Sol–gel catalysts immobilized on stainless steel meshes for Ba²⁺ removal in a continuous flow process: An experimental design

The need to increase the production of food and industrial inputs, in order to serve the market, has generated waste and an increase in the pollutant load in water, air, and soil. In this polluting load are organic compounds (emerging pollutants and industrial plants) (Khadim et al. 2022) and inorganic compounds (heavy metals) (Jabbar et al. 2022).

Barium is present in effluents from chemical, petrochemical, automotive, and metallurgical industries, among others. In addition to effluents, groundwater is also susceptible to contamination by this metal, due to some urban activities such as domestic slurry production, septic tanks, leaking municipal sewage networks, and leaking gas stations. Also, agricultural activities, in general, by the inadequate use of mineral and organic fertilizers, herbicides, and pesticides, cause metal contamination. However, there are few studies in the literature concerned with the Ba pollution potential. However, due to the new and growing barium applications in industrial processes and its consequent increase in human exposure, there is a need to develop effective processes for the removal of barium(II) from the environment (Pepe et al. 2013; Kravchenko et al. 2014; Zhao et al. 2022).

In this context, many methods have been studied to remove contaminating compounds, for example, adsorption (Atiyah et al. 2022; Khadim et al. 2022), ozonation (Graça et al. 2020), photocatalysis (Fidelis et al. 2019; Abbood et al. 2023), photocatalytic ozonation (Fidelis et al. 2023), UV (Bueno et al. 2023), UV/TiO₂, electrochemical processing (Kadhum et al. 2021), Fenton (Ali et al. 2023), and UV/H₂O₂ process (Ali et al. 2022b).

In particular, the photocatalytic process has attracted a lot of attention due to efficient contaminant removal. However, application of photocatalysts usually occurs in suspension, which guarantees a large surface area for the reaction and facilitates a mass transfer process (Fontana et al. 2018; Abreu et al. 2021). The use of photocatalysts in suspension, especially in continuous flow processes, requires a subsequent separation process to remove the catalyst from the final effluent. Nevertheless, most semiconductors are fine powders, requiring additional procedures for their recovery. In this way, the cost of the process increases, making its application on an industrial scale unfeasible. Thus, there is a need for the development of catalysts that can be easily recovered and reused.

The use of immobilized catalysts can reduce costs and simplify the process. Many attempts to immobilize TiO₂ photocatalyst on different support structures were made. These studies sought to increase the surface/volume ratio at the same time, which consequently increases the photocatalytic activity. Materials being studied include encapsulated (Atiyah et al. 2022), biopolymer (Lenzi et al. 2022), magnetic (Oliveira et al. 2023), among others. Some of these materials can be affected by process parameters such as pH, temperature, and contaminant concentration.

Semiconductors can be immobilized on different materials to form mechanically resistant layers (Rachel et al. 2002; Miranda-García et al. 2010; Shan et al. 2010; Panniello et al. 2012; Bet-moushoul et al. 2016; Manassero et al. 2017; Cunha et al. 2018; Karaolia et al. 2018). Although many studies point out that catalytic systems with immobilized catalyst are less efficient, some authors such as Gar Alalm et al. (2016), Vaez et al. (2012), Zhang et al. (2017), and Adamek et al. (2019) achieved greater catalytic activity for catalysts on a suitable support, compared to suspended catalysts.

El-Kalliny et al. (2014) used TiO₂ immobilized on stainless steel meshes for water purification. The study indicated that stainless steel wire was a good photocatalyst substrate due to its large surface area and the possibility for light to be effectively distributed through it.

Thus, the contribution of the manuscript was the synthesis of TiO₂ structured catalysts using a sol–gel method and immobilizing them on stainless steel meshes. These catalysts were applied in barium removal, in a continuous system. To better evaluate the process, a study of parameters involved in the process was carried out through the experimental design. The catalyst was characterized by scanning electron microscopy (SEM) and point-of-zero charge (PZC).

The following chemicals were used in the experiments: glacial acetic acid (Neon); titanium(IV) isopropoxide 97% (Aldrich); nitric acid PA 65% (Synth); absolute ethanol PA (Dinâmica); barium nitrate PA (Perquim); formic acid 0.5 g/L (Synth); and TiO₂ P25 (Degussa).

The synthesis of the sol–gel TiO₂ precursor solution started with the addition of 10 mL of glacial acetic acid to an Erlenmeyer flask that is stirred using a magnetic stirrer, followed by the addition of 50 mL of titanium(IV) isopropoxide, kept under stirring for 5 min. Then, 200 mL of deionized water was added dropwise, keeping the solution stirred for 1 h. Subsequently, 6 mL of nitric acid was added to the solution, which was kept under stirring for 12 h at room temperature. After this period, it was submitted to the evaporation process in order to reach a more viscous consistency that provides a better adhesion to the support (Denisov et al. 2017; Martino 2022).

The TiO₂ catalyst synthesized by the sol–gel method was immobilized on 304 stainless steel nets in three different mesh sizes as shown in Table 1.

