BiOI/Magnetic Nanocomposites Derived from Mine Tailings for Photocatalytic Degradation of Phenolic Compounds (Caffeic Acid) in Winery Wastewater

Abstract

The development of advanced photocatalysts that are efficient, recyclable and sustainable represents a significant challenge in the face of the growing presence of persistent organic contaminants in industrial wastewaters. This paper presents a novel approach based on the design of new heterostructures synthesized from BiOI and magnetic materials, using not only synthetic magnetite, but also magnetic compounds extracted from mine tailings, transforming environmental liabilities in active supporting materials through valorization strategies in line with the circular economy. Through precise control of composition, it was established that a proportion of 6% by mass of the magnetic phase allows the formation of a heterostructure that is highly photocatalytically efficient. These compounds were evaluated using caffeic acid, an organic contaminant of agroindustrial origin, as a target compound. Experiments were carried out under simulated solar radiation for 120 min. Among the materials synthesized, the BiOI/MMA heterostructure, derived from industrial tailing A, displayed an outstanding photodegradation efficiency of over 89.4 ± 0.25%, attributed to an effective separation of photoinduced charges, a broad active surface and a synergic interface interaction between its constituent phases. Furthermore, BiOI/MMA exhibited excellent structural stability and magnetic recovery capacity, which allowed for its reuse through two consecutive cycles without any significant losses to its photocatalytic performance. Thus, this study constitutes a significant contribution to the design of functional photocatalysts derived from industrial tailings, thus promoting clean, technological solutions for the treatment of wastewater and reinforcing the link between environmental remediation and circular economy.

1. Introduction

The growing concern over the pollution of water resources has prompted a strong international response through the development of novel, sustainable technologies for the effective treatment of industrial wastewaters, particularly from the pharmaceutical, textile, food, and wine sectors [1,2]. Among these non-conventional technologies, Advanced Oxidation Processes (AOPs) stand out due to their ability to generate highly reactive species, enabling the degradation of persistent organic pollutants [3,4]. Within this category, heterogeneous photocatalysis has gained prominence for its high efficiency, low environmental impact, and operation under mild conditions such as the use of solar radiation [2,5].

HP employs the use of photoactive semiconductors, with bismuth oxyiodide (BiOI) being one of the most promising ones, as several studies have shown its high photocatalytic efficiency under both the visible spectrum as well as simulated solar radiation [6]. It is also worth mentioning that BiOI has a ~1.95 eV band gap, which allows it to use up to 95% of solar radiation input [7]. These properties have enabled this compound to become one of the most used photocatalysts for HP in recent years. However, despite its high efficiency in degradation processes such as the mineralization of organic compounds, the nanometric size of BiOI makes its separation and subsequent recovery from the aqueous phase difficult after treatment processes of wastewater, which in turn makes it difficult to use in large-scale processes [8].

In order to circumvent this issue, it becomes necessary to fabricate heterostructures that allow for a complete and efficient separation of the photocatalyst from the aqueous phase, by the use of, as an example, external magnetic fields in the form of magnets [9].

The concept of a heterostructure references a hybrid material wherein two or more semiconductors, with different electronic and structural properties, are intimately combined, generating interphases that facilitate the separation of photoinduced charges and improve photocatalytic activity [10,11]. Although the literature also employs the terms “heterojunction” and “composite”, the term heterostructure will be uniformly employed throughout this manuscript to describe the nature of the synthesized materials and their photocatalytic performance in a more precise manner.

This strategy not only allows for an easier recovery and recycling of the material, but also contributes to decreasing the recombination rate of electron–hole pairs thanks to the effective coupling between the components in the system, which in turn generates a stable heterostructure. These heterostructures induce internal electric fields, which favors the separation of charge carriers, thus significantly increasing the photocatalyst’s efficiency [10].

In this regard, magnetite (Fe₃O₄) has been reported in several studies [11,12,13] as an ideal candidate to act as a magnetic support for BiOI. The reasoning is that magnetite is an n-type semiconductor with an inverse spinel structure, which is highly valued in the context of green chemistry and heterogeneous catalysis [14]. Although its photocatalytic efficiency is limited, magnetite has a high photochemical stability, and more importantly, excellent magnetic properties, which make it easy to retrieve from an aqueous phase through an external magnet, which is paramount for applications where the goal is to retrieve and recycle the catalyst. Moreover, magnetite can either be artificially synthesized in the lab or found naturally in mineral deposits [15].

The formation of a three dimensional p–n-type heterostructure between the bismuth oxyiodide (BiOI) and magnetite (Fe₃O₄) has shown to be highly effective in increasing the photocatalytic performance under UV–Visible radiation. This structure favors a directional transference between the photoinduced charge carriers, where the electrons originating in the magnetite are promoted to its conduction band and subsequently transferred to the conduction band of the BiOI. At the same time, the holes formed on the BiOI remain on the valence band, which enables an efficient separation between charges and significantly reduces electron–hole recombination, a critical factor in the photocatalytic efficiency of the process [16,17].

The synergy between both semiconductors not only widens their absorption on the UV–Visible range, but also improves the generation of reactive oxygen species (ROS). The presence of dissolved oxygen, in particular, facilitates the formation of superoxide radicals (•O₂⁻), which play a crucial role in the advanced oxidation of organic contaminants, such as caffeic acid, promoting their degradation and eventual mineralization [18,19].

Recent studies show that composite materials based on BiOI and magnetite possess excellent structural stability and can function repeatedly after multiple cycles of usage. Specifically, these materials can be reused efficiently through at least five consecutive photocatalysis cycles without any significant decrease in their activity, which highlights their potential as sustainable photocatalysts in the treatment of wastewater. This particular feature is enhanced by the magnetic properties of magnetite, which allow for a fast and efficient recovery of the material through the use of an external magnetic field [20,21].

Furthermore, both the morphology and composition of the photocatalyst have a significant effect on the structural auto-organization of the system, influencing the formation of hierarchical architectures such as microspheres or lamellar aggregates. These morphological aspects have direct implications on the efficiency of charge transport and accessibility of active sites during the photocatalytic process [22,23].

Although there are several pathways to the synthesis of the BiOI/Fe₃O₄ heterostructure, the solvothermal method is one of the most commonly used as it yields materials with high crystallinity, phase purity and morphological control. This method, usually classified as a solvothermally assisted sol–gel, uses ethylene glycol as a solvent, and precursors such as bismuth (III) nitrate pentahydrate (Bi(NO₃)₃·5H₂O) and potassium iodide (KI) are added in stoichiometric amounts [21,24,25]. However, this technique has significant limitations from the sustainability viewpoint, including the use of toxic organic solvents and the necessity of high temperatures and pressures during long reaction times, thus demanding a high energy input [26,27].

In response to these limitations, assisted co-precipitation has been proposed as a synthetic pathway that is more sustainable and efficient. This technique yields composite materials at lower temperatures, without the need of pressurized equipment, reducing energy costs as well as minimizing the usage of dangerous solvents. Furthermore, it prevents interference in later stages of the process, like in evaluating the mineralization of contaminants, which is a crucial advantage for environmental applications [28,29,30]. Co-precipitation, on top of being more environmentally friendly, can be escalated and allows for fine tuning on stoichiometry and morphology through the adjustment of certain parameters such as pH, reactive addition speed, and the Fe/Bi ratio [22,31].

On the other hand, the usage of magnetic material recovered from mine tailings represents an innovative strategy with a high potential in regard to the circular economy and environmental remediation. This not only offers a sustainable alternative to synthesize photocatalysts, but it also promotes giving value to environmental liabilities produced by the extractive mining industry. Each year the mining industry, particularly in the case of copper mining, produces a large volume of solid waste known as tailings as a result of the flotation and concentration processes that are unable to completely extract the target metals. The resulting tailings are a complex matrix of waste minerals, heavy metals, excess chemical reagents and water, becoming a significant environmental hazard due to their toxicity, physicochemical stability and large volume [32,33]. Therefore, mine tailings correspond to a complex matrix which, besides containing usable iron oxides, may also have heavy metals such as copper, lead and arsenic, as well residual reactants used during the flotation process. These compounds represent a relevant environmental hazard, as their release can negatively impact the quality of soils and surrounding water [34]. In this sense, any strategy that aims to valorize tailings must not only consider the potential exploitation of their useful mineral phases, but also keep in mind the risks associated with the presence of toxic elements. Under this framework, the recovery of the magnetic fraction as a functional support offers an attractive alternative, as long as it is accompanied by proper cleaning and characterization procedures in order to guarantee environmental safety.

The prolonged storage of this waste requires extensive land spaces and constitutes a latent threat for the surrounding ecosystems, especially in arid and semiarid regions such as northern Chile, where long term evaporation increases the concentration of hazardous substances within these deposits [35,36]. Chile—the lead copper producer in the world—currently possesses over 740 tailing deposits, out of which approximately 65.8% are located in the Atacama and Coquimbo regions, meaning there is a total of 487 of these deposits in these regions alone [37,38].

Iron in amongst the most abundant elements that can be found in this type of waste, most commonly in the form of various iron oxides such as hematites (α-Fe₂O₂), magnetite (Fe₃O₄) and goethite (α-FeOOH). These compounds could potentially be recovered and transformed into magnetic materials, creating an opportunity to increase their value as functional tools for environmental technologies [39,40,41]. Such materials can be easily functionalized or integrated into semiconductor heterostructures applied to advanced decontamination processes, such as heterogeneous photocatalysis or heterogeneous photo-Fenton, with advantages such as easy recovery, structural stability and activity under solar and visible radiation [42,43].

