Novel solar simulated photocatalytic heterolysis of pharmaceutical wastewater via slag nanocomposite immobilization: Optimization using response surface methodology

The pharmaceutical sector has recently expanded due to the rising demand for its many goods, which are used to treat animals and humans (Alalm et al. 2015). Pharmaceutical industrial wastewater includes several bio-recalcitrant organics that are challenging for traditional wastewater treatment methods to eliminate (Gar Alalm et al. 2016). Not only do industries produce highly concentrated wastewater in medicines, but groundwater, surface water, and municipal wastewater have also been documented to contain small amounts of various pharmaceuticals (Shi et al. 2019). Due to their widespread use and intake in humans and animals, medicines and antibiotics are regarded as the most important pharmacological chemicals (Alalm et al. 2015; Shi et al. 2019). The pharmaceuticals are also widespread in the sewage and must be removed because of their possible negative effects on human health and ecosystems, like the following references clearly highlight (Di Marcantonio et al. 2021, 2023).

The steelmaking industry generates a huge quantity of byproducts (Kang et al. 2015). Slag is a solid byproduct produced during the steelmaking process (Proctor et al. 2000). The yearly global production of steel byproducts in the form of slag exceeds 50 million tons (Tsakiridis et al. 2008). SiO₂, Fe₂O₃, Al₂O₃, P₂O₅, CaO, MnO, MgO, and TiO₂ are the major constituents of steel slag (Tossavainen et al. 2007). The storage of this slag necessitates the use of a vast quantity of land and results in significant contamination of the environment (Wang et al. 2014; Safo et al. 2022a). Appropriate treatment and the use of this slag to reduce its pollution impact on the environment are becoming increasingly important.

A few recent researches have shown that transforming slag into a nanocomposite may boost the efficiency of the photodegradation reaction due to the enhanced dispersion of iron in water and the potential of iron reuse (Safo et al. 2022a, 2022b). Numerous water treatment applications have made use of steel slag that has been ground into a nanocomposite. However, there are limits to the amount that may be collected and reused (Mensah et al. 2022; Onen et al. 2023). In addition, the issue of collection may be resolved by coating the supporting plate with a catalyst. However, typical immobilization procedures such as heat bonding are energy-intensive (Khataee et al. 2009). Therefore, in this study, immobilization was done using Polysiloxane and aluminum plates without expending any energy. Polysiloxane may form strong bonds between the plates and catalyst particles (Samy et al. 2020). Moreover, acid treatment was used to convert the slag into a nanocomposite.

For the first time, steel waste was synthesized into nanocomposites steel slag (NCSS) using the solvothermal preparation technique and applied under a novel photo reactor for photodegradation of paracetamol (PC) in a reusable immobilize mode. The impacts of operational factors, such as catalyst dosage, pH, PC concentration (PC), and reaction time, were investigated through response surface methodology (RSM). The reusability of NCSS was also assessed to study the stability of the produced nanocomposite.

In this investigation, slag from Egypt's Ezz steel was examined. Nitric acid (69% HNO₃), hydrochloric acid (37% HCl), and PC were purchased from the United Kingdom (Fisher Scientific Company).

The preparation of the nanocomposite steel slag (NCSS) was done based on our previous studies (Safo et al. 2022a). In brief, the steel slag was added to 10 mL of HNO₃ in a glass beaker and continuously stirred at 500 rpm at room temperature for 5 h. The solution was poured into a glass dish and dried at 70 °C for a day (24 h). The dry samples were mashed and calcined for 60 min at 1,000 °C. The final product was named nanocomposite steel slag (NCSS).

In addition, the coating was done using the method reported by Gar Alalm et al. (2018). In brief, the immobilization of the catalyst was done by coating the surface of a 3003-grade aluminum plate of thickness 2 mm and diameter 60 mm containing 1 mm thickness of polysiloxane with the optimum amount of NCSS from the RSM. The plate was then dried at room temperature for 720 min. Finally, the plate was washed with distilled water for several times to remove unbonded particles and to validate the tight connection between the aluminum plates and the nanocomposite.

The composition, including the chemical and elemental components of the prepared samples, was characterized by energy-dispersive X-ray spectroscopy (JEOL, JEM-2100F, Japan) and X-ray fluorescence (Shimadzu XRF, Japan). The prepared sample's microstructure, morphology, and nanostructure were analyzed using a scanning electron microscope (SEM) (JEOL JSM-6010LV SEM, Japan). Transmission electron microscope (TEM) was used to obtain high-resolution images. The bond and the functional groups of samples were also examined using Fourier Transform Infrared Spectrometer (Shimadzu FTIR-8400s, Japan). The crystallographic information of the raw slag and NCSS were also examined using X-ray diffraction with a 1.54 Å, Cu-Kɑ radiation beam (Shimadzu XRD-6100, Japan).

The test was done based on the matrix parameters from the RSM, as shown in Table 1, at varying pH from 3 to 12 and PC concentrations of 10–30 ppm under the novel photo rector.

