Aluminum-Loaded Bifunctional Resins for Efficient Fluoride Removal from Aqueous Solutions

Abstract

The deep defluorination of water remains a significant environmental challenge. In this work, aluminum was loaded onto the bifunctional resin S957 containing a phosphoric-sulfonic acid difunctional group for efficient fluoride removal. Al-S957 demonstrated excellent fluoride removal performance across a broad pH range. When anions and organics coexisted, Al-S957 exhibited significantly better fluoride adsorption performance compared to aluminum-loaded monofunctional resins. The adsorption followed an endothermic chemisorption process on a monolayer surface. FTIR and XPS analyses further revealed that fluoride removal relied on a ligand exchange mechanism. Column adsorption conducted over five cycles highlighted the strong practical potential of Al-S957. The results suggested that Al-S957 exhibits significant potential for practical applications.

1. Introduction

Fluoride contamination in drinking water is a growing global concern [1]. The World Health Organization recommends that fluoride levels in drinking water should not exceed 1.5 mg/L for optimal health [2]. Excess fluoride can lead to severe health issues, including rickets, osteoporosis, arthritis, neurological disorders, and cancer [3,4,5]. Fluoride enters natural water sources primarily through the dissolution of fluoride-rich rocks and minerals such as cryolite, topaz, and fluorite [6]. Additionally, various untreated industrial wastewaters, such as those from semiconductor, polytetrafluoroethylene, glass, fertilizer, and pesticide industries, are also the main sources of fluoride [5]. These sources together aggravated the threat of fluoride effects. Elevated fluoride levels in water have been documented in over 20 countries across Africa, Europe, and the Americas [3]. Consequently, it is imperative to find an effective and robust fluoride remediation technology.

A number of defluorination techniques, including precipitation, coagulation, membrane separation, and adsorption, have been employed for the removal of excess fluoride [6]. Among these, adsorption has emerged as the most efficacious approach for deep fluoride removal, primarily attributable to its advantages of efficiency, convenience, operability, simplicity of design, and economy [7,8,9]. Some studies have shown that metal modification can substantially enhance the fluoride removal performance of materials [10,11,12]. By utilizing metal ions as adsorption centers and exploiting mechanisms such as ligand exchange and electrostatic adsorption, efficient fluoride removal can be achieved from low-concentration fluorinated water [13,14]. Among various modifying metals, aluminum is widely recognized as a cost-effective and readily available metal for adsorbent modification [15]. According to Lewis acid–base theory, hard basic fluorides exhibit a strong inherent affinity for hard acidic aluminum [16]. This is further demonstrated by the first-step coordination constant of Al-F complexes (Log K₁ = 6.10), surpassing those of other common metal–fluoride systems, such as Fe-F (Log K₁ = 5.28), Mg-F (Log K₁ = 1.3), and La-F (Log K₁ = 2.77) [17]. The superior thermodynamic stability of aluminum–fluoride coordination complexes ensures reliable and efficient fluoride removal, making aluminum-based materials highly effective for water defluorination applications.

Various carriers have been explored for aluminum loading, including zeolite, ion exchange resins, activated carbon, and metal–organic frameworks [14,18,19,20]. Among these, ion exchange resins stand out for their notable advantages, including good stability and ease of operation [21]. However, while aluminum-loaded monofunctional resins have been reported for efficient fluoride removal, they have limitations such as low adsorption capacity, metal loss, and susceptibility to interference by coexisting ions [22,23]. In contrast, multifunctional group materials have been shown to have fast adsorption kinetics and good regeneration properties [24]. The chelating resin S957 with bifunctional sulfonic acid and phosphoric acid groups has also been demonstrated to adsorb aluminum effectively [25].

Therefore, this study involved loading aluminum onto bifunctional resin S957 and employing aluminum-loaded monofunctional resins (phosphate-based resin D412 and sulfonic acid-based resin D001) for comparative evaluation. The adsorption properties of Al-S957 were extensively investigated, including its kinetics, isotherms, and thermodynamics. The effects of solution pH, coexisting anions, and organic compounds on fluoride adsorption were also examined. The adsorption mechanism was explored, and fixed-bed column experiments with real photovoltaic wastewater were conducted to assess its practical performance.

