Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2026 Sep 17;11(38):57394–57407. doi: 10.1021/acsomega.6c05790

Efficient Fluoride Removal Using Cerium-Modified Red Mud: Adsorption Mechanism, Interference Resistance and Environmental Safety

Chunyan Xu †,‡, Qingshuang Wei †, Zikai Yu †, Ziao Xiao †, Tingyu Wang †, Dating Huang §,*, Yizhen Liu §, Jianguang Liu ∥, Lang Liu †,‡
PMCID: PMC13628995  PMID: 42825063

Abstract

A cerium-modified red mud adsorbent (Ce-RM) was prepared from industrial red mud waste by acid activation, cerium-salt treatment and calcination, and applied to fluoride removal from aqueous solutions. Batch adsorption tests examined the effects of initial fluoride concentration, pH, adsorbent dosage, contact time and coexisting ions. SEM, BET, XRD, FTIR and XPS were used to characterize the morphology, pore structure, crystalline phases and surface chemistry of Ce-RM and relate these features to fluoride uptake. Ce-RM removed more than 97% of fluoride at an initial F– concentration of 20 mg·L–1, pH 3 and a dosage of 0.4 g·L–1, with an equilibrium adsorption capacity of about 56 mg·g–1. The adsorption kinetics followed the pseudo-second-order model. The equilibrium data fitted both Langmuir and Freundlich models, with Freundlich giving the better fit, indicating heterogeneous surface adsorption dominated by multilayer uptake with partial monolayer adsorption. FTIR and XPS results suggested that ligand exchange, Ce–F complexation and electrostatic attraction contributed to fluoride removal. Real wastewater tests showed good resistance to interference, supporting Ce-RM as a promising waste-derived adsorbent for fluoride-containing industrial wastewater.

graphic file with name ao6c05790_0016.webp

graphic file with name ao6c05790_0014.webp

1. Introduction

Fluoride is a trace element of physiological relevance, and appropriate intake plays an important role in maintaining human health. However, long-term exposure to excessive fluoride can cause serious adverse effects, including dental fluorosis and skeletal fluorosis. According to the World Health Organization (WHO) guidelines, the fluoride concentration in drinking water should not exceed 1.5 mg·L–1, while the standard in China is set at less than 1.0 mg·L–1. According to China’s Integrated Wastewater Discharge Standard (GB8978-1996), the permissible fluoride concentration in effluent ranges from 10 to 20 mg·L–1, depending on the discharge category. Industrial wastewater containing fluoride represents a major anthropogenic source of environmental fluoride pollution. Effluents generated from lithium-ion battery manufacturing and recycling, , aluminum fluoride production, semiconductor processing, electroplating, metallurgy and ceramic industries typically contain elevated fluoride concentrations. Upon discharge into aquatic environments, these effluents can substantially increase fluoride loading in receiving water bodies and, in certain cases, exceed the World Health Organization guideline value for fluoride in drinking water (1.5 mg·L–1) by a considerable margin, thereby posing significant risks to human health. Consequently, fluoride pollution has emerged as a critical global environmental challenge that demands immediate and sustained attention.

A range of technologies has been developed for fluoride removal from fluoride-containing wastewater, including ion exchange, electrodialysis, precipitation, coagulation, membrane separation, and adsorption. , Compared with other treatment technologies, adsorption has attracted considerable attention as a promising fluoride-removal strategy, owing to its cost-effectiveness, operational simplicity, high efficiency and applicability to low-concentration fluoride wastewater. In recent years, the performance benchmarks for absorptive defluoridation materials have continued to improve. MgO/biochar composites have achieved maximum fluoride adsorption capacities of 63.47–141.98 mg·g–1; nano-MgO-loaded activated carbon has reached an adsorption capacity of 121.1 mg·g–1 over a broad pH range; and activated-carbon-supported Ce–Al oxides have shown a Langmuir maximum adsorption capacity of 31.65 mg·g–1. − However, many high-capacity adsorbents still depend on synthetic carbon supports, engineered metal oxides, or relatively complex and relatively elaborate fabrication processes. In contrast, low-cost red mud-based adsorbents generally remain limited by their adsorption capacity: activated red mud increased fluoride uptake only from 0.18 to 2.6 mg·g–1, while sodium alginate-red mud showed an adsorption capacity of only 0.533 mg·g–1 under its optimal pH condition. , These comparisons indicate that the key challenge is not simply whether fluoride can be adsorbed, but how high fluoride affinity, low cost and solid-waste valorisation can be achieved within a single material.

