New Approach of the Oxidant Peroxo Method (OPM) Route to Obtain Ti(OH)4 Nanoparticles with High Photocatalytic Activity under Visible Radiation

Brazilian Nanotechnology National Laboratory (LNNano), Brazilian Center for Research in Energy and Materials (CNPEM), Zip Code 13083-970 Campinas, São Paulo, Brazil National Laboratory of Nanotechnology for Agrobusiness (LNNA), EMBRAPA-Brazilian Agricultural Research Corporation, Rua XV de Novembro, 1452, São Carlos, SP 13560-970, Brazil Interdisciplinary Laboratory of Electrochemistry and Ceramics (LIEC), Department of Chemistry, Federal University of São Carlos, Rod. Washington Luis km 235, CP 676 São Carlos, SP 13565-905, Brazil FACET-Department of Chemistry, Federal University of Grande Dourados, Dourados, Mato Grosso do Sul 79804-970, Brazil


Introduction
Photoactivated processes are receiving enormous attention due to important applications as solar cells, CO 2 photoreduction, water splitting for hydrogen production, for the degradation of organic compounds, and others [1,2]. During these processes, semiconductors are irradiated with UV or visible radiation to excite electrons from the valence band (VB) to the conduction band (CB) creating electron/hole pairs. Most of the pairs recombine, but the remaining, via oxidation-reduction reactions, promote the oxidation of water, producing hydroxyl radicals (OH ⋅ ) from the positive holes in the VB and superoxide anions (O 2 − ) from the electrons in the CB. In addition, the electrons in the CB can reduce the carbon dioxide (CO 2 ) generating hydrocarbons with a greater added value. These powerful oxidant radicals react with organic and toxic compounds, converting them to water, carbon dioxide, and other simple substances [3,4].
In the photocatalytic process, two events can occur simultaneously: one involves the oxidation of adsorbed water in the photogenerated holes in the VB, generating hydroxyl radicals (OH ⋅ ) that have a high oxidation power of organic compounds and the second is related to the reduction of an acceptor of electrons (usually dissolved O 2 or CO 2 ) that were excited to CB.
The photogenerated electrons in the CB react with CO 2 , reducing them to carbon monoxide, methane, ethanol, etc. [5]. This catalytic CO 2 photoreduction process, also called artificial photosynthesis, is a branch of heterogeneous photocatalysis.
The most widely used semiconductor in photocatalytic processes is TiO 2 ; however, due to its wide band gap (3.2 eV for the anatase phase), it can be activated only by near-UV light (k < 385 nm), which represents a small fraction of solar light (about 3-4%). The inactivity under visible light strongly limits the practical application of the TiO 2 photocatalyst [6]. In recent years, much research has been focused on improving the photocatalytic property of TiO 2 , trying to minimize the two main drawbacks associated to the use of TiO 2 : (i) high electron and hole recombination rates and (ii) low efficiency under visible irradiation [7].
In order to utilize visible light from solar energy and enhance the photocatalytic reactions, efforts have been focused on exploring novel methods to modify semiconductors such as the process of doping, sensitization, and formation of heterostructures [8]. Photosensitization can be achieved by a photosensitizer that absorbs light energy transforming it into chemical energy, which is transferred to substrates under favorable conditions. The photosensitizers may be adsorbed on the semiconductor surface by an electrostatic or chemical interaction and, upon excitation, inject electrons into the conducting band of the semiconductor. Based on the reported studies in the literature, organic dyes and coordination metal complexes are very effective sensitizers [9][10][11].
In this work, the OPM route was used with a new approach to obtain nanoparticles of Ti(OH) 4 with peroxo groups on the surface, in which its photocatalytic activity was evaluated both in the degradation process of the methylene blue dye and the amiloride drug, and in the process of CO 2 photoreduction, with particular emphasis on the effect of the groups on the optical properties and the photocatalytic efficiency of the catalyst.  [12][13][14]. The peroxytitanate complex solution was then heated to 80°C under stirring until the formation of a yellow gel that was dried at 60°C for 24 h to form the Ti(OH) 4 with the surface modified with the peroxo group. The Ti(OH) 4 was calcined at different temperatures between 300 and 500°C for 1 h at a heating rate of 10°C min −1 in closed alumina boats. The different samples calcined 300, 400, and 500°C are referred as TiO 2 -300°C, TiO 2 -400°C, and TiO 2 -500°C, respectively.

