Influence of doping with Sm3+ on photocatalytic reuse of ZnO thin films obtained by spin coating

The difficulty in the powder reuse favors the study of materials in the form of thin films. ZnO based films have high photocatalytic potential. In this work, ZnO:xSm3+ (x = 0, 1, 2 and 4 %mol) thin films were prepared by spin coating method. The resins obtained to manufacture the thin films were prepared by complex polymerization method. The samples were characterized by X-ray diffraction (XRD), atomic force microscopy (AFM), field emission scanning electron microscopy (FE-SEM) and spectroscopy UV-Vis. XRD patterns indicate that doping with 4%Sm forms Sm2O3 as the secondary phase. FE-SEM images of the cross section of thin films indicated a uniform thickness ranging from 354 to 367 nm between samples. The incorporation of Sm3+ ions in the ZnO lattice increases the Egap of the films. The photocatalytic performance of the films was tested with the photodegradation of methylene blue dye. Pure ZnO and ZnO:1%Sm3+ exhibited the best activity in the photodegradation. Thin films of pure ZnO exhibit the best photocatalytic results for the first cycle, but considerably reduce their efficiency with reuse. Sm3+ incorporation, without secondary phase formation, was efficient for the maintenance of the photocatalytic property after 3 cycles.


1.INTRODUCTION
In recent years, semiconducting oxide photocatalysts, such as zinc oxide (ZnO), have increasingly received high attentions as the materials offer great potentials in the degradation of various environmental pollutants [1,2], such as pesticides, heavy metals and dyes [3,4]. In addition, the ZnO is known as one of the most important active photocatalysts due to its advantages, including high initial activity rates, many active sites with high surface reactivity, low price and environmental safety [5]. However, the photocatalytic activity of ZnO is limited to the ultraviolet region and the rapid recombination of photocatalytic loads (h+/e-) decreases the efficiency of the process, making it unfavorable to use as commercial ZnO photocatalyst [6]. Recently, it has been reported that doping of rare earth ions on ZnO improves photocatalytic performance [5]. Some studies have reported that rare earth ions are useful, since rare earth ions produce energy levels of impurities in the gap and their response to visible light. In addition, photogenerated charge carriers are produced, thus accelerating the interfacial transfer of the charge and inhibiting the recombination of the electron-hole pairs [7,8].
The ZnO powder catalysts show tendency to be difficult to recover and recycle. One substitutive form is thin films of ZnO which comprises dispersed immobilized ZnO particles or layers on a surface [2]. The ZnO thin films can be manufactured by various techniques, such as pulsed laser deposition (PLD) [9], molecular beam epitaxy (MBE) [10], chemical vapor deposition [11], sol gel process [12], spray pyrolysis [13] and Pechini [14]. Although physical methods have been widely used in thin film technology, chemical methods such as the Complex Polymerization Method (CPM) produce low-cost, simple and highly controlled films [15].
In this study, the effect of the photocatalytic properties of ZnO:xSm3+ (x = 0, 1, 2 and 4 %mol) thin films produced by the resin obtained by CPM was evaluated. The thin films were characterized by X-ray diffraction (XRD), UV-Vis spectroscopy, field emission scanning electron microscopy (FEG-SEM) and atomic force microscopy (AFM). The methylene blue dye was used in the degradation test is a batch of product in the textile industry. Three photocatalytic tests were performed without any treatment on the film.

2.MATERIALS AND METHODS
Polymer resins were prepared using citric acid/metal cations in the ratio of 6:1. Ethylene glycol was added in a ratio of 40/60 (%mass) relative to citric acid, keeping this constant value in all compositions. Firstly, Citric acid (C6H8O7.H2O) and Zn(NO3)2.6H2O were dissolved in 80 mL of water under stirring at 70°C, ethylene glycol was added after 1 hour and the temperature was raised to 90°C to form a homogeneous and transparent resin. The samarium resins were obtained in a similar form, where samarium nitrate was added stoichiometrically in the molar ratios of 1, 2 and 4%.
