Enhanced Photocatalytic Activity of Pure Anatase Tio 2 and Pt-Tio 2 Nanoparticles Synthesized by Green Microwave Assisted Route

*e-mail: giuseppe.ciccarella@unisalento.it


Introduction
Titanium dioxide (TiO 2 ) nanoparticles have been intensively studied as catalyst for the photodegradation of organic compounds due to their unique physical and chemical properties [1][2][3] .Their photocatalytic performance is greatly determined by the particle's crystal phase, crystallinity, phase stability, size and surface chemical properties that largely depend on their synthetic route 4 .TiO 2 has three nature crystallographic phases: anatase, rutile, and brookite.The physicochemical properties of the three phases are very different from each other, and they are closely related to the synthesis conditions.Anatase is the most thermodynamically stable among the three nano crystalline phases if the size of the particles is less than 11nm, brookite is the most stable phase between 11 nm and 35 nm and rutile is the most stable when all the sizes are larger than 35 nm 5 .Moreover, anatase is generally considered to be more active than rutile phase as a photocatalyst 6 .Anatase with higher crystallinity is preferred for photocatalysis, due to that the higher crystallinity offers fewer defects acting as recombination sites between photo-generated electrons and holes 7 .
Many efforts have dedicated to the modification of TiO 2 nanoparticles in their morphology and phase structures through various methods, such as the hydrothermal synthesis method, sol-gel, anodization and template method 8 .Among these means, the hydrothermal synthesis method has received particular attention owing to its technical convenience in obtaining well-structured TiO 2 nanostructures.Hydrothermal synthesis, in which chemical reactions can occur in aqueous or organic media under the self-produced pressure at low temperature (usually lower than 250°C), can solve those problems encountered during sol-gel process (particles size growth, reduction of specific surface area of particles, phase transition) 9 .
However, the conventional hydrothermal method usually needs high pressure and temperature, long duration and complex procedures for preparation of TiO 2 nanocrystals 10 .Moreover, the general synthesis of TiO 2 nanoparticles usually need high temperature calcinations or complex chemical processes to convert amorphous TiO 2 into crystallines, which generally result in grain growth of the particles and hence the loss of surface area, finally deteriorate the photocatalytic activity 11 .Thus, it would be greatly desirable to get well high yield, crystalline anatase TiO 2 nanoparticles in one-step synthesis procedure.The rapidly growing number of inorganic particles that were prepared by nonaqueous and/or nonhydrolytic processes clearly indicated that synthesis routes in organic solvents under exclusion of water represent a versatile alternative to aqueous methods 12 .
More recently, microwave irradiation has been reported to effectively enhance the efficiency of hydrothermal method for the preparation of inorganic materials 13,14 .
The microwave-assisted method offers a clean, cost effective, energy efficient, eco-friendly, rapid and convenient method of heating, which results in higher yields in shorter reaction times 15 .No further heat treatment is required for sample production.Microwaves are also used in synthesis to improve the nanometer size fraction of particles 16 and to enhance structural and morphological properties for the nanomaterials 17,18 .In addition, microwaves appear today as a new technology for the development of the green chemistry, with synthesis methods involving preparation procedures based on solvent-free or reduced amount of reactants 14 .
Organic surfactants commonly used during synthesis process are found on the surface of the nanoparticles in trace amounts in spite of the extensive cleaning procedures to which they are subjected.Such compounds passivate the particle surface, so that nanoparticles lose their active sites and therefore affect the photocatalytic process 19 .
In order to remove these substances, plasma cleaning 20 , calcinations 21 , super-hydride reaction 17 , heat 22 and high-temperature hydrogen treatments 23 have been reported in literature.These techniques, however, either require expensive special equipment or can compromise the morphological integrity of the nanoparticles due to harsh treatment conditions.UV-ozone oxidation (UVO) is a useful alternative capable of yielding nearly atomically clean surfaces 24 .The method typically uses ultraviolet light that includes wavelengths of 185 and 257 nm, where the former generates ozone upon interacting with molecular oxygen and the latter excites the organic contaminating molecules 24 .The UV-ozone oxidation process involves the simultaneous action at room temperature of ozone and ultraviolet light, which are responsible for the oxidation of the carbon-containing compounds into carbon dioxide and water.
In this study, we developed of a rapid, non-aqueous and industrially viable relatively-low temperature synthesis method for pure anatase TiO 2 nanoparticles via microwave technology using Ti (IV) isopropoxide as precursor, benzyl alcohol as solvent and acetic acid as additive reagent.Furthermore, we proposed for the first time to the best of our knowledge, the use of ozone gas alone for the successful removal of the organic capping layer (acetic acid) from titania nanoparticles surface.The physicochemical characteristics of nanocrystalline ozone-treated TiO 2 powder in terms of morphology, crystallization, surface areas were carefully characterized.
Platinum nanoparticles were then photo-deposited onto TiO 2 nanoparticles and the obtained sample was labeled as Pt-TiO 2 .
The photocatalytic activity of TiO 2 and Pt-TiO 2 samples under visible light was evaluated by degradation Rhodamine B aqueous solution as a model pollutant.A considerably enhanced photocatalytic activity was found for TiO 2 and Pt-TiO 2 nanoparticles in comparison with commercial Degussa P25 and Pt-P25, respectively, used as sample references.

