oxides a high concentration of defects are found that would facilitate electron-gap recombination, reducing their photocatalytic activity [75,76]. In contrast, the amorphous phases of Nb2O5 and ZrO2 have exhibited high photocatalytic activity [77,78], greater than their crystalline counterparts, on evaluating them in aqueous methanol. Among the studies that stand out on this subject is the study conducted by Othani et al. [79] who obtained TiO2 powders containing different compositions of the amorphous and anatase phases. These samples were prepared using a precursor of amorphous TiO2 and treating it at different temperatures between 573 and 1073 °K. The samples obtained were used in three types of photocatalytic reaction, in aqueous suspensions and in all of these the photocatalytic activity of the amorphous TiO2 (non-hydrated TiO2, Ti(OH)4 or TiO(OH)2) was negligible, but increased with the presence of anatase phase. The researchers justified the low activity of amorphous TiO2 taking into consideration the rapid recombination of the photoexcited electron and hole, an action facilitated by the existence of localized defects on the surface and in the bulk of the material. Zhang and Maggard carried out work to determine the photocatalytic activity of hydrated forms of amorphous titania [74]. They synthesized at low temperatures hydrated and amorphous forms of TiO2. These showed significant photocatalysis velocities in aqueous solutions of methanol. The researchers associated the high values for the velocity of photocatalysis shown by these hydrated forms of amorphous TiO2 with the large amount of water adsorbed on the surface and to the presence of methanol species in the active sites, thereby facilitating a high concentration of Ti3+ on the surface on illuminating the samples with the 400WXe arc lamp (200 to N1000 nm). Wu et al. [80] recently formed films of hydro-oxygenated amorphous titanium (a-TiOx:OH) using plasm enhanced chemical vapor deposition (PECVD) and carried out a systematic investigation of the photocatalytic activity on them, looking at the porosity and the hydroxyl (OH) groups present in the films. They found that the photocatalytic activity of the films that showed anatase type structure (films thermally treated from 200 to 400 °C) was much higher than that of the amorphous films with porous structure. From the analysis of the results of this study the authors concluded that the photocatalytic activity of the amorphous TiO2 films was determined by the porosity originating from the appearance of the OH group. Because the results listed above and those obtained in other studies showed a negligible photocatalytic activity for amorphous TiO2 [75,76], there was no reason to conduct studies, taking appropriate care, of the bactericidal activity of this material. As such, the results obtained in this study are of interest in gaining a better understanding of the toxicity of nanoparticles of amorphous TiO2 and their potential use in environmental remediation processes. The studies that have been carried out to determine efficiency in the inactivation of bacteria have shown that the anatase phase TiO2 is the most efficient, the rutile TiO2 showing a lower efficiency [75,81–83] Furthermore, normally during bacteria inactivation tests, the TiO2 together with the bacterial strains are continuously illuminated with UV light, so that the action of the high energy photons from the UV radiation are superimposed on the action of the free radicals produced by the TiO2 and the percentage of the action of each cannot be reliably quantified. It is therefore important to separate


the action of bacterial inactivation promoted by the high energy photons from that caused by the free radicals. This paper describes the synthesis of amorphous TiO2 nanoparticles (TiO2–ANPs), using the sol-gel method, and their characterization. Considering the nature of the synthesis process a mechanism was proposed to explain the formation of these nanoparticles. Examining a potential application of the TiO2-ANPs, this article presents the results obtained upon carrying out a systematic study of the bactericidal action of amorphous TiO2 synthesized. To determine the effect of this material on the inactivation of E. coli bacteria, a predetermined amount of TiO2 was activated beforehand, exposing it to the action of UV radiation for 1 h and immediately adding it to cultures of E. coli suitably arranged for carrying out the testing. Using this methodology, the actual bactericidal effect of TiO2 on the strain of E. coli could be determined, while avoiding the simultaneous action of the UV radiation.

195

amorphousTiO2 previously activated using UV radiation, for which the material was exposed to this radiation for 1 h. To have a negative control as a reference for bacterial growth, a test was performed using the antibiotic Trimetroprime (Sigma 98%). 2.3.1. Preparation of the culture medium The culture medium used was Mueller Hinton broth. To 250 ml of distilled water in an Erlenmeyer flask, 5.25 g of Mueller Hinton Broth (powder) was added and the mixture stirred manually until completely dissolved. This culture medium was then placed in an autoclave at 140 °C for 40 min. The bacterial inoculum used in this process was obtained through the methodology outlined in a pre-defined protocol [84]. The bacterial inoculum placed in the tubes was much lower, in colony count, to the inoculum initially prepared (McFarland 0.5); and to obtain this, a serial dilution of the original inoculum was performed.

2. Materials and methods 2.1. Synthesis and characterization of amorphous TiO2 Stable TiO2 sols were obtained by the reactions of the hydrolysis and polycondensation of titanium tetrabutoxide (TBT-Across), Ti(OBu)4. To this ethyl alcohol (38.17 ml) was used as a solvent to dissolve the TBT (10.98 ml), taking care that the titanium precursor was not hydrolyzed upon contact with the environment. The solution containing TBT and ethyl alcohol was stirred for 20 min and finally distilled water (0.85 ml) as added, continuing to stir for a further 20 min. To determine the effect of the heat treatment on the structure and properties of the TiO2 obtained the sols were subjected initially to a drying program at 80 °C for about 3 days until the ceramic powders were obtained. The dry solid sample underwent a heat treatment that involved a temperature increase of 3 °C/min, from ambient to 150 °C, to evaporate both the physisorbed water and the solvent present in the sample. Considering the results of the thermic analysis performed on the synthesized solids, the powders were treated at different temperatures, up to 380 °C for 2 h. The sample dried at 80 °C was treated at temperatures of 300, 350 and 380 °C, at a heat rate of 5 °C/min. The solid material obtained from the coloidal suspensión, after drying at 80 °C, was characterized using thermal analysis with a DSC Q100-TA instrument. Solid samples were taken at different temperatures and characterized using infrared spectroscopy (Thermo Electron Nicolet IR200), UV–Visible adsorption spectroscopy (Spectronic Genesys 5), XRD (Philips PW1710 using Kα radiation from Cu (λ = 1.54 Å), in the 2θ range between 10° and 70°, at a scanning speed of 0.04° s−1), and electron microscopy both transmission (JEOLJEM 1200 EX) and scan (JEOL JSM-6340F). To break up the clusters that exist in the samples, they were subjected to attrition milling for 1 h, using ethanol as solvent. These samples, dried at 80 °C, were those observed with TEM.

2.3.2. Preparation of the nutrient agar The nutrient agar used was blood agar base. In an Erlenmeyer flask containing 400 ml of distilled water, 16 g of blood agar base (BBL Blood Agar Base - Infusion Agar) was added, keeping the mixture constantly stirred at a temperature of 300 °C. When the agar was completely dissolved, it was placed in an autoclave at a temperature of 140 °C for 40 min. Once this process was complete, 28 ml of human blood was added to mixture, stirring manually and then poured into petri dishes; this procedure was performed in a laminar flow cabinet. 2.3.3. Preparation of the TiO2 suspensions The synthesized amorphous TiO2, once in powdered form, was dispersed in dimethyl sulfoxide (DMSO). The TiO2 slurry was prepared at a concentration of 1000 ppm that according to the reported literature [85,86] is a suitable concentration for bacterial inactivation. As such, in order to do the testing and take account of the variables studied, there had to be at least 3 reaction tubes: ► one test tube containing the inoculums, to be used as positive control (normal growth) ► a tube containing the inoculum and that was subjected to the action of UV radiation ► and finally, a tube containing the inoculum and the amorphous TiO2 slurry, the oxide that was previously activated by subjecting it to UV radiation for 1 h. To prepare each of these tubes, the methodology described in a protocol pre-defined was used. The samples prepared for consideration of each of the variables mentioned above were placed in petri dishes previously conditioned for the purpose to determine the effect on the cultures at 30, 60 and 150 min after beginning the assays; reading the colony forming units was done 24 h after the start of the test, after incubating them at 37 °C.