In this context, the following paper proposes a strategy to increase the value of mine tailings as functional support in the synthesis of photocatalytic composite materials through the integration of iron oxides extracted from such tailings as a magnetic phase to be used in BiOI/magnetic material-type heterostructures. Such an application would not only promote a circular economy in the mining industry, but it would also make the recovery of the photocatalyst much easier via magnetic separation, which is a significant advantage regarding operating conditions in the treatment of wastewater through HP.

To this end, magnetic material was recovered from two tailing deposits found in Chile’s Coquimbo Region, known as “El Culebrón” and “Dina”, both of which possess high iron contents, primarily in the form of magnetite and hematite according to past geometallurgic studies [39]. The first step in this work involved the optimization of the proportion of synthetic magnetite in the synthesis of the BiOI/Fe₃O₄ heterostructure using the co-precipitation method, determining a 6% mass content of Fe₃O₄ in respect to BiOI. This ratio provided the highest photocatalytic efficiency in the degradation of phenolic compound caffeic acid under simulated solar radiation, whilst also minimizing charge recombination and favoring magnetic recovery.

With these results, the synthesis was repeated using the same mass percentage (6%) of magnetic material but substituting the synthetic magnetite with the magnetic material extracted from the mine tailings. This resulted in the synthesis of two new structures, named BiOI/MMA and BiOI/MMB, corresponding to the magnetic material recovered from “El Culebrón” (MMA) and “Dina” (MMB), respectively. Both materials were comprehensively characterized using structural, morphological, and photocatalytic analysis techniques to evaluate their performance in the degradation of the target contaminant, caffeic acid (CA), in aqueous solution.

Selecting CA as a target in this photocatalytic performance study is justified by CA’s environmental relevance and chemical behavior. This phenolic compound is commonly found in wastewater originating from agroindustrial activity, particularly in effluents from wine distilleries, olive oil plants, and general industries that process plant-based biomass that is rich in polyphenols [44,45,46].

From an environmental standpoint, CA has been classified as a persistent contaminant, given its high chemical stability, low biodegradability, and resistance to conventional biological treatments [47]. Its presence in wastewater may negatively impact the microbiological activity of biological treatment systems, inhibiting processes such as aerobic and anaerobic digestion, which is a considerable challenge for traditional treatment technologies [48,49].

2. Results and Discussion

2.1. Determining the Optimal Ratio of Magnetite in the BiOI/Magnetite (Fe₃O₄) Heterostructure

With the objective of determining the optimal ratio of magnetite within the BiOI/Fe₃O₄ heterostructure that allows for the maximum photocatalytic efficiency in the degradation of caffeic acid, different materials were synthesized, each with a different mass percentage of magnetite: 2, 4, 6 and 8%. After initial testing, considering the materials with 4% and 6% magnetite presented high degradation percentages, it became necessary to synthesize additional materials containing 5% magnetite, with the goal of fine-tuning the optimal amount of magnetite in this particular interval.

Figure 1 shows the degradation percentages achieved by each of the heterostructures 30 min into the photocatalytic reaction. Although each experiment was extended to 60 min, kinetic profiles indicate the existence of two clearly differentiated regimes. CA concentration decreases considerably and peaks during the first 30 min, with degradation slowing down past this mark, tending to stabilize by the end of the experiment at 60 min.

On the 30 min mark, the BiOI/Fe₃O₄ possessing 5% magnetite achieved a CA degradation of 71.5%, slightly superior to the 6% (71.2%) and 4% (68.2%) materials. However, at the end of the 60 min experiment, the material with 6% magnetite had the highest photocatalytic performance, degrading 79.9% of the target chemical, compared to the 5% (77.2%) and 4% (73.4%) formulations. This behavior suggests that, although a smaller amount of magnetite favors fast kinetics at the beginning, a slightly higher amount of magnetite helps maintain the efficiency of the process during longer reaction times. Such behavior also agrees with previous studies of BiOI/Fe₃O₄-type heterostructures, where the optimal composition for such a structure has been described to lie between 5 and 6% mass of magnetite, wherein photocatalytic efficiency is maximized as effective charge separation light absorption, and available active surfaces are at an equilibrium [49,50,51]. Studies on core/shell-type structures and tridimensional heterostructures have demonstrated that a moderate proportion of Fe₃O₄ enables high degradation percentages of dyes and phenolic compounds under visible light, whilst maintaining the material’s recyclability and operational stability [52].

Figure 1 also shows that progressively increasing the magnetite content up to 6% translates to an increase in photocatalytic efficiency. However, when increasing the magnetite content up to 8%, a significant decrease in photocatalytic efficiency can be observed, which indicates that increasing the magnetite ratio beyond the optimal range compromises performance.

Such a decrease in photocatalytic efficiency observed in the materials with high magnetite content can be explained by different factors. Excess magnetite might induce an optical shielding effect, partially blocking light absorption from the BiOI, and thus reducing the generation of photoinduced electron–hole pairs [49]. On the other hand, excessive surface covering of BiOI by magnetite particles may decrease the active surface available to adsorb both the contaminant and redox reactions, aside from interfering with the interfacial charge transfer, having a direct effect on the efficient separation of charge carriers [50]. Additionally, although magnetite could ease charge-separate by acting as an electron trap, an excess of it may transform into a recombination center instead, thus reducing the efficiency of the photocatalytic process [52,53,54].

Therefore, after evaluating the performance of the materials with different mass percentages of magnetite (2–8%), the optimal amount of magnetite in the BiOI/Fe₃O₄ heterostructure was determined to be 6%, as this composition achieves the ideal synergy between light absorption, charge separation and surface transfer efficiency. Going beyond this ratio brings forth counterproductive effects that decrease the overall photocatalytic performance.

Additionally, all data obtained confirms that the BiOI/Fe₃O₄ materials that were synthesized in this study offer a significant improvement on the photocatalytic activity towards CA compared to the standalone components of the heterostructure, which highlights the synergic effect derived from the formation of this structure.

2.2. Photocatalytic Evaluation of the BiOI/Magnetic Material (MM) Heterostructures

Once the optimal proportion of magnetite for a high-photocatalytic-efficiency BiOI/Fe₃O₄ heterostructure was determined, this being 6% by weight, the synthesis of the BiOI/MM was carried out using the magnetic material extracted from tailings A and B in a 6% ratio, and under the same experimental conditions used in the synthesis of the BiOI/Fe₃O₄ heterostructure.

Figure 2a shows the evolution of the relative concentration (C/C₀) as a function of time, while Table S1 shows the degradation percentages of caffeic acid achieved by the heterostructures synthesized with the magnetic materials obtained from tailings A and B, respectively.

It is important to highlight that the degradation percentages reported after 30 min correspond to the initial stage of the photocatalytic reaction, characterized by higher reaction rates. Kinetic profiling for the degradation of caffeic acid using the synthesized material evidenced two clearly defined regimes: first, a fast drop in concentration during the first 30 min, and a subsequent phase, where degradation slows down and tends to stabilize at around 60 min. The choice of this intermediate point allows for a more accurate comparison between the initial efficiencies of each heterostructure, as well as evaluating the effect of the magnetic material on the process kinetics.

A first-order kinetic model was also applied to evaluate the photocatalytic performance of the BiOI-based systems, according to Equation (1), where C₀ is the initial concentration of the contaminant after the dark adsorption step, C is the concentration at reaction time t, and k is the apparent rate constant (min⁻¹).

Ln (C₀/C) = k·t

(1)

The obtained k values from the linear regression of the experimental data (Figure 2a) were 0.027, 0.022 and 0.017 min⁻¹ for BiOI/Fe₃O₄ (6%), BiOI/MMA and BiOI/MMB. These results confirm that the incorporation of the magnetic phase significantly improves degradation kinetics the BiOI/Fe₃O₄ (6%) system being the most efficient, closely followed by the BiOI/MMA heterostructure. The correlation coefficients (R² > 0.95) indicate that the pseudo-first-order model satisfactorily describes the kinetic behavior of the degradation process within the studied time interval.

According to the data in Table S1, the BiOI/MM obtained from tailing A achieved a photocatalytic efficiency of 75.1% after 60 min, surpassing the BiOI/MMB, which only achieved 68.8 ± 0.5% under the same timeframe. It is important to highlight that the performance demonstrated by the BiOI/MMA is comparable to that of the BiOI/Fe₃O₄ (6%) material that was previously obtained with synthetic magnetite, which achieved 79.9 ± 1.1% and 94.3 ± 0.80% after 60 and 120 min, respectively, which points to MMA being a potential functional substitute of synthetic magnetite in photocatalytic degradation processes using this type of heterostructure.

Figure 2b shows the degradation percentages of CA after 60 min under simulated solar radiation, using the individual components (BiOI, Fe₃O₄) and optimized BiOI/Fe₃O₄ (6%), BiOI/MMA and BiOI/MMB.

Standalone BiOI displayed a degradation efficiency of 64.4%, while synthetic magnetite presented very limited activity on its own, achieving a degradation percentage of only 11.3%, which confirms its low photocatalytic efficiency when used in isolation. However, the 6% BiOI/Fe₃O₄ material shows a significant improvement in comparison, achieving a degradation percentage of 79.9%, evidencing the synergic effect between both compounds.

Moreover, the magnetic materials extracted from mine tailings presented an even more limited photocatalytic activity when tested in isolation: MMA achieved a 29.9% while MMB barely made it to 15.0%, which suggests there are differences in their mineralogical composition or in their adsorption and charge transfer capabilities. However, by forming heterostructures with BiOI, a substantial improvement on their photocatalytic efficiency was observed: BiOI/MMA achieved a 75.1% while BiOI/MMB a 68.8%. These results show that, although magnetic material derived from tailings have lower intrinsic activity compared to synthetic magnetite, combining them with BiOI allows the formation of active heterostructures, particularly in the case of tailing A, which performed similarly to the BiOI/Fe₃O₄ (6%) system.