Material and methods
Titanium (IV) isopropoxide TTIP, benzyl alcohol, acetic acid, chloroplatinic acid solution 8 wt.% in H 2 O, Rhodamine B (purity 99%), TiO 2 Degussa P-25 (containing anatase and rutile phases in a ratio of about 3 : 1) were purchased from Sigma Aldrich (USA).All chemicals and solvents were of analytical grade and used without further purification.Rhodamine B consists of reddish-violet powder, its molecular formula is C 24 H 31 ClN 2 O 3 (molecular weight: 479.02).Double distilled water was filtered through a Millipore membrane filter before use and used throughout the experiments.
The samples were characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), selected area diffraction pattern (SAD), BET surface area analyzer and FT-IR spectroscopy.Pt content was determined by inductively coupled plasma-atomic emission spectroscopy (ICP-AES).
X-ray diffraction measurements were carried out in the reflection mode on a Rigaku model diffractometer with Cu Kα radiation (λ = 0.154 nm).The X-ray diffraction data were collected at a scanning rate of 0.02°/s in 2θ ranging from 10° to 80°.TEM images were taken using a TEM operated at 100 kV (Jeol Jem 1011).Specimens for TEM analysis were prepared by drop-casting a single drop samples in ethanol solution onto standard carbon supported 300-mesh copper grids and drying slowly in air naturally.The nitrogen adsorption-desorption isotherms were obtained at 77 K using Brunauer-Emmett-Teller (BET) technique using a Quantachrome ASiQwin system.The FT-IR spectra of the TiO 2 powders were collected using a Jasco FT/IR-6300 spectrometer with a resolution of 0.07 cm −1 .UV-visible absorption spectroscopy was performed using a Varian Cary 300 Scan spectrophotometer with a 10-mm path length quartz cuvette.

Microwave-assisted synthesis of acetic acid capped TiO 2 nanoparticles
In a typical synthesis procedure, 1ml of TTIP (3 mmol) was slowly added into 10 ml of benzyl alcohol in a vessel.Then, acetic acid (60 mmol) was added dropwise under vigorous magnetic stirring.The vessel was sealed and exposed to high intense microwave irradiation using a microwave digestion system (model Mars, CEM, Matthews, NC).The system uses a 2.45 GHz microwave frequency and it is controlled by temperature.The reaction mixture was maintained at 200°C for 45 min under medium stirring.The resulting product was collected and washed several times with methanol and finally dried under vacuum at 40°C.