2.2.
Characterization. The XPS spectra were performed using an XPS VG Microtech ESCA 3000 (MgKα and AlKα radiations) at an operating pressure of 3 · 10 −10 mbar. The binding energies were corrected for the charging effect by assuming a constant binding energy for the adventitious O1s peak. The LabRAM microspectrometer (HORIBA JobinYvon) was used to obtain the Raman spectra of the catalysts. The excitation wavelength was a 514.5 nm line of a 5.9 mW He-Ne laser as the excitation source is through an Olympus TM BX41 microscope. Thermogravimetric analysis of the catalyst was carried out using a TGA Q500 thermogravimetric analyzer (TA Instruments, New Castle, DE) in air flow and recorded from room temperature to 600°C, at a constant heating rate of 10°C/min. The Ti(OH) 4 was characterized by differential scanning calorimetry (DSC 404C controlled by TASC 424/3A, Netzsch, Germany) between 25 and 550°C using an aluminum crucible and a constant heating/cooling rate of 10°C/min with a flux of 0.50 cm 3 /min. A Rigaku D/ MAX 200 diffractometer was used to determine crystal structure of the catalysts, with CuKα radiation and scanned for 2θ values from 5 to 80°. Infrared absorption spectra were recorded by a Fourier transform infrared spectrometer (FTIR) (Bruker EQUIXOX 55, Ettlingen, Germany). Spectra were recorded for KBr disks containing a powdered sample and obtained in the 400-1000 cm −1 range, with 64 scans and 4 cm −1 resolution. The surface morphology of the catalysts was observed by field emission scanning electron microscopy (FEGSEM, ZEISS model-SUPRA 35). The band gap of the catalysts was determined according to the method proposed by Wood and Tauc [15] and Tolvaj et al. [16]. The spectra were recorded using the UV-visible diffuse reflectance spectroscopy (DRS) at room temperature between 200 and 800 nm using a Varian model Cary 5G in the diffuse reflectance mode (R). The specific surface areas of the catalysts were measured with a Micromeritics ASAP-2020 instrument and calculated by the Brunauer-Emmett-Teller (BET) method. The catalysts were pretreated (degasification) by heating at 70°C under vacuum until reaching a degassing pressure of less than 20 mmHg.

Photodegradation.
The photocatalytic activity of the catalyst was evaluated using twenty milliliters of a 10 mg L −1 solution of MB dye or AML drug, mixed with 10 mg of the catalyst, and irradiated with visible (six 15 W Osram lamps, with maximum intensity at 440 nm) and UV (six 15 W Phillips TUV UVC lamps, with maximum intensity at 254 nm) light inside a thermostated photoreactor at 25 ± 3°C. To investigate 2 International Journal of Photoenergy the influence of radiation on the degradation process of organic compounds in the photocatalysis process, the same experiment was carried out without the presence of the catalyst. For the catalyst adsorption capacity test, we used the same conditions except the presence of radiation. Specified amount of reaction mixture was collected at regular time intervals and analyzed in a UV-Vis spectrophotometer (Shimadzu UV-1601 PC) in the absorbance mode monitor.
2.4. CO 2 Photoreduction. The CO 2 photoreduction tests were performed using 300 mg of the catalyst suspended in 300 mL of water in a 500 mL steel cylindrical reactor, covered with borosilicate glass. Ultrapure CO 2 was bubbled through the reactor for at least 10 min to ensure that all dissolved oxygen was eliminated. The illumination system included a UVC lamp (PHILIPS 11 W) with a wavelength of 253.7 nm in the center of the reactor, and the intensity of the incident light was 9.1 mW/cm 2 . The reaction progress was monitored by collecting and analyzing the sample at regular intervals. Gaseous products were determined with the help of GC-TCD and GC-FID (Varian, CP-3800) using a packed column (HayeSep N (0.5 m × 1.8 ″ )) at a flow rate of 30 mL min −1 for H 2 , 300 mL min −1 for air, and 30 mL min −1 for N 2 , injector temperature of 150°C, TCD detector temperature of 200°C, and FID detector temperature of 150°C. For determining the gaseous products, a 2 μL sample was injected in the GC and then the yield was correlated by injecting a standard gaseous mixture. Blank reactions were carried out to ensure that the CH 4 originated were photoreduction products of CO 2 . In the blank reaction, no catalyst was added and all other conditions were maintained the same.