Red mud, a high-volume alkaline solid waste generated during Bayer alumina production, is rich in Fe, Al and Si oxides and contains porous structures and surface hydroxyl sites, making it a potential low-cost substrate for treating fluoride-containing wastewater. However, unmodified red mud provides a limited number of adsorption sites and exhibits poor selectivity toward F–, making it difficult to achieve deep defluoridation. Cerium modification offers a more targeted route for chemical regulation, as cerium oxides and hydroxides can bind fluoride ions through Ce–F coordination and surface hydroxyl exchange. A recent study reported a fluoride adsorption capacity of 14.74 mg·g–1 for Ce-modified red mud, highlighting the promise of this strategy while leaving room for further improvement in removal efficiency and validation in real industrial wastewater.

On this basis, we prepared a Ce-RM adsorbent using inexpensive red mud as the main matrix and a small amount of Ce as the surface-active component, aiming to enhance red-mud valorisation and fluoride-removal selectivity. Single-factor batch adsorption experiments were conducted to investigate the effects of initial F– concentration, solution pH, adsorption time and coexisting ions on fluoride removal performance. The adsorption behavior and underlying removal mechanism of Ce-RM were further elucidated through adsorbent characterization, thermodynamic analysis, adsorption isotherm fitting and kinetic modeling. Finally, the practical applicability of Ce-RM was evaluated using aluminum fluoride wastewater to assess its suitability for treating this type of fluoride-containing effluent.

2. Materials and Methods

2.1. Materials

All chemicals in this study were analytical grade and used without further purification. Red mud (RM) is sourced from an alumina plant in Guangxi, China. Sulfuric acid (H2SO4, 98 wt %) was purchased from Sinopharm Chemical Reagent Co., Ltd. Cerium chloride heptahydrate (CeCl3·7H2O), sodium fluoride (NaF), sodium phosphate (Na3PO4), sodium sulfate (Na2SO4), and sodium chloride (NaCl) were supplied by Macklin Reagent Co., Ltd. Deionized water (resistivity 18.2 MΩ·cm) was used in all experiments to prepare solutions and rinse samples. A fluoride stock solution (100 mg·L–1) was made by dissolving NaF in deionized water and was then diluted to the needed concentrations for adsorption tests.

2.2. Method for Preparing Ce-RM Adsorbent

Red mud (RM) was dried at 378 K, ground and sieved through a 100-mesh standard sieve with a nominal aperture of approximately 150 μm. The sieved RM was acid-washed by immersion in 0.1 mol·L–1 dilute H2SO4 for 120 min. The acid-treated RM was then repeatedly washed with deionized water until the washing solution became neutral, and subsequently separated by centrifugation. The recovered solid was dried at 378 K for 360 min to constant weight and then calcined at 873 K for 120 min to complete pretreatment. The pretreated red mud was subsequently impregnated with 0.5 mol·L–1 CeCl3 solution at a solid-to-liquid ratio of 1:2 under stirring at 313 K for 120 min to promote sufficient liquid–solid contact and relatively uniform dispersion of Ce species, followed by oven drying at 378 K for 180 min. The dried sample was calcined in a muffle furnace at 873 K for 120 min to give the cerium-modified red mud adsorbent Ce-RM.

2.3. Characterization of Adsorbents Methods

Scanning electron microscopy (SEM, JSM-5800, Japan) were observed the surface shape, pore structure, and particle spread of Ce-RM. A surface area and porosimetry analyzer (Micromeritics ASAP 2020) was employed to determine the specific surface area and pore size distribution of the adsorbent. X-ray diffraction (XRD, Rigaku Ultima IV, Japan) gave the phase composition, crystallinity, crystal structure, and mineral phases of the adsorbent. Qualitative analysis of functional groups on the adsorbent surface was performed using Fourier transform infrared spectroscopy (FT-IR, Thermo Nicolet IS5, USA). X-ray photoelectron spectroscopy (XPS, Rigaku Ultima IV, Japan) showed the elemental makeup, chemical states, and surface chemical setting of the samples.

2.4. Batch Adsorption Experiments

A 100 mL fluoride solution (made from NaF) went into a 100 mL screw-cap reagent bottle. The solution pH was set with dilute H2SO4, Ce-RM was added, the effects of solution pH (1–7), initial fluoride concentration (10–150 mg·L–1), contact time (0–100 min), and coexisting ions on fluoride removal were tested. Except for thermodynamic experiments, all adsorption tests were conducted in a thermostatic water bath at 298 K with a shaking speed of 200 rpm for 1 h. After equilibrium, the suspension was filtered through a 0.45 μm membrane, and the residual fluoride concentration was determined using a fluoride ion-selective electrode (PF-2-01). The fluoride removal efficiency (R) and the adsorption capacity at equilibrium (qe) were calculated according to eqs and , respectively

R=(C0−Ce)C0×100% 1
qe=(C0−Ce)Vm 2

where C 0 and C e (mg·L–1) represent the initial fluoride concentration in the simulated wastewater and the equilibrium fluoride concentration in the supernatant after adsorption, respectively; V (L) is the volume of the fluoride solution; m (g) is the dosage of the Ce-RM adsorbent; q e (mg·g–1) is the adsorption capacity at equilibrium; and R (%) is the fluoride removal efficiency.