The stainless steel nets were previously cut into a rectangle shape of 20 × 4 cm, and then turned into a cylindrical shape. The nets were washed and dried at 343 K for 2 h in an oven with air circulation and renewal. The coating process was started using the wash-coating method, the nets were dipped in the precursor solution for 1 min and left to dry at room temperature. After the drying period, the nets were calcined in a heating ramp at 623, 823, and 1,023 K at a rate of 3.2 K min⁻¹. Then, the structured catalyst was cooled to room temperature by washing under running water, and then sonicated for 15 min in order to remove excess semiconductors. Subsequently, the nets were again dried at 343 K for 2 h.

The powder catalyst was characterized by: (i) PZC determination. The PZC was determined by applying a batch equilibration experiment (Guilarduci et al. 2006). In Erlenmyer flasks, 50 mg of catalyst and 50 ml of ultrapure water were mixed, under different initial pH values. Followed by constant stirring for 24 h in a shaker at 298 K and 120 rpm. The final pH measurements of the suspensions were performed. The pH_PZC corresponds to the average of the final pH values which tended to be a constant value, regardless of the initial pH value. (ii) SEM associated with energy dispersive spectroscopy (EDS): SEM characterization was performed using a VEJA 3 LMU – TESCAN microscope with a 30 kV filament, 3.0-nm resolution SE, and retractable BSE (higher energy electrons) detectors, low-vacuum mode (500 Pa), chamber with 230 mm of inner diameter, and CCD camera for preview of samples. The microscope is also equipped with an EDS detector, model AZTec Energy X-Act, resolution of 130 eV, brand Oxford. Before performing the analyses, all catalysts were metallized with gold for 10 min using the IC-50 ION COATER-SHIMADZU equipment.

The calibration curve was performed with Ba²⁺ at the following concentrations: 5, 10, 15, 20, 25, 30, 40, and 50 ppm. The correlation coefficient obtained was 0.998.

A peristaltic pump was used to deliver the solution to the system. The concentrations of barium(II) and formic acid were the same for the batch tests. In addition to the flow, the effect of pH was also observed.

Fontana et al. (2018) carried out photocatalytic tests with different catalysts in suspension for the removal of barium(II). The approximate results for 90 min of reaction for the catalysts are as follows: commercial TiO₂, TiO₂ (Anatase), TiO₂ P25, and Nb₂O₅ were 6, 22, 30, and 34%, respectively.

Comparing the batch results, it is possible to see that the synthesized sol–gel catalyst behaves similar to the TiO₂ P25 and Nb₂O₅ catalysts in 90 min of reaction.

An adsorption reaction was also carried out under the same conditions as the photocatalytic reaction but in the absence of light. The results indicated that there was no adsorption on the catalyst surface, which is in agreement with Fontana et al. (2018). The literature describes the differences obtained for specific surface areas of catalysts due to preparation methods, calcination temperature, and addition of promoters (Ali et al. 2022a, 2022b; Fuziki et al. 2023). In particular, TiO₂ prepared by the sol–gel method has higher values for surface area, 311 m²/g, and commercial, 19.78 m²/g (Lenzi et al. 2011).

At 623 and 1,023 K, barium(II) removal was seen only with the thin net (TF), and also this net immobilized more catalysts. On the other hand, TM and TG nets, both calcined at 623 and 1,023 K, did not remove Ba²⁺ ions, within the detection limit of the equipment. As shown by Denisov et al. (2017), the increase in temperature in the material generates more phase anatase in the structure, where the photoreactions are more efficient. However, throughout the reaction, it led to a more active stainless steel substrate component deposition on the titanium dioxide film surface, obscuring the catalyst photocatalytic properties. Thus, higher calcination temperatures on structures like this are not as good as when using the catalyst in suspension. Also, Barati et al. (2009) got better results when calcined their catalysts at intermediary temperatures, nearby 823 K. Therefore, 823 K was the chosen temperature to carry out successive experiments, while the mesh considered for successive experiments was the medium one, TM, that came up with the same results as TG, but TM is lighter and malleable.

From the results presented in Figure 4, the calcination temperature was set at 823 K and the mesh TM for the successive tests. For each run (Table 2), the pH and flow rate effect analysis were made by a simple experiment design, with two levels and triplicate on a central point – the last one for the eligibility of pure error – pH was changed so that the original pH value of the barium synthetic solution (2.8) remained within the chosen range for the experimental design levels, while the flow rate was lower to increase the residence time and, consequently, increase the contact of the solution with the catalyst. Table 2 shows the experimental design matrix.

Thus, a new factorial experimental design on two levels was made at pH = 8.2 that presented a better result on the ion removal, and was chosen as the new central point. As seen before, the flow rate does not have a significant role in influencing the results, so it remained at the same previous levels and a new experimental design was generated (Table 4).