2.2. E. coli culture The E. coli used in this work was provided by the laboratory for teaching in Biology of the University of Cauca. Twenty four hours prior to the micro dilution test, the E. coli strains were reseeded in Mueller Hinton agar, taking them from three to five well-isolated colonies of similar size and morphology. This system was incubated for 12 h at a temperature of 37 °C, under constant agitation until a turbidity equivalent to that of tube No. 0.5 on the McFarland scale was reached. 2.3. Inactivation of the E. coli Lethality curves were used to determine the inactivation of amorphous TiO2 on E. coli strains. The following variables were analyzed: direct action of UV radiation on a culture with no TiO2, phase present in the solid, and contact time of bacteria with

2.3.4. Colony count after incubation Readings were carried out by “naked eye” at 30, 60 and 150 min, recognizing the turbidity of the tubes and making the dilutions necessary for the carrying out of the count, dilutions that are listed below, according to whether turbidity is observed or not. ► For time zero (t = 0) of the lethality curve: To measure the concentration of the control tube at t = 0, two dilutions were carried out, one at 1/100 (0.05 ml from the positive control tube in 4.95 ml of Mueller Hinton (MH)) and another at 1/ 10 of the previous solution, taking 0.5 ml in 4.5 ml of MH broth to obtain a concentration 1/1000. From both tubes, 0.05 ml aliquots were taken and dripped onto the surface of Petri plates containing the appropriate medium for the growth of the microorganism

196


under study; they were then spread using a Drigalski spatula. After incubating for 24 h at 37 °C in a suitable atmosphere (containing 5% CO2), counting of colonies on the plates was done and this number was multiplied by the dilution used, i.e. by 2 × 103 in the first and by 2 × 104 in the latter. The counts obtained were assessed according to the following criteria: from the counts obtained in both plates, the one that recorded intermediate values between 50 and 200 CFU per plate was chosen, because if it was b50 the statistical error would be significant given that E(error) = 2/(n × 100)1/2, where n is the number of colonies counted, while if the count is N200, the systematic error (error in counting and overlapping colonies) would be considerable. ► For the times of 30, 60 and 150 min of the lethality curve: After 30, 60 and 150 min from the start of the test, for those tubes that remained clear (tubes in which the TiO 2 has a bacteriostatic action on the microorganism), the following serial dilutions were made in order to carry out the respective microorganism counts:

3. Results and discussion 3.1. Material characteristics 3.1.1. Thermal analysis of the solid obtained from the colloidal suspension. In the Thermogravimetric Analysis (TGA) thermogram, Fig. 1, it can be seen that the mass of the sample at 205 °C was reduced by 18.89% due to water loss, and between 205 and 250 °C it decreased 2.43% due to the oxidation of the organic phase that existed in the solid. Finally, between 250 and 420 °C there was a significant weight loss of 9.13%, which can be accounted for by considering the elimination of the organic phase that was still present in the sample. For temperatures above 450 °C, the weight loss was less evident and the mass of the solid tended to remain constant, reaching at 1197 °C a total loss of mass of the material of ~33.03%, compared to the initial mass of the sample. Moreover, observing the Differential Thermal Analysis (DTA) curve (Fig. 1), a broad endothermic peak can be seen around 100 °C, which corresponds to the evaporation of the water adsorbed in the material. Moreover, the curve has two exothermic peaks: one around 235 °C, associated with the oxidation of part of the organic phase that existed in the titanium solid, and another at 390 °C that indicates the complete combustion of the organic phase and the crystallization of the oxide.

1) 1/10 (0.1 ml from the “turbid” tube (growth tube) in 0.9 ml of MH broth) 2) 1/10 (0.1 ml of this resulting suspension in 0.9 ml of MH broth) 3) 1/10 (0.1 ml of this resulting suspension in 0.9 ml of MH broth) 4) 1/10 (0.1 ml of this resulting suspension in 0.9 ml of MH broth) From the original tube and the four dilutions above, 0.01 ml were seeded in a petri dish with the appropriate medium for the growth of the organism under study and spread using a Drigalsky spatula. After 24 h of incubation at 37 °C in the appropriate atmosphere in each case, the colonies were counted and the result multiplied by the reciprocal of the dilution, as follows: the first by 1 × 102, the second by 1 × 103, the third by 1x104, the fourth by 1 × 105 and the last dilution by 1 × 106.

3.1.2. Crystalline structure of the synthesized solids. On obtaining the ceramic powders by the sol-gel method, characterization of them was begun. The obtained solids, on drying the samples at 80 °C, underwent a number of pre-defined heat treatments based on the results of the thermal analysis (Fig. 1); the temperatures used for the treatment were: 250, 300, 350 and 380 °C, for 2 h. In Fig. 2, the X-ray diffractograms are shown for the solids of the titanium system, which indicate that even at a temperature of 350 °C, the material remains amorphous and that the transition from amorphous to anatase phase (PDF 21-1272) should occur between 350 and 380 °C.

Fig. 1. DTA/TG curves corresponding to the solid obtained from the process of drying the titanium colloidal suspension.


197

Fig. 2. X-ray diffractograms of the titanium system solids heat treated at various temperatures for 2 h, highlighting that up to 350 °C, the sample remains amorphous.

3.1.3. IR spectroscopy. Fig. 3 shows the IR spectra for the samples initially characterized by XRD. In these spectra the typical bands of water are apparent at ~3340 cm−1 (modes stretching) and ~1630 cm−1 (modes bending), indicating that the treatment performed at 150 °C did not completely eliminate all of the water present, but only that physisorbed by means of hydrogen bonds. The band around 1400 cm− 1 can be associated with the vibration of residual organic C\\H groups [87]. Looking specifically at the region of the spectrum between 1000 and 400 cm−1, where bands associated with Ti\\O, Ti\\OH and Ti\\O\\C bonds are mainly located [88], it shows a band of large area that tapers off as the temperature is increased. Bands corresponding to the Ti\\O\\Ti vibrations, at 500 cm− 1, and Ti\\O vibrations, at 600 cm− 1 are found in all the IR spectra in Fig. 3. For more detailed information on the local structure of the solids in the amorfo-anatase transition, between 350 and 380 °C according to Xray results (Fig. 2), deconvolution of the IR spectra corresponding to samples heat treated at these temperatures was performed. The region of interest, at which the deconvolution process was carried out, was that where there are mainly the bands associated with the Ti\\O, O\\Ti\\O and Ti\\OH bonds, i.e. between 1000 and 400 cm−1 (Fig. 3). Looking at

Fig. 3. IR spectra corresponding to titanium system solids heat treated at different temperatures, for 2 h.

the shape of the IR spectrum in this range, Fig. 4, the difference is evident. The spectrum of the amorphous TiO2 sample, Fig. 4(a), has a completely different shape to that of the sample that contains anatase as the predominant phase, Fig. 4(b), more so when the presence is taken into account of the Ti\\O\\C bond [88], for example, due to the fact that the organic phase has not been completely removed (see Fig. 3). In this range of the IR spectrum, the bands at 603 and 657 cm− 1 are clearly differentiated and could be associated with the species of symmetry E3u y A2u [89] related to the functional groups that involve Ti\\O bonds. Also present are the bands at 811 cm− 1 (especie A2u [90]) and that located at 930 cm−1, which only appears in this spectrum. The spectra of those samples for which the presence of anatase as the only crystalline phase was shown by XRD, Fig. 3, had a different shape, similar to that indicated in the Farmer's book [90]. In this spectrum, Fig. 4(b), bands are found located at ~ 622 cm− 1 (E3u species [89]), ~ 842 cm− 1 (A2u – LO + acoustic) [90] and at 473 cm−1 (E2u – LO [90]). 3.1.4. Morphology, particle size and agglomeration of the synthesized TiO2 particles. Fig. 5(a) shows the micrograph obtained with SEM of the amorphous TiO2 powders synthesized by sol-gel, treated at 350 °C for 2 h. Highlighted by the photograph is the presence of different sizes of agglomerates, some larger than 10 μm. Fig. 5(b) shows the photograph taken by TEM of the same sample in powder form. This amorphous TiO2 was found to have a primary particle size of ~100 nm, shape spheroidal (Fig. 5(b)) and a specific surface of 17.445 ± 0.03 m2/g. 3.1.5. UV-Visible spectroscopy To obtain the UV–Vis absorption spectra suspensions in millimolar concentrations of nanoparticles (~100 nm, Fig. 5(b)) were used. These were dispersed in ultrasound to reduce the effect of scattering of the radiation caused by them, which made it possible to obtain the spectra shown in Fig. 6. Observing carefully the UV–visible spectra in Fig. 6, changes are observed in the location of the absorption maxima, so that for amorphous TiO2, the maximum was found at 261.7 nm and a shoulder around ~216 nm, while for anatase phase TiO2, these were located at 215 and 273.8, respectively. According to the literature [49a,b], bands between 205 and 215 nm can be attributed to isolated dissolved species of titania coordinated tetrahedrally, while those located at λ N 250 nm can be associated with centers of Ti(IV) in a penta and octahedral arrangement. Given the small size of the particles, their solubility ought to increase and the isolated dissolved species would correspond to the product of

