Performance Evaluation of the SrZrxSn1-xO3 Photocatalytic System for Remazol Yellow Dye Degradation Employing Box-Behnken Design Avaliação de Desempenho do Sistema Fotocatalítico SrZrXSn1-XO3 para Degradação de Corante Amarelo de Remazol Utilizando o Modelo Box-Behnken Evaluación del Sistema Fotocatalítico SrZrXSn1-XO3 Utilizado para la Degradación del Colorante Aamrillo Remazol Empleando Diseño Box-Behnken

Contamination of effluents often occurs due to improper disposal of textile dyes or their by-products. These can often be carcinogenic and/or mutagenic to the biome. Given the above, the need for effective methods for treating effluents is clear. This treatment occurs by biological, physical, and/or chemical processes. Regarding chemical processes, heterogeneous photocatalysis stands out, mainly because it guarantees an effective degradation of contaminants. In this sense, mixed metal oxides, act as photocatalysts and constitute structures capable of producing a large family of solids with physical properties suitable for the degradation of many pollutants. Modified ABO3 perovskites, as in the case of the SrZrxSn1-xO3 semiconductor system, are effective in the degradation of textile dyes in effluents. The present work aimed to use the Box-Behnken model to evaluate the performance of the oxides resulting from the structural modifications of the perovskite SrZrxSn1-xO3 system, concerning the discoloration of the golden yellow dye remazol. The synthesized oxides were characterized by instrumental techniques and a Box-Behnken 3 project was developed. From this, the influence of some factors such as structural modification, oxide mass, exposure time, and the number of UVC lamps was evaluated. The discoloration of the dye was monitored from the attenuation of the absorbance at the wavelength 411 nm. According to the results obtained, the highest percentage of discoloration was obtained using the modified oxide SrZr0,25Sn0,75O3 for an approximate time of 6 hours in contact with 1 UVC lamp.


Introduction
Population growth has caused an increasing demand for consumer goods resulting in expanded industrial development of the textile sector. With the industrial advance the dyes used for dyeing have become synthetic and potentially harmful to the environment. The contamination of effluents during the dyeing process has become a serious problem. The textile industry generates large volumes of contaminated effluent daily, which if not treated and disposed of correctly can cause serious environmental problems (Catanho, Malpass, Motheo, 2006, Soutsas, et al. 2010. For this reason, the removal of these contaminants has become a serious object of study (Mahmouda, et al. 2017, Sainia, et al. 2017, Fabbricino, Pontoni, 2016. In the literature, there are reports on works involving the degradation of this type of contaminant by biological (Correa, et al. 2009), physical (Filtration, coagulation / flocculation), chemical (chemical precipitation, chemical oxidation, adsorption) (Kuns, et al.2002, Dotto, et al. 2011, Vasques, et al. 2011) and photocatalytic processes (Catanho, Malpass, Motheo, 2006, Ramo, et al 2020. Among the chemical processes, Advanced Oxidative Processes (AOPs) have been studied and widely used to ensure an effective degradation of the contaminants (Maeda, Eguchi, Oshima, 2014, Wang, Tadé, Shao, 2015. These AOPs generate hydroxyl radicals that consist of a powerful oxidizing agent. Since the hydroxyl radicals generated are highly reactive and not selective, they can act on the chemical oxidation of a wide range of substances (Peralta-Zamora, et al., 1999, Malato, et al. 2002. Among the most commonly used AOPs is heterogeneous photocatalysis based on the activation of a semiconductor (catalyst) by sunlight or artificial light. The most applied catalysts are: TiO2, ZnO, FeO3, SiO2, ZnS, all presenting promising results in the degradation of dye in water (Costa, et al. 2004, Cervantes, Zaia, Santana, 2009. The metal oxides of the general formula ABO3, also called perovskite (Toniolo, et al., 2012), such as SrSnO3 and CaSnO3, act as photocatalysts ( Barros Neto, Scarminio, Bruns, 2010). These compounds have versatile structures capable of originating a large family of solids. They were initially studied because they have excellent physical properties such as ferroelectricity, piezoelectricity, pyroelectricity and magnetic effects (Toniolo, et al., 2012). Modified ABO3-type perovskites, such as the SrZrxSn1-xO3 semiconductor system, have shown good results for the degradation of textile dyes in effluents, indicating their suitability in the treatment of industrial effluents. In order to evaluate the degradability of these perovskites, however, it is necessary some experimental parameters must be identified, such as structural modification, quantity of oxide to be used, exposure to UVC light and time of contact of the oxide with the dye.
To evaluate the influence of these parameters, we carried out photocatalytic tests using experimental designs, which are statistical tools that enable the determination of the influence of independent variables and their multivariate and simultaneous interactions ( Barros Neto, Scarminio, Bruns, 2010). Indeed, the use of chemometric tolls help to optimize and undertand the catalytical processes and other important processes related to solid materials (Yücela, Yücelb, Durakb, 2016, Meinratha, 1998, Meinratha, Lisd, Elbanowskid, 2004).