Together, these results demonstrate the potential in valorizing mining waste as a source of functional materials in photocatalytic applications, and highlight the importance of properly characterizing the composition of the magnetic material extracted from them in order to optimize their performance.

2.3. Characterizing Materials

The techniques used in the characterization of the different materials involved in this study allowed for a deeper understanding of the relationship between their structure and functional behavior. Results obtained from these techniques become particularly useful to explain the differences observed between the photocatalytic efficiencies of the synthetic materials made with magnetite and those made from secondary sources such as mine tailings.

The structural characterization of the materials in this study was performed using X-ray diffraction, and their patterns are shown in Figure 3a. Standalone BiOI presents sharp, well-defined peaks, typical of its tetragonal phase, in accordance with JCPDS card No. 10-0445 (ICDD, 2001a), which evidences its high crystallinity. Synthetic magnetite shows intense peaks, typical of a cubic spinel phase (Fe₃O₄), which corresponds to the JCPDS card No. 19-0629 (ICDD, 2001b), especially at 2θ ≈ 30.1°, 35.5°, 43.1°, 53.4°, 57.0° and 62.6°, confirming the effective formation of iron oxide. Magnetic material obtained from mine tailings (MMA and MMB) showed signals that coincide with those of magnetite, particularly for MMA, as its peaks are more intense, suggesting a higher crystallinity compared to MMB. Moreover, MMB presents peaks attributed to common residual phases found in mine tailings such as quartz (JCPDS No. 46-1045; ICDD, 2001c) and hematite (JCPDS No. 33-0664; ICDD, 2001d). The diffractograms of the BiOI/MMA and BiOI/MMB heterostructures show the coexistence of the constituent phases without evidence of the formation of new secondary phases, which indicates an effective structural integration. BiOI peaks remain evident in the new materials (for example, in 2θ ≈ 29.5°, 31.7°, 45.5°). Magnetite peaks are also present. Relative intensities of said peaks do not change significantly, which suggests high stability in each phase of the heterostructure. These observations evidence the structural stability of the composite materials and justify their subsequent analysis in regard to their magnetic properties, which will allow for the correlation between their structure with their photocatalytic activity.

With the goal of identifying the crystal phases in the diffractograms, Raman spectroscopy was performed on the BiOI, MMA, MMB, BiOI/MMA and BiOI/MMB samples. Figure 3b shows the normalized Raman spectra of the aforementioned materials in a range of 100 to 800 cm⁻¹.

The Raman spectra for standalone BiOI shows intense, well-defined bands around ~147, ~293 and ~519 cm⁻¹, consistent with the vibration patterns A₁ and E found in tetragonal-structure BiOI in the literature [49]. The lack of peaks attributed to iron oxides supports its high structural purity and is in line with the data obtained from XRD.

In the case of the magnetic material MMA, a dominant band can be found at ~668 cm⁻¹, attributed to the distinctive A₁g pattern of magnetite, as has been widely reported in the literature [55,56]. The absence of peaks associated with BiOI suggests that the sample is predominantly composed of magnetite as a dominant crystal phase, which correlates with the intense magnetite patterns observed in the XRD analysis. The sharpness and intensity of this band suggest low oxidation and a good structural order within the magnetic phase. In contrast, Raman spectra for MMB evidences the coexistence of bands around ~226, ~291 and ~410 cm⁻¹, a distinctive feature of hematite (Fe₃O₄), and a weaker band at ~668 cm⁻¹, corresponding to magnetite (Fe₃O₄). This combination indicates partial oxidation from magnetite to hematite, which could negatively influence the material’s magnetic properties and thus affect its performance as a photocatalytic support [57].

In BiOI/MMA, the Raman spectra show signals that correspond exclusively to BiOI, without the presence of bands associated with magnetic phases. This behavior suggests an effective covering of the magnetic material’s surface with the photocatalyst, which could favor the charge transfer efficiency in the heterogeneous interface by eliminating recombination centers associated with iron oxides.

Conversely, the Raman spectra for BiOI/MMB present bands that can be attributed to both BiOI and hematite, which confirms the formation of a heterostructure between the photocatalyst and the partially oxidized supporting structure. The coexistence of both phases also indicates partial structural integration. However, the presence of hematite could reduce the material’s magnetic properties and compromise the efficiency of separation through external methods such as magnetism, aside from modifying the electronic transfer trajectory during the photocatalytic process. Altogether, these Raman results complement previous structural analyses and contribute key vibrational evidence that allows for comprehension of the relationship between structure, surface composition, and functional interaction of the heterostructures that were synthesized.

Figure 4 shows SEM images representative of the materials synthesized in this study. In Figure 4a, corresponding to BiOI, a tridimensional morphology can be observed, which is characterized by porous microspheres composed of assembled nanoplates. These nanoplates, which are two-dimensional structures, stack and assemble radially, creating hierarchical spheres, which coincides with what has been reported in the literature for BiOI-based materials obtained through assisted precipitation methods [58,59]. This architecture favors a high specific surface area and facilitates charge separation, which are critical aspects in photocatalytic applications. In Figure 4b, magnetite (Fe₃O₄) presents a homogeneous, nanospherical morphology with an average size of around 280 nm. The spheres exhibit a porous surface, which is consistent with what has been previously reported by other authors [60]. Figure 4c, which corresponds to the BiOI/Fe₃O₄ (6%) composite material, reveals hierarchical microspheres similar to those of standalone BiOI but with a denser surface texture. This “ball of yarn”-like arrangement suggests a decrease in surface porosity, possibly due to the partial coalescence of magnetite nanoparticles in the cavities within BiOI, which can impact the accessibility to active sites and affect the adsorption process of contaminants [49].

Figure 4d shows MMB materials extracted from tailing B. Amorphous structures can be identified, with irregular 3D forms and dense aggregates without a clear crystal phase. This morphology suggests a heterogeneous mix of mineral phases, likely dominated by poorly crystallized iron oxides, which would limit its efficiency as a photocatalytic support [61]. Figure 4e, corresponding to the BiOI/MMB heterostructure, evidences the absence of well-defined microspheres, showcasing amorphous domino-like structures, which are lacking the characteristic nanoplate morphology associated with BiOI. This lackluster integration between BiOI and the iron oxides found in MMB could explain the low efficiency observed in the photodegradation experiments, impeding effective charge transfer between both phases [62]. Conversely, Figure 4f, which shows the material MMA obtained from tailing A, while still presenting an irregular morphology, implies a higher crystallinity and purity of the magnetic phase compared to MMB, as MMA particles appear to be finer and follow a more compact, well-knit distribution in comparison. This difference in crystallinity between both materials is also in agreement with the XRD analysis. Lastly, Figure 4g shows the BiOI/MMA heterostructure. Well-formed, tridimensional microspheres can be observed, with a higher surface density of nanoplates compared to standalone BiOI. This structure suggests a good interphase interaction between BiOI and the magnetic phase derived from MMA, which correlates to the higher photocatalytic efficiency presented by this material, by facilitating the transfer of photoinduced charges and enhancing surface adsorption [63].

Figure S1 shows the elemental distribution maps obtained through energy-dispersive spectroscopy (EDS) for the standalone components and their corresponding heterostructures. Figure S1a, corresponding to BiOI, presents a homogeneous distribution of the elements Bi, O, and I, a well-known characteristic of this semiconductor in its tetragonal phase. This homogeneity suggests the successful synthesis and high crystallinity of the material, in agreement with the structural analysis previously discussed.

Figure S1b,c show the EDS analyses for the magnetic materials obtained from mine tailings materials MMA and MMB, respectively. In the case of MMA, intense signals for Fe and O can be observed, alongside the presence of secondary elements such as Al and Si. This composition indicates that MMA is predominantly composed of iron oxides, possibly magnetite (Fe₃O₄), silicate-based minerals that were present in the original matrix. Likewise, Figure S1d MMB also exhibits predominantly Fe and O, with the additional presence of Mg and weaker signals for Si, which suggests a smaller amount of non-magnetic impurities.

The heterostructures synthesized from these materials are presented in Figure S1d,e. The BiOI/MMA heterostructure in Figure S1d reveals a uniform distribution of the elements Bi, O and I over the surface of the magnetic material, which indicates an effective covering of the bismuth compound through the surface of the iron oxides found in MMA. Conversely, for the BiOI/MMB heterostructure in Figure S1e, although it also presents a homogenous distribution of Bi, O, and I, there are specific regions wherein the only element detected is Fe. This suggests that these areas were not completely covered by BiOI, meaning that there are hollows that leave MMB’s surface, rich in iron oxides such as hematites, exposed. This disruption on the BiOI layer could explain the lower photocatalytic efficiency observed for this material, as it would limit the effective active area that can interact with contaminants.

It is important to mention that EDS analysis is not available for synthetic magnetite nor for the BiOI/magnetite (6%) heterostructure due to experimental restrictions. However, XRD analyses for these materials confirm their composition, as the magnetite sample presents clear peaks consistent with the cubic spinel phase corresponding to Fe₃O₄ (JCPDS No. 19-0629), without the presence of other secondary phases. The BiOI/Fe₃O₄(6%) heterostructure shows the coexistence of the tetragonal phase corresponding to BiOI (JCPDS No. 10-0445) and the cubic spinel of magnetite, without evidence of other phases.

Furthermore, previous studies have reported that EDS spectra for pure magnetite exhibit strong signals for Fe O, and that on BiOI/magnetite-type heterostructures, even at low proportions (5–10%), elemental maps composed of Bi, I, O, and Fe are predominant, with the elemental distribution being dependent on the synthesis method and the cover efficiency [25,53,64]. This data supports the theory that the materials in this study possess similar composition characteristics, validating the structural and morphological analyses discussed in previous sections.