198


Fig. 4. Deconvolution of the IR spectra, in the region between 400 and 1000 cm−1, of solid samples of the titanium system heat treated at different temperatures for 2 h: (a) 350 °C and (b) 380 °C.

this dissolution. What the results absorption of UV–Vis absorption spectroscopy indicate (Fig. 6) is that the two samples of titanium amorphous and anatase - give a different result so that their optical behavior under UV–Vis radiation should not be the same. 3.1.6. Suggested mechanism for the formation of amorphous TiO2 nanoparticles (TiO2-ANPs). In this work, the sol-gel method was used to synthesize the TiO2ANPs because it is a versatile process and allows the reproducibility of the obtained ceramic powders reliably to be ensured [14,43,72,73,91, 92]. In a typical sol-gel process, initially a colloidal suspension or sol is formed due to the hydrolysis and condensation reactions undergone by the precursor, titanium tetrabutoxide - (TiOBu)4 in this work, on being immersed in a solvent, ethanol (EtOH) and water that is gradually added to the system in this case. Complete polymerization and the loss of the solvent leads to the transition from a liquid sol to a solid-type gel phase. The sol-gel process, aqueous or non-aqueous, has proved extremely useful in synthesizing titania nanostructures that have a high surface area and that are often amorphous [44]. Bearing in mind the process used in this work to synthesize the TiO2ANPs, no catalyst, either acidic or basic was added, considering the results of the work of Zhang et al. [93], where an irregular, amorphous product with a low concentration of acid was obtained. On adding Ti(OBu)4 to the ethanol, given the high reactivity of the alkoxide, an exchange is favored with the alcohol molecules, which contain active hydrogen atoms, as well as the formation of an intermediate compound, through the following reaction [94]: TiðOBuÞ4 þ EtOH↔TiðOBuÞ3 ðOEtÞ þ BuOH

a reversible reaction with an equilibrium determined by the concentration of the reactants and the reactivity of titanium alkoxide. The reaction mechanism in this system can be explained proposing a coordinationtype intermediate with the following structure:

The addition of the distilled water to the Ti(OBu)4 – EtOH system promotes hydrolysis and dehydration reactions of the intermediate compound that can be represented diagramatically very simply as follows [95]:

ð4Þ Considering the air humidity, the butyl-ethyl titanate would be hydrolyzed and dehydrated according to the following diagrammatic process [95]:

The hydrolysis of butyl-ethyl titanate and its subsequent dehydration, on adding water, can therefore be summarized as follows: Fig. 5. Micrographs obtained with scanning electron microscopy - SEM (a) and transmission - TEM (b) of titanium oxide synthesized by the sol-gel method and thermally treated to 350 °C for 2 h.

nTiðOBuÞ3−x ðOEtÞx þ 4nH2 O→nTiðOHÞ4 þ ð3−xÞnBuOH þ nxEtOH þ nH ð5Þ


199

Fig. 6. UV–visible absorption spectra corresponding to different concentrations of TiO2 synthesized in this study: (a) amorphous (350 °C) and (b) anatase (380 °C).

nTiðOHÞ4 →ðTiO2 Þn þ 2nH2 O

ð6Þ

leading to the formation of amorphous Ti(OH)4 precipitate as a result of the rapid hydrolysis reaction of the titanium intermediate. To better understand the hydrolysis and condensation reactions that occur in the Ti(OBu)4 – EtOH – H2O system, and that would lead to the formation of the TiO2-ANPs, it is necessary to consider the characteristics of the molecular structure of Ti(OBu)4. The results of the work of Babonneau et al. [96] indicated that both Ti(OEt)4 and Ti(OBu)4 had an oligomeric structure (trimeric) since they contained terminal and bridge OR groups (R = Et, Bu). If we considered the structural model proposed for Ti(OEt)4 and the similarity of the XANES and EXAFS results of this titanium alkoxide with Ti(OBu)4 [96], the structure of the latter would consist of three equivalent Ti sites in five-fold coordination, so that each titanium atom would be surrounded by three short and two long Ti\\O bonds. Furthermore, since the Ti(OBu)4 was dissolved in EtOH, before favoring the hydrolysis of the compound Butyl-Ethyl titanate through the addition of the distilled water, this dilution ought to lead to lower association, so that Ti(OBu)4 ought not to remain trimeric, more so considering that EtOH is a polar solvent [97]. Bearing in mind the nucleophilic properties of the alcohol, ligand exchange, dissociation and solvation of the oligomer would occur as shown in the following reaction:

characteristic of the titanium alkoxides since titanium always seeks to satisfy its coordination number, expanding its original coordination through undergoing the hydrolysis reactions. It is worth pointing out that due to the fact that the trimeric structure proposed for the Ti(OR)4 (R = Et, Bu) in which the titanium atom has a sixfold coordination [98–101] does not match the XANES - EXAFS experimental results, which suggest a five-fold coordination [96]. The presence of Ti6O4(OEt)16, the first hydrolysis product proposed from Bradley's model has not been observed in the studies of X-rays carried out by other groups, but rather Ti7O4(OEt)20 [102], so that it is expected that in the system used in this work to obtain the TiO2-ANPs, the compound resulting from reaction Eq. (8) taking place is obtained as the first product of hydrolysis. The state that ought to mark the boundary of soluble oxideethoxides is the formation of an infinitely long triple-chain molecule of the type [Ti3O4(OBu)4 − x(OEt)x]∞ that ought to provoke a fundamental change in the structure of the system, leading to the production of insoluble giant molecules [72,100]. Obtaining these polymers chain can be described by means of the following simplified diagram:

2 Ti3 ðOBuÞ12 þ 6EtOH→3 Ti2 ðOBuÞ8−x ðOEtÞx ; ð2−xÞEtOH; xBuOH ð7Þ Taking into account the molecular structural similarity of the Ti(OEt)4 and the Ti(OBu)4 [96], and the results of the studies done by Bradley et al. [98–100], it is expected that in the system used to obtain the TiO2-ANPs, compounds with the general formula [Ti3(y + 1)O4y(OBu)4(y + 3)−x(OEt)x] are formed, with y = 0, 1, 2, 3, …, in relation to how much water is added to the Ti(OBu)4-EtOH starting solution. Meanwhile, assuming that the metal alkoxides should adopt the smallest possible structural unit consistent with all metal atoms, to obtain the largest coordination number limited by the condition that the coordination number of oxygen does not exceed the value of four, it is to be expected that the hydrolysis of titanium alkoxides generate titanium oxide alkoxides TiOh(OR)4 − 2h [101], specifically the analog to the hydrolysis product reported by Watenpaugh and Caughlan [102], which might be formed through the reaction: 3TiðOBuÞ4−x ðOEtÞx þ 4TiðOBuÞ3−x ðOEtÞx ðOHÞ→Ti7 O4 ðOBuÞ20−x ðOEtÞx þ ð4−xÞBuOH þ xEtOH ð8Þ indicating that the condensation of hydrolyzed alkoxides should proceed via alkoxolation rather than oxolation [72], the principal

Polymers of the type (Ti3O4(OBu)4 were obtained initially by Winter and Boyd [103,104]. The experimental results show that the fourth alkoxy group is very difficult to remove via hydrolysis or alcoxolation [95,103–106] so that the condensation of butyl-ethyl titanate should occur preferentially via olation because the load conditions required (δ(OH) b b0, δ(Ti)N N 0 and N-4N N 0) are fulfilled [72]. The formation of olated polymers would be explained if it is considered that during the aging of the system, subjected during its aging to a treatment at 80 °C for 3 days, the solvent (EtOH in this work) would be released via syneresis and evaporation. On adding all the water to the Ti(OBu)4-EtOH system, and during the drying of the suspension with the subsequent heat treatment (T b 350 °C) to which the precipitate was subjected, the formation of the solid was favored that would contain TiO2-ANPs, analogous to that observed in the work of Barringer and Bowen [105], in which powders of TiO2 were obtained via controlled precipitation of Ti(OEt)4, with particles that will have an amorphous structure, inside, and that were coated with (5–10 nm) crystalline precipitates (anatase crystal structure), as a result of water washing.