Methodology
The methodology for the development of this work is structuralist since from the study of the photocatalytic process of removing the yellow dye from remazol and the evaluation of the influence of some experimental parameters, a statistical model was built capable of predicting the percentage of degradation of the dye ahead variations in these experimental parameters.

Synthesis and characterization of the SrZrxSn1-xO3
Oxides in the SrZrxSn1-xO3 modified system (x = 0.25, 0.50 and 0.75) were synthesized based on the Pechini method (Pechini, 1967). The method consists of forming a chelate between the metal cations and a carboxylic acid, such as citric acid.
The chelate produced was polymerized using ethylene glycol to form a polyester (polymer chain) that was subsequently thermally treated to obtain the crystalline material. Tin and Zirconia citrates were obtained as described in the literature (Cavalcante, et al., 2007, Alves, et al., 2009).

X-ray diffraction
Samples were analyzed using an XRD-6000 X-ray diffraction spectrometer from SHIMADZU, with a power of 2 kVA, a voltage of 30 kV and a current of 30 mA. The slits used were: 1 ° divergence, 1 ° dispersion and 0.3 mm reception slit, with sweeps in the range of 2θ = 10-90 °.

Raman spectroscopy
Measurements were performed using a Raman spectrophotometer with a photoluminescence accessory, coupled to a Renishaw Raman microscope equipped with a solid state laser diode operating at an exposure time of 5 minutes.

UV-visible spectroscopy
Absorption spectra in the UV-Visible region, solid oxide and solutions, were recorded in the 900 to 190 nm region using a spectrophotometer, SHIMADZU, model UV-2550. Solid oxide spectra were used to calculate the gap energy according to the TAUC method (Wood, Tauc, 1972). And the spectra of the solutions were used to evaluate the attenuation of the radiation absorption in the 411nm wavelength.

Photocatalytic Tests Using Experimental Designs
To evaluate the photocatalytic performance of the oxides resulting from the structural modifications of the SrZrxSn1-xO3 perovskite system, experimental conditions such as structural modification, amount of oxide to be used, time of exposure and intensity of UV radiation (number of lamps) were identified with the aid of techniques of planning and optimization of experiments. To evaluate the significance of the 4 factors studied, as well as the interactions among them, a Box-Behnken design was used, according to Table1. in water was set at 5mg.L -1 . Each experiment was performed in triplicate. The dye used was remazol golden yellow RNL.
Each experimental system defined by the Box-Benhken model was represented by a beaker 100 mL of the solution containing the dye and the appropriate mass of the oxide (defined by the experimental design). Then, each beaker was placed inside a photocatalytic reactor to expose it to the appropriate amount of UV radiation for the time indicated experimental design. The intensity of the UV radiation was varied according to the number of 30W UV Phillips UVC (254 nm) lamps placed inside the photocatalytic reactor. The dependent variable was the attenuation of the absorbance at the characteristic wavelength of the chromophore responsible for the dye color used (411 nm), related to photocatalytic discoloration of the dye (Ajmal, 2016). This variable was monitored by recording spectra in the UV-vis region.
The experimental design as well as the statistical treatment and evaluation of the data obtained were performed using software Statistica®, Statsoft, version 10. Figure 1 shows the XRD patterns of the SrZrxSn1-xO3 system calcined at 700 ° C for 2h under air. We observed welldefined peaks at 700 ° C related to the orthorhombic perovskite structure, according to crystallographic sheets JCPDS 00-044-0161 (ZrSnO3) and 01-077-1798 (SrSnO3) for pure and substituted samples. We can also observe low intensity peaks related to strontium carbonate (#), around 25,3 °; 36.2 °; 44.2 ° and 50.0 °; According to the JCPDS form (01-077-198).