2. Materials and Methods

2.1. Materials

Sodium fluoride (NaF) was purchased from Macklin Reagent Co., Ltd. (Shanghai, China). Aluminum chloride hexahydrate (AlCl₃·6H₂O), sodium nitrate (NaNO₃), sodium bicarbonate (NaHCO₃), sodium sulfate (Na₂SO₄), sodium dihydrogen phosphate (NaH₂PO₄), anhydrous ethanol (C₂H₆O), sodium chloride (NaCl), and sodium alginate (SA) were supplied by Sinopharm Group Chemical Reagent Co., Ltd. (Shanghai, China). Humic acid (HA), tannic acid (TA), sodium dodecyl benzene sulfonate (SDBS), and L-tryptophan (Trp) were purchased from Aladdin Reagent Co., Ltd. (Shanghai, China). All chemicals are of analytical regent grade.

D412 and S957 were purchased from Tianjin Shuanglian Sci. & Tech. Co., Ltd. (Tianjin, China). D001 was purchased from Zhengzhou Hecheng New Material Sci. & Tech. Co., Ltd. (Zhengzhou, China). Table S1 provides the basic information about the three resins. All resins are pretreated with hydrochloric acid (HCl, 1 M), rinsed to neutrality with deionized water, washed with ethanol to remove any residual contaminants, and dried to a constant weight.

2.2. Loading Aluminum onto Resins

S957 was reacted with a 12.5% solution of AlCl₃·6H₂O at 65 °C for 4 h, with a solid–liquid ratio of 1:50. Following the loading process, the resin was filtered, washed to neutrality with deionized water, and dried at 50 °C for 12 h. The resulting aluminum-loaded resin was referred to as Al-S957. Aluminum loading onto D412 and D001 was performed using the same procedure.

2.3. Analysis and Characterization

The fluoride content in the solution was measured using the fluoride ion-selective electrode method (GB 7484-87), employing the 232-01 reference electrode and the PF-2-01 fluoride ion electrode from Leici. The pore size distribution and specific surface area of Al-S957 were determined through nitrogen adsorption–desorption tests performed using a microporosity analyzer (BET, Micromeritics 3Flex; Micromeritics Instrument Corporation, Norcross, GA, USA). The surface morphology of the material was analyzed through scanning electron microscopy (SEM, Hitachi Regulus 8100; Hitachi Limited, Tokyo, Japan). The aluminum content loaded onto the resin was quantified using inductively coupled plasma emission spectrometry (ICP-OES, Thermo Fisher iCAP PRO; Thermo Fisher Scientific, Waltham, MA, USA). X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha; Thermo Fisher Scientific, Waltham, MA, USA) and Fourier transform infrared spectroscopy (FTIR, Thermo Fisher Scientific Nicolet iS20; Thermo Fisher Scientific, Waltham, MA, USA) were utilized to examine chemical bonding and functional group species before and after fluoride adsorption.

2.4. Batch Adsorption Experiments

The adsorption of fluoride was carried out by adding 50 mg of Al-S957 in 50 mL of fluoride solution (4 mg/L), which was shaken thoroughly at 298 K to reach the adsorption equilibrium. The effect of pH on fluoride adsorption was evaluated by adjusting the initial pH of the feed solution from 3 to 11. The selective fluoride removal was examined in the presence of coexisting anions, including Cl⁻, NO₃⁻, HCO₃⁻, SO₄²⁻, and PO₄³⁻. Humic acid (HA), tannic acid (TA), sodium dodecyl benzene sulfonate (SDBS), L-tryptophan (Trp), and sodium alginate (SA) were introduced to assess their influence on the fluoride removal capacity. Experiments on the effects of coexisting anions and organic matter were also conducted for Al-D412 and Al-D001. Adsorption isotherms of fluoride were explored at different initial concentrations (2 to 30 mg/L) in the temperature range of 283, 298, and 313 K. To determine the adsorption kinetics, 0.5 g of Al-S957 was contacted with 500 mL of fluoride solution, and 1 mL of solution was extracted and analyzed at various time intervals for 10 h. Adsorption experiments were performed three times, with the average value taken as the adsorption capacity. The error bars were calculated using the standard error formula.

2.5. Fixed-Bed Column Experiments

To evaluate the performance of Al-S957 in practical applications, a fixed-bed column test was conducted by loading 7 mL of Al-S957 onto a glass column (14 mm diameter, 120 mm length). Real wastewater from a photovoltaic industrial park in Jiangsu province was passed through the column, with a peristaltic pump (Lange-YZ1515X, Baoding, China) used to maintain a steady flow rate. After adsorption saturation, Al-S957 was regenerated using 10 wt% AlCl₃·6H₂O solution and rinsed with deionized water until the effluent pH was neutral, preparing it for the next adsorption–regeneration cycle. For comparison, the same fixed-bed column experiments were also performed using Al-D412 and Al-D001 under identical conditions.