2.5. Adsorption Kinetics and Isotherms

Under an initial fluoride concentration of 20 mg·L–1, solution pH of 3 and Ce-RM dosage of 0.4 g·L–1, adsorption was conducted under constant-temperature shaking for 60 min. Samples were collected at 5, 10, 15, 20, 30, 40, 50, and 60 min, filtered through a 0.45 μm membrane, and analyzed to determine the residual F– concentration in solution. The experimental data were fitted using the pseudo-first-order and pseudo-second-order kinetic models, described by eqs and , respectively, to further elucidate the rate-controlling steps and equilibrium distribution characteristics of the adsorption process.

qt=qe(1−e−k1t) 3
qt=k2qe2t1+k2qet 4

To elucidate the adsorption behavior and distribution characteristics of F– on the Ce-RM surface, Simulated water samples with concentration gradients ranging from 10 to 120 mg·L–1 were prepared. After adsorption equilibrium was reached, the suspensions were filtered through a 0.45 μm membrane, and the residual F– concentration was measured to calculate the equilibrium adsorption capacity. The obtained equilibrium data were fitted using the Langmuir and Freundlich adsorption isotherm models, as described by eqs and , respectively.

qe=qmKLCe1+KLCe 5
qe=KFCe1/n 6

here, t represents the contact time, and q t (mg·g–1) is the adsorption capacity at time t. K 1 (min–1) is the pseudo-first-order rate constant, and k 2 is the pseudo-second-order rate constant. In the isotherm models, C e (mg·L–1) denotes the equilibrium concentration of the adsorbate, q m (mg·g–1) represents the theoretical monolayer adsorption capacity in the Langmuir model, and K L (L·mg–1) is the Langmuir constant related to the adsorption energy. By contrast, K F [(mg·g–1)·(L·mg–1)1/n] and n are the Freundlich empirical constants that describe the adsorption capacity and adsorption intensity, respectively.

3. Results and Discussion

3.1. Characterization of Adsorbents

To evaluate the morphological evolution induced by cerium modification and fluoride adsorption, the surface microstructures of pristine red mud (RM), cerium-modified red mud prepared through acid activation, cerium-salt impregnation and calcination (Ce-RM), and fluoride-loaded Ce-RM (Ce-RM-F–) were characterized by scanning electron microscopy (SEM). Figure a shows that pristine red mud (RM) has characterized by dense aggregates of block-like and rod-like particles with a broad particle size distribution. The compact and rough surface, together with well-defined rod-like crystalline features, indicates the characteristic mineral aggregation of untreated red mud. Similar mineral aggregation and limited intrinsic adsorption capacity have also been reported for red mud-based fluoride adsorbents. After acid leaching, calcination, and subsequent cerium modification, the surface morphology of Ce-RM undergoes significant changes, as shown in Figure b. Compared with RM, Ce-RM shows a rougher and more loosely aggregated surface, with visible interparticle voids and improved particle dispersion. Fine particles are uniformly anchored on the substrate surface, producing abundant surface wrinkles and external surface defects. This structural reorganization increases the accessible surface area and exposes additional active sites, thereby contributing to the enhanced fluoride removal performance of Ce-RM. The SEM image of fluoride-loaded Ce-RM (Ce-RM-F–) is shown in Figure c. After fluoride adsorption, Ce-RM retains its rough and loosely aggregated morphology, and no obvious structural collapse is observed. Compared with Ce-RM, Ce-RM-F– shows slightly stronger local particle aggregation and surface coarsening, with some visible interparticle voids partially covered by fine particles. These changes suggest that fluoride adsorption does not destroy the overall structure of Ce-RM, but may induce local surface coverage or rearrangement during the adsorption process.

1.

1

Open in a new tab

SEM images of native red mud (a); Ce-RM before (b) and after (c) adsorption.