200


3.2. Assay of cell viability of Escherichia coli strain To perform this assay, amorphous TiO2 synthesized by the sol gel method and heat treated at 350 °C for 2 h was used (Fig. 2). To determine the effect of the photocatalytic activity of amorphous TiO2 on bacterial growth, the oxide was previously submitted to the action of radiation for 1 h, placing it in a DMSO suspension, to favor the suspension of the nanocrystals considering that this compound will not have any additional physical-chemical effect on the nanocristales, and this was then placed on an orbital shaker that inside a laminar flow hood with overhead UV lamps (λ ~254 nm). The intensity of the light falling on the surface of the TiO2 slurry was about 30 Wm−2. Although on treating amorphous TiO2 at 350 °C during its synthesis, the water molecules are removed, subjecting it to the action of UV radiation would induce in it a super-hydrophilic behavior similar to that observed by Gao et al. in thin films of amorphous TiO2 excited using light [107]. The super-hydrophilic behavior of TiO2 is associated with the chemisorption of water molecules at Ti3+ generated by the photo reduction of the surface Ti4+ which is closely related to the oxygen bridge sites in the crystal face [108]. The existence of Ti3+ after irradiation with UV is important for the formation of the superhydrophilicity, just as Wang et al. [109] concluded after measuring a single crystal. This hydrophilicity can be maintained with sunlight. As Gao et al. [107] proposed in their work, on treating amorphous TiO2 at 350 °C, Ti\\OH bonds would be cut and generate Ti\\O\\Ti bonds. If this heat treated amorphous TiO2 is subjected to the action of UV radiation, for even a short time (in the work by Gao et al. [107] it was 5 min) and exposed to ambient atmosphere, Ti\\OH groups should be generated; this is the action performed in this work after activating the amorphous TiO2 synthesized for dispersal in cultures of E. coli. In actual fact, according to Gao et al. [107], more than the Ti\\O\\Ti groups it is the dangling bond formation that promotes the generation of photoinduced superhydrophilicity, such that a water molecule would easily be absorbed in the dangling bond for it then to become dissociated again [110], the amorphous TiO2 particles exhibiting this characteristic of superhydrophilicity as occurs with the amorphous TiO2 films [60,61]. By means of this phenomenon, the presence of water or hydroxyls on the surface of the particles of amorphous TiO2 synthesized is always guaranteed, which constitute the major traps for the photo-excited holes [111], thus promoting the production of active radicals that have an important role in the inactivation processes. The loss of bacterial viability was examined by the viable colony count method. In this research, in one of the tests as indicated above, the bacteria of E. coli were suspended with the amorphous pre-activated TiO2 and the culture was not radiated with UV light, as is usually done [112–114], in order to determine the sole effect of previously irradiated amorphous TiO2 and avoid the effect of UV radiation on the bacterial strain. At the same time, two cultures of E. coli were formed without TiO2: one to study the proliferation of bacteria in darkness, taken as a reference for the growth of the cell population, and the other was

illuminated with UV light to determine the effect of radiation on the strain of E. coli. Samples from the three cultures were taken simultaneously at 30, 60 and 150 min into the test. All the plates were incubated at 37 °C for 24 h and the viable colony count was performed, as already mentioned, on blood agar base plates, after making serial dilutions of the sample in MH broth as indicated in item 2.2. The starter bacterial suspension was formed at a concentration of 2.65 × 106 UFC/ml. Fig. 7 shows the photographs obtained of the cultures used for the study of E. coli inactivity by the amorphous TiO2 synthesized in the course of this work. In these photographs, the number of colony forming units (CFU/ml) in the starter culture at time zero can be seen, as well as the population development of these after 30 min, given a suitable medium for growth. Furthermore, it shows their development on being subjected to the action of radiation and in the presence of activated amorphous TiO2. In Fig. 7 it can be seen that in the initial culture a high quantity of CFU/ml of E. coli is present and that this underwent significant changes, at first glance, when placed in contact with the amorphous TiO2 previously exposed to the action of UV radiation for 1 h before adding it to the bacterial culture. This oxide produces a complete bacterial inactivation within 30 min of coming into contact with the bacteria. Fig. 8 shows photographs of the cultures after 60 (Fig. 8(a)) and 150 (Fig. 8(b)) min from the experiment beginning, illustrating both the growth of the E. coli strains and the effects of UV radiation and the presence of activated amorphous TiO2 on them. Behavior similar to that observed at 30 min (Fig. 7) was observed in the cultures - the complete elimination of bacteria when in direct contact with activated amorphous TiO2. The loss of viability of cells of E. coli through the action of UV radiation and through the presence of pre-activated amorphous TiO2 were determined by counting the number of colonies present in the cultures after 24 h of incubation. The curves of CFU/ml versus time for the different conditions of culture are shown in Fig. 9. In this figure shows the normal bacterial growth, starting from approximately 2.65 × 106 UFC/ml, as well as the effect of an antibiotic, trimethoprim (Sigma 98%), and that of the radiation on its proliferation over time. When the bacteria are subjected to the action of the antibiotic for 30 min, a significant decrease occurs in the viability of the strain, which then increases its proliferation again slightly over time. Meanwhile, when E. coli cells (2.65 × 106 CFU/ml) were subjected to the action of UV radiation for 30 min, the population was significantly reduced and continued to drop slowly until 150 min of exposure without achieving the complete elimination of the bacteria. Further, in Fig. 9 the curves indicate that the E. coli bacteria in the presence of 1 mg ml−1 [1000 ppm] of amorphous TiO2 previously irradiated with UV light for 1 h, lost it viability after 30 min, a result that was maintained throughout the duration of the test. Thus, complete inactivation of E. coli was evident following 30 min of exposure to activated TiO2. The population numerical data obtained in this test are shown in Table 1.

Fig. 7. Monitoring of the small-size phenotype colonies in relation to the total of colonies on LB medium during exposure of E. coli bacteria to UV radiation and amorphous TiO2 previously irradiated, after 30 min the test beginning.


201

Fig. 8. Monitoring of the E. coli cultures at 60 (a) and 150 (b) min into the test, evaluating the effect of the radiation and of the presence of previously activated amorphous TiO2 on the strains.

Fig. 9. Curves showing the variation in the number of colony forming units (CFU) versus time, to indicate the population variation of E. coli on subjecting the cultures to the action of an antibiotic (Trimethoprim), UV radiation, as well as to exposure to previously activated amorphous TiO2. The initial concentration of cultured cells was 2.65 × 106 UFC/ml.