Results and Discussion
3.1. Characterization. The synthesis of Ti(OH) 4 was carried out following the same principle of peroxo complex hydrolysis by the formation of water through the oxidation of H 2 O 2 as reported by Ribeiro et al. and Camargo et al. [12][13][14]. However, in this work, instead of the oxidation process of H 2 O 2 occurring through a redox reaction, the solution of peroxo-titanium complex was heated, leading to the decomposition of H 2 O 2 and the formation of O 2 and H 2 O, causing hydrolysis of peroxo complex and the consequent formation of the precipitate [17][18][19][20].
The surface of the Ti(OH) 4 was characterized by X-ray photoelectron spectroscopy (XPS), in which the XPS spectrum in the O1s region showed two peaks (Figure 1(a)). The main peak at 529.9 eV is related to the oxygen anions, O 2− , bound to the metal cations in the lattice, and the second peak at 532.9 eV could be attributed to the formation of peroxo groups on the Ti(OH) 4 [20]. Raman spectra of Ti(OH) 4 and calcined materials are reported in Figure 1(b). The Ti(OH) 4 spectrum is dominated by a broad band between 880 and 940 cm −1 , related to the O-O stretching vibration, which is typical for peroxo groups on the surface of titanium oxide as reported by Zou et al. [21]. It was observed that calcination temperatures above 300°C cause a reduction in the amount of peroxo groups on the catalyst surface and in higher temperatures, the presence of this band no longer occurs showing the complete elimination of the groups on the material surface. Figure 2(a) shows the TG-DTG curves of Ti(OH) 4 . The TG curve for the Ti(OH) 4 showed weight loss in two stages. The first was observed between 25°C and 150°C, and the second was recorded at a temperature range from 250°C to 300°C. The weight loss in the first temperature interval could be attributed to the removal of adsorbed water in the catalyst surface. At temperatures between 250°C and 300°C, there were weight losses of 24%, which could be attributed to the peroxo groups bonded to catalyst particles.
The two weight loss regions observed in the TG are associated with one endothermic DSC peak and a few overlapped exothermic peaks (Figure 2(b)). The first weight loss of 28% was attributed to the loss of physically adsorbed water, and it could be manifested by a clear endothermic DSC peak at The XRD patterns of the Ti(OH) 4 , TiO 2 -300°C, TiO 2 -400°C, and TiO 2 -500°C are shown in Figure 3. It can be seen from Figure 3(a) that the Ti(OH) 4 and TiO 2 -300°C patterns show low crystalline with peaks at 2θ = 26.5°related to the rutile phase (JCPDS 21-1276) and 2θ = 48.7°and 62.2°related to the anatase phase; however, the materials calcined at 400 and 500°C presented peaks related only to the anatase phase (JCDS 21-1272). The crystallite size of the particles has been estimated from Debye-Scherrer's equation using the XRD line broadening of the (1 0 1) plane diffraction peak. In temperatures higher than 400°C, the TiO 2 starts to crystallize resulting in larger crystallites, which can be attributed to the thermally promoted crystallite growth. The obtained crystallite sizes are shown in Table 1.
The structural properties of the materials were further investigated by Raman spectroscopy (Figure 3(b)). The samples calcined at 400 and 500°C exhibit the characteristic Raman-active modes of the TiO 2 anatase phase, in agreement with the phase composition of the materials determined by XRD [22].
In order to provide additional evidence and to confirm the effect of calcination temperatures on the Ti(OH) 4 , FTIR characterizations were performed. The infrared spectra of Ti(OH) 4 and samples calcined at different temperatures are presented in Figure 4. The broad band at 2700-3650 cm −1 and the band at 1636 cm −1 correspond to the surfaceadsorbed water and hydroxyl groups, respectively. The main band at 400-800 cm −1 was attributed to Ti-O stretching and Ti-O-Ti bridging stretching modes [23]. Notably, increasing   International Journal of Photoenergy the temperature, the surface-adsorbed water and hydroxyl groups decreased slightly. There is no obvious characteristic band of peroxo groups on the catalyst surface, which may be limited by the detection resolution of IR. Figure 5 shows the FEGSEM images of Ti(OH) 4 and catalysts calcined at 300 to 500°C. It can be seen that the Ti(OH) 4 particles are irregular in shape and have a wide particle size distribution. Most are larger than 200 nm, and it is observed that the treatment with temperature did not change the size and morphology.
The optical properties of the Ti(OH) 4 , commercial TiO 2 (TiO 2 -COM), and catalysts calcined at 300 to 500°C were examined using diffuse reflectance UV-Vis spectroscopy, and the band gap values were calculated by the Tauc method, as shown in Figure 6 [15,16].
It is observed in the UV-Vis spectra that the increase of the calcination temperature caused a shift of the absorption edge to higher values of energy, as can be observed Figure 6. All the catalysts synthesized presented a lower band gap value than commercial titanium dioxide (Table 1).
This variation of band gap energy with increasing calcination temperature may be related to the different densities of structural and surface defects present in each of the samples. Because defects cannot be controlled in the materials, their presence only provides a change in the amount and distribution of intermediate energy levels within the band gap region.
Ti(OH) 4 presented a band gap energy value of 2.15 eV, with an onset of absorption from 576 nm, indicating that this catalyst is sensitive to visible light, which represents a possible photoactivity under visible radiation.

