In our previous work (Ponce et al. 2022) a re-utilizable low-cost eco-friendly material based on TiO2 and agarose was optimized for the degradation of Methylene Blue and Rhodamine B pigments. Taking the obtained conclusions into account, the photocatalytic agent was shaped as Spheres/ellipsoids using 5% wt. agarose solutions aiming to degrade the two drugs objective of this study.
The first results indicated good degradation rates for both drugs, better for A90 (PRM C/Co: 0.29 and IBU C/Co: 0.24) than for A60 pieces (PRM C/Co: 0.39 and IBU C/Co: 0.32) after a one-week lasting experiment. However, a white turbidity was progressively generated during the experiment, more intensively when ibuprofen solutions were degraded in the presence of the highly loaded composites (A90). The precipitate deposited was identified by X-ray diffraction (data not shown) as anatase, the photocatalytic agent introduced in the beads. Thermogravimetry analysis was employed to try to quantify the amount lost, the examination of the beads used in degradation experiments showed that less than a 5% of the ceramic component had been lost. Despite this scarce percentage, different strategies were proposed to minimize the TiO2 leakage by reinforcing the gelling matrix (Fig. 1). The first idea consisted on increasing the agarose CONTENT to try to entrap more efficiently the ceramic particles. However, the leakage did not completely disappear and the fabrication of the pieces was hindered due to the higher viscosity of the agarose gels. The COATING or BLENDING with chitosan, alginate or gelatin did not eliminate completely the turbidity generation and complicated the beads manufacture. Fortunately, the use of citric acid to CROSSLINK the agarose resulted in less permeable matrixes that did not allow the TiO2 particles migration as deduced from the absence of turbidity. Figure 2 shows the degradation results obtained for crosslinked vs. non treated materials; in addition, freshly prepared samples, treated with citric ac., were, as well, included in this preliminary photocatalytic characterization. These results confirmed a superior photocatalytic efficiency for those samples containing the higher ceramic percentage (A90), and shoed very similar degradation rates for the different series considered: non-treated lyophilized, and crosslinked lyophilized and non-lyophilized. The employment of beads without a drying preservation method had not been included among the strategies depicted in Fig. 1 but it was considered taking into account the possible alteration of the matrix integrity during the freeze-drying and rehydration process.
The use of hydrated beads, i.e. without being preserved through a drying procedure, was considered as a plausible and practical alternative once that it was observed that they could be easily stored in water during years. These materials did not evidence any shape alterations, morphology modifications or ceramic particles leakage and maintained the photocatalytic efficiency. This work focuses on the use of this type of “hydrated or fresh” spheres thus avoiding the requirement of a drying step by means of a freeze-drying infrastructure, not always available. Indeed, following the objective of this research of minimizing the complexity of the synthetic and fabrication routes, aiming to facilitate the scale-up of this technology, the citric acid treatment was productively replicated under sunlight in a summer day in a greenhouse. In this sense, the alternative gelling agents: gellan, agargel™ and agar were chosen, besides their aggregation capacity, as a function of their renewal origin as well as, in the last case, lower cost and higher availability.
Manipulable and stable beads containing the two already studied percentages (60, 90) were successfully prepared by employing the four gelling agents. All these materials show the same macroscopic aspect exposed in the inset of Fig. 2. Once crosslinked, the FTIR characterization allows to confirm the formation of an ester bond as evidenced, more easily in the absence of the ceramic component, by the band observed at 1720 cm− 1 (Fig. 3) (Awadhiya et al. 2016, Mali et al. 2017, Ponnusamy et al. 2022, Simões et al. 2020). Despite the simplicity of this procedure: low temperatures and the absence of activators, it ensures a sufficient crosslinking to support the aggressive media generated during the drug degradation experiments. In addition, it can be, even, applied out of a laboratory contrastingly to other more effective and controlled routes. Moreover, citric acid has been termed as a “green” cross-linker due to its non-toxic nature, readily availability and inexpensively that has been successfully employed in the biomaterials field (Moon et al. 2019, Salihu et al. 2021). A scheme of the agarose crosslinking due to the presence of the citric acid has been depicted in Fig. 1.