202


The evolution of the cultures, Figs. 7 and 9, and the results shown in Fig. 9 and Table 1, on the inactivation of the bacteria, provide evidence that irradiating the amorphous TiO2 previously, to activate it, produces a strong biocidal action on the E. coli. To explain this behavior of the oxide, it should be taken note of that after the TiO2 is subjected to the action of UV radiation, electrons are produced in the conduction band and holes in the valence band through the reaction [115,116]: þ

TiO2 þ hν⇔hvb þ ecb −

ð9Þ

“Trapping” may then occur of holes in the surface states due to the nanoparticles have “a high surface to bulk ratio” encouraging a greater number of defects in the surface than in the bulk. Furthermore, since the amorphous TiO2 synthesized in this work was obtained using the sol-gel method, with the presence of an aqueous phase and later exposed to the air, it can be considered that the most abundant traps in the surface of the solid should have the form of adsorbed water or of hydroxyl groups, as a result of the dissociative chemisorption of water, which would favor the trapping of holes through the reactions [116,117]: þ

ð10Þ

þ

ð11Þ

hvb þ H2 Oads ⇔Hads þ þ OHads hvb þ OHads − ⇔ OHads

producing “free-radical” •OHads, “intermediates” of photocatalytic reactions, species that have been detected using the “spin-trap” technique [118]. These trapped holes, outlined by the above reactions, are apparently very stable as demonstrated by Bickley and Stone in their work [117]. This process of “trapping” would enter into direct competition with the electron-hole recombination process, such that through the development of reactions (7) and (8) the availability of holes for recombination is reduced, an action that diminishes the likelihood of recombination and therefore the speed of this process is reduced. The importance of the trapping of holes in the TiO2 has been found to depend on the crystalline phase that the oxide presents [115]. Both the •OH and the HO2 radicals, which are produced when there are ions from the adsorbed oxygen molecule present and reaction is favored [117]: OH þ O2ads − → HO2 þ Oads −

ð12Þ

have been detected in photocatalysis testing with anatase phase TiO2 but not with the rutile phase. Therefore, while in the anatase phase the entrapment of minority charge carriers (holes) takes place quickly, in the rutile phase the lifetime of the electrons is short. In other words, for the latter the electron-hole recombination process is the most important [115]. It is expected, therefore, that given the value of the energy gap (about ~3.4–3.5 eV [74], close to that of the anatase (~3.2 eV)), the high concentration of surface defects and the manner in which the amorphous TiO2 was synthesized in this work, in it the hole trapping Eqs. (10) and (11) predominate rather than the recombination process. Another structural aspect to consider is the presence of localized Anderson states in the amorphous TiO2 [119] that could increase the lifetimes of e− - h+ and therefore increase the photocatalytic activity of this oxide [79,116].

Table 1 Numerical data obtained using the lethality curves method on the population of E. coli strains considering the different variables of interest, at 30, 60 and 150 min into the test. Time (min)

Growth (CFU/ml)

Antibiotic Trimetroprima (CFU/ml)

Radiation (CFU/ml)

Activated amorphous TiO2 (CFU/ml)

0 30 60 150

2.65E6 3,75E6 4,6E6 4,8E6

2.65E6 7.25E5 8,55E5 8,57E5

2.65E6 4.2E5 2E5 3.6E4

2.65E6 0 0 0

During tests conducted to determine the bactericidal effect of synthesized amorphous TiO2, lighting stages were implemented, when the oxide was subjected to the direct action of UV radiation for 1 h, followed by a short dark interval, while the activated material was placed in the respective cultures of E. coli. The lighting stage caused the development of reactions (9) to (11) to be favored, while in the dark interval, considering the stability of the trapped holes and therefore that of the hydroxyl radicals, •OH, it is possible that electron capture by the perhydroxyl radical •HO2 (hydrogen dioxide radical) is favored, obtained through reaction (12) to produce the perhydroxyl ion HO2 (hydrogen dioxide (− 1) anion or hydrogen peroxide (− 1)) to be established in a potential trap for the gaps and would favor the reaction [117]: þ

HO2 − þ h →HO2

ð13Þ

This reaction would increase the separation of the photo-excited charges. As Bickley and Stone indicate [117], it is very likely that during the illumination stage, and subsequent “dark interval”, species are formed that are more efficient than the original hydroxyl ions in trapping holes and increasing the separation of the photo-excited charge. The scientific literature [117,120,121] indicates that the presence of water or water vapor increases “the ability” of the TiO2 to participate in the photoprocess, such that the residual hydroxyl would be those responsible for increasing its photoactivity. Meanwhile, the decrease in the particle size should result in a greater photon efficiency from a higher interfacial charge carrier transfer rate [122]. For ultrafine particles, as is the case in the present study (Fig. 5), many e−/h+ pairs should be generated very close to the surface so that the process of trapping holes by the hydroxyl groups, reactions (10) and (11), would be favored. Additionally, it is expected that the amorphous TiO2 has a surface with a lot of defects, mainly oxygen vacancies, which would favor the dissociation of the water, forming two hydroxyl groups per vacancy [123]. Additionally, various ways of generating reactive oxygen species (ROS) by the nanoparticles have been proposed [124,125]. Among these, the possibility has been considered that the small particles could induce the spontaneous generation of ROS, given the physico-chemical characteristics of the surface [125,126]. Just as occurs in other systems [127], a mechanism may be shown that would allow the production of ROS, even in the dark, considering the action of the superoxide species generated, mainly, by the action of the existing surface defects in the TiO2-ANPs (ionized oxygen vacancies, for example) on the O2 present in the environment, and through the following reactions, species such as hydroxyl radical, •OH, and hydrogen peroxide, H2O2, would occur [127]: O2 þ e− → O2 −

ð14Þ

O2 − þ H2 O→ HO2 þ OH−

ð15Þ

HO2 þ HO2 →H2 O2

ð16Þ

H2 O2 þ O2 − →O2 þ OH þ OH−

ð17Þ

In this scheme, the oxygen in the atmosphere may react with electrons from the surface of the amorphous TiO2 to form the superoxide radical (reaction 14), which, in the presence of water, is solvated, forming hydroperoxyl radicals (reaction 15). These hydroperoxyl radicals may be recombined to form H2O2 (reaction 16) and the H2O2 can react with the superoxide radical to form hydroxyl ions and hydroxyl radicals (reaction 17). On the other hand, in relation to the generation of ROS by amorphous TiO2, Jiang et al. [128] studied the correlation between crystal phase and oxidant capacity using TiO2 nanoparticles of 11 different crystal phase combinations but similar size. They found that the ability of


different crystal phases of TiO2 nanoparticles to generate ROS was highest for amorphous, followed by anatase, and then anatase/rutile mixtures, and lowest for rutile samples. Finally, although the mechanism that justifies the bactericidal action of activated amorphous TiO2 is not known, it can be considered that the attack by the •OH and HO2 radicals produced by means of reactions (10) to (12), and which arise from the previous action of UV radiation on the TiO2 nanoparticles, would affect the membrane permeability of the bacteria, responsible for example for their respiratory activity, causing the loss of this function and therefore their death, a behavior that has been the subject of studies by a number of researchers [82,86,116,129, 130]. Hydroxyl radicals, •OH, and hydrogen peroxide (H2O2) produced by means of reactions (14) to (17) would also affect the bacteria membrane. The damage to the cell wall could be generated when E. coli bacteria come into contact with the amorphous TiO2, due to the fact that the barrier imposed by the outer lipid membrane (from 6 to 18 nm thick) and the peptidogly can layer are ruptured severely by the ongoing reactions, photocatalytic or otherwise (surface reaction), during the first 30 min of the test, generating progressive cytoplasmic membrane damage [116], such that the bacteria becomes permeable, even to large molecules. Additionally, Kumar et al. [126] propose that the production of a substantial quantity of ROS is possible through the interaction of the nanoparticles with the cell components since according to the authors, the ROS generated in the absence of cells (cell-free system) was negligible and independent of nanoparticle concentration. As the aforementioned phenomena take place, the cytoplasmic membrane would be seen to be severely affected, an occurrence that would explain the irreversible loss of viability such as is shown in Figs. 7 and 8, and indicates the behavior of the curves in Fig. 9 and the data in Table 1. The possibility should not be ruled out that the particles of amorphous TiO2 initially are adsorbed on the surface of bacteria, an action favored by electrostatic attractive interactions since the point of zero charge for the E. coli is pH 2–3 and for the TiO2 pH 3.56, for it to then gain entry to the bacteria [82]. Recent work on this topic using eukaryotic cells [131,132] indicates that the absorption of the TiO2 particles may result from phagocytosis [132]. Once the TiO2 particles have entered the cells, and the cytoplasmic membrane has deteriorated, the intracellular components would be more exposed to direct attack. Previous studies indicate that the TiO2 shows no selectivity in attacking the cellular components [133,134]. In other words, on entering, it indiscriminately attacks all the intracellular components present. In summary, the bacteria in their normal growth media (2.65 × 106 UFC/ml) have their intracellular contents intact, protected by the cell wall and the cytoplasmic membrane. After 30 min, however, of exposure to amorphous TiO2 previously activated with UV light, the integrity of the cell wall and the cytoplasmic membrane is compromised by the action of oxide, either through leakage of the cell contents or the entry of TiO2 particles. In the aforementioned mechanism of bacterial inactivation, modification of the permeability of the cell wall and cytoplasmic membrane-the main targets of attack by TiO2 particles - is important. When the cell wall breaks, the radicals, mainly the hydroxyl radical •OH, which has the most bactericidal role, and H2O2 (2OH → H2O2), begin to act significantly in bacterial inactivation, entering the damaged cell and directly attacking the enzymes [82,112] and DNA [82,130]. As suggested by Kumar et al. [126] an increase in lipid peroxidation is expected on putting the nanoparticles of amorphous TiO2 in contact with the cells, and that the peroxidation occurs like a chain reaction, propagating through an intermediate peroxy radical for as long as unsaturated lipid molecules are available in the cell. Considering the results of this work and those obtained by other researchers that evaluate the antibacterial behavior of the nanoparticles of some oxides [125,135,136], can be concluded that, although is necessary to performed a more detailed and rigorous study of the antibacterial capacity of the amorphous TiO2 and compare it with the commercial TiO2, the TiO2-ANPs presented a promising antibacterial functionality and