3. Results and Discussion

3.1. Material Characterization

As shown in Figure 1a,b, the SEM image revealed that the surface roughness of Al-S957 increases, which is beneficial for enhancing adsorption performance [26]. ICP-OES analysis showed that the aluminum contents of Al-S957, Al-D412, and Al-D001 were 1.05, 0.71, and 1.18 mmol/g, respectively. Notably, the aluminum loading capacity of Al-S957 exceeds that of Al-D412, which can be attributed to the presence of sulfonic acid groups in S957. These functional groups enhance the hydrophilicity of the resin and improve mass transfer kinetics during the loading process, thereby facilitating the chelation of a greater number of aluminum compared to D412 [11]. Although Al-D001 achieves an even higher aluminum loading, it exhibits substantial aluminum leaching during subsequent adsorption experiments, as discussed in detail in Section 3.3.

Nitrogen adsorption–desorption isotherms of S957 and Al-S957 are shown in Figure 2a,b. According to the IUPAC classification, the isotherms of both samples are type IV with an H3 hysteresis loop, indicating the presence of fissure-type pores [27,28]. The pore size distribution calculated by the BJH method shows that both S957 and Al-S957 possess predominantly mesoporous structures (Figure 2c,d), which are favorable for adsorption [29].

Table 1. Physical properties of S957 and Al-S957.

Adsorbents	BET Surface Area (m2/g)	Average Pore Diameter (nm)	Total Pore Volume (cm3/g)
S957	31.38	32.66	0.25
Al-S957	40.53	26.39	0.23

provides the physical properties of S957 and Al-S957. Following aluminum loading, the BET surface area of S957 increases significantly, facilitating better contact between fluoride and the active adsorption sites of Al-S957. Additionally, compared with S957, the average pore diameter and total pore volume of Al-S957 exhibited negligible changes, with a marginal decrease. This phenomenon is primarily attributed to the modification process, during which the hydrogen form of S957 is converted to the aluminum form. The incorporation of aluminum introduces additional positive charges into the resin framework, inducing a contraction of the matrix structure and thereby resulting in a minor reduction in pore size [30].

3.2. Effect of Initial pH on Fluoride Adsorption

Figure 3 illustrates the fluoride removal performance of Al-S957 across a pH range of 3 to 11. At pH 3, fluoride predominantly exists as HF that hinders the removal of fluoride [26]. Above pH 4, the fluoride adsorption capacity decreases gradually with increasing pH and declines sharply at pH 11. This decline is primarily attributed to the competition between OH⁻ and fluoride for adsorption sites [31]. Overall, Al-S957 has been shown to have a decent fluoride removal behavior throughout a wide pH range, which aligns with the typical pH of fluoride-contaminated water, indicating its potential for practical applications [32,33].

As shown in Figure 3, the equilibrium pH after adsorption was lower than the initial value, primarily due to the hydrolysis of aluminum on Al-S957 [8]. The isoelectric point of Al-S957 was determined (Figure S1), and the pH_pzc value was 2.85, indicating that the surface of Al-S957 was negatively charged over the tested pH range. However, the adsorption performance did not significantly decrease with pH exceeding pH_pzc, suggesting that electrostatic interactions are not the fundamental mechanism of fluoride removal by Al-S957. This conclusion is further supported by subsequent XPS characterization. In addition, the high selectivity of S957 observed in the coexisting anion experiments indicates that the adsorption mechanism involves specific ligand exchange rather than non-specific electrostatic attraction.

3.3. Effect of Coexisting Anions and Organics on Fluoride Adsorption

The composition of actual water is relatively complex, with inorganic salts and organic compounds influencing the defluorination efficiency of the adsorbent. To assess the selectivity of Al-S957, Al-D412, and Al-D001, common coexisting anions and organics were selected for comparative analysis.

Figure 4a shows the effects of various anions on Al-S957; these can be observed in the following order: HCO₃⁻ > PO₄³⁻ > SO₄²⁻ > Cl⁻, and NO₃⁻. The negligible impact of Cl⁻ and NO₃⁻ on the adsorption capacity of Al-S957 indicates that inner-sphere complexation is the predominant mechanism for fluoride removal. SO₄²⁻ may partially form an inner-sphere complex with aluminum, potentially competing with fluoride for active sites. PO₄³⁻, like fluoride, significantly inhibits fluoride removal due to its affinity for aluminum [34]. HCO₃⁻ exhibits the most pronounced inhibition, as hydrolysis-generated OH⁻ raises solution alkalinity, suppressing fluoride adsorption. Additionally, the similar ionic radii of HCO₃⁻ and fluoride contribute to competitive adsorption [35].