Using high-performance liquid chromatography (HPLC), the PC was measured for the initial and final degradation processess after the samples were filtered using a syringe filter of porosity 0.2 um into a C-18 column. As a mobile phase, a combination of ultrapure water and methanol with corresponding proportions of (60:40) was utilized to detect the signal peak of the PC. The injection process was done at a flow rate of 0.9 ml/min, 40 °C temperature, and 25 ml volume.

Here, C₀ signifies the concentration at a zero time, and ΔC is the difference between the initial concentration and the concentration at any given time, t.

The optimization of operational parameters such as A; NCSS dosage, B; PC, and C; reaction time, was done using RSM. As shown in Table 1, the response matrix was generated using a Box–Behnken design (BBD). The BBD was used because it does not need models where all factors are simultaneously performed at their maximum or minimum values but rather gives the same information as a complete factorial design with fewer experiments (Safo et al. 2023). Consequently, this approach is advantageous for avoiding trials under severe circumstances that might provide unsatisfactory findings (Tetteh et al. 2021).

It also emerged in the raw-slag for both the stretching vibration of the O–H peak at 3,436 cm⁻¹ and the bending vibration of the H–O–H absorption peak at 1,638 cm⁻¹, as shown in Figure 2(b). The former and the latter might be associated with chemically bonded water (Bai et al. 2018). These showed that crystallization and adsorption water were present in the raw-slag, and the same occurrences were seen in the NCSS nanocomposite produced, as shown in Figure 2(b). Furthermore, there was a distinctive absorption peak of about 1,487 cm⁻¹ whose intensity was reduced, and a certain displacement occurred. This might be ascribed to the raw-slag chemical and physical interaction with the HNO₃ and the heat treatment as described in a previous study (Safo et al. 2023) which diminished the raw-slag lattice deformation and vibration, reducing the corresponding stretching vibration C–O modes, which generally arise if a sample increases in CaO content (Ma et al. 2019; Jeon et al. 2021). Additionally, the absorption peak of 1,002 and 710 cm⁻¹ are features of Si–O and Si–O–Al stretching modes whose vibration stretch increases, indicating an increase in silica and silicate content (Bai et al. 2018; Sanad et al. 2021). The bond around 519 cm⁻¹ is a vibration of Fe–O, Mg–O, Ti–O, and other metal-oxides (Sarkar et al. 2019).

The EdX analysis of the NCSS as shown in Figure 3(b) reveals that the NCSS contains around 26.3 wt% of O, 32.4 wt% of Fe, 11.94 wt% of Si, 0.27 wt% of Ti, 3.26 wt% of Al, 2.93 wt% of Mg, 20.43 wt% of Ca, and 2.47 wt% of Mn.

The samples' X-ray diffraction patterns were examined using the Hanawalt System for Joint Committee for Powder Diffraction Standards (JCPDS). Figure 4(b) illustrates the XRD patterns of the raw-slag and the prepared NCSS which show the complex composition of raw-slag, which is made up of Hematite (Fe₂O₃, PDF No. 00-024-0072), Larnite (Ca₂SiO₄, PDF No. 00-006-0476), Calcite (CaCO₃ PDF No.00-003-0569), Laihunite (Fe_1.5O₄Si, PDF No. 01-078-1435), Esseneite (AlCaFeO₆Si PDF No. 00-040-0496) and Andradite (Ca₃Fe₂O₁₂Si₃, PDF No. 00-003-1136. In the NCSS the crystallization diffraction peak centered at 2Ө of 47° for the Calcite (CaCO₃ PDF No.00-003-0569) vanished with an increase in the Hematite (Fe₂O₃, PDF No. 00-024-0072) peak at 35.56°. Indicating some degree of physical and chemical interaction between the raw-slag and HNO₃, as well as a lattice change in crystallinity after heat treatment at 1,000 °C. As a result, the bonding strength of raw-slag was increased, which corresponds with the FTIR and EDX results (Ali et al. 2020; Safo et al. 2022a).

The concentration of PC remained constant throughout the time, demonstrating that PC is non-degradable under visible light without a catalyst (Feng et al. 2013). The result in Figure 5 shows the degradation ability of NCSS over a time range of 10–60 min at different pH. The degradation was very low at the lower time of 10 but kept rising from 10 to 60 min time interval. This can be attributed to the transfer of photons from the light to the active site of the NCSS, which was very slow at the initial time, but as the time increased, more photons were transferred to the NCSS surface, which generated an electron hole capable of undergoing a redox reaction to breakdown the structure of the molecules of PC (Zia & Riaz 2020). Moreover, Figure 5 also shows that the photodegradation efficiencies of PC were 24.5, 41.2, 68.3, 89.2, and 97.92% at a pH of 12, 10, 7, 5, and 3, respectively, at a time of 40 min. The difference in the photodegradation efficiency at varying pH is associated with the nature of the protonation and deprotonation of the NCSS catalyst and PC molecules in a solution. At a pH of 3, the surface of NCSS is positively charged, which adsorbs and degrades more of PC (Alalm et al. 2015). Moreover, at high pH, there is oxidation of the ions in the hematite content into iron hydroxides, which act as a scavenger for hydroxyl radicals which prevent the degradation of the molecules of PC (Yang et al. 2009).