The pore structures of pristine red mud (RM), cerium-modified red mud (Ce-RM), and fluoride-loaded Ce-RM (Ce-RM-F–) were systematically characterized using N2 adsorption–desorption analysis with an automated specific surface area analyzer, as shown in Figure . The N2 adsorption–desorption isotherms in Figure a indicate that RM, Ce-RM, and Ce-RM-F– all exhibit distinct mesoporous adsorption characteristics. For all three samples, the adsorbed volume increases gradually in the low relative-pressure region, indicating that micropores make only a limited contribution to the overall pore structure. When the relative pressure (P/P0) exceeds 0.8, the adsorption volume increases sharply and a distinct adsorption–desorption hysteresis loop appears, suggesting the presence of mesoporous/macroporous structures formed by particle aggregation, interlayer voids and irregular fissures within the materials. Relative to pristine red mud, Ce-RM shows a higher overall N2 adsorption uptake, suggesting that acid washing, calcination and cerium incorporation disrupted the compact red mud matrix. This structural modification removed some soluble or inert components, exposed additional pores and surface sites, and enhanced pore accessibility. After fluoride adsorption, Ce-RM-F– exhibits a further increase in N2 uptake in the high relative-pressure region, indicating that F– adsorption does not cause obvious pore blockage or framework collapse. The material retains a good pore structure. Figure b shows that the pore-size distribution is consistent with the evolution of the adsorption–desorption isotherms. RM exhibits a relatively low pore volume and an indistinct pore-size distribution. After cerium modification, Ce-RM displays a markedly increased pore volume in the mesoporous region, together with a broader pore-size distribution, indicating that the modification promotes the development of a more evolved hierarchical pore structure. For Ce-RM-F–, the pore volume in the small-mesopore range is further enhanced while the broad mesopore-macropore distribution is largely retained. This suggests that F– is more likely immobilized on Ce-containing active sites and pore-wall surfaces, rather than blocking the pore channels through simple physical deposition. These findings suggest that cerium modification enriches red mud with active sites capable of coordination exchange, electrostatic adsorption and surface complexation with fluoride ions, while also enhancing interparticle mass-transfer channels within the material. The larger open pores facilitate the diffusion of F– from the solution into the interior of the material, while the abundant mesoporous structure increases the adsorption interface and effective contact area. Therefore, the BET results provide pore-structure-level evidence that acid washing-calcination and cerium modification markedly enhance the pore accessibility and surface reactivity of red mud, thereby supporting its improved defluorination performance.

2.

2

Open in a new tab

N2 adsorption/desorption isotherms (a) and pore size distribution curves (b) of RM, Ce-RM, Ce-RM-F–.

Fourier-transform infrared spectroscopy (FTIR) was used to analyze the surface functional groups of pristine red mud (RM), Ce-modified red mud prepared through acid activation, cerium-salt impregnation and calcination (Ce-RM), and fluoride-loaded Ce-RM (Ce-RM-F–). As shown in Figure a, the broad band around 3433 cm–1 is assigned to the stretching vibrations of surface hydroxyl groups and adsorbed water molecules. In this context, Fe–OH and Al–OH refer to hydroxylated surface sites coordinated to Fe or Al centers on oxide/oxyhydroxide surfaces, rather than to lattice oxygen in the oxide phase or to separate bulk hydroxide phases. The peak near 1633 cm–1 corresponds to the H–O–H bending vibration of physically adsorbed water. In pristine red mud, the band at around 1448 cm–1 is assigned to CO3 2– stretching vibrations in carbonate minerals such as calcite, whereas the peak near 991 cm–1 corresponds mainly to Si–O, Si–O–Al or Al–O vibrations in aluminosilicate phases. , These features suggest that untreated red mud contains carbonate minerals and aluminosilicate frameworks. In Ce-RM, prepared by acid washing, calcination and cerium modification, the carbonate band near 1440 cm–1 is substantially attenuated, reflecting the partial dissolution or removal of carbonate species during acid treatment. The enhanced band at 1018–1033 cm–1 indicates a change in the local bonding environment of Si–O/Al–O and surface metal–oxygen/hydroxyl groups after activation and Ce loading, rather than directly proving the formation of new adsorption sites. Together, these changes suggest that acid activation removes part of the carbonate impurities and modifies the surface chemical environment of red mud. After fluoride adsorption, several characteristic bands of Ce-RM-F– change measurably. The hydroxyl stretching vibration band is slightly weakened and shifted, which may result from hydrogen-bond rearrangement and/or interaction between F– and surface hydroxyl groups.

3.

3

Open in a new tab

Ftir (a) and XRD (b) of RM, Ce-RM, and Ce-RM-F–.