Removal of the organic capping layer from TiO 2 nanoparticles.
As a routine procedure, freshly lab-prepared colloidal nanoparticles were repeatedly washed by a large amount of solvent (methanol) and collected by centrifugation.Although excess capping agents could be stripped off from nanoparticles by extensive washing, there are always residuals resistant to removal.Generally, the presence of residual capping agents brings extra complexity to the system, such as uncertain coverage density of capping molecules and non-covalent Enhanced Photocatalytic Activity of Pure Anatase TiO 2 and Pt-TiO 2 Nanoparticles Synthesized by Green Microwave Assisted Route interaction between capping molecules and reactants, as well as charge transfer at the organic-metal interface 25 .To simplify the uncontrollable variables and to enhance their catalytic activity, nanoparticles with "clean surface" are therefore favored.We have developed a simple strategy to deal with this problem.Dried TiO 2 samples (200 mg) were placed in a spherical glass container with cylindrical necks and they were treated, under stirring, with an ozone (O 3 ) flow for 1 h.The ozone oxidation process involved the action of ozone alone which is responsible for the oxidation of the carbon containing compounds into carbon dioxide and water.At the same time, the oxidation products were swept off by O 3 gas flow.The successful removal of organic capping layer was proved through FT-IR spectroscopy.

Preparation of Pt loaded TiO 2 nanoparticles.
The loading of Pt nanoparticles was performed on commercial P25 and ozone-treated TiO 2 samples.The obtained samples were labeled as Pt-P25 and Pt-TiO 2 .
In a typical procedure, in a 100 ml beaker, titania nanoparticles (2.0 g) were dispersed in 50 ml of methanol.Then, H 2 PtCl 6 (0.50 ml, 8 wt% aqueous solution) was added.The reaction mixture was irradiated by a 150 W Xe lamp (without an optical filter) for 30 min.The milky white suspensions turned light grayish with the deposition of Pt.The resultant Pt-titania composites were retrieved by centrifugation with 6000 rpm rotation rate, washed five times with excess methanol and dried under vacuum.
Pt contents in the catalyst were determined by ICP-AES.The catalysts were digested in aqua regia and standard solutions of known concentrations were adopted as calibration standards.ICP measurements revealed that TiO 2 nanoparticles were loaded with 1.0 wt% Pt metal.

Photodegradation of Rhodamine B
The photocatalytic degradation of Rhodamine B was investigated in the presence of catalysts: TiO 2 , Pt-TiO 2 , P25 and Pt-P25.The photodegradation of Rhodamine B (RhB) for the different samples was investigated using a 150 W Xe lamp (LOT Oriel Gimbh) as source for artificial sunlight.
The catalyst sample (50 mg) was sonicated in 10 ml distilled water (15 min) to ensure complete dispersion of nanoparticles in solution.The catalyst was mixed with a 10 -5 M solution of RB in distilled water (50 ml) for 30 min in the dark to reach an adsorption/desorption equilibrium.The mixture was continuously stirred in air and irradiated under Xe lamp light with a 420 nm cutoff filter to remove UV radiation.The distance between the lamp and the bottom of the solution was about 10 cm.The progress of the reaction was monitored by recording the absorption spectra of irradiated samples after every 5 min.The initial pH values of the catalyst-RhB solutions were in the range 5.2-5.8.

Sample structural and morphological characterization
X-ray diffraction technique was performed to analyze and identify phase formation and crystallographic information of the samples.Figure 1 shows the X-ray diffraction patterns of TiO 2 and P25 samples.The diffraction patterns and peak positions of TiO 2 nanoparticles (Figure 1a) match very well with those of tetragonal TiO 2 anatase with cell constants a=b=0.37710nm, c= 0.9430 nm and α=β=γ=90°, in agreement with the standard diffraction data (JCDS 21-1272).No lines corresponding to any secondary and/or impurity phases were observed, indicating the high purity of the samples.Broad XRD peaks indicate formation of nanosized particles.XRD spectrum of the P25 (Figure 1b) confirmed the presence of anatase along with rutile phase, in agreement with the standard diffraction data (JCDS 21-1272 and 21-1276) as certified by Sigma Aldrich producer.
Transmission electron microscopy (TEM) and selected area electron diffraction pattern (SAED) have been employed to characterize the size, shape and structure of the samples.In Figure 2 a typical micrograph of the synthesized TiO 2 sample, with the corresponding selected area diffraction pattern is presented.
To determine the size and aspect ratio of nanomaterials, digitized TEM images obtained from randomly chosen fields, were processed with image software and the obtained data were analyzed.In such a way, more than 100 particles were analyzed.The lengths L A and widths L B were measured as the maximal length and width of the particles, respectively.The average particle length and width were 13.8 ± 5.5 nm and 9.0 ±1.2 nm, respectively.The aspect ratio R is defined as R=L A /L B .The aspect ratio distribution reported in the inset of Figure 2a clearly evidenced that the nanoparticles have an aspect ratio in the range 1-3.The SAED pattern (Figure 2b) recorded on the area shown in Figure 2a exhibited diffraction spots/rings with interplanar spacing that could be indexed according to tetragonal TiO 2 anatase, as reported in Table 1.A perfect agreement in the d values calculated from experimental diffraction ring patterns and literature data is evident.This result confirmed the phase purity of the obtained products is high.
TEM observation of P25 sample evidenced that particles exhibited a relatively uniform particle size distribution, with an average size of about 24.1 ± 9.5 nm (Supplementary Information, Figure S1).