Table 1. Texture and optical properties of the standalone materials and their corresponding heterostructures.

Material	Specific Surface (m2/g)	Average Pore Diameter (nm)	Pore Volume (cm3/g)	Eg Value (eV)
BiOI		16.3	0.31	1.98
Magnetite (Fe3O4)	19	23.0	0.13	1.47
BiOI/Fe3O4 (6%)	15	17.0	0.067	1.80
MMA	1	-	0.006	2.00
MMB	1	-	0.001	2.00
BiOI/MMA	13	19.5	0.064	1.98
BiOI/MMB	11	18.6	0.052	2.00

shows the specific surface (BET) data, pore volume, and average pore diameter of the standalone materials and their respective heterostructures. It can be observed that surface area increases significantly through the formation of BiOI/MM-type heterostructures, both in the case of the synthetic materials as well as those derived from mine tailings, with the exception of pure magnetite.

In the case of standalone BiOI, the starting surface area of 7 m²/g increases to 11 m²/g and 13 m²/g for the BiOI/MMA and BiOI/MMB heterostructures, respectively. This increase in surface area can be attributed to the generation of tridimensional structures with high porosity, promoted by the aggregation of nanoplate-type particles over the irregular surface of the magnetic materials. The SEM images support this lamellar aggregate morphology, which is characteristic of mesoporous materials [65,66].

Although the difference in specific surface area between BiOI/MMA (13 m²/g) and BiOI/MMB (11 m²/g), determined via BET, is relatively small, the results obtained from photocatalytic degradation up to 60 min (75.1% and 68.8%, respectively) evidence that this parameter is not the dominant factor in the performance of each material. BiOI/MMA’s high efficiency is attributed to a combination of factors that act synergistically, among which the following stand out: (I) a more efficient separation and transfer of photoinduced charges, (II) the nature of the BiOI-magnetic material interphase interaction, (III) slight variations in the absorption of visible light and the generation of reactive oxygen species (ROS), and (IV) the specific surface and electronic properties of the material extracted from tailing A. In this sense, although specific surface area contributes to the performance of the materials synthesized in this study, it is important to highlight that it is the joint effect of interphase, optical, and electronic characteristics that better explains BiOI/MMA’s higher photocatalytic efficiency compared to that of BiOI/MMB.

MMA and MMB, on the other hand, present extremely low specific surfaces (~1 m²/g), which evidences their lack of meso- and micropores. This low porosity prevented the determination of average pore diameter through BJH, suggesting that the mineralized, dense, and poorly textured nature of these wastes severely limits their adsorptive capabilities.

Conversely, synthetic magnetite exhibits a higher specific surface (19 m²/g) and a higher pore diameter (23.0 nm), indicating a structure with relatively broad pores. However, by forming the BiOI/magnetite (6%) heterostructure, a slight decrease in surface area can be observed (15 m²/g), which can be explained by the partial covering of the magnetite pores by the BiOI layer [67].

The BiOI/MMA and BiOI/MMB heterostructures not only increase their specific surface compared to their components derived from mine tailings, but they also present pore diameters in the mesoporous range (19.5 nm and 18.6 nm, respectively) and relatively high pore volumes. These textural properties are typical of materials possessing IV(a)-type isotherms and H3 hysteresis loops, which are characteristics that have been widely associated with porous structures formed by lamellar aggregates or non-rigid plates [68]. This type of interparticle porosity promotes the diffusion of reactive species, increases the exposition of active sites, and enhances the interaction between the catalyst’s surface and the contaminants, thus amplifying photocatalytic efficiency.

Lastly, it is important to highlight that the BiOI/MMA, apart from having a higher specific surface area, also presents a slightly higher pore diameter compared to BiOI/MMB, which suggests a better mass accessibility to internal active sites. This could be directly related to the higher photocatalytic efficiency of BiOI/MMA on the degradation of CA, compared with its counterpart derived from tailing B.

The forbidden band energy (Eg) of the synthesized materials was estimated from the DRS spectra through the Tauc method 69] (Figure S2). In all cases, an indirect electron transfer (n = 1/2) was considered, based on theoretical and experimental evidence that indicates that both BiOI and iron oxides present indirect optical transitions. BiOI, in particular, has been widely reported as an indirect transition semiconductor, according to calculations based on the density functional theory (DFT) and UV-Vis spectroscopy [68,69,70,71]. Likewise, magnetite (Fe₃O₄) and hematite (α-Fe₂O₃) have been characterized as indirect transition semiconductors in several structural and optical studies [72,73].

According to the data shown in Table 1, standalone BiOI presents an Eg value of 1.98 eV, meaning that its activation occurs under visible radiation. Synthetic magnetite exhibited an Eg value of 1.47 eV, which evidences its narrow-gap nature and its capacity to efficiently absorb in the visible range, possibly favored by electron transfer mechanisms between the Fe²⁺ and Fe³⁺ states.

The tailing-derived materials MMA and MMB exhibited similar Eg values close to 1.98 and 2.00 eV, suggesting the existence of a multiphase composition, including iron oxides and silicate minerals. These samples show a complex mineralogy, with non-stoichiometric phases which could modify their band structure. Complementary analyses through elemental mapping (Figure S1) confirm the coexistence of iron oxides alongside different phases corresponding to other types of oxides, supporting this hypothesis.

In the case of the heterostructures, an Eg value of 1.80 eV was determined for BiOI/Fe₃O₄(6%), while the values obtained for BiOI/MMA and BiOI/MMB were 1.90 and 2.00 eV, respectively. The Eg reduction in the BiOI/magnetite system can be attributed to the interface interaction between both components, which favors an ideal band alignment for the separation of photoinduced charges. This alignment aids on the interface charge transfer and enhances photocatalytic activity under simulated solar radiation. This behavior is beneficial from a photocatalytic perspective, as it can improve the capture of visible radiation and the separation of electron–hole pairs, without significantly increasing the required energy of excitation. The similarity of the Eg values of the BiOI/MMA and BiOI/MMB heterostructures suggests a common interface interaction effect rather than a structural dominance dependent on the type of tailing. Likewise, the Eg value of 1.98 eV obtained for the BiOI/MMA heterostructure is due to the thorough covering of MMA’s surface with the BiOI semiconductor, which is not the case for the BiOI/MMB heterostructure, possessing hollows devoid of BiOI according to previous analyses, therefore granting it an Eg value similar to that of the standalone tailing materials at 2.00 eV.

These Eg results, in the case of heterostructures, depend heavily on interface interaction and the covering of the semiconductor through the surface of the magnetic support compound. To better understand these interactions, a deeper analysis of surface composition as well as oxidation states present in the materials was performed through X-ray photospectroscopy (XPS), allowing us to correlate the changes observed on the electronic structure with the chemical nature of the heterogeneous interfaces of the materials analyzed (Figure 5).

High-resolution XPS spectra of the Bi 4f region in Figure 5a reveal spin–orbit doublets strongly associated with the +3 oxidation state of bismuth, located at ~158.5 eV (Bi 4f_7/2) and ~163.8 eV (Bi 4f_5/2), corresponding to the Bi present in the BiOI network [71,72,73]. The coinciding positioning of the peaks for all three heterostructures indicates that there is no significant changes in the chemical properties of bismuth, which suggests a high stability of the BiOI phase regardless of the type of support used. There are, however, clear differences in the intensities of said peaks. BiOI/MMA exhibits the highest intensity, which suggests a higher surface exposure of BiOI or a more thorough covering of the supporting material with the photocatalyst. BiOI/MMB presents an intermediate intensity, while BiOI/Fe₃O₄ (6%) exhibits the lowest intensity, possibly due to a thicker BiOI layer or lower exposure of the active material.

These differences remain in agreement with the morphological and structural results obtained from SEM, XRD and EDS, and have a direct correlation to photocatalytic performance. In fact, BiOI/MMA exhibits a higher photocatalytic activity than BiOI/MMB, which suggests that both the amount and quality of the BiOI layer over the supporting material heavily influence the efficiency of the process. The adequate dispersion of BiOI on MMA could be promoting a higher charge separation and a more effective interaction between the semiconductor and the target contaminant [68,74].

In Figure 5b, XPS spectra for the I region show that BiOI/Fe₃O₄ (6%), BiOI/MMA and BiOI/MMB present two well-defined peaks centered around ~623 eV and ~634.5 eV, corresponding to I 3d_5/2 and I 3d_3/2, respectively. These spin–orbit doublets confirm the presence of iodide (I⁻) within the BiOI network [75]. Slight displacements in the bond energy between samples was detected, with BiOI/Fe₃O₄ (6%) presenting a slightly higher bond energy compared to the heterostructures derived from mine tailings. These differences suggest subtle electron interactions between BiOI and its supporting material, possibly associated with interface charge transfer or coordination effects with either magnetite or other metallic species found in tailings [69,76].

Regarding peak intensity, the following trend was observed: BiOI/MMB > BiOI/MMA > BiOI/Fe₃O₄ (6%). The higher intensity registered for BiOI/MMB suggests a higher surface exposure of BiOI, possibly due to its dispersion or to the low chemical interference of the supporting material. The lower intensity for BiOI/Fe₃O₄ (6%) could be attributed to a lower surface load of BiOI or to stronger interactions between the semiconductor and the iron oxide support, which could in turn cause a shielding effect or a local modification around the iodide’s surroundings. Morphological factors such as crystal orientation, surface rugosity and system porosity can also have an influence on photoelectronic detection efficiency. The consistent spin–orbit separation (~11.5 eV) between the doublets confirms that the BiOI structure remains stable despite the compositional differences. Altogether, these findings evidence that the nature of the supporting material has a direct effect on XPS signals, the surface distribution of BiOI, and, ultimately, its photoelectrochemical behavior [73,74].

