Role of Platinum Deposited on TiO₂ in Photocatalytic Methanol Oxidation and Dehydrogenation Reactions

1. Introduction

Titanium dioxide is regarded to be one of the most common photocatalysts, having a wide range of properties, such as a strong resistance to chemical and photocorrosion, strong oxidation capability, low operational temperature, low-cost, being and nontoxic [1]. These properties make TiO₂ an attractive candidate for its utilization as a photocatalyst in the photocatalytic processes. TiO₂ has been extensively studied and demonstrated to be suitable for numerous applications such as, destruction of microorganisms [2–5], inactivation of cancer cells [6, 7], protection of the skin from the sun [8–11], photocatalytic water splitting to produce hydrogen gas [12–14], manufacture of some drug types [15–17], degradation of toxic organic pollutants in water [18–20], and self-cleaning of glass and ceramic surfaces [21]. Even though TiO₂ is the most used semiconductor material, it exhibits some disadvantages, such as low surface area and fast recombination rate between the photogenerated charge carriers and the maximum absorption in the ultraviolet light region.

Different attempts have been performed to improve the efficiency of TiO₂ depressing the recombination process of the photoelectron-hole pairs. Some of them include the modification of TiO₂ surface with other semiconductors to alter the charge-transfer properties between TiO₂ and the surrounding environment [22, 23], sensitizing TiO₂ with colored inorganic or organic compounds improving its optical absorption in the visible light region [24–28], bulk modification by cation and anion doping [29–38], and fabrication of TiO₂ surface from polyhedral to produce hallow TiO₂ [39, 40]. TiO₂ nanoparticles are considered to be more active photocatalysts as compared with the bulk powder. The ratio of surface area to volume of nanoparticles has a significant effect on nanoparticles properties. This leads to a higher chemical activity and loss of magnetism and dispersibility [41].

This work was focused on the characterization of the prepared Pt-loaded TiO₂ (Hombikat UV 100) samples. Moreover, the photocatalytic oxidation and photocatalytic dehydrogenation of methanol have been studied employing both the bare and Pt-loaded TiO₂ in the O₂ and N₂ atmosphere. The methanal formation was determined using Nash method at a wavelength of 412 nm.

2. Materials and Methods

A known weight (2 g) of TiO₂ (Hombikat UV 100, Sachtleben, Germany) was suspended under continuous stirring at 250 rpm in a solution containing 40 cm³ of 40% aqueous methanal (Chemanol), 10 cm³ of methanol (Hayman), and the appropriate volume of hexachloroplatinic acid (Riedel-De-Haen AG) dissolved in HCl. The reaction mixture was maintained at 303 K, purged with nitrogen gas (20 cm³/min) and irradiated by UV-A light employing Philips Hg lamp (90 W) with the light intensity of 3.49 mW/cm² (Efbe-Schon 6 lamps) for 4 h. This period of irradiation time was found to be the most sufficient time for the complete photodeposition process of metallic platinum. The concentration of platinum was monitoring by the atomic absorption spectroscopy (Shimadzu-AA-6300, Japan). The milky white suspension turns to the pale grey colour with the deposition of Pt. The suspension solution was filtered and washed by absolute methanol, throwing in a desecrater overnight. At the end the product was dried in an oven at 100°C for 2 h [31, 32]. Band gap energies of bare and Pt (0.5)-loaded on TiO₂ surface were determined, via the measurement of reflectance data R by (Cary 100 Scan) UV-visible spectrophotometer system. It is equipped, with using a Labsphere integrating sphere diffuse reflectance accessory for diffuse reflectance spectra over a range of 300–500 nm by employing BaSO₄ as reference material.

In all photocatalytic experiments, 100 cm³ of 40 mM aqueous methanol solution (HPLC grade, Sd fine-CHEM limited) was mixed with certain weight of bare TiO₂ or platinized TiO₂ and was suspended using a magnetic stirrer at 500 rpm. At different time of intervals 2.5 cm³ of reaction mixture was collected in a plastic test tube and centrifuged (4000 rpm, 15 minutes) in an 800 B centrifuge. The supernatant solution was carefully removed by a syringe to a new plastic test tube and centrifuged again to remove the fine particles of bare TiO₂ or platinized TiO₂. The concentration of formed methanal was determined spectrophotometrically at 412 nm following Nash method [42, 43] using UV-visible spectrophotometer (T80+, PG Instruments Limited, England).