X-ray difraction
We observed a widening of the peaks and displacement to higher values of θ as a function of the increase of the Zr 4+ concentration in the SrZrxSn1-xO3 system. This behavior may be related to the rearrangement of the cation substitution at the octahedron site of the crystalline system, generating greater long-range disorder (Tarrida, Larguem, Madon, 2009).  Figure 2 shows the Raman spectra of the SrZrxSn1-xO3 substituted system. A change of the profile in the region between 147 and 180 cm -1 , also known as the lattice modifier region, can be observed in relation to the lattice vibrational modes (dodecahedron site, cation A of ABO3 structure) (Tarrida, Larguem, Madon, 2009, Zhang, Tang, Ye, 2006. The substitution of Sn 4+ (lower cation) by Zr 4+ (higher cation) in the SrZrxSn1-xO3 system promotes significant changes in relation to the active modes in the Raman spectrum of these samples, especially in relation to the sample SrZr0.25Sn0.75O3 (See Figure 2B). This indicates significant changes relative to the region of the lattice former (octahedron site, Research, Society andDevelopment, v. 10, n. 2, e48610212328, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i2.12328 6 cation B of the ABO3 structure), mainly in relation to the 223 cm 1 region (binding mode, BO), thus presenting a greater number of active modes in the Raman spectrum (See Table 2). Table 2: Frequencies (cm -1 ) of the Raman absorption bands and designations for the SrZrxSn1-xO3 system.

Source: Authors,
As can be observed in Figure 2 and Table 2, there are variations in the profiles of the bands in relation to the region between 407 and 412 cm -1 (torsional mode, B-O3) and 558 to 560 cm -1 (stretch mode, BO) for all samples. In addition, we can observe the presence of bands in the region of 698-750 cm -1 (Figure 2), which according to the literature are related to secondorder vibrations, resulting from overlays of several active modes (Tarrida, Larguem, Madon, 2009, Zhang, Tang, Ye, 2006, Nakamoto, 1986. The behavior of the Raman spectra of the samples in the analyzed SrZrxSn1-xO3 system, showed that the composition SrZr0.25Sn0.75O3 is able to have a better potential for photodegradation of the dye in relation to the other compositions, because it has a greater number of bands in the Raman spectra. According to the literature, the greater the number of bands observed in the Raman spectra, the greater the degree of distortion of the orthorhombic perovskite structure, the greater polarization of the molecule and consequently the greater migration of electrons in the crystalline lattice (Zhang, Tang, Ye, 2006, Nakamoto, 1986.

UV-Vis spectroscopy
An increase in the amount of the substituent cation Zr 4+ in the SrZrxSn1-xO3 system (Table 2) promotes the increase of band gap energy values of the substituted materials, corroborating the behavior observed in the Raman spectra of these materials (See Figure 2), where the composition SrZr0.25Sn0.75O3 presents a greater degree of asymmetry due to the presence of a greater number of active modes in the Raman spectrum (Tarrida, Larguem, Madon, 2009, Zhang, Tang, Ye, 2006 Research, Society andDevelopment, v. 10, n. 2, e48610212328, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i2.12328 7 cations at the octahedron sites promotes different levels of electron polarization in the orthorhombic perovskite structure (Zhang, Tang, Ye, 2006, Mountstevens, Attfield, Redfern,2003, which is clear from the different profiles and band numbers presented in the Raman spectra (Figure 2), as well as band gap values obtained from the absorption curves in the visible region of these materials (see Figure 3 and Table 3).
Source: Authors, Table 3: Band gap values calculated from the absorption spectra of the SrZrxSn1-xO3 system by the Wood and Tauc method (Wood, Tauc, 1972).