Photodegradation.
Photocatalytic studies were carried out on Ti(OH) 4 and materials calcined at 300, 400, and 500°C and were compared with the TiO 2 -COM nanopowder (Sigma-Aldrich 99.8%) (Figure 7). To test the photocatalytic activity of the catalysts, two organic compounds were used as the pollutant model, the methylene blue dye (maximum absorption at 554 nm), and the amiloride drug (maximum absorption 286 nm) (Figure 8).
The photolysis test of the MB was carried out under UV and visible radiation under the same conditions of the photocatalysis reaction, except without the presence of a catalyst. In this study, we can observe that under UV radiation the dye presents a small degradation by this process, which indicates that the catalyst plays a key role in the degradation of MB. The catalysts were investigated by the adsorption capacities under the same conditions as the photoreduction but without the presence of radiation, in order to distinguish between the adsorption and photocatalytic phenomena ( Figure S1). Ti(OH) 4 presented the highest adsorption capacity of approximately 19% in relation to other materials; however, it is observed that the photocatalytic effect of Ti(OH) 4 is much higher than that of other materials under visible radiation (see Supplementary Materials). It is important to note that the results are due to the combined effect of the adsorption and photocatalysis process. Figure 7(a) shows that Ti(OH) 4 and TiO 2 -COM were the two catalysts that showed the highest photocatalytic activity, with 100% discoloration of the solution in 60 min of reaction under UV radiation. The degradation rate becomes less pronounced when the calcination temperature used for the production of materials is increased from 300 to 500°C. On the other hand, when the reaction was conducted under visible radiation, the Ti(OH) 4 was much more active than the other catalysts, which was expected since it has a higher absorption in the visible region compared to the other catalysts (Figure 7(b)).
One of the problems of testing photocatalytic activity of a catalyst using dye is that the sensitization process can occur [24]. This mechanism occurs when a molecule that has absorption in the visible region, such as dyes, is adsorbed on the surface of a semiconductor and is excited from its ground state (HOMO) to the excited state (LUMO), transferring these electrons to the semiconductor spontaneously, being the first stage of its oxidation. Furthermore, it can be found that amiloride is very stable under visible light radiation (Figure 8(b)), revealing that degradation does not occur without a photocatalyst. It is observed that Ti(OH) 4 had the best photocatalytic activity, presenting 90% removal with 20 min of reaction under  These results show that the absorption edge shift of Ti(OH) 4 , due to the presence of the peroxo groups on the surface, had a great influence on the catalytic activity of the catalyst under visible radiation, where it was observed that the removal of these peroxo groups in the surface by the calcination process decreased the photocatalytic activity of these materials under visible radiation, probably due to their low absorption capacity in the visible region.
The order of reaction with respect to MB and AML degradation was determined by plotting the reaction time as a function of ln C / C 0 according to the following equation for the materials: where C 0 and C represent the concentration of the substrate in solution at time zero and the time of illumination,  International Journal of Photoenergy respectively, and k represents the apparent rate constant (min −1 ). The apparent rate constants are summarized in Tables 2 and 3. The results show that the reaction followed first-order kinetics.
The stability of the catalyst over a photodegradation is a critical factor for its practical application. Therefore, the photostability of the Ti(OH) 4 sample, which showed the best photoactivity, was evaluated by performing recycling experiments under UV irradiation ( Figure S2). After each reaction cycle, the sample was separated from the MB solution by centrifugation and placed immediately in contact with a freshly prepared MB solution. This procedure was repeated four times under the same conditions used in the photocatalytic tests. As shown in Figure S2, a slight decrease (ca. 4%) occurs after the first photocatalysis cycle. In addition, the peroxo groups were analyzed after the 7 International Journal of Photoenergy photocatalysis reaction, in which it was observed that after the reaction the groups were maintained ( Figure S3).
The formation of the stable metal-peroxo group, which may be partially in equilibrium with radical metal-O 2 ⋅ species, has been recognized to play an important role in the photocatalytic process, as these groups can inhibit the recombination of photogeneration charges and interact with photogenerated holes [25,26]. Photogenerated valence band holes (h + ) are trapped at the surface oxygen to form the ⋅ OH radical. On the other hand, the electrons in the conduction band can react with H 2 O to form the ⋅ OOH radical (Figure 9) [27].
Zhang and Nosaka studied the formation of the ⋅ OH radical from TiO 2 and demonstrated that the mechanism of ⋅ OH generation occurs differently depending on the crystalline phase of TiO 2 ( Figure 9). They observed that on the surface of the rutile, ⋅ OH is generated by the oxidation of the water through the h + photogenerated with the peroxo group on the surface; on the other hand, in the anatase phase, the OH is generated by the reaction of the metal-O 2 ⋅ radical [28,29]. This difference in the mechanism of ⋅ OH generation occurs due to the packaging of the crystalline structure, in which the rutile phase presents a greater packaging with respect to the anatase phase, better stabilizing the peroxo group in the surface. Since the peroxo group is equivalent to the adsorbed H 2 O 2 , the presence of the rutile phase in Ti(OH) 4 may be favoring an increase in the generation of the ⋅ OH radical that consequently increases its photocatalytic activity.