XPS spectra corresponding to the Fe 2p region for the BiOI/Fe₃O₄ (6%), BiOI/MMA and BiOI/MMB (Figure 5c) show the characteristic spin–orbit doublets around ~710.8 eV (Fe 2p_3/2) and ~724.5 eV (Fe 2p_1/2), consistent with the presence of ferric species in each sample’s surface. In the case of BiOI/magnetite (6%) and BiOI/MMA these peaks are mainly attributed to magnetite, which suggests a controlled integration of the oxide surface with the BiOI phase [77,78]. BiOI/MMB, on the other hand, exhibits broad, intense peaks with possible secondary components, which indicates the coexistence of other types of iron oxides such as hematites and goethite. This interpretation is supported by the DRX results previously obtained, where crystal phases corresponding to these oxides were detected, as well as the EDS analysis, which showed a higher relative proportion of surface iron on BiOI/MMB compared to the other heterostructures [68].

These differences in composition as well as structures can be correlated to the photocatalytic behavior observed for each material: while BiOI/Fe₃O₄ (6%) and BiOI/MMB exhibit similar efficiencies for the degradation of the target contaminant, BiOI/MMB offers significantly lower performance. Despite its higher Fe 2p signal, the deficient covering of the supporting material by the BiOI layer, evidenced in the SEM images, suggests a less homogeneous interface and a higher exposure of the mineral support’s surface, which could favor the recombination of electron–hole pairs or limit the charge transfer process from the semiconductor [74]. Furthermore, the presence of additional iron oxides could act as unwanted recombination centers or interfere with the energy coupling between both phases [54]. These findings highlight the relevance of interface interactions, morphological distribution and the nature of the supporting material in the design of high-efficiency photocatalytic heterostructures.

High-resolution XPS spectra for O 1s show notable differences in the surface chemistry of oxygen between the BiOI/Fe₃O₄ (6%), BiOI/MMA and BiOI/MMB materials. Figure 5d and

Table 2. Deconvolution of XPS O 1s spectra for BiOI/Fe₃O₄ (6%), BiOI/MMA and BiOI/MMB.

Sample	Peak (eV)	Assignation	Peak (eV)	Assignation	Peak (eV)	Assignation
BiOI/Fe3O4 (6%)	529.7	O2− in Bi-O or Fe-O	531.0	OH− surface/defects	532.3	Surface water/organics
BiOI/MMA	529.8	O2− in Bi-O or Fe-O	531.0	OH− surface/defects	532.2	Surface water/organics
BiOI/MMB	529.6	O2− in Bi-O or Fe-O	531.0	OH− surface/defects	532.2	Surface water/organics

shows the deconvolution of high-resolution XPS spectra corresponding to the O 1s region for the heterostructures. On all cases, spectra were deconvoluted in three main components: the dominant signal, centered around ~529.7–529.8 eV, is attributed to the oxide ion (O²⁻), characteristic of Bi-O and Fe-O bonds [75,76]. A second contribution, found between ~531.0 and 531.6 eV, is associated with hydroxyl groups (OH⁻) or defects in surface oxygen [77,79], while the third and weakest signal, found at ~532.2–532.3 eV, is correlated to adsorbed water molecules (H₂O) or organic species on the surface such as carbonyl groups (C=O) [60]. This data is consistent with what has been previously reported in the literature regarding bismuth-based semiconductors, where the presence of surface oxygen and structural defects has a direct impact on photocatalytic activity [78,80].

BiOI/Fe₃O₄ (6%) shows a dominant contribution of structural oxygen, which suggests an effective integration between BiOI and the supporting material, as well as a low density of defects or adsorbed species on its surface [47]. A similar behavior can be observed on BiOI/MMA, although with a slightly higher intensity on the surface component, possibly associated with a higher rugosity or compositional heterogeneity of the supporting material. Conversely, BiOI/MMB exhibits a significantly higher proportion of surface oxygen (~531.6 eV), suggesting that its surface is richer in adsorbed OH⁻ and H₂O groups. This difference could be attributed to the irregular covering of BiOI over the magnetic support, leaving the tailings’ surface exposed with a high affinity for oxygenated species, which has been corroborated by SEM, XRD and EDS analyses [48,81,82].

These observations support the hypothesis about both interface interaction as well as the quality of the BiOI layer over the supporting materials being a determining factor on surface chemical stability and photocatalytic performance. In particular, a high proportion of surface oxygen could act as electron recombination centers, or imply lower crystallinity and dispersion of the semiconductor layer, which would explain the lower photocatalytic efficiency observed on the BiOI/MMB heterostructure [81,82].

The survey XPS spectra (Figure 5e) obtained for BiOI/Fe₃O₄ (6%), BiOI/MMA, and BiOI/MMB show the presence of all expected elements found on BiOI-based materials: bismuth (Bi 4f), iodine (I 3d), and oxygen (O 1s), as well as iron (Fe 2p), which comes from the supporting materials. Furthermore, a signal for carbon (C 1s) can be observed, attributed mainly to the conductive carbon tape used to set down samples, alongside organic contaminants commonly seen in XPS analyses [82].

Altogether, XPS analysis confirms the presence of the expected elements and highlights differences in intensity and distribution of surface species, which is associated with the quality of the photocatalyst covering and its subsequent activity, with BiOI/MMA and BiOI/magnetite (6%) demonstrating a higher photocatalytic performance compared to BiOI/MMB, possibly because their surface homogeneity and the quality of their BiOI covering are superior.

2.4. Magnetic Properties

Magnetic properties of the synthesized materials were evaluated through hysteresis cycle measurements at room temperature, as shown in Figure 6a. BiOI/Fe₃O₄ (6%) exhibited a faint ferromagnetic behavior, with a magnetization saturation (Ms) of 0.1412 emu approximately, a remanent magnetization (Mr) of 0.00143 emu, and a coercivity (Hc) of 8.478 Oe, attributed to the predominant presence of magnetite nanoparticles, which exhibit soft ferromagnetism and superparamagnetism when finely dispersed [83].

The BiOI/MMA material shows a significantly sharper magnetic response, with a magnetization saturation (Ms) of 0.3729 emu, a remanence (Mr) of 0.02476 emu, and a coercivity (Hc) of 64.81 Oe. This data indicates a considerable input of magnetic phases in the composite material, most likely derived from iron residues possessing magnetic coupling properties. The high magnetization and coercivity displayed by this material could facilitate its recovery through external magnetic fields, on top of contributing a higher operation stability on catalytic recirculation processes [84].

BiOI/MMB presented a magnetization saturation of 0.04969 emu, a remanence of 0.00425 and a coercivity of 88.04 Oe. Although with a lower magnetization, this material displayed a higher coercivity between the three heterostructures, which could be related to a high magnetic anisotropy or a cleaving effect over the magnetic particles. This behavior suggests that this material presents a high stability when faced with field inversions, an important feature in photocatalytic applications which require multiple cycles [78].

Altogether, these results indicate that all three heterostructures display useful magnetic properties that enable them to be recovered through external magnetic fields, which represents an advantage in their use on heterogeneous photocatalytic systems. Moreover, both BiOI/MMA and BiOI/MMB show high coercivity, which could translate into a higher structural stability during long-term operations.

From a morphological standpoint, hysteresis loops obtained for all three heterostructures display a closed, symmetrical form, which confirms their ferromagnetic nature. In the case of BiOI/Fe₃O₄ (6%), the loop is narrow and steep near the origin, typical of materials with low coercivity and reduced remanent magnetization, reflecting a soft, reversible magnetic behavior. Meanwhile, the heterostructures based on mine tailings present wider and higher loops, indicative of a higher magnetic retentivity once the external magnetic field was removed. The BIOI/MMA loop, in particular, reveals both a high saturation magnetism and a considerable remanence, which suggests a strong magnetic interaction between the active phases of the composite material. These results support the hypothesis that industrial tailings incorporate functional magnetic phases that not only increase the recoverability of the material, but also may have a positive influence on the charge separation processes in photocatalytic applications [78,79].

Furthermore, magnetic properties were also evaluated through low-frequency magnetic susceptibility measurements (Figure 6b) at 15.6 kHz, thus obtaining the real (K′) and imaginary (K″) components normalized by mass. These parameters allow for the quantification of the material’s static magnetic response (K′) and its energy losses through magnetic relaxation (K″), which are relevant in magnetic separation and catalytic activation applications [84].

The BiOI/MMA sample exhibited the highest real susceptibility (K′ ≈ 2.21 × 10⁻⁴ emu/g) and the highest imaginary component (K″ ≈ 2.8 × 10⁻⁵ emu/g), which suggests a high concentration of ferromagnetic phases, presumably iron oxides. This behavior may favor both the recovery of the catalyst through magnetic separation and its activation in processes coupled with external magnetic fields [85].

In contrast, BiOI/MMB presented the lowest magnetic response (K′ ≈ 1.9 × 10⁻⁵ emu/g; K″ ≈ 2.0 × 10⁻⁶ emu/g), which indicates a lower amount of active magnetic species or a heterogeneous distribution of said species on the BiOI matrix. Meanwhile, the BiOI/Fe₃O₄ (6%) showed intermediate values (K′ ≈ 3.3 × 10⁻⁵ emu/g; K″ ≈ 4.0 × 10⁻⁶ emu/g), consistent with a soft ferromagnetic behavior, characteristic of systems possessing finely dispersed magnetite particles [86]. This type of response is adequate to keep a suitable dispersion of the photocatalyst particles in suspension, without compromising its subsequent recovery through magnetic separation [87].