Before the deposition of the resin, silicon substrates (0.7 cm x 0.7 cm) underwent two-step cleaning, were first washed in ultrasonic bath in isopropyl alcohol and distilled water. The substrates were dried directly in spin coating. The polymeric precursor was then deposited on the substrates by spin coating to 7000 rpm for 30 sec, four layers being deposited. For each deposited layer was annealed at 350°C for 1 hour and 700°C for 2h at a heating rate of 1oC/minute in air.
The thin films were characterized by XRD using a Shimadzu diffractometer (Model XRD-7000, CuK radiation (λ = 1.54 Å), 40 kV and 30 mA and 2θ from 30° to 40°. The morphology was observed by SEM (Carl Zeiss Supra 35 VP-Model, Germany). Surface of the thin films was investigated by atomic force microscope (AFM). The reflectance of the films was obtained by the UV-Vis-NIR spectrophotometer (Shimadzu, UV 2600) over the spectral wavelength ranges from 200 nm to 900 nm. All the measurements were taken at room temperature. The total amount of organic carbon was used for mineralization analysis. For this, the TOC analyzer Shimadzu 500 was used.
Photocatalytic properties of ZnO:xSm3+ thin films (x = 0, 1, 2 and 4 %mol) were evaluated by measuring the photodegradation of MB (methylene blue) dye with a molecular formula C16H18N3SCl (99.5% purity) in an aqueous solution were illuminated by UVC lamps (Philips TUV 15 W, with maximum intensity at 254 nm = 4.9 eV). Samples with dimensions of 0.7 cm2 were placed in a quartz cell under visible light MB containing solution (1x10-5 mol.L-1). These suspensions were 1 hour in contact MB dye before the test is to allow the saturated absorption of MB onto the catalyst. The variations of the maximum absorption band of supernatant solutions were monitored at intervals of 30 minutes by UV-visible absorbance spectra measurements using a Shimadzu (UV-2600 model) spectrophotometer. The absorbance variation was observed for 420 minutes. All measurements were taken at room temperature. The reuse occurred in a similar manner to that described above, where after each cycle the thin films were air dried and the aqueous solution of MB (1x10-5 mol.L-1) was replaced with the aqueous solution of 420 minutes. The photodegradation was calculated by the equation 1 [9], given below: Where: C0 is the concentration of MB before irradiation (concentration of MB after 1 hour adsorption in dark) and C is the concentration of MB after the irradiation certain team. . 1 shows the XRD patterns obtained for the thin films of ZnO:xSm 3+ (x = 1, 2 and 4 %mol). All thin films show the characteristic peaks of ZnO layer type structure with hexagonal Wurtizita (JCPD 36-1431) were three diffraction peaks are related to reflection planes (100), (002) and (101). Thin films of ZnO:xSm with x = 1 and 2% did not present the formation of secondary phases. The presence of Sm2O3 phase was observed in ZnO:4%Sm thin film. According to Fig. 1, it is observed that the incorporation of Sm 3+ in the ZnO lattice provides a shift of the peaks to smaller angles, indicating that the doping occurred successfully. The shift in the peaks occurs due to the difference between the ionic radius of the cations, being 0.096 and 0.074 nm for the Sm 3+ and Zn 2+ , respectively [16].

Fig
It is interesting to note that the formation of secondary phase Sm2O3 caused reduction in intensity of peaks, as observed in the crystal planes (100), (020) and (011). The same happened in Rani et work. al. in his research depositing doped ZnO films with Nd 3+ was performed by synthesis by spray pyrolysis, XRD patterns showed that when a new peak by doping 5 mol% of Nd 3+ was the formation of oxide neodymium, and the intensities of the diffraction peaks (200) and (110) the doped ZnO film decreases significantly compared to pure ZnO sample [17]. The reduction in intensity occurred along with increasing the mean peak height width. This fact is associated with a reduction in crystallite size [18]. For more information on crystallite size, the Scherrer's equation was used, as shown in equation 2 [19]. Where β is the full width at half maximum of the main peak in XRD pattern. The estimated values for crystallite size are shown in table 1.