203

comparable with other oxides. Specifically, the work of Azam et al. [135] showed that the nanoparticles of ZnO, CuO and Fe2O3 were more efficient against Gram-positive bacterial strains compared to Gram-negative bacterial strains, standing out the ZnO, behavior that contrast the favorable results obtained in this work of the amorphous TiO2 about the E. coli. In general, the antibacterial behavior of the nanoparticles would be determinate by their particle size, their chemical composition and immobilization strategy [136] as well as their crystallization degree, as suggested by the results of this work. 4. Conclusions In this study, in a controlled and reproducible way using a simple sol-gel method, amorphous TiO2 nanoparticles (TiO2-ANPs) were synthesized to have a size of b 100 nm and a spheroidal morphology. Considering the nature of the synthesis process used and on previous work carried out by other research groups, a possible mechanism was proposed that may explain the formation of these nanoparticles. Moreover, in the course of this research it was shown that E. coli bacteria exposed to the action of amorphous TiO2 that has been previously activated using UV light lose their viability, degrading completely within 30 min of the experiment starting. The methodology, comprising stages of lighting followed by dark intervals, revealed, in so separating them, the true effect of activated amorphous TiO2 and of UV radiation on E. coli strains. The “memory effect” with amorphous TiO2, in which the specimen recovers its activity in the dark, is explained by taking into account the formation of apparently quite stable hole traps, mainly hydroxyl groups (OH−) and perhydroxyl (HO− 2 ) ions, which would generate stable hydroxyl radicals (•OH) that would have the most bactericidial role in the inactivation processes. The nano-sized particles with a high concentration of surface defects favoring the dissociative chemisorption of the water, encouraging a high concentration of residual hydroxyls, as well as the amorphous nature of the oxide, which guarantees the presence of localized Anderson states, are aspects which would increase the photocatalytic activity of the specimen and would explain the substantial bactericidal action of amorphous TiO2 on strains of E. coli. The results obtained in this work enable a better understanding of the toxicity of amorphous TiO2 nanoparticles and their potential use in the inactivation of bacteria, specifically strains of E. coli. Acknowledgements We are grateful to the University of Cauca for making their laboratory facilities available for carrying out this work and to VRI – Unicauca for all logistical support. We are especially grateful to Colin McLachlan for suggestions relating to the English text. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]

J.F. Banfield, D.R. Veblen, Am. Mineral. 77 (1992) 545. W.H. Bauer, Acta Crystallogr. B 35 (1979) 530. G.V. Samsonov, The Oxide Handbook, IFI/Plenum Press, New York, 1982. X. Bokhimi, A. Morales, M. Aguilar, J.A. Toledo-Antonio, F. Pedraza, Int. J. Hydrog. Energy 26 (2001) 1279. F.A. Grant, Rev. Mod. Phys. 31 (1959) 646. L. Brohan, A. Verbaere, M. Tournoux, G. Demazeau, Mater. Res. Bull. 17 (1982) 355. A.G. Dylla, G. Henkelman, K.J. Stevenson, Acc. Chem. Res. 46 (5) (2013) 1104. L.S. Dubrovinsky, N.A. Dubrovinskaia, V. Swamy, N.M. Muscat, R.A. Harrison, B. Holm, B. Johansson, Nature 410 (2001) 653. J. Akimoto, Y. Gotoh, Y. Oosawa, N. Nonose, T. Kumagai, K. Aoki, H. Takei, J. Solid State Chem. 113 (1994) 27. M. Mattesini, J.S.D. Almeida, L. Dubrovinsky, N. Dubrovinskaia, B. Johansson, R. Ahuja, Phys. Rev. B 70 (2004) 212110. M. Latroche, L. Brohan, R. Marchand, M. Tournoux, J. Solid State Chem. 81 (1989) 78. J.K. Dewhurst, J.E. Lowther, Phys. Rev. B 54 (1996) R3673. M. Gopel, W.J. Moberly Chan, L.C. De Jonghe, J. Mater. Sci. 32 (1997) 6001. S.G. Kumar, K.S.R.K. Rao, Nanoscale 6 (2014) 11574. V.V. Hong, H. Zung, N.H.B. Trong, Eur. Phys. J. D 44 (2007) 515. H. Zhang, B. Chen, J.F. Banfield, Phys. Rev. B 78 (2008) 214106.