Magnetic susceptibility analyses and hysteresis cycles confirmed that the three heterostructures that were synthesized (BiOI/Fe₃O₄ (6%), BiOI/MMA, and BiOI/MMB) presented ferromagnetic behavior, with notable differences in their intensities and stabilities depending on their composition. BiOI/MMA, in particular, presented the most intense magnetic response, attributed to a higher concentration of crystalline iron oxides found in its base industrial tailing such as magnetites (Fe₃O₄) and possibly hematite (Fe₂O₃).

The BiOI/magnetite (6%) material exhibited a soft ferromagnetic behavior, associated with the controlled presence of finely dispersed magnetite nanoparticles, granting it a suitable magnetic response without compromising the dispersion of the suspended catalyst [86]. Meanwhile, BiOI/MMB, although it had the lowest magnetic content, presented the highest coercivity, possibly related to a high magnetic anisotropy or the interlocking with particles in non-conductive phases [78].

The symmetrical and closed hysteresis loops confirmed the ferromagnetic nature of all three heterostructures. Altogether, these results highlight how the crystalline composition as well as the distribution of iron oxides such as magnetites and hematite have a significant role in the material’s magnetic response. This functionality, when integrated in a controlled manner through industrial tailings or synthetic precursors, upgrades the design of active, stable and easily recoverable photocatalysts, which are key components for high-impact and environmentally sustainable applications in heterogeneous photocatalysis processes [79,87]. Lastly, both techniques performed in this study have given a complete vision of the dynamic magnetic behavior displayed by the synthesized materials, especially for the BiOI/MMA heterostructure, which is essential in the evaluation of its applicability in separation, recovery, or catalytic activation in liquid media.

Although the results obtained in this study demonstrate that it is possible to transform mine tailings into functional materials for high-value photocatalytic applications, thus simultaneously contributing to environmental remediation and the circular economy, it is important to highlight that mine tailings may contain heavy metals such as lead and arsenic, among others, in trace amounts that could limit their direct application on a large scale (see Tables S2 and S3). In this work, the magnetic fraction was previously subjected to a cleaning process using ethanol in order to remove surface impurities and soluble residues (Section 3.2), and structural (XRD), morphological (SEM-EDS), and surface (XPS) analyses confirm the absence of secondary phases attributed to heavy metals. But in the case of industrial scaling of the BiOI/MMA material, it will become necessary to complement its use in wastewater treatment with selective lixiviation processes or immobilization, with the goal of ensuring the technological and environmental viability of the valorization of tailings in photocatalytic applications.

2.5. Reutilization of the BiOI/MMA Heterostructure

Stability and recycling capability is a critical component in photocatalytic materials aimed to be applied in real environmental processes. In this study, the BiOI/MMA material was evaluated in two consecutive usage cycles. It can be observed in Figure 7, however, the results obtained show that the material has a high stability with a low decrease in degradation efficiency from 89.4% to 84.3%, corresponding to a decrease in photocatalytic efficiency of only 5.1% after the second cycle, which agrees with what has been reported in the literature for BiOI/magnetic material composites, which hold efficiencies above 80% even after several cycles [19,71,86]. These findings reinforce the potential of the BiOI/MMA heterojunction as a sustainable and recyclable photocatalyst, and it is proposed as a future work to increase the number of cycles in order to more rigorously evaluate its long-term stability in environmental remediation processes.

This behavior agrees with previous studies that report high photocatalytic stabilities for BiOI-type materials in composites with magnetic oxides, which preserve efficiencies above 80% after multiple usage cycles [21].

The decrease in photocatalytic efficiency observed during the second reuse cycle of the BiOI/MMA material could be attributed to the partial saturation of active sites on the photocatalyst’s surface, due to the strong adsorption of the contaminants or intermediate subproducts which block these sites, reducing the availability of reactive areas for new photocatalytic processes [88,89]. These results allow for the consideration of the BiOI/MMA material as a promising candidate for real, high-scale applications, where efficiency, recoverability and reuse of the photocatalytic materials are key aspects.

2.6. Proposed Photocatalytic Mechanism

Results shown in Figure 8 evidence that the most significant inhibition is produced by oxalic acid, which indicates that photogenerated holes (h⁺) are the dominant reactive species in the degradation of CA. The marked decrease in activity in the presence of p-benzoquinone means that superoxide radicals (•O₂⁻) also play an important role in the system. In contrast, the inhibition observed in presence of isopropanol was minimal, suggesting that hydroxyl (•OH) radicals play a secondary role.

These results are consistent with what has been reported in the literature for BiOI-based materials, as well as its heterostructures with magnetite, where it has been demonstrated that photogenerated holes and superoxide radicals are the main reactive species in photodegradation processes, while hydroxyl radicals provide minimal contribution. More specifically, recent works indicate that in BiOI/Fe₃O₄ heterostructures, the effective separation of photogenerated charges favors the accumulation of holes in the valence band of BiOI, and the generation of •O₂⁻ from elections in the conduction band, enhancing photocatalytic efficiency and magnetic reusability of the material [89,90].

Considering the results shown in Figure 8 regarding the tests with radical scavengers and what has been reported in the literature [89,90,91,92,93,94], a reasoned degradation pathway for CA using BiOI/MMA has been proposed (Figure 9), which considers the following:

I.

Hydroxylation of the aromatic ring, mainly caused by h⁺ and, to a lesser degree, •OH.

II.

Opening of the aromatic ring, favored by h⁺ and •O₂⁻.

III.

Formation of short chain carboxylic acids (maleic, oxalic and acetic acids) and partial oxidation intermediates.

IV.

Final mineralization to CO₂ to H₂O.

The proposed pathway aligns with what has been described for phenolic compounds in systems based on BiOI and BiOI/magnetite heterostructures, where the combination of direct oxidation by h⁺ and reactions mediated by •O₂⁻ forms the main degradation pathway.

In summary, results yielded by the tests performed with scavengers confirm that the high efficiency of BiOI/MMA is attributed to the synergy between the separation of photoinduced charges facilitated by the magnetic phase and the predominant participation of h⁺ and •O₂⁻ in CA degradation, with a secondary contribution of •OH.

Table 3. Comparison between the BiOI/MMA material with BiOI/synthetic magnetic material photocatalysts.

Material	Magnetic Composition	Target Contaminant	Eg eV	Photocatalytic Efficiency (%)	Irradiation Time (min)	Reuse (N° of Cycles)	Photocatalytic Loss (%)	Reference
BiOI/MMA	MMA (Tailing A)	CA	1.98	89.4	120	2	5.10	This work
BiOI/Fe3O4	Fe3O4	CA	1.97	77.2	180	0	---	[95]
BiOI/Fe3O4	Fe3O4	Phenol	1.78	90.0	300	5	16.0	[22]
BiOI/rGO/Fe3O4	Fe3O4	RhB	1.40	95.9	480	10	13.6	[77]
NiFe2O4/BiOI	NiFe2O4	RhB CV MB	1.84	91–96	120	5	8.00	[96]
ZnFe2O4/BiOI	ZnFe2O4	MO	1.87	81.2	60	---	---	[97]

provides a comparative overview of the main results reported for different BiOI-based photocatalytic systems coupled with magnetic materials [22,77,95,96,97] including heterostructures such as NiFe₂O₄/BiOI, BiOI/rGO/Fe₃O₄, and CoFe₂O₄/BiOI. These systems exhibit a variety of strengths: for example, NiFe₂O₄/BiOI composites often reach very high photocatalytic efficiencies (>90% for dyes such as RhB, CV, and MB within 120 min), while BiOI/rGO/Fe₃O₄ has demonstrated remarkable stability, maintaining ~82% efficiency for RhB even after 10 reuse cycles. Other systems, such as BiOI/Fe₃O₄, have shown good activity toward phenolic compounds (90% phenol removal in 300 min, five cycles), though with higher photocatalytic loss (16%).

In comparison, the BiOI/MMA system reported in this work demonstrates a unique balance of performance indicators. Specifically, it achieved 89.4% degradation of caffeic acid in 120 min with a band gap of 1.98 eV, while retaining 84.3% efficiency in the second cycle with only 5.1% photocatalytic loss. These values are competitive with, and in some cases superior to, those of BiOI coupled with synthetic ferrites (e.g., NiFe₂O₄/BiOI or ZnFe₂O₄/BiOI). More importantly, the BiOI/MMA system introduces an unprecedented dimension of sustainability, as the magnetic phase is directly obtained from mine tailings, an environmental liability, instead of relying on energy-intensive synthesis of ferrite nanoparticles.

In Table 3 it can be observed that the BiOI/MMA heterostructure presents an Eg value of 1.98 eV, which is practically identical to that of pure BiOI, while that of BiOI/Fe₃O₄ (6%) decreases down to 1.80 eV. This difference may be explained by the nature of the interphase and the coverage of the BiOI layer in each system. In the case of BiOI/Fe₃O₄ (6 %), synthetic magnetite possesses high purity and crystallinity, which favor a more uniform interphase interaction and a more efficient band alignment between BiOI and Fe₃O₄, resulting in a narrowing of the band gap. In contrast, although structural analysis (SEM, EDS and XRD) confirms the integration of BiOI and MMA, the existence of possible secondary phases, such as silicates and other minerals, interrupts the continuity of the interphase, which causes the optical response to be primarily dominated by BiOI, maintaining its characteristic value of ~1.98 eV. Nevertheless, even though the Eg value of BiOI/MMA is similar to that of BiOI, photocatalytic tests demonstrate that the BiOI/MMA interphase effectively contributes to charge separation and improves degradation efficiency, which indicates that the interphase interaction is functional despite not significantly displacing the Eg value into the visible range.