Removal of organic capping layer
TiO 2 surface characterization was based on the FT-IR spectra (Figure 3).The as-synthesized nanoparticles (Figure 3a) have two typical absorption bands at 1451 cm −1 and 1531 cm −1 related to the symmetric and asymmetric stretching vibrations of the carboxylate groups.The two sharp absorption bands at 2852 cm −1 and 2938 cm −1 are attributed to the symmetric and asymmetric CH 2 stretching vibrations of molecules with long alkyl chains, respectively.The carbonyl band originating from the free ester at 1742 cm −1 clearly shows that the non-hydrolytic sol-gel reaction of acetic acid and TTIP forms the ester 26 .Moreover, the small peaks between 3000 cm −1 and 3100 cm −1 can arise from the C-H stretching of the aromatic ring of benzyl alcohol, 1235 cm −1 being typical of the C-C vibration.The treatment of the samples under ozone flow removed the organic capping layer on the surface of the nanocrystals and this phenomenon could be observed by verifying in the FT-IR spectra (Figure 3b) the disappearance of the characteristic absorption peaks at 1531 cm −1 and 1451 cm −1 of the chelating bidentate carboxylate groups binding to the surface of the nanocrystals 17 .

Porosity and surface area characterization
The structural and textural properties of the ozone-treated TiO 2 samples were investigated by N 2 adsorption-desorption isotherms at 77 K. Figure 4 shows the obtained results    analyzed by BET method for surface area and BJH pore size distribution as inset.The relative isotherm of the sample was of type IV, which is characteristic of mesoporous materials.
The Brunauer-Emmett-Teller (BET) specific surface area analysis result was 89.46 m 2 g −1 , denoting a significant increase of the surface area compared to P25 (56 m 2 g −1 ).The BJH pore size distributions (inset of Figure 4) evidenced an average pore diameter of about 2.9 nm.The obtained high and uniform porosity could ensure enhanced reaction rates due to the high level of interaction of the reactants with the active sites.

Pt loading onto TiO 2 nanoparticles
XRD spectrum acquired after Pt photodeposition didn't show any difference with respect to un-loaded sample.When the impurity level onto the surface of any sample, Pt in this case, remains less than 5%, it is less likely to be detected by XRD but the presence of Pt onto the surface of TiO 2 was confirmed by TEM, EDX and ICP analysis (Supplementary Information, Figure S2).
TEM observations of platinum loaded sample (Figure 5) evidenced that the estimated range of Pt nanoparticles deposited onto the surface of TiO 2 nanoparticles was found to be 1-2 nm.Small black dots in TEM images are platinum particles.
It is well known that there is no chemical bonding between Pt and TiO 2 surface, but Pt remains adhered onto the oxide surface by a well known interaction called strong metal (Pt)-support (TiO 2 ) interaction (SMSI) 27 .The photodeposition of platinum nanoparticles was carried out in acidic environment in order to avoid possible agglomeration of platinum.In fact, in acidic environment the titania particles surface is positively charged, since zero point of charge (pH zpc ) is at a pH of 6.25.The positively charged TiO 2 surface will facilitate adsorption of PtCl 2− 6 and its hydrolyzed ions (the concentration of anions will be higher on the surface), thereby increasing the entrapment of excited conduction band electrons resulting in a high deposition rate of platinum nanoparticles on the support 28 .