204


[17] V.V. Hong, Amorphous nanoparticles, in: K.S. Sattler (Ed.), Handbook of Nanophysics: Nanoparticles and Quantum Dots, CRC Press, New York 2010, pp. 1-1–1-10. [18] V.V. Hong, Chem. Phys. Res. J. 4 (1–2) (2011) 43. [19] V.V. Hong, D. Ganguli, Phys. Rep. 518 (2012) 81. [20] L.X. Chen, T. Rajh, Z. Wang, M.C. Turnauer, J. Phys. Chem. B 101 (1997) 10688. [21] K.L. Yeung, A.J. Maira, J. Stol, E. Hung, N.K.-C. Ho, A.C.W.J. Soria, K.-J. Chao, P.L. Yue, J. Phys. Chem. B 106 (2002) 4608. [22] V.V. Hong, Phys. Status Solidi B 244 (2007) 1280. [23] V. Petkov, G. Holzhuter, U. Troge, T. Gerber, B. Himmel, J. Non-Cryst. Solids 231 (1998) 17. [24] U. Diebold, Surf. Sci. Rep. 48 (2003) 53. [25] U. Bach, Y. Tachinaba, J.E. Moser, S.A. Haque, J.R. Durrant, M. Graetzel, D.R. Klug, J. Am. Chem. Soc. 121 (1999) 7445. [26] N.G. Park, J. Van de Lagemaat, A.J. Frank, J. Phys. Chem. B 104 (2000) 8989. [27] M. Kaneko, I. Okura (Eds.), Photocatalysis: Science and Technology, KodanshaSpring Verlag, New York, 2002. [28] M. Anpo, P.V. Kamat (Eds.), Environmentally Benign Photocatalysts, Springer Science + Bussines Media, New York, 2010. [29] G. Nogami, R. Shiratsuchi, S. Ohkubo, J. Electrochem. Soc. 138 (3) (1991) 751. [30] E. Topoglidis, A.E. Cass, G. Gilardi, S. Sadeghi, N. Beaumont, J.R. Durrant, Anal. Chem. 70 (1998) 5111. [31] J. Geserick, T. Froeschl, N. Huesing, G. Kucerova, M. Makosch, T. Diemant, S. Ecckle, R.J. Behm, Dalton Trans. 40 (2011) 3269. [32] Z. Zhang, A. Kladi, X.E. Verykios, J. Phys. Chem. 98 (1994) 6804. [33] M.M. Shubert, V. Plzak, J. Garche, R.J. Behm, Catal. Lett. 76 (2001) 143. [34] W.P. Hsu, R. Yu, E. Matijevic, J. Colloid Interface Sci. 156 (1993) 56. [35] P. Kubiak, T. Froeschl, N. Huesing, U. Hoermann, U. Kaiser, R. Schiller, C.K.C.K. Weiss, J. Landfester, M. Wohlfahrt-Mhrens, Small 7 (2011) 1690. [36] L. Kavan, Chem. Rev. 12 (2012) 131. [37] A. Mills, H.R. Davis, D. Worsley, Chem. Soc. Rev. 22 (1993) 417. [38] P. Pichat (Ed.), Photocatalysis and Water Purification, Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim, 2013. [39](a) F. Hossain, O.J. Perales-Pérez, S. Hwang, F. Román, Sci. Total Environ. 1047 (2014) 466–467; (b) M. Zimbane, M.A. Bucheri, G. Cacciato, R. Sanz, G. Rappazo, S. Boninelli, R. Reitano, L. Romano, V. Privitera, M.G. Grimaldi, Appl. Catal. B Environ. 165 (2015) 487. [40] N. Cioffi, M. Rai (Eds.), Nano-antimicrobials: Progress and Prospects, Spring-verlag. Berlin, Heidelberg, 2012. [41] Y. Paz, Z. Luo, L. Rabenberg, A. Heller, J. Mater. Res. 10 (1995) 2842. [42](a) G. Sheveglieri (Ed.), Gas Sensors, Kluwer Academic Publishers, Dordrecht, 1992; (b) G. Eranna, Metal Oxide Nanostructures as Gas Sensing Devices, Taylor & Francis Group, LLC, Boca Raton, 2012. [43] X. Chen, S.S. Mao, Chem. Rev. 107 (2007) 2891. [44] T. Fröschl, U. Hörmann, P. Kubiak, G. Kucerová, M. Pfanzelt, C.K. Eiss, R.J. Behm, N. Hüsing, U. Kaiser, K. Landfester, M. Wohlfahrt-Mehrens, Chem. Soc. Rev. 41 (2012) 5313. [45] M. Cargnello, T.R. Gordon, C.B. Murray, Chem. Rev. 114 (19) (2014) 9319. [46] L. Qiao, M.T. Swihart, Adv. Colloid Interf. Sci. (2016) In press http://dx.doi.org/10. 1016/j.cis.2016.01.005. [47] H. Zhang, J.F. Banfield, Chem. Mater. 14 (2002) 4145. [48] Y. Gao, Y. Masuda, W.-S. Seo, H. Ohta, K. Koumoto, Ceram. Int. 30 (2004) 1365. [49](a) L. Marchese, E. Gianotti, V. Dellarocca, T. Maschmeyer, F. Rey, S. Coluccia, J.M. Thomas, Phys. Chem. Chem. Phys. 1 (1999) 585; (b) M.E. Raimondi, E. Gianotti, L. Marchese, G. Martra, T. Maschmeyer, J.M. Seddon, S. Coluccia, J. Phys. Chem. B 104 (2000) 7102. [50] J. Tang, F. Redl, Y. Zhu, T. Siegrist, L.E. Brus, M.L. Steigerwald, Nano Lett. 5 (2005) 543. [51] L. Zhao, J. Yu, J. Colloid Interface Sci. 304 (2006) 84. [52] C.-I. Wu, J.-W. Huang, Y.-L. Wen, S.-B. Wen, Y.-H. Shen, M.Y. Yeh, Mater. Lett. 62 (2008) 1923. [53] V.V. Hong, Nanotechnology 19 (2008) 105706. [54] G. Madras, B.J. McCoy, Cryst. Growth Des. 7 (2007) 250. [55] V. Luca, S. Djajanti, R.F. Howe, J. Phys. Chem. B 102 (1998) 10650. [56] L. Gao, Q. Zhang, Scr. Mater. 44 (2001) 1195. [57] H. Usui, O. Miyamoto, T. Nomiyama, Y. Horie, T. Miyazaki, Sol. Energy Mater. Sol. Cells 86 (2005) 123. [58] N. Sakai, A. Fujishima, T. Watanabe, K. Hashimoto, J. Electrochem. Soc. 148 (2001) E395. [59] S. Scarel, C.R. Aita, H. Tanaka, K. Hisano, J. Non-Cryst. Solids 303 (2002) 50. [60] Y. Gao, Y. Masuda, K. Koumoto, Langmuir 20 (2004) 3188. [61] Y.-M. Jeong, J.-K. Lee, H.-W. Jun, G.-R. Kim, Y. Choe, J. Ind. Eng. Chem. 15 (2009) 202. [62] H. Sabbah, Mater. Express 3 (2) (2013) 171. [63] H. Xiang, M.D. Slater, M. Balasubramaniam, C.S. Johnson, T. Rajh, J. Phys. Chem. Lett. 2 (2011) 2560. [64] H.F. Lu, F. Li, G. Liu, Z.-G. Chen, D.-W. Wang, H.-T. Fang, G.Q. Lu, Z.H. Jiang, H.-M. Cheng, Nanotechnology 19 (2008) 405504. [65] C. Randorn, J.T.S. Irvine, P. Robertson, Int. J. Photoenergy 2008 (2008), 426872. http://dx.doi.org/10.1155/2008/426872. [66] E. Omurzak, T. Mashimo, C. Iwamoto, Y. Matsumoto, S. Sulaimankulova, J. Nanosci. Nanotechnol. 9 (11) (2009) 6372.