3. Results and Discussion

3.1. Characterisation of Bare and Platinized TiO₂

3.1.1. FTIR Analysis

The Fourier transform infrared spectra of bare and platinized TiO₂ are depicted in Figure 1. The illustrated peaks at 3350–3450 cm⁻¹ correspond to the stretching vibration mode of O–H bonds of free water molecules and at 1620–1630 cm⁻¹ correspond to the bending vibration mode of O–H bond of chemisorbed water molecules. The absorption intensity of surface O–H groups in TiO₂ is regularly increased with the increasing of the percentage of metals content. These findings are in a good agreement with the literature data [44–46]. The broad intense band below 1200 cm⁻¹ is due to Ti–O–Ti bridging stretching mode in the crystal. This peak appeared as unsymmetrical valley with the increasing of metal loading (or content) on TiO₂ exhibiting a maximum at 580 cm⁻¹. This change is related to the formation of Ti–O–M vibrations [47, 48]. The intense bands at 3621, 3645, and 3696 cm⁻¹ in all spectra are attributed to the characteristic tetrahedral coordinated vacancies of ₄Ti⁴⁺–OH besides two bands at 3765 and 3840 cm⁻¹. These revealed that the octahedral vacancies designated as ₆Ti³⁺–OH are found. In the presence of metal loaded on TiO₂ the peaks of ₆Ti³⁺–OH are not observed. This is because the metal acts as an electron trapper, mainly preventing the formation of Ti³⁺–OH species [49]:

$$ ()

$$ ()

3.1.2. XRD Analysis

The XRD patterns of different TiO₂ samples (bare and platinum loaded) are shown in Figure 2. The mean crystallite size (L) of samples was calculated by Scherrer′s equation (3) and the crystallite size (Ĺ) of samples can be estimated from plotting the modified Scherrer’s formula (4) [50] as shown in Figure 3. The corresponding values are listed in Table 1:

$$ ()

$$ ()

In (3) and (4), k is Scherer’s constant depending on shape of particles (0.94), λ is the wavelength of the X-ray radiation (0.15418 nm for CuK_α), β is the full width of half maximum (FWHM) intensity (in degree which converted to radian), and θ is the diffraction (Bragg) angle [50, 51].

Table 1. Mean crystallite sizes and crystallite sizes of bare TiO₂ and Pt-loaded on TiO₂.

Crystal components	Pt %	Mean crystallite sizes (L)/nm	Crystallite sizes (Ĺ)/nm
TiO2 Hombikat (UV 100)	0.000	11.487	10.132
Pt-TiO2	0.250	10.799	10.021
Pt-TiO2	0.500	9.355	9.503
Pt-TiO2	0.750	10.221	9.589
Pt-TiO2	1.000	10.475	8.262

No peak was observed for Pt (0.25 wt%)/TiO₂ sample at θ = 46.5^°. This result is in good agreement with the previous findings [52]. However, θ = 46.6^° which is related to Pt appeared very weak band with Pt (0.5%) loading and increased for Pt (1.0%) as shown in Figure 2. The mean crystallite size of both bare and platinized TiO₂ decreased from 11.487 nm to 9.355 nm, respectively. The crystallite size of bare TiO₂ was found to be equal to 10.132 nm. This value was decreased with the increasing of Pt content on TiO₂. The decreasing of the mean particle size of platinized TiO₂ is attributed to the location and incorporation of Pt(IV) with Ti(III) in TiO₂ lattice. Moreover, the ionic radius of Pt(IV) (0.63 Å) is relatively smaller than that of Ti(III) (0.67 Å) [53, 54].

3.1.3. AFM Analysis

Figure 4 shows the three-dimensional AFM images of bare and Pt-loaded TiO₂ surface which were used to measure the particle sizes. AFM images indicate that the shapes of both bare and platinized TiO₂ are spherical. The results summarized in Tables 1 and 2 indicate that the particle sizes for all samples are found to be bigger than the values found for crystallite size. This indicates that each particle consists of several crystals (polycrystals) [55]. The values of crystal size and particle size for bare TiO₂ are more than those values for metalized TiO₂. This is related to the increasing of the number of located of Pt⁴⁺ ions in TiO₂ lattice, which depresses the growth of TiO₂ Hombikate (UV 100) nanocrystals [54]. The results show that each particle consists of about 9 to 11 crystals, according to the results obtained from the calculation of Crystallinity Index values by employing the following equation [56]:

$$ ()

where D_p is the particle size which is measured by AFM analysis and L and $$ are the corresponding mean crystallite size and the crystallite size calculated by the Scherrer equation and the modified Scherrer equation employing XRD data, respectively.

Table 2. Particle size measured by AFM and Crystallinity values of bare TiO₂ and platinized TiO₂.

Samples	Particle size/nm	*Crystallinity Index	**Crystallinity Index	Average Crystallinity Index
TiO2	80.940	7.046	7.988	7.517
Pt(0.25)/TiO2	63.600	5.889	6.346	6.117
Pt(0.50)/TiO2	77.020	8.233	8.104	8.168
Pt(0.75)/TiO2	54.890	5.370	5.724	5.547
Pt(1.00)/TiO2	73.130	6.981	8.851	7.916

• *Crystallinity Index calculated by divided particle size on mean crystallite size and **Crystallinity Index calculated by divided particle size on crystallite size.