Formation of anatase nanoparticles.
TiO 2 nanoparticles synthesis method presented in this work is based on a microwave-assisted non-aqueous sol-gel process, where titanium tetraisopropoxide is used as titanium oxide precursor, benzyl alcohol as solvent and acetic acid as additive reagent.Benzyl alcohol is a mildly coordinating solvent, with a high boiling point, benign in nature and several reports have documented the non-aqueous synthesis of monodisperse and well separated TiO 2 nanoparticles, where benzyl alcohol has been employed as a solvent.Niederberger et al. [29] found that the reaction between benzyl alcohol and TiCl 4 could provide a versatile method for non-aqueous preparation of TiO 2 nanocrystals: 5 nm sized anatase nanoparticles were obtained by means of a synthesis performed at 40 °C for 7-14 days 29 .Melcarne et al. 30 and Zimmermann et al. 31 proposed a modification of this non-aqueous sol-gel process by the use of TTIP as titanium oxide precursor.However, the proposed synthetic approach required still long reaction time (12 h and 48 h, respectively).In our microwave-assisted synthetic approach, microwave irradiation with respect to conventional heating greatly accelerates nanoparticle formation by facilitating the dissolution of the precursor in the solvent and increasing the rate constants for the esterification reaction, resulting in faster production of monomer and consequently in an earlier nucleation event 32 .Moreover, acetic acid further catalyses crystallization, and due to the fast nucleation and slow growth and coarsening kinetics, the number of particles is large 33 , thus limiting the particle radius to about 10-15 nm.
It is well known that anatase is formed due to the lower surface energy as compared to the other phases.The lower surface energy affects phase formation in terms of thermodynamics and kinetics.Thermodynamically, anatase is the structure with the lowest total energy if the nanoparticles are sufficiently small 34 .
It has been remarked that this method does not require water as solvent because the esterification reaction between the benzyl alcohol and the unreacted acetic acid gradually provides water into the reaction environment 30 .It constitutes, then, a simple and fast procedure for the preparation of uniform size TiO 2 anatase nanoparticles.It is particularly fit to be scaled-up on large volumes in the perspective of a potential industrialization.
Photocatalytic degradation of Rhodamine B (RhB) was evaluated using TiO 2 , P25, Pt-TiO 2 and Pt-P25 samples in suspensions.All experiments were conducted at the natural pH of the dye.
Blank experiments were carried out on a RhB solution containing no catalyst and they proved the occurrence of negligible degradation of the dye upon direct visible photolysis.This may be due to the natural self-fading of the dye.In addition, RhB-titania samples kept in the dark didn't show any appreciable absorbance changes over time.
The variation of concentration of RhB in TiO 2 and Pt-TiO 2 catalyzed systems as function of time was evaluated and compared with P25 and Pt-P25.It was evident that UV-Vis characteristic absorption peak at 554 nm of the RhB solution gradually decreases during illumination time for all samples (Supplementary Information, Figure S3).The possible photodegradation of RhB includes four main processes: N-deethylation, chromophore cleavage, ring opening and mineralization.
The curves of changes in relative concentration of RhB (calculated from the absorption at 554 nm) versus irradiation time for each sample are demonstrated in Figure 6a.
It is clear that our TiO 2 nanoparticles sample shows significant improvement in the photodegradation of RhB compared to P25 and also to Pt-P25.The degradation is fastest and the concentration follows an exponential decay.It can be seen that RhB solution was almost degraded completely within 20 min using Pt-TiO 2 , within 40 min using TiO 2 samples.Comparing the synthesized samples with the reference ones (P25and Pt-P25), the degradation process slows down significantly at longer reaction times.
According to numerous literature works, the influence of the initial concentration of the solute on the photocatalytic degradation rate of the most organic compounds is described by pseudo-first order kinetics 35 − = dC kC dt (1)   where k (min -1 ) is the apparent rate constant and is affected by dyes stuff concentration.Integration of Equation 1 will lead to the expected relation The plot of Equation 2 versus t for the different studied catalyst is shown in Figure 6b.The figure shows that the photocatalytic degradation follows the pseudo-first order.The apparent rate constants for all studied samples is determined from the slopes of ln(C 0 /C) versus irradiation time and the obtained values with the corresponding standard error (SE) are summarized in Table 2.
Remarkably, the reaction rate constant of TiO 2 and Pt-TiO 2 samples are about 8 and 6 times, respectively as high as those of commercial P25 and Pt-P25.
The enhanced photocatalytic activity of prepared TiO 2 sample with respect to P25 can be attributed the larger BET surface area, which can offer more active adsorption sites and photocatalytic reaction centers and to the presence of only pure anatase phase, which is known to be more active than rutile phase for TiO 2 photocatalyst.
The enhancement in decomposition rate of organic pollutants in the presence of metalized TiO 2 has been discussed previously by several authors.The enhancement in decomposition rate of organic pollutant in the presence of metalized metal oxides may be mainly attributed to the formation of a Schottky barrier between the metal and the semiconductor 36 .So, when the platinum particles are deposited onto the surface of TiO 2 , a Schottky barrier formed at the interface of TiO 2 and Pt metal particles, resulting in an efficient channeling of excited conduction band electrons from the bulk of TiO 2 to the newly formed interface.As a result, the electron density in TiO 2 particles decreases which in turn prevents the electron-hole pair recombination.As a result, higher photocatalytic activity was observed.