=
(2)   Fig. 2 shows the micrographs obtained by AFM. It is possible to observe the formation of grains grown and homogeneously distributed on the surface of the thin films, where this growth is associated to the time and temperature used in calcination [20]. The increase of the samarium concentration promotes the growth of the particles on the surface of the films. Fig. 3 shows the histograms referring to the average diameters of the ZnO particles on the surface of the films, where the mean particle size of ZnO follows the order: ZnO:1%Sm < ZnO < ZnO:2%Sm < ZnO:4%Sm. The addition of samarium at concentrations above 1% provides a greater number of defects, which imply the local energetic increase that acts as a driving force for growth with the calcination temperature [21].  The surface and cross aspects of the films were examined by FE-SEM. Fig. 4a presents the surface image of the thin film of pure ZnO, where the presence of small pores on its surface is noticed and that the grains present hexagonal aspect. The appearance of hexagonal grains on the surface of ZnO thin films is reported in other works [22]. According Nayeri et al. [23], the grain growth on the surface occurs due to the high energy applied to the system, where the larger grains grow through the coalescence of smaller grains. The measured thicknesses of the thin films ZnO:xSm (with x = 0, 1, 2 and 4 %mol) are shown in table 1. Fig. 4b shows the cross section of the ZnO:4%Sm, indicating that the thin film obtained by spin coating and calcined at 700°C is uniform and dense. As can be seen in table 1, the films have thicknesses in the range of 354.5 to 367.7 nm. The optical gap was obtained by extrapolating the linear region of the curve according to the Wood and Tauc method [24]. The main changes in the value of the energy of the optical band gap (Egap) can be correlated with the reduction or creation of structural defects or states located within the prohibited area, which may decrease or increase the intermediate energy levels. The band gap energy values Egap were calculated and shown in Fig. 5. The calculated band gap energy of ZnO films varies from 3.57 eV to 3.60 eV, indicating that with increasing concentration of the dopant, there was a slight increase in the energy band of these films. Chen et al. [25] reported that the band gap effects are directly related to the film thickness and the crystallite size of the material. As shown, the increase in Sm concentration reduces the crystallite size and the thickness of the thin films, increase their gap energy. The reductions in crystallite size and film thickness with the incorporation of rare earth ions, followed by the small increase in gap energy is widely discussed in the literature [26,27]. Generally, in the optical absorption, the electrons near the edge of the valence band become excited to the bottom of the conduction band due to gap energy [28]. During this transition process, if these electrons encounter disorder, it causes density of their states ρ(hν), where hν is the photon energy, tailing into the energy gap. This tail of ρ(hν) extending into the energy band gap is termed as Urbach tail [29]. We used a linear curve fitting to obtain the Urbach tail, which is defined as the width of the localized states available in the optical band gap that affects the optical band gap structure and optical transitions. The Urbach tail is determined by the following equation 3 [29]: Where: α is optical absorption, E is the photon energy, () is constant and EU is the Urbach energy, which refers to the width of the exponential absorption edge [29]. The value of EU was calculated from the slope, and the obtained values are given in Table 2, which indicates that no major changes occurred, as well as band gap values. ZnO is a class of semiconductor enriched with lattice defects, such as zinc vacancy (VZn), zinc interstitial (Zni), oxygen vacancy (VO), and oxygen interstitial (Oi). Similarly, the dopants can introduce defects to the ZnO [30]. The cationic substitution of Zn 2+ by Sm 3+ can cause many structural, morphological, optical and photocatalytic changes, such as the lattice parameter, gap, grain size, active defect sites and visible light absorption. It is known that the catalytic activity of ZnO nanoparticles doped with lanthanide ions is greater than