For Al-D412 (Figure 4b), a similar trend was observed. However, the fluoride removal of Al-D001 (Figure 4c) is significantly hindered by SO₄²⁻, Cl⁻, andNO₃⁻. The observed effect was speculated to result from the relatively weak binding of aluminum to D001, making it susceptible to interference from these anions, potentially leading to aluminum loss. To verify this hypothesis, an aluminum leakage experiment was conducted. Figure S2 reveals that aluminum leakage from Al-D001 increased notably with NaCl concentration, exceeding that from Al-S957. This is attributed to the relatively weak electrostatic interaction between the sulfonic acid groups of D001 and aluminum, whereas the incorporation of phosphate groups in S957 enhances its binding affinity.

Further static adsorption–desorption cycles with a 10 wt% AlCl₃·6H₂O solution confirmed the stability of Al-S957. The aluminum concentration decreased from 0.3215 mg/L after the first cycle to 0.3047 mg/L after the third, indicating consistent chelation performance. The pronounced difference in aluminum leakage also implies that sulfonic acid groups in S957 likely do not interact strongly with aluminum, with phosphonic acid groups being primarily responsible for chelation, consistent with prior findings [25].

The influence of SO₄²⁻ and NO₃⁻ on Al-D001 resembled that of Cl⁻, while the effects of HCO₃⁻ and PO₄³⁻ were less severe compared to Al-S957, possibly due to precipitate formation between leaked aluminum and these anions, indirectly aiding fluoride removal. In conclusion, Al-S957 demonstrates superior selectivity and is more suitable for the treatment of high-salt fluorinated water.

As for organic interference (Figure 4d–f), Al-S957 consistently maintained excellent defluorination performance across various organic concentrations. Benefiting from the hydrophilicity of the sulfonic acid group, Al-S957 inherits the excellent defluorination performance of Al-D001 with little effect of organic matter on it, while the performance of Al-D412 is more significantly impacted. Additionally, under the combined action of sulfonic acid groups and chelating groups, Al-S957 exhibits the highest adsorption capacity and strong resistance to organic interference.

Overall, these results demonstrate that Al-S957 effectively integrates the aluminum-chelating capability of Al-D412 and the organic resistance of Al-D001. As a result, Al-S957 exhibits enhanced fluoride adsorption efficiency and excellent selectivity in the presence of coexisting anions and organic matter. Due to its superior performance, subsequent experiments and data analyses were primarily conducted using Al-S957, while the fixed-bed tests also included Al-D001 and Al-D412.

3.4. Adsorption Kinetics

Figure 5 illustrates the fluoride adsorption capacities of Al-S957 over different contact times. A rapid adsorption phase occurred within the first 120 min, after which the rate gradually slowed, reaching equilibrium at approximately 240 min. To further elucidate the adsorption kinetics, both pseudo-first-order kinetic and pseudo-second-order kinetic models were applied to the experimental data. The corresponding equations are presented in Equations (1) and (2):

$$Q_{\mathrm{t}}=Q_{\mathrm{e}}\left(1-{\mathrm{e}}^{-k_{1}t}\right)$$

(1)

$$Q_{\mathrm{t}}={\left({\displaystyle \frac{1}{Q_{\mathrm{e}}}}+{\displaystyle \frac{1}{k_2{Q}_{\mathrm{e}}^{2}t}}\right)}^{-1}$$

(2)

where Q_t (mg·g⁻¹) represents the fluoride adsorption capacity at time = t, Q_e (mg·g⁻¹) represents the equilibrium adsorption capacity, and k₁ (min⁻¹) and k₂ (g·mg⁻¹·min⁻¹) represent the adsorption rate constants for the pseudo-first-order kinetic and pseudo-second-order kinetic models, respectively.

The fitting results and corresponding parameters are displayed in Figure 5 and Table S2. The data reveal that the pseudo-second-order kinetic model offers a superior fit, with a strong correlation coefficient greater than 0.99, indicating that chemisorption likely governs the fluoride adsorption process on Al-S957 [36].