Conclusion
A facile and rapid synthesis of phase-pure anatase TiO 2 nanoparticles through a nonaqueous microwave assisted process has been reported, using Ti(IV) isopropoxide as precursor, benzyl alcohol as solvent and acetic acid as additive reagent.Slightly elongated nanoparticles with a length of 13.8 ± 5.5 nm and a diameter of 9.0 ±1.2 nm were obtained after a reaction period of 45 min.
Since capping agents bring extra complexities for the catalytic study of colloidal nanoparticles, in order to simplify the uncontrollable variables and to enhance nanoparticles catalytic activity, the residual organics onto nanoparticles surfaces were completely removed through a new low-cost treatment under ozone flow, at room temperature in air.The successful removal of organic capping layer has been proved through FT-IR spectroscopy.Experimental results evidenced that synthesized anatase titania nanoparticles exhibited higher photocatalytic activity compared to P25 and Pt-P25.Moreover, the photocatalytic activity of the synthesized nanoparticles was further increased by Pt loading.The improved photocatalytic performance could be attributed to the anatase phase purity, small size, large specific surface area and clean surfaces of the prepared nanoparticles.In fact, anatase is generally considered to be more active than rutile phase for TiO 2 photocatalyst.The larger surface area of TiO 2 nanoparticles with cleaned surfaces is beneficial for the RhB adsorption, further leading to more photoexcited electrons transferred from RhB to TiO 2 and accompanied with higher dye degradation efficiency.

Figure 1 .
Figure 1.Typical XRD patterns of a) synthesized TiO 2 and b) P25 nanoparticles.A and R stand for anatase and rutile phase, respectively.

Figure 3 .
Figure 3. FT-IR spectra of TiO 2 nanoparticles a) before and b) after treatment under ozone flow for 1 h.

Figure 2 .
Figure 2. (a) Typical TEM image of the synthesized TiO 2 nanoparticles, with its corresponding (b) selected area diffraction pattern.Inset: aspect ratio distribution of the nanoparticles.

Figure 6 .
Figure 6.a) Photocatatalytic degradation of RhB solution by synthesized TiO 2 and Pt-TiO 2 and reference P25 and Pt-P25 samples under visible light irradiation; b) first order kinetic plot for the photodegradation of RhB

Figure S3 .
Figure S3.Change of the characteristic absorbance intensity of the RhB solution (at 554 nm) as a function of irradiation time for a) P25, b) Pt-P25, c) TiO 2 and d) Pt-TiO 2 samples.

Table 1 .
Interplanar spacings deduced from electron diffraction pattern reported in Figure2btogether with the corresponding ones obtained from the literature data (JCPDS 21-1272).

Table 2 .
Apparent rate constant k values for the different samples.