The maximum value of average Crystallinity index for Pt (0.5)/TiO₂ is found to be 8.168. That referred to the suppression of the crystal defects number through decreasing the amorphous phase present in TiO₂ and overall enhancing the photocatalytic activity of TiO₂ [57].

3.1.4. UV-Visible Diffuse Reflectance Spectra

The UV-vis absorbance spectra of the bare TiO₂ and platinized TiO₂ (0.5% Pt) powders were also measured to confirm the Pt-loading trend and to measure the effect of Pt loading. The results from UV-visible reflectance spectra as plotted in Figure 5 clearly show the shift of absorption edge towards longer wavelength for platinized TiO₂. These results indicate that the excitation of metalized TiO₂ occurs with the narrowing and red shift of the band gap energy (E_g) peak [58]. These results were subsequently agreed with the increasing of the average Crystallinity Index [57].

3.2. Effect of the Metal Loading on Photocatalytic Activity of Methanol Solution

The photocatalytic activity of the platinized titanium dioxide was first increased with the increasing of the metal loading until a maximum was reached with the following decrease in the activity. Figures 6 and 7 show the results obtained with the samples containing different amount of platinum. The highest photocatalytic activity was observed with the Pt loading of 0.5 wt%. This loading percentage may give the most efficient separation of photogenerated electron-hole pairs [59]. The presence of Pt on the TiO₂ surface leads to an increase of the surface barrier and the space charge region becomes narrower. As a result of the metal loading the space charge region becomes narrower leading to an increase of the efficiency of the electron-hole separation [60] and formation of the Schottky barrier by the electron transfer from the conduction band of TiO₂ to the conduction band of Pt. Thereby the recombination process is suppressed according to the following equations [31, 32, 61]:

$$ ()

$$ ()

$$ ()

Platinum acts as electron scavenger hindering the recombination of the charge carriers and ultimately exhibiting the enhancement of the photoreactivity as shown in the following equation [31, 32, 62, 63]:

$$ ()

However, when the percentage of the metal reached maximum, the additional amount leads to making the space charge layer very narrow. As a result the penetration depth of light exceeds the space charge layer. The recombination of the electron-hole pairs will be favorable and the photocatalytic activity will be reduced [60]. Moreover, the presence of metal on the TiO₂ surface reduces the number of the surface hydroxyl groups leading to the reduction of the photoreactivity [64]. This means that the metal on the TiO₂ surface acts both as an efficient trap site and as a recombination center at the same time [65]. Hence the rate of the methanal (HCHO) formation will be slower while the conversion of methanal to formic acid (HCOOH) is a faster process. On the other hand, with the increasing of the metal amount, TiO₂ samples will become more grey in color. Thus, the changed optical properties of the samples could lead to the screening of the light towards the TiO₂ and suppression of the electrons excitation to the conduction band [31, 66].

Two mechanisms for the photocatalytic oxidation (in the presence of O₂) and photocatalytic dehydrogenation (in the presence of N₂) of methanol with Pt (0.5)/TiO₂ are suggested as shown in Scheme 1. The scheme shows the differences between the mechanism of photooxidation and photodehydrogenation of methanol on platinized titanium dioxide. The formation of CaCO₃ in photooxidation process was indicated by passing the outlet gas in Ca(OH)₂ solution. However, no CO₂ formation was indicated in photodehydrogenation of methanol.

Differences in experimental conditions, such as, experimental equipment, type of photocatalyst, position of band edges of semiconductor compared to redox potential of O₂/$$ and ⁻OH/^•OH, and type and concentration of organic pollutant, cause difficulties in the comparison of photocatalytic activity of different materials. Xiang et al. [67] measured the formation rates of hydroxyl free radical for various semiconductor photocatalysts at the same experimental conditions. They discussed the difference of rates formation of hydroxyl free radical on various semiconductors. In another study Xiang et al. [68] showed that hydroxyl radicals are one of active species and indeed participate in photocatalytic reactions. They also found that the photocatalytic activity of Ag-TiO₂ exceeds that of P25 by a factor of more than 2. Our results are in good agreement with these findings.

The different yields that are suggested in the two mechanisms are HCOH and H₂O in the absence of oxygen (photocatalytic dehydrogenation of methanol) and HCOOH in the presence of oxygen (photocatalytic oxidation of methanol). The pH of the reaction suspension after one hour of irradiation was found 6.93 in dehydrogenation process while it was 4.82 in photooxidation process. This indicates the further oxidation of the formed formaldehyde to formic acid.

4. Conclusions

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.