[67] M. Kulkami, A. Mazare, E. Gongadze, S. Perutkova, V. Kralj-Iglic, I. Milosev, P. Schmuki, A. Iglic, M. Mozetic, Nanotechnology 26 (2015) 062002. [68] Y.-D. Wang, C.-L. Ma, X.-D. Sun, H.-D. Li, J. Non-Cryst. Solids 319 (2003) 109. [69] Z. Li, Y. Zhu, J. Wang, Q. Guo, J. Li, Ceram. Int. 41 (2015) 9057. [70] Q. Li, B. Liu, L. Wang, D. Li, R. Liu, B. Zou, T. Cui, G. Zou, J. Phys. Chem. Lett. 1 (2010) 309. [71] Q. Li, R. Liu, T. Wang, K. Xu, Q. Dong, B. Liu, J. Liu, B.I. Liu, AIP Advances, 5, 2015 097128. [72] J. Livaje, M. Henry, C. Sanchez, Prog. Solid State Chem. 18 (1988) 259. [73] M. Kakihana, J. Sol-Gel Sci. Technol. 6 (1996) 7. [74] Z. Zhang, P.A. Maggard, J. Photochem. Photobiol. A Chem. 186 (2007) 8. [75] K. Tanaka, M.F.V. Capule, T. Hinasag, Chem. Phys. Lett. 187 (1991) 73. [76] H. Jeensen, K.D. Joensen, J.E. Jorgensen, J.S. Pedersen, E.G. Sogaard, J. Nanopart. Res. 6 (2004) 519. [77] H. Kominami, K. Oki, M. Khono, S.I. Onoue, Y. Kera, B. Ohtani, J. Mater. Chem. 11 (2001) 604. [78] C.Y. Wu, X.P. Zhao, Y.J. Ren, Y.H. Yue, W.M. Hua, Y. Cao, Y. Tang, Z. Gao, J. Mol. Catal. A Chem. 229 (2005) 233. [79] B. Othani, Y. Ogawa, S. Nishimoto, J. Phys. Chem. B 101 (1997) 3746. [80] C.Y. Wu, B.S. Chiang, S. Chang, D.S. Liu, Appl. Surf. Sci. 257 (2011) 1893. [81] S.T.L. Suib (Ed.), New and Future Developments in Catalysis: Solar Photocatalysis, Chapter 10, Elsevier B. V., Amsterdam, The Netherlands, 2013. [82] V.H. Grassian (Ed.), Nanoscience and Nanotechnology: Environmental and Health Impacts, John Wiley & Sons, Inc., publication, Hoboken - New Jersey, 2008 (chap. 13). [83] A. Rincón, C. Pulgarin, Appl. Catal. B Environ. 49 (2004) 99. [84](a) G.J. Tortera, B.R. Funke, C.L. Case, Introducción a la microbiología, Editorial Médica Panamericana, Bogotá, 2007; (b) Manual de pruebas de susceptibilidad antimicrobial, American Society for Microbiology, Washington DC, in: M.B. Coyle (Ed.), Methodology to Obtain the Bacterial Inoculum, 2005 (https://nepis.epa.gov and www.analizacalidad.com/ docftp/fi148anmic.pdf). [85] M. Bekbolet, Water Sci. Technol. 35 (11–12) (1997) 95. [86] M. Cho, H. Chung, W. Choi, J. Yoon, Appl. Environ. Microbiol. 71 (1) (2005) 270. [87] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Part A, B, Wiley-Interscience, New Jersey, 2009. [88] V.A. Zeitler, C.A. Brown, J. Phys. Chem. 6 (9) (1957) 1174–1177. [89] J.T. Luxon, R. Summitt, J. Chem. Phys. 50 (1969) 1366. [90] V.C. Farmer (Ed.), The Infrared Spectra of Minerals, Mineralogical Society Monograph 4, Adlard & Sons Ltd., at the Bartholomew Press, Dorking, Surrey 1974, p. 194. [91] C.J. Brinker, G.W. Scherer, Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing, Academic Press, Inc. San Diego, 1990. [92] A.C. Pierre, Introduction to Sol-Gel Processing, Kluwer Academic Publishers, Norwell, Massachusetts, 1998. [93] D. Zhang, L. Qi, J. Ma, H. Cheng, J. Mater. Chem. 12 (2002) 3677. [94] H.H. Beacham, Adv. Chem. Ser. 23 (1959) 282. [95] T. Ishino, S.-I. Minami, Technol. Rep. Osaka Univ. 3 (1953) 357. [96] F. Babonneau, S. Doeuff, A. Leaustic, C. Sanchez, C. Cartier, M. Verdaguer, Inorg. Chem. 27 (1988) 3166. [97] D.C. Bradley, R.C. Mehrotra, D.P. Gaur, Metal Alkoxides, Academic Press, London, 1978. [98] D.C. Bradley, R. Gaze, W. Wardlaw, J. Chem. Soc. (1955) 721–726, http://dx.doi.org/ 10.1030/JR9550000721. [99] D.C. Bradley, R. Gaze, W. Wardlaw, J. Chem. Soc. (1955) 3977–73982, http://dx.doi. org/10.1030/JR9550003977. [100] D.C. Bradley, R. Gaze, W. Wardlaw, J. Chem. Soc. (1957) 469–478, http://dx.doi.org/ 10.1039/JR9570000469. [101] D.C. Bradley, Nature 182 (1) (1958) 1211. [102] K. Watenpaugh, C.N. Caughlan, Chem. Commun. (1967) 76–77, http://dx.doi.org/ 10.1039/C19670000076. [103] G. Winter, J. Oil Color Chemists' Assoc. 34 (1951) 30. [104] T. Boyd, J. Polym. Sci. 7 (6) (1951) 591. [105] E.A. Barringer, H.K. Bowen, Langmuir 1 (1985) 414. [106] E.A. Barringer, H.K. Bowen, Langmuir 1 (1985) 420. [107] Y. Gao, Y. Masuda, K. Koumoto, Langmuir 20 (2004) 3188. [108] R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojima, A. Kitamura, M. Shimohigoshi, T. Watanabe, Nature 388 (1997) 431. [109] R. Wang, N. Sakai, A. Fujishima, T. Watanabe, K. Hashimoto, J. Phys. Chem. B 103 (1999) 2188. [110] J.P. Jolivet, M. Henry, J. Livage, Metal Oxide Chemistry and Synthesis, John Wiley & Sons, Chichester, UK, 2000. [111] C.S. Turchi, D.F. Ollis, J. Catal. 122 (1990) 178. [112] T. Matsunaga, R. Tomoda, T. Nakajima, H. Wake, FEMS Microbiol. Lett. 29 (1985) 211. [113] K. Sunada, T. Watanabe, K. Hashimoto, J. Photochem. Photobiol., A 156 (2003) 227. [114] Z. Huang, P.C. Maness, D.M. Blake, E.J. Wolfrum, S. Smolinski, W. Jacoby, J. Photochem. Photobiol., A 130 (2000) 163. [115] K.M. Schindler, M. Kunst, J. Phys. Chem. 94 (1990) 8222. [116] R.I. Bickley, T. Gonzales-Carreño, J.S. Lees, L. Palmisano, R.J.D. Tilley, J. Solid State Chem. 92 (1991) 178. [117] R.I. Bickey, F.S. Stone, J. Catal. 31 (1973) 389. [118] C.D. Jaeger, A.J. Bard, J. Phys. Chem. 83 (24) (1979) 3146. [119] N.F. Mott, E.A. Davis, Electronic Processes in Non-crystalline Materials, Clarendon Press, New York, 1979 210.

M.A. Vargas, J.E. Rodríguez-Páez / Journal of Non-Crystalline Solids 459 (2017) 192–205 [120] V.F. Kiselev, O.V. Krylov, Electronic phenomena in adsorption and catalysis: on semiconductors and dielectrics, Springer Ser. Surf. Sci. 7 (1987) (Springer Verlag, Berlin Heidelberg, chap. 9). [121] S.R. Morrison, The Chemical Physics of Surfaces, 2th Edition, Plenum Press, New York, 1990 (chap. 9). [122] Z. Zhang, C.-C. Wang, R. Zakaria, J.Y. Ying, J. Phys. Chem. B 102 (1998) 10871–10878. [123] X. Pan, M.Q. Yang, X. Fu, N. Zhang, Y.J. Xu, Nanoscale 5 (2013) 3601. [124] J. Song, H. Kong, J. Jang, Chem. Commun. 36 (2009) 5418. [125] T. Xia, M. Kovochich, J. Brant, M. Hotze, J. Sempf, T. Oberley, C. Sioutas, J.I. Yeh, M.R. Wiesner, A.E. Nel, Nano Lett. 6 (2006) 1794. [126] A. Kumar, A.K. Pandey, S.S. Sinh, R. Shanker, A. Dhawan, Free Radic. Biol. Med. 51 (2011) (1872). [127] V.L. Prasanna, R. Vijayaraghavan, Langmuir 31 (33) (2015) 9155. [128] J. Jiang, G. Oberdorster, A. Elder, R. Gelein, P. Mercer, P. Biswas, Nanotoxicology 2 (1) (2008) 33.

205

[129] P.C. Maness, S. Smolinski, D.M. Blake, Z. Huang, E.J. Wolfrum, W.A. Jacoby, Appl. Environ. Microbiol. 65 (1999) 4094. [130] H.A. Foster, I.B. Ditta, S. Varghese, A. Steele, Appl. Microbiol. Biotechnol. 90 (6) (2011) 1847. [131] H. Sakai, E. Ito, R. Cai, T. Yoshioka, Y. Kubota, K. Hashimoto, A. Fujishima, Biochim. Biophys. Acta 1201 (1994) 259. [132] R. Cai, R.K. Hashimoto, K. Itoh, Y. Kubota, A. Fujishima, Bull. Chem. Soc. Jpn. 64 (1991) 1268. [133] H. Hidaka, S. Horikoshi, K. Ajisaka, J. Zhao, N. Serpone, J. Photochem. Photobiol., A 108 (1997) 197. [134] H. Hidaka, S. Horikoshi, N. Serpone, J. Knowland, J. Photochem. Photobiol., A 111 (1997) 205. [135] A. Azam, A. Ahmed, M. Oves, M.S. Khan, S.S. Habib, A. Memic, Int. J. Nanomedicine 7 (2012) 6008. [136] M. Moritz, M. Geszke-Moritz, Chem. Eng. J. 228 (2013) 596.