CO 2 Photoreduction.
Another aspect of the photocatalysis process is the CO 2 photoreduction also known as artificial photosynthesis. Unlike the photodegradation process, which generates oxidizing radicals ( ⋅ OH e ⋅ OOH), the photoreproduction process happens with the electrons photoexcited to the CB that are captured by the molecules of CO 2 adsorbed on the surface of the catalyst forming carbon dioxide (CO 2 ⋅ ) radicals, which will later lead to the formation of compounds with a higher added value such as CH 4 , ethanol, and formic acid [30].
Thus, to verify the Ti(OH) 4 activity in CO 2 photoreduction, tests were performed under UV radiation in the aqueous medium ( Figure 10).
A series of control experiments were conducted for quality assurance and to ascertain correct results. The first tests were performed in the dark with and without a catalyst, under the same experimental conditions as the CO 2 photoreduction. No products were detected in the two control experiments, indicating that CO 2 conversion, as described below, is a true photocatalytic reduction process.
Analysis of the gas samples indicated that only CH 4 was formed using Ti(OH) 4 and the yield was 6.81 μmol·L −1 ·g −1 , which was almost 3 times higher than the result obtained using P25 (2.18 μmol·L −1 ·g −1 ), which is the conventional catalyst used in photocatalytic processes. The concentration of CO 2 present in the gaseous medium varied during the reaction (Figure 10(b)); this variation is related to the equilibrium displacement over the photoreduction process. During the photoreduction process, the CO 2 is consumed from the aqueous medium, and in order to maintain the equilibrium, the part of the gas moves to the liquid medium causing that variation in the concentration of CO 2 in the gas phase during the reaction.
It is important to note that the CO 2 concentration variation in the gas phase was approximately 140-130 mmol·L −1 ·g −1 for all materials, indicating that, despite the small variation observed, CO 2 concentration may be considered constant during the reaction time.

Conclusions
The OPM route proved to be efficient in obtaining Ti(OH) 4 nanoparticles with the surface modified by peroxo groups. These groups had a great influence on the processes of photocatalysis, both in the oxidation process of organic compounds and in the photoreduction of CO 2 , not only increasing the absorption of the radiation in the visible region but also providing an effective separation of photogenerated charges. We investigated the effect of the calcination temperature on the photocatalytic activity of Ti(OH) 4 , and the results showed that the best result was achieved with uncalcined Ti(OH) 4 , exhibiting 98% degradation of AML and MB under irradiation of visible light, being more than 10 times more active than TiO 2 -COM. In addition, Ti(OH) 4 with the modified surface proved to be a good candidate for applications in CO 2 photoreduction processes.
It is important to note that the OPM method can be used to obtain other oxides such as niobium oxide, vanadium oxide, and tungsten oxide that are good candidates for applications in photocatalytic processes.

Data Availability
No data were used to support this study.

Conflicts of Interest
The authors declare no conflict of interest.   Figure 9: Schematic illustration of OH generation at rutile and anatase TiO 2 . 9 International Journal of Photoenergy