Additionally, to investigate the influence of coexisting anions on adsorption kinetics, experiments were conducted at varying Cl⁻ concentrations. As shown in Figure 5 and Table S2, the rate constants derived from the pseudo-second-order model gradually decreased with increasing Cl⁻ concentration. This trend is attributed to the enhanced ionic strength, which increases the solution’s viscosity and impedes the diffusion of fluoride toward the active sites [37].

3.5. Adsorption Isotherms

The Langmuir and Freundlich isothermal adsorption models were utilized to fit the defluorination process, and the related equations are presented in Equations (3) and (4):

$$Q_{\mathrm{e}}={\displaystyle \frac{Q_{\mathrm{m}}{K}_{\mathrm{L}}{C}_{\mathrm{e}}}{1+K_{\mathrm{L}}{C}_{\mathrm{e}}}}$$

(3)

$$Q_{\mathrm{e}}=K_{\mathrm{F}}{C}_{\mathrm{e}}^{{\displaystyle \frac{1}{n}}}$$

(4)

where Q_m (mg·g⁻¹) denotes the maximum adsorption capacity, C_e (mg·L⁻¹) represents the equilibrium fluoride concentration, and n is the Freundlich linear index. K_L (L·mg⁻¹) and K_F (L·mg⁻¹) represent the equilibrium constants for the Langmuir and Freundlich model, respectively.

Figure 6 shows the adsorption isotherms, and the fitting parameters are summarized in Table S3. As the initial fluoride concentration increases, the adsorption capacity of Al-S957 also rises. At low fluoride concentrations, abundant adsorption sites on the adsorbent surface lead to a rapid increase in adsorption capacity. However, as the fluoride concentration continues to increase, adsorption sites gradually become saturated, causing the isotherm to level off.

Furthermore, the adsorption capacity increases with temperature, indicating that the defluorination process is endothermic. Among the two models, the Langmuir model fits the adsorption data better, suggesting a monolayer adsorption on a homogeneous surface [38]. The fitted K_L values consistently fall between 0 and 1, confirming that the adsorption is favorable [39]. The maximum adsorption capacity of Al-S957 at 298 K reaches 15.49 mg/g, surpassing many metal-modified adsorbents reported in the literature (

Table 2. The adsorption properties of metal-modified materials.

Materials	Qmax (mg/g)	Reference
La-loaded chelating resin	3.38	Robshaw et al. [40]
Activated carbon fibers modified with zirconium	6.12	Prathibha et al. [41]
La-modified biochar (La-TBC)	7.55	Li et al. [42]
Zr-impregnated magnetic chitosan graphene oxide (Zr–MCGO)	8.8	Liu et al. [43]
La-modified biosynthetic crystal (BC-La)	10.92	Zhou et al. [44]
Al-loaded TP207 resin	12.05	Rodríguez-Iglesias et al. [22]
Zr-loaded TP260 resin	12.1	Shin et al. [45]
Zr impregnated highly functional binary biopolymeric composite	12.13	Preethi et al. [46]
Al-loaded S957 resin	15.49	In this work

).

3.6. Thermodynamic Study

In order to further elucidate the adsorption characteristics, the thermodynamics of adsorption was investigated using isotherm data at different temperatures. Entropy change (ΔS), enthalpy change (ΔH), and free-energy change (ΔG) were calculated using Equations (5)–(7):

$$∆G=∆H-T∆S$$

(5)

$$∆G=-\mathrm{R}T\mathrm{l}\mathrm{n}{K}_{\mathrm{L}}$$

(6)

$$\mathrm{l}\mathrm{n}{K}_{\mathrm{L}}={\displaystyle \frac{∆S}{\mathrm{R}}}-{\displaystyle \frac{∆H}{\mathrm{R}T}}$$

(7)

where T represents the reaction temperature (K), K_L represents the adsorption equilibrium constant as defined in the Langmuir isotherm equation, and R (8.314 J·mol⁻¹·K⁻¹) is the universal gas constant.

The linear relationship between lnK and 1/T(K⁻¹) for the adsorption of fluoride by Al-S957 at different temperatures is depicted in Figure S3. The thermodynamic parameters are listed in Table S4. All negative values of ΔG indicate that the defluorination process is spontaneous [47]. In addition, with temperature increasing, the higher negative value of ΔG denotes a more energetically favorable adsorption [42]. The positive ΔH value verifies that the process is endothermic, while the positive ΔS value reflects increased freedom at the adsorption interface [48].