A potential concern may arise about how this treatment may affect to the morphology and microstructure of these composites. The characterization techniques employed (SEM, TG-DTA and MIP) allows to conclude that these materials, independently on the gelling agent used, can still be described as a hydrogel matrix in which the TiO2 particles are completely embedded, even for the higher ceramic loads. The polysaccharide chains, whose mobility has been restricted due to the crosslinking, do not only entrap the anatase particles but also generates a pore architecture that allow a fluids migration throughout the materials that ensure the intimate contact between the dissolved drugs and the photocatalytic agent. This structure can be envisaged by means of the Mercury Intrusion Porosimetry (MIP), depicted in Fig. 4, that shows two different assemblies depending on the amount of ceramic included. Independently on the gelling agent, the highly loaded spheres (90%) showed an almost monodispersal pore distribution centered between 0,1 − 0,05 µm that can be attributed to the space left among the ceramic particles. On the contrary, in those samples with less amount of TiO2, the role of the hydrogel component increases leading to the generation of a porosity, around 100–200 µm when freeze-drying is applied, which otherwise would be occupied by water. This much opened structure resulted more effective to degrade the dyes studied in our previous work (Ponce et al. 2022) but, taking into account the degradation results already mentioned and those described hereinafter, the amount of photocatalytic agent is more critical in this case. Figure 5, that depicts the SEM micrographs, show scarce differences between the different gelling agents (a-d), all showing the characteristic honeycomb structure based on the 100–200 µm pore distribution already detected by MIP. This figure demonstrates, as well, the perfect integration of the ceramic particles within the agarose and agar polysaccharides as can be observed, even, within the thin walls that compose the cells of the honeycomb (e,f). In order to discern whether this embedding may affect to the catalytic surface exposed, N2 adsorption measurements were carried out. As shown in Fig. 6, very similar isotherms and specific surface area values (7 m2/g) can be observed for the TiO2 powder and the A60 and A90 composites.
The thermogravimetric characterization allows, by examining the amount of the ceramic residue that remains after calcining at 600oC, to confirm the scarce differences of the prepared materials, in terms of composition, when compared to the theoretical values shown in table 1. Moreover, Fig. 7a depicts the thermal decomposition of lyophilized samples as well as those hydrated; the last ones experiment a significant weight loss around 100oC, and a second one more extended, in terms of temperature, that can be ascribed to water and agarose elimination, respectively. The hydrogel calcination can be more precisely outlined in dried samples; the agarose is completely removed at 450oC thus allowing to calculate the organic (polysaccharide) and inorganic (TiO2) percentages. In addition, the detection, by means of the Differential Thermal analysis, of endo and exothermic transitions during the polysaccharide’s elimination allows to deduce modifications attributable to chemical changes after the crosslinking, interaction with the titanium oxide or spatial distribution of the organic component among the ceramic particles. The curves observed after the crosslinking treatment are characterized by a unique endothermic peak around 400-450oC that supposes a slight temperature displacement (agarose, gellan) or a simplification on the decomposition pattern (agar, agargel™). However, much considerable changes are detected due to the interaction with the ceramic: decomposition patterns with less intense peaks located at 300 and 400oC. Finally, Fig. 7b includes, as well, the curves of samples employed in a degradation experiment under light (A90 L) or dark (A90 D) conditions; the null changes observed when compared with the as prepared sample (A90) reveals that the interaction with the drugs do not alter the photocatalytic composites.