It has been discovered that PC photodegradation using NCSS is equivalent to the performance of other catalysts previously utilized in photocatalytic degradation reactions. For instance, after 60 min of using Ag/Ag₂O-PTh catalyst, the highest degradation of PC was 98% (Zia & Riaz 2020). In another work, the photodegradation efficiency of PC after 3.5 h of oxidation using TiO₂/cellulosic fiber was only 90% (Yanyan et al. 2017). Moreover, the PC photodegradation achieved by Au/TiO₂ composites at 180 min was 68.8% (Nasr et al. 2019). In addition, the highest PC photodegradation achieved by TiO₂-graphite composite after 180 min was 88% (Vaiano et al. 2018).

The high values of R² (98.58%) and adjusted R² (95.39%), as well as the standard deviation of 2.13 and the little discrepancy between predicted and actual PC photodegradation efficiency at a pH of 3, provide evidence that the model was suitable for the application, as shown in Table 3.

In order to determine whether or not the model is significant, an ANOVA test for PC photodegradation efficiency was carried out, and the results are shown in Table 4. According to the F-values and p-values, the NCSS catalyst dosage and the irradiation time had a more substantial impact on the photodegradation efficiency of PC, while the PC had the least influence. Moreover, the low P-value indicates the significance of the prediction model.

According to the findings in Figure 6(b), a rise in the PC at the beginning of the experiment had an adverse effect on the amount of PC that was successfully degraded. The increased number of PC molecules that get adsorbed on the surface of NCSS in the presence of increasing concentrations of PC causes a reduction in the number of active sites, which in turn causes a drop in the photodegradation efficiency (Alalm et al. 2015). The reaction interaction effect between NCSS dosage and PC can be further described according to the analysis of variance in Table 4. There was a significant relationship between NCSS dosage and PC. This relationship is also depicted in the plot of 3D and contour in Figure 6(c), where increasing NCSS dosage increased degradation efficiency at low PC concentrations.

A numerical optimization study was conducted where the optimum operating values were 56.395 min, 48.96 mg/L, and 10.76 ppm for irradiation time, NCSS catalyst dosage, and PC, respectively, as shown in Figure 7 with a photodegradation efficiency of 96.96%.

By continuously producing and radicals to undergo oxidation reactions with PC molecules, there will be a complete convention of transformation products into CO₂ and H₂O.

The photodegradation efficiencies of the repeating cycle were 96.23, 94.13, 92.78, 89.33, and 87.24%, respectively. The minor reduction in photodegradation efficiencies in subsequent runs was caused by the restricted amount of particles exiting the plate due to the magnet's quick movement. Furthermore, in each run, the PC molecules may cover the active sites of the coated NCSS catalyst particles. The reduction in active sites led to a drop in the photodegradation efficiency in the last two repeating cycles. However, the high photodegradation efficiencies in the five repeating cycles confirm the stability and reusability of the NCSS catalyst for both coated and suspended modes, making it a viable catalyst for industrial wastewater treatment.

In this work, a raw waste slag was synthesized successfully and utilized for the photodegradation of PC in an immobilized mode. The characterization of raw-slag and the produced NCSS catalyst demonstrated that the produced catalyst is stable and can function as a nanocomposite with different crystalline peaks. NCSS achieved complete degradation of PC molecules after 60 min at a pH of 3. The photodegradation was optimized using RSM, resulting in 96.96% degradation efficiency at a pH of 3, an NCSS dosage of 48.96 mg/L, PC concentration of 10.77 ppm, and a time of 56.4 min. Hydroxyl free radicals were observed to be the most deriving reactive oxidation species for the degradation process. The PC degradation efficiency in five repeating cycles at the optimum parameters were 96.23, 94.13, 92.78, 89.33, and 87.24%, respectively. This result indicates that the final effluent of the pharmaceutical wastewater could be treated with an NCSS catalyst produced from slag waste before discharge into water bodies to reduce the risk posed to humans and the environment.

The authors want to express their appreciation to the Japan International Cooperation Agency (JICA) and Egypt-Japan University of Science and Technology (E-JUST) for providing financial and technical support, respectively. Additionally, the first author, K.S., wishes to thank TICAD 7 for the scholarship opportunity for the master's in chemical and Petrochemical Engineering and also the support of Mavis Berko, Elsa Odira and Abigail Oteng.

K. S. conceptualized the whole article, developed the methodology, validated the article, conducted a formal analysis, investigated the article, rendered support in data curation, wrote the original draft, wrote the-review and edited the article. H. N. conceptualized the article, rendered support in data curation, wrote the review and edited the article, and supervised the work. M. M. rendered support in data curation, wrote the review and edited the article, and supervised the work. H. N. rendered support in data curation, wrote the review and edited the article, and supervised the work. A. H. E. rendered support in data curation, wrote the review and edited the article, and supervised the work.

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

The authors declare there is no conflict.