Therefore, the novelty of this work lies not only in its ability to match the performance of previously reported BiOI/magnetic composites but also in advancing the principles of the circular economy and environmental remediation. By valorizing mining residues as functional magnetic supports, the BiOI/MMA system transforms a waste material into a high-value component for green technologies. This differentiating factor represents a significant step forward in photocatalyst design, situating the present study at the intersection of high photocatalytic performance, magnetic recyclability, and sustainable resource utilization.

Finally, the results obtained in this study establish a clear correlation between the nature of the magnetic material used as support (in this case synthetic magnetite and magnetic material derived from mine tailings), the quality of the BiOI layer, and the observed photocatalytic performance. Heterostructures synthesized from magnetic materials derived from industrial tailings, particularly in the case of the BiOI/MMA heterostructure, displayed efficiencies comparable to those obtained from the synthetic magnetite-based heterostructure, which evidences the potential of MMA as a sustainable alternative in the design of recyclable photocatalysts. This behavior is directly associated with the chemical composition, crystallinity and morphology of the supporting material, as well as its interface interaction with BiOI, which are factors that have an influence on the generation and separation of charge carriers, accessibility to active sites, and recovery post photocatalytic reaction. In addition, structural, optical, magnetic and surface characterization studies have allowed for the validation of the effective formation of the BiOI/MMA heterostructure as stable and functional, reinforcing the hypothesis that mine tailings can be transformed into functional materials of high added value in environmental decontamination applications under simulated solar radiation. This approach is part of a circular economy strategy applied to the mining industry, with significant implications for the development of sustainable technologies for wastewater treatment.

Future Perspective: Thanks to the development of BiOI/MM heterostructures, they have great potential as sustainable photocatalysts, as is evidenced by the results obtained in this study. Future research should focus on achieving a good covering of the magnetic phase (MM) with BiOI, obtaining a mesoporous texture and further decreasing the proportion of non-magnetic impurities, and make it so the band gap is found in the visible range (1.8–2.0 eV) to promote the use of solar radiation. Compared to industrial photocatalysts such as TiO₂ and ZnO, among others, BiOI/MM systems present clear advantages, among which is their activation under visible light, their recyclability, and their coherence with circular economy strategies through the valorization of mining waste. However, certain limitations persist, such as lower long-term stability compared to the traditional semiconductor TiO₂ and the possible formation of deposits of reaction intermediates on the photocatalyst’s surface. Addressing these challenges will be essential to driving the development of BiOI/MM heterostructures towards large-scale environmental remediation applications.

3. Materials and Methods

3.1. Sample Preparation

Samples were collected from 2 mine tailings located in the Coquimbo Region, making sure that they were representative and homogeneous. The tailing known as “El Culebrón” was labeled as tailing A, and the “Dina” tailing was labeled tailing B. Both of these tailings are located in different areas in this region, and the collection process was performed randomly, ensuring that sampling conditions would minimize compositional variability. For sampling tailings, eight samples were taken in triplicate by excavating a 40 × 40 × 80 cm pit, taking the material from bottom up with polyethylene shovels. In the lab, solid samples were dried at 130 °C over 24 h and subsequently sieved through a 100-mesh screen (W.S. Tyler, Mentor, OH, USA), following the procedure described by Zamarreño and Diaz [98]. Afterwards, samples were mixed together and homogenized in order to obtain a representative composite sample, which was used in analyses and synthesis of the BiOI/MM nanocomposites.

Furthermore, Figure S3a,b have been incorporated into the manuscript, where the tailings dams and sampling points are shown in order to provide better clarity. This information has been added to the Supplementary Materials, improving the transparency and reproducibility of this study.

The samples consisted of fine mineral powders that were not sieved, and magnetic material (MM) was extracted so it could be used in the synthesis of the BiOI/MMA and BiOI/MMB heterostructures. Tables S2 and S3 show the chemical composition of each tailing [99,100].

3.2. Extracting Magnetic Material from Mine Tailings

This procedure is illustrated below in Figure 10. Once samples from both mine tailings (A and B) were collected (Figure 10, step 1), their corresponding magnetic material (MM) was separated from them, which was performed through the use of permanent magnets of different sizes and intensities (Figure 10, step 2), with the goal of maximizing the recovery of both paramagnetic and ferromagnetic phases from these environmental liabilities.The MM obtained from each tailing was then subjected to a washing process using analytical-grade ethanol in order to eliminate surface impurities (Figure 10, step 3), mineral residue adsorbates and any other compound that could interfere in subsequent stages of this study. The clean materials were then dried in a vacuum oven at 60 °C for 60 min (Figure 10, step 4), ensuring the removal of solvent traces without risking any thermally induced structural alterations. Finally, the MM is obtained (Figure 10, step 5).

3.3. Synthesizing Magnetite

In order to standardize the amount of MM from each tailing in the final heterostructures, it became necessary to synthesize magnetite, which was performed through the solvothermal method; 5.4 g of iron (III) chloride hexahydrate (FeCl₃ · 6H₂O), which was previously dried in a vacuum oven, was mixed with 15.99 g of sodium acetate and added to 160 mL of ethylene glycol. The mixture was put in an ultrasonic bath for 25 min, manually mixing it with a glass rod every 5 min to aid in its complete dissolution. The solution was then transferred to a 100 mL Teflon-coated autoclave reactor and heated to 200 °C for 12 h. After the 12 h were completed, the reactor was left to slowly cool down to room temperature. The solid residue formed in the reactor was then removed from the liquid phase with the use of a magnet, and subsequently washed with 100 mL of analytical-grade ethanol in several cycles until the ethylene glycol was completely removed. Finally, the solid material obtained was dried in a vacuum oven for 60 min and stored in a dry amber glass vial inside a desiccator.

3.4. Optimizing the Amount of Magnetite in the Photocatalyst

The synthesis of the BiOI/magnetite heterostructure was performed through the co-precipitation method using different mass percentages of synthetic magnetite: 2, 4, 5, 6 and 8%. This was implemented in order to determine the ideal amount of synthetic magnetite needed to achieve the most optimized photocatalytic activity in the degradation of caffeic acid (CA) in water.

An amount of 0.3276 g of bismuth (III) nitrate pentahydrate [Bi(NO₃)₃ · 5H₂O] was dissolved in 20 mL of analytical-grade ethanol and stirred for 60 min with a magnetic stirrer. The necessary amount of synthetic magnetite was weighed according to the percentages that were previously specified (2, 4, 5, 6 and 8%) and added to the bismuth (III) nitrate pentahydrate solution, which was then stirred mechanically for an additional 60 min.

At the same time, 0.112 g of potassium iodide (KI) was dissolved in 20 mL of deionized water and stirred magnetically for 20 min. This solution was carefully added dropwise to the bismuth nitrate solution after the latter had been stirred for 60 min. The resulting solution’s pH was then immediately adjusted to 8.5 using a 22% ammonia solution and stirred continuously for an additional 3 h at 145 rpm in a mechanical stirrer.

The precipitate that formed after the synthesis procedure was finished was subjected to a large magnet with a high-intensity magnetic field, which allowed for the selective recovery of any solid material that had magnetic properties. This solid was then washed with 15 mL of analytical grade ethanol, and this washing procedure was subsequently repeated for three consecutive cycles with the goal of removing residual impurities and reaction subproducts. The semiconductor material was dried in a Vacucell vacuum oven for 60 min, weighed and stored in an Eppendorf vial inside a desiccator until usage in photocatalytic tests.

BiOI was synthesized following the same procedure as described above, but forgoing the addition of magnetite during the early synthesis steps.

3.5. Synthesizing BiOI/Magnetic Material (BiOI/MM)

In order to synthesize each of the heterostructures, the co-precipitation method was followed as described in Section 3.3, substituting the synthetic magnetite for the magnetic material obtained from each of the mine tailings, and taking into consideration the optimized amount of magnetite determined in Section 3.4.

Photocatalytic tests were subsequently carried out using the newly synthesized BiOI/MMA and BiOI/MMB heterostructures, as well as their base components, following the experimental procedure described in Section 3.6. However, reaction times were extended to 80, 100, and 120 min. This extension was specifically applied to the heterostructures derived from magnetite and mine tailings, with the former acting as a control experiment that would serve as a guideline to measure whether the materials derived from tailings are suitable for photocatalytic applications.

In this stage, degradation percentages were determined, and photocatalytic efficiencies were compared between both materials. The effect of the magnetic material, originating from tailings A and B, on the photocatalytic performance of each of the novel photocatalysts can thus be evaluated.

3.6. Determining Photocatalytic Efficiencies

The explanation of this procedure is shown in Figure 11. The photocatalytic experiments were performed by adding 250 mL of a 10 ppm caffeic acid (CA) solution to a batch reactor, where 0.1 g of BiOI/magnetite (BiOI/MMA or BiOI/MMB) was added (Figure 11, step 1). The material was stirred in the CA solution for 40 min in the dark, allowing it to reach the adsorption equilibrium (Figure 11, step 2). Once the 40 min had passed, a xenon lamp was then turned on, simulating solar radiation(Figure 11, step 3). Before each photocatalytic experiment, the lamp was turned on for 10 min to guarantee a stable irradiation and a consistent spectral distribution.

The emission spectrum of the UV–visible lamp employed in the photocatalytic experiments is shown in Figure S4. The lamp exhibits a broad emission band in the range of approximately 300–900 nm, with a maximum intensity centered at around 600 nm, along with secondary peaks at shorter wavelengths.

The intensity of the xenon lamp used in this study was not determined. However, all photocatalytic experiments were performed using the same xenon lamp at a fixed distance from the reactor and with a constant operating power. These conditions guarantee that photocatalytic experiments were carried out in a reproducible and consistent manner.