3.7. Adsorption Mechanism of Al-S957 for Fluoride

Figure 7a shows the infrared spectra before and after fluoride adsorption. After adsorption, the peaks at 3414 cm⁻¹ and 1169 cm⁻¹ were weakened, which was attributed to the stretching vibration of −OH [49]. This indicates that the adsorption process is accompanied by the breaking of −OH [50]. At the same time, the bending vibration of Al-O at 702 cm⁻¹ was weakened, while a new Al-F peak appeared at 838 cm⁻¹ after adsorption, indicating that the fluoride and Al-O bond interacted to form a new complex [51].

The XPS wide-scan spectra of Al-S957 before and after adsorption, shown in Figure 7b, confirmed the presence of F 1s, indicating fluoride adsorption onto the material [52,53]. Furthermore, the detection of S 2p and P 2p peaks confirms the coexistence of sulfonic and phosphonic acid groups, verifying the bifunctional nature of the S957 resin. As shown in Figure 7c, after adsorption, a new peak corresponding to Al-F appeared at 75.23 eV, while the binding energy of the Al-O peak shifted to 74.73 eV, and the corresponding peak ratios decreased [54]. As shown in Figure 7d, the O 1s peak of pristine Al-S957 was resolved into three distinct components at 530.81 eV, 531.93 eV, and 533.15 eV, corresponding to Al-O, −OH, and H₂O, respectively [55]. After fluoride adsorption, the binding energy of the O 1s peak decreased by 0.12 eV. The Al-O peak area percentage declined from 14.85% to 12.36%, attributed to the formation of Al-F compounds, where fluoride occupies the metal-binding sites [56]. This is consistent with the analysis of FTIR spectra. Additionally, the −OH peak area decreased from 60.45% to 48.27%, providing further evidence of −OH ligand exchange as a key mechanism in fluoride adsorption [8].

Overall, the superior fluoride adsorption performance of Al-S957 is primarily attributed to the strong complexation between aluminum and fluoride, which effectively displaces the original −OH ligands. This proposed adsorption mechanism, illustrated in Figure 8, is in good agreement with previous related studies [10,23,57].

3.8. Application Performance and Scalability of Al-S957

To evaluate the application potential of Al-S957, real photovoltaic industry wastewater was studied, with its composition shown in Figure 9. When the effluent fluoride concentration is controlled below 1 mg/L, Al-S957 can treat 170 BV of industrial wastewater, with only a minimal reduction in treated water volume after five adsorption–regeneration cycles. At the same time, the 10 wt% AlCl₃·6H₂O solution is demonstrated to be an effective desorption agent. For comparison, the fixed-bed column experimental results for Al-D001 and Al-D412 are presented in Figure S4. During the initial adsorption cycle, Al-D412 and Al-D001 treated only 80 BV and 7 BV of the real wastewater. Moreover, after multiple regeneration cycles, the treated water volumes for both materials also declined. These findings further confirm the superior fluoride removal performance of Al-S957 in treating real industrial wastewater.

To further assess the scalability and practical applicability of Al-S957 for industrial water treatment, a preliminary evaluation of its synthesis was performed. The required raw materials are commercially available, and a detailed cost analysis (Table S5) estimates the treatment cost at approximately USD 0.52 per liter of photovoltaic wastewater, confirming the economic viability. Environmentally, the synthesis avoids the use of toxic or hazardous reagents, and the aluminum species involved pose minimal environmental risk. Regarding reproducibility, the S957 resin consistently exhibits stable aluminum ion chelation performance across multiple synthesis batches. Taken together, these factors suggest that Al-S957 is a promising candidate for practical industrial-scale water treatment applications.

4. Conclusions

This study employed S957, which contains a phosphoric–sulfonic acid difunctional group, as the carrier of aluminum to achieve deep defluorination. Al-S957 has a maximum fluoride adsorption capacity of 15.49 mg/g at 298 K. The adsorption is a spontaneous endothermic process that occurs rapidly within 2 h. The process was minimally affected by pH, demonstrating consistent performance across a broad pH range. Even in the presence of competing anions and organics, it maintained high fluoride selectivity. Notably, under identical conditions, Al-D001 exhibited poor aluminum-binding stability, while Al-D412 was significantly affected by organic interference. Al-S957 has been found to efficiently remove fluoride through a ligand exchange mechanism. Furthermore, Al-S957 also can remove fluoride efficiently from actual photovoltaic wastewater and retain high removal efficiency after multiple adsorption–regeneration cycles. These results demonstrate that Al-S957, with superior adsorption selectivity and strong resistance to organic interference, has excellent fluoride removal potential in complex water.