Absorption spectra in the UV-vis of the remazol golden yellow dye
The color of an azo dye, such as Remazol, is a result of the interaction between an azo group (-N = N-) and an aromatic species (Sauer et al., 2005, ZIlle, Gornacka, Rehorek, 2005 The UV-vis absorption spectra of the remazol golden yellow dye at the 5 ppm concentration is shown in Figure 4, where the absorption band of the azo compound (411 nm) and the aromatic ring (238 and 293 nm) can be observed. The efficiency of the decolorizing capacity of the oxide studied was related to absorbance attenuation at wavelength 411 nm. Source: Authors,

Box-Behnken design
To evaluate of the performance of the synthesized oxides as to the decolorization capacity of the remazol golden yellow dye, a Box-Behnken model was used owing to its ability to evaluate factors at three levels using few experiments. The Box-Behnken model was used to evaluate the performance of the synthesized oxides in relation to their ability to discolor the dye.

Box Behnken statistical analysis
The Initially, a simple model was developed where interactions were not included, i.e., only the main factors were considered. The coefficient of determination (R 2 ) for this model was 0.75085 and the adjusted R 2 was 0.6401, indicating that 24.91% of the total variation was not explained by the model. The value of adjusted R 2 less than R 2 is related to the small sample size and the amount of terms in the model (Yetilmezsoy, Demirel, Vanderbei, 2009). We tested the model that took into account the linear and quadratic interactions between the factors, except for the quadratic effect of the oxide composition, because this is a stoichiometric ratio. In this case the model had a R 2 = 1.0000 and an adjusted R 2 = 0.9999. The high value of R 2 demonstrates the high significance of the model and the correlation between the dependent and the independent variables.
The RS graph in Figure 6 shows the relationship between time and UVC radiation intensity. Source: Authors, According to this graph the lowest absorbance values are reached at times longer than 6 hours, using one lamp. In this case, the composition was fixed at the intermediate level and mass at the minimum level. The evaluation of the individual experiments confirmed the conclusions obtained through the statistical analysis of the data. The percentage of discoloration for these experiments was approximately 97%, calculated according to the following equation (Weng, Tao, 2015, Sales, et al., 2014.

Final Considerations
Perovskite SrZrxSn1-xO3 was successfully obtained with the polymer precursor method, based on the Pechini method.
The photocatalytic performance of the oxides resulting from the structural modifications of the SrZrxSn1-xO3 type perovskite system relative to the discoloration of remazol golden yellow dye was evaluated using a Box Beckhen experimental design. This permitted a performance evaluation of the oxides resulting from structural modifications. In this experimental design, four independent variables were evaluated: structural modification, amount of oxide used (grams), time of exposure (hours) and intensity of UV radiation (number of lamps). The dependent variable evaluated was the attenuation of the absorbance at the characteristic wavelength (411 nm) of the chromophore responsible color related to photocatalytic discoloration of the dye. The evaluation of the Box-Behnken model showed that the oxides that presented the best discoloration performances were SrZr0.25Sn0.50O3 or SrZr0.50Sn0.50O3, for exposure times longer than 6 hours and using 1 lamp.
The model explains the efficiency of the degradation of the oxide.
Thus, according to the data obtained, we can affirm that the application of the modified oxides for the degradation of remazol golden yellow dye is a viable treatment for water reuse and reduces pollution of the water bodies by textile industries.
And the Box Behnken planning model is an adequate tool for the evaluation of the photocatalytic capacity of these oxides providing multivariate and scientific interpretation of the action of individual factors and their interactions.
Chemistry, like most sciences, has a great potential to meet the demands of society as well as improve living and health conditions. For the continuation of this research and for it to be effectively applied to society's problems, studies will be carried out on real samples, that is, effluents from the textile industry.