In order to test the degradation capacity, as well as the possibility of reutilizing several times the photocatalytic beads, an experiment, consisting on exposing these materials to three cycles in which the ibuprofen or paracetamol solution was renewed, was carried out. The already used spheres were rinsed in deionized water before initiating a new cycle. Figure 8, which depicts the UV spectra at the different times considered in each cycle, demonstrates a progressive diminution on the intensity of the most characteristic bands, including those employed to quantify the degradation of paracetamol (245 nm) and ibuprofen (222 nm) The degradation results depicted in Figs. 9 (paracetamol) and 10 (ibuprofen) permit to draw the following tendencies: 1) Agar and agarose manifest the best results when compared with gellan or agargel 2) Higher ceramic contents ensure better catalyzation 3) Similar degradation rates are observed for both drugs only differentiating in the faster initial catalysis followed by a much-sustained step of ibuprofen while paracetamol is destroyed more progressively in a linear way 4) Degradation rates slightly increase with the recycling cycles.
The observed degradation percentages reach, in the most efficient experiments (agarose and agar with the highest ceramic amount), up to 70–75% of drug destruction in the first series of essays. This degradation increases in a 5% after a new experiment reusing the photocatalytic beads thus demonstrating a stability and regenerating capacity that has been scarcely described for photocatalytic materials used in drug treatment (Choina et al. 2014, Jallouli et al. 2017, Khedr et al. 2019). Lower destruction rates were obtained when gellan alone or blended with agar (Agargel™) were employed. Despite these slightly lower results these gelling agents have demonstrated the capacity to obtain durable and stable materials with similar microstructure to those containing agar or agarose. In fact, gellan containing TiO2 particles has been proposed to be applied as wound dressing as well as for diverse cleaning and disinfecting purposes (De Filpo et al. 2015, Ismail et al. 2019, Razali et al. 2020). The excellent results obtained with agar supposes a cheaper and even more ecofriendly (no extraction process of agarose is required) alternative to the use of this polysaccharide as a gelling agent.
Concerning the role of the ceramic content, it can be clearly observed that any of drugs degradation improves with the amount of anatase, around a 10% when comparing samples containing a 60% in the solid content to the equivalent inclosing a 90%. Conversely, in the previously studied dye degradation, lower ceramic percentages, which ensures a higher porosity and a faster and more facile interaction with the surrounding media, yield higher degradation rates. This difference can be attributed to the slower degradation rate observed for the drugs essayed when compared to the pigments studied (methylene blue and rhodamine). As it has been demonstrated by means of the texture and porosity characterization (Figs. 4, 5), the drug-polluted media reaches in no time the photocatalytic particles whose exposed surface is not altered by the shaping action of the gelling agent. At this point the higher amount of titanium dioxide interacting with the drugs to be degraded explains these higher depletion percentages.
The degradation rates, very comparable for both drugs but slightly higher for ibuprofen, can be explained considering the analogous structure based on an aromatic ring that undergoes an initial hydroxylization of the side chain groups followed by the ring opening which finally gives out CO2 and H2O. Both degradation mechanisms, initiated by the HO• radicals formed by the photogenerated e− and h+ pairs when the anatase is illuminated by solar light, have been extensively discussed in the literature: the different intermediates formed during the paracetamol (Moctezuma et al. 2012, Phong Vo et al. 2019, Yang et al. 2008, Zhang et al. 2008) and ibuprofen (Brillas 2022, Choina et al. 2014, Ferrando-Climent et al. 2012, Jiménez-Salcedo et al. 2019, Miranda et al. 2021) destruction explain the dissimilar degradation kinetics by which the initially ibuprofen fast catalyzation is followed by a much steady step that equals the results obtained for paracetamol.
Taking these results into account, only agar and agarose-based beads were employed in the experiment designed (higher volumes of tap water and exposition to sunlight) to approach more closely to the conditions employed in actual decontamination processes. The so obtained results, depicted in Fig. 11, show analogous tendencies to those observed in the indoor essays: similar degradation rates for both drugs and gelling agents as well as greater for beads containing the highest amount of ceramic. The only difference lies in the faintly better destruction percentages that almost reach a 90% under these conditions.