As shown by Majidnia & Idris (2015) and Fontana et al. (2018), barium(II) removal increases with pH increase to a certain extent, but increasing the solution pH beyond that did not provide any improvement. Chen & Ray (1998) discussed that although the high concentration of OH⁻ ions in the medium increases the electron promotion and the formation of positive holes, the carbon dioxide generated is trapped in the solution, resulting in the formation of bicarbonate and carbonate, which due to their high reaction constants with the hydroxyl radicals end up eliminating them. This may also explain why the pH variation in a range above 6 in these tests did not have a significant effect on barium(II) removal.

As higher pH levels did not have a significant impact on test answers, the best option is to use it in a neutral range, close to 7.0, because it does not require the addition of much reagent, saving resources.

About the flow rate, there was no evidence regarding its effect on the results. Thus, in this chosen gradient (level −1 to +1 or 0.72–1.44 mL/min), the flow variation has little impact on the response of the experiments. However, the same cannot be said for much higher flow rates as the residence time would be very low, which would provide little contact between the element and catalyst. Meanwhile, in much lower flow rates, there would be an increase in the contact time with the catalyst. Although in this study there was no evidence of increase in barium(II) removal, instead the lower the flow rates, the lower was the ion removal. Perhaps by passing the solution very slowly through the reactor, it did not have enough movement to come into full contact with the catalyst, thus reducing the process efficiency.

The corresponding pH values for Ka₁ and Ka₂ were 4.9 and 7.91, respectively. The pH_PCZ found for the catalyst synthesized was 6.4. This result was in agreement with the PCZ value found by Andronic et al. (2016) for titanium dioxide. Also, these results are close to the values found by other authors who used different analysis methods, such as the surface potential measurement by Chou & Liao (2005) or the zeta potential measurement (magnitude of repulsion or attraction of charges between particles) made by Arlos et al. (2016). Therefore, in the reaction medium with a pH lower than 6.4, the photocatalyst studied has a positively charged surface, generating affinity for molecules with a negative charge, while in the reaction medium with a pH greater than 6.4, the surface has a negative charge, thus having more affinity for positively charged molecules. This confirms and explains why the best degradation rates of the photocatalytic assays occurred at pH values above 6.0, since the barium molecules in the reaction medium have a positive charge (Ba²⁺) and are, therefore, attracted to the catalyst surface when it is negatively charged, favoring the degradation process. Also, in media with low pH, there is a higher concentration of hydronium ions (H₃O⁺) near the surface of the catalyst, producing repulsive forces and also compete with barium ions.

This behavior was observed by Majidnia & Idris (2015), who, when evaluating the effect of pH on the degradation of barium in radioactive wastewater using supported TiO₂ found that as the pH increased, the sorption capacity of Ba(II) increased steadily until reaching pH 8, but increasing the pH of the solution beyond this value did not show any improvement. Also, Fontana et al. (2018) had similar results when degrading barium using titanium dioxide as a photocatalyst in suspension, where the best degradation rates were observed at pH values greater than 7.

However, both in the work by Majidnia & Idris (2015) similar to the work by Fontana et al. (2018), it was observed that at pH values higher than the PCZ of titanium dioxide, it did not obtain higher degradation rates, remaining constant or even lower. This effect was explained by Chen & Ray (1998), when they observed in their experiment that the best pH value for degradation was close to the PCZ of TiO₂ because, despite the high concentration of OH⁻ ions in the medium, it increased the promotion of electrons and the formation of positive vacancies trapping the carbon dioxide generated in the solution, resulting in the formation of bicarbonate and carbonate, which due to their high reaction constants with hydroxyl radicals, end up eliminating them. This may also explain why the pH variation in a range above 6, in the tests carried out for the present work, did not show a significant effect on degradation.

Comparing the results of the TiO₂ and P25 sol–gel catalysts in suspension, we observed that the synthesized catalyst is as efficient as the commercial catalyst (P25). When we use the continuous process, with the structured catalyst, there is a loss of activity due to the immobilization of the catalyst on the meshes, causing a decrease in active sites for the catalytic reaction to occur. In addition, effects of the calcination temperature of the meshes were observed. Calcination at 623 and 1,023 K causes a negative effect on the Ba(II) removal performance when stainless steel screens are used, in fine thick special meshes (TF and TG). On the other hand, at an intermediate temperature, 823 K, similar results were obtained for all mesh sizes (TF, TM, and TG), about 20.83% of Ba(II) removal at pH = 6.0 with residence time of 63 min in a continuous flow, which is a significant value compared to the removal of 24% of Ba(II) from the batch test at 63 min of reaction. The results obtained, under the conditions studied for pH, indicate that the best range is around 7, while the volumetric flow rate does not have a great effect on the removal of Ba²⁺ ions.

The authors are thankful to the Brazilian agencies CNPq and CAPES for financial support of this work, C²MMa.

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by L.N.B., G.G.L., A.M.T., E.A. and M.Z.F. The first draft of the manuscript was written by L.N.B., G.G.L., A.M.T., and M.Z.F. Also, the authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.