that of pure ZnO [31]. The doping process with lanthanide ions effectively suppress electron-hole recombination and produce excess free radicals necessary for degradation [6]. These radicals are strong oxidizing agents to degrade refractory organic compounds such as organic dyes and convert them into harmless materials such as H2O and CO2 [32]. The catalytic activities of Sm doped ZnO were evaluated by the degradation of aqueous methylene blue. Methylene blue dye is widely used in the textile industry [4]. Fig. 6 shows photocatalytic activity of ZnO:xSm 3+ (x = 0, 1, 2 and 4 %mol). The photolysis curves were plotted without the presence of catalyst. The degradation efficiency of pure and Sm doped ZnO films for an irradiation time of 420 min was measured to be ZnO = 89.3%, ZnO:1%Sm 3+ = 90.4%, ZnO:2%Sm 3+ = 88.1% and ZnO:4%Sm 3+ = 68.2% while photolysis reduced only 3.1%. The increase in photocatalytic activity for the ZnO:1%Sm 3+ sample is associated with the introduction of intermediate levels in the ZnO conduction band [33]. The reduction of photocatalytic activity for doping above 1% Sm 3+ may be associated with reduction of anionic vacancy and excess cationic vacancy generated by the substitution of Zn 2+ by Sm 3+ . In addition, Sm2O3 phase formation reduces the photocatalytic activity of ZnO. The photocatalytic activity may also be associated with the concentration and size of the ZnO particles on the films surface. As observed through the AFM images, the ZnO:1%Sm 3+ sample has the smallest mean particle size, consistent with the photocatalytic results. The TOC tests indicated that ZnO, ZnO:1%Sm 3+ , ZnO:2%Sm 3+ , ZnO:4%Sm 3+ samples and the photolysis reduced the amount of organic carbon by 80.1, 84.7, 78.3 and 1.4%. Recycles and maintaining high photocatalytic activity are critical issues for long-term use in practical applications of the catalyst. Therefore, two criteria are required to be considered: (i) the stability of the catalyst to maintain its high activity over time [28] and (ii) the ease with which the catalyst could be recycled from solution [34]. Fig. 7 shows the reuse curves for pure and Sm doped ZnO films. According to Fig. 7, the photocatalytic activity of ZnO reduces its efficiency with the course of the reuse. Efficiency reduces from 89.3 to 77% at the end of the third cycle. Yassitepe et al. [1] reported that the photocatalytic activity of ZnO film gradually decreased, but if the photocatalyst are stored in the dark overnight, the photocatalyst activity can be recovered. Doping with Sm gives a lower efficiency reduction with the course of the reuse. The film with 1% Sm reduces from 90.4 to 83.4% and the film with 2% Sm reduces from 88.1 to 84.4%. Thus, the sample with 2% Sm presents the best photocatalytic activity after the third cycle of use, indicating that the incorporation of Sm increases the stability of the catalyst. As shown in Fig. 2, the increase in Sm concentration generates larger grains on the surface of the films. Larger grains have lower surface energy, interacting less with the dye molecules, maintaining the photocatalytic activity after more cycles [35]. The use of films instead of particles facilitates their reuse, avoiding the need for complex treatments for the non-release of particulates as residue. In addition, nanoparticles are difficult to separate, making the process impracticable on larger scales.

4.CONCLUSIONS
ZnO:xSm3+ (x = 0, 1, 2 and 4 %mol) thin films were successfully obtained by the Complex Polymerization Method (CPM) and spin coating. XRD analysis is showed that the thin film samples have wurtzite structure, presenting the formation of secondary phase only for ZnO:4%Sm sample. FE-SEM illustrate that the thin films of ZnO produced by CPM method have a homogeneous and dense characteristic. The incorporation of Sm3+ into the ZnO lattice provides the reduction of the crystallite size and thickness of the thin films, in addition to generating a slight increase in gap energy. The sample with 1% Sm shows a smaller particle size on the surface of the film, thus having the best photocatalytic activity. Larger particles on the surface of thin films make them more stable, where the sample with 2% Sm presented better reuse property. The appearance of the Sm2O3 phase reduced the activity of the thin film obtained with 4% Sm.