In order to determine the photocatalytic efficiency of the different BiOI/magnetite materials, 10 mL samples of the solution were taken at specific time frames of 5, 10, 15, 20, 30, 40 and 60 min under simulated solar radiation. Each sample was filtered through a 0.22 mm syringe nylon filter (Figure 11, step 4).

Samples were analyzed through UV–Visible spectroscopy with a Thermo Scientific Evolution 220 spectrophotometer (Waltham, MA, USA), and through chromatography using an Agilent 1201 HPLC-DAD system (Santa Clara, CA, USA). The mobile phase consisted of a buffer/ACN 87:13 v/v, with the buffer being a pH 2.50 formic acid solution. All solvents used were HPLC-grade solvent purchased from Merck (Rahway, NJ, USA). CA was detected at two of its absorption wavelengths of 293 and 325 nm, and a retention time of 9 min after an 11 min runtime (Figure 11, step 5). The results yielded from these analyses were used to calculate the CA concentration in each sample, and through this, the degradation percentages associated with each BiOI/magnetite material (Figure 11, step 6).

3.7. Experiments with Reactive Species Scavengers

In order to determine the photocatalytic mechanism in this study, trapping experiments were performed using selective agents with the goal of identifying the reactive species involved in the degradation of CA using the BiOI/MMA material. A 10 mg L⁻¹ CA solution was prepared, and 1 L of this solution was divided into four aliquots amounting to 250 mL each. Each aliquot received a different treatment: 136 µL of isopropanol (IPA, •OH scavenger), oxalic acid (OA, h⁺ scavenger) to 20 mmol L⁻¹, and p-benzoquinone (BQ, •O₂⁻ scavenger) to 10 mmol L⁻¹. Kinetics experiments with and without scavengers were performed in triplicate for 120 min under simulated solar radiation. Determination of the photocatalytic efficiency as a degradation percentage was performed through the methodology described in Section 3.6.

3.8. Characterizing the Synthesized Materials

In order to establish a correlation between the material’s structural, morphological, optical, electronic and magnetic characteristics and its photocatalytic performance, it became necessary to characterize its individual components such as BiOI, Fe₃O₄, and magnetic materials extracted from mine tailings magnetic material A and B (MMA and MMB), alongside the BiOI/Fe₃O₄ (6%), BiOI/MMA and BiOI/MMB heterostructures.

Crystal phases were analyzed through X-ray diffraction (XRD) complemented with Raman spectroscopy (Bruker, Karlsruhe, Germany), allowing for the identification of phases and detection of possible residual impurities, which were then identified through using a Bruker D4 dust diffractometer (Ettlingen, Germany) equipped with a Lynxeye detector and a Cu-Kα radiation source with a nickel filter to eliminate Kβ radiation. The analysis was carried out in an angle range between 3° and 70° (2θ), with a light path of 0.020°.

Morphology was evaluated through scanning electron microscopy (SEM) using a Hitachi SU-3500 microscope (Hitachi, Chiyoda City, Japan) operating at 15 kV and a 60 Pa vacuum, complemented with energy-dispersive X-ray spectroscopy (EDS) to determine elemental composition. A Physical Electronics VersaProbe II spectrometer (HORIBA, Palaiseau, France) was used to determine surface composition and oxidation states, using monochromatic Al-Kα (hν = 1486.6 eV) radiation. High-resolution spectra were corrected using the C 1s (284.8 eV) as a reference.

Specific surface area was determined using adsorption–desorption nitrogen isotherms at −196 °C, which were registered in a NOVA 1000e analyzer (QuantaChrome, Boynton Beach, FL, USA). Samples were previously degassed for 16 h under vacuum at 150 °C. Specific surface areas were calculated according to the Brunauer–Emmett–Teller model (BET). Pore size distribution was determined using the desorption branch of the nitrogen isotherms using the Barrett–Joyner–Halenda (BJH) model. Optical properties were analyzed through diffuse reflectance spectroscopy (DRS) using a UV-Vis Thermo Scientific Evolution 220 spectrophotometer (Madison, WI, USA) equipped with an integrating sphere. Magnetic properties were evaluated by measuring magnetic hysteresis cycles using a Lake Shore 7400 vibrating-sample magnetometer (VSM) (Westerville, OH, USA), applying a magnetic field ranging from −10,000 to +10,000 Oe at room temperature. Magnetic susceptibility was measured with a Bartington MS2 susceptometer operating at 0.465 kHz (Oxfordshire, UK). Magnetic susceptibility was measured with a Bartington MS2 susceptometer operating at 0.465 kHz. Surface chemistry and oxidation states of key elements (Bi, I, Fe and O) were analyzed through X-ray photospectroscopy (XPS).

Furthermore, surface composition and oxidation states for the elements Bi, I, Fe and O were analyzed through X-ray photoelectron spectrometry (XPS) (Staib Instrument, Williamsburg, VA, USA). Analyses were performed on Surface Analysis Station 1 spectroscopy equipment, model XPS RQ300/2, equipped with an Al Kα (λ = 1486.6 eV) radiation source, operating at 15 kV and 10 mA. A DESA 150 detector was used (New York, NY, USA), operating at 2700 V. Spectra were acquired in an energy range between 1000 and 0 eV, with a pass size of 1.0 and 0.2 eV and an acquisition per point of 0.2 to 0.5 s. No secondary monochromator was used. Peak assignment and identification of chemical states was performed using references from the scientific literature as well as specialized databases.

Magnetic properties for the BiOI/magnetite (6%), BiOI/MMA and BiOI/MMB materials were evaluated through hysteresis cycles, using a Princeton Measurements Corp. alternating gradient magnetometer, model MicroMag 2009. Measurements were carried out at room temperature under a continuous sweep mode, with an applied magnetic field up to ±10 kOe and a field increase of 80 Oe. The average time per point was 100 ms, and the equipment was operated with the MicroMag AGM-VSM (software version 2014/11/04) for system control and data acquisition.

Key magnetic parameters, such as saturation magnetization (Ms), magnetic remanence (Mr) and coercivity (Hc), were determined through the hysteresis cycle measurements, with the goal of characterizing the magnetic behavior of the BiOI/magnetic material composites derived from mine tailings.

Magnetic susceptibility of the heterostructures was evaluated through a model MFK1-FA alternating current susceptometer (AGICO, Brno, Czech Republic), operating at a frequency of 15.6 kHz and an applied magnetic field of 200 A/m. During each measurement, the equipment automatically registered the real (K′) and imaginary (K″) components of magnetic susceptibility, which were normalized by sample mass and reported as K′/g and K″/g, respectively.

The real component (K′) stands for the part of the magnetization in phase with the applied field, and represents the capacity of the material to be magnetized in a reversible form, while the imaginary component (K″) is correlated with the internal energy loss associated with hysteresis or magnetic relaxation processes [89]. Each sample was measured at least three times to ensure reproducibility, and results were expressed as averages with their respective standard deviations.

This characterization allowed for the establishment of comparisons between the synthesized materials in regard to the intensity, efficiency and stability of their magnetic response, evidencing the differences attributed to the nature and proportion of the magnetic component incorporated into the structure.

4. Conclusions

The results obtained in this paper clearly demonstrate the technical and environmental feasibility of valorizing mine tailings as non-conventional sources of magnetic materials for the design of photocatalytic structures based on BiOI. This strategy represents an original approach in the heterogeneous photocatalysis field, by integrating principles of both circular economy and environmental remediation through the usage of environmental residues originating from the mining industry in the synthesis of advanced functional materials.

In this study, an optimal ratio of 6% by mass of magnetic material was established in order to maximize the photocatalytic performance of the heterostructures by achieving an efficient equilibrium between the absorption of visible radiation, the separation of photoinduced charges, and the accessibility of active sites. The BiOI/MMA heterostructure, derived from tailing A, displayed exceptional behavior, achieving a degradation of caffeic acid above 89% after 120 min under simulated solar radiation, a performance comparable to the control system BiOI/magnetite (6%). Such high performance is attributed to the excellent structural integration between BiOI and the supporting material, alongside the formation of active interfaces that favor an efficient charge transfer and decrease electron–hole recombination. Furthermore, magnetic properties associated with the supporting material allowed for the recovery and reuse of the photocatalyst without any significant loss to its activity, providing robustness to the system.

Structural, optical and magnetic analyses revealed that the mineralogic composition and crystallinity of the magnetic materials recovered from industrial tailings have a direct effect on photocatalytic performance, which reinforces the importance of a detailed characterization of the tailings at the time of their valorization. The functional connection between BiOI and the magnetic material derived from mine tailings allows for a synergic combination of the capacity of photocatalytic absorption and the ease of magnetic recovery, thus creating a novel system that is efficient, reusable and environmentally sustainable.

Although this study offers structural, morphological, optical, surface-chemical, textural and photocatalytic evidence that supports the effective formation of a heterojunction between BiOI and magnetic materials obtained from mine tailings, future works will include a thorough photoelectrochemical characterization, such as transient measurements of photocurrent response, in order to more precisely validate and quantify the efficiency of charge separation displayed by these materials for applications in environmental remediation.

The experiments performed with reactive species scavengers show that the photocatalytic degradation of caffeic acid with the composite system BiOI/MMA is mainly dominated by the action of photogenerated holes (h⁺) and superoxide radicals (•O₂⁻), with a secondary contribution by hydroxyl radicals (•OH), which is in agreement with mechanisms previously described in the literature for BiOI and BiOI/magnetite systems.

Altogether, this work constitutes a significant contribution to the development of more sustainable decontamination technologies, proposing an innovative solution that can also be applied on a large scale to transform industrial waste into active materials for the treatment of wastewater. The evidence presented in this paper showcases the potential of mine tailings as strategic inputs in the design of photocatalysts in the future, opening up new perspectives for the implementation of sustainable processes in the context of a circular economy.