The crystalline structures of the samples were examined by XRD Figure b. Pristine red mud shows multiple diffraction peaks assigned to hematite, calcite, sodium aluminosilicates, hydrogarnet-type phases and silica, confirming the coexistence of iron oxides, aluminosilicates and carbonates in the Bayer red mud matrix. These phases are commonly reported for Bayer-process red mud. , After acid washing, calcination and cerium incorporation, the diffraction pattern of Ce-RM becomes weaker and broader than that of RM. Several carbonate- and sodium aluminosilicate-related reflections are attenuated, indicating partial dissolution of carbonate species and partial disruption of the original mineral framework during acid activation. Weak reflections assigned to Ce-containing oxide/silicate species are also observed, suggesting that cerium species were introduced into the red mud matrix, although part of the Ce-containing phase may be poorly crystalline or highly dispersed. After fluoride adsorption, the XRD pattern of Ce-RM-F– changes markedly. Sharp and intense reflections appear, mainly corresponding to hematite and Ce-containing oxide/silicate phases, rather than showing only limited changes in diffraction intensity. This suggests that contact with the fluoride solution changes the relative contribution or apparent crystallinity of the mineral phases in the solid residue. Such changes may arise from selective dissolution of less crystalline components, surface deposition, or enrichment of Fe- and Ce-containing crystalline domains.

X-ray photoelectron spectroscopy (XPS) was further employed to probe the changes in surface chemical states before and after fluoride adsorption, thereby elucidating the F– removal mechanism on Ce-RM. As shown in the XPS survey spectra (Figure a), Ce, O, Fe, Al, Ti, Ca and C are detected on the Ce-RM surface. After fluoride adsorption, a distinct F 1s signal appears in the survey spectrum, indicating that fluoride is retained on the adsorbent surface. The O 1s spectra were fitted into M-O, M–OH and adsorbed H2O/oxygen-containing components (Figure b). Before adsorption, the O 1s spectrum is dominated by M–OH, indicating abundant hydroxylated metal sites on Ce-RM. After fluoride adsorption, the relative contribution of M–OH decreases from 61.08% to 42.93%, while the adsorbed H2O/oxygen-containing component increases from 14.78% to 44.81%. This redistribution indicates a change in the surface oxygen environment during fluoride uptake. Together with the FTIR shifts in the hydroxyl and metal–oxygen vibration regions, the O 1s result suggests that surface hydroxyl groups participate in fluoride adsorption, most likely through hydroxyl exchange and local hydrogen-bond rearrangement. The high-resolution Ce 3d spectra (Figure c) show that Ce was present as mixed Ce3+/Ce4+ species on the Ce-RM surface before adsorption. The fitted Ce3+ components are assigned to the peaks at approximately 885.13 and 903.04 eV, while the Ce4+ components are assigned to the peaks at approximately 881.88, 888.60, 897.80, 900.46, 907.13, and 916.06 eV. After fluoride adsorption, the Ce 3d envelope changes, with the Ce4+-related components appearing at approximately 884.49, 900.50, 903.58, 909.57, and 918.52 eV, together with Ce3+-related components at approximately 888.13, 891.19, and 903.04 eV. The change in the Ce3+/Ce4+ distribution after fluoride uptake indicates that the local electronic environment of Ce was altered during adsorption. This variation may be associated with electron redistribution caused by fluoride coordination with Ce-containing surface sites. , The F 1s spectrum provides direct evidence for surface fluoride binding (Figure d). No clear F 1s signal is observed before adsorption, whereas Ce-RM after adsorption shows a peak at approximately 686.85 eV. This peak is assigned to surface-bound fluoride associated with metal-fluoride interactions. Combined with the Ce 3d and O 1s results, the F 1s signal indicates that fluoride is immobilized mainly through coordination with Ce-containing sites and exchange with surface hydroxyl groups. Similar fluoride-binding pathways have been reported for Fe–Al–Ce hydrous oxides and rare-earth-modified alumina. The Fe 2p spectra show Fe 2p3/2 and Fe 2p1/2 envelopes with satellite features (Figure e), which are characteristic of Fe-containing oxide or oxyhydroxide phases. This assignment is consistent with the hematite-related reflections observed in the XRD pattern. Because the Fe 2p region contains overlapping multiplet and satellite structures, it is used here for qualitative phase identification rather than for quantitative Fe2+/Fe3+ analysis. The Al 2p spectrum before adsorption is dominated by an Al–O component at approximately 73.79 eV (Figure f). After fluoride adsorption, the Al 2p envelope shifts to higher binding energy and a component at approximately 76.66 eV appears, which can be associated with a fluorinated Al environment, while the residual Al–O component remains near 74.51 eV. This suggests that Al-related surface sites are also affected by fluoride uptake. Taken together, the spectroscopic and structural results indicate that fluoride uptake by Ce-RM is governed mainly by surface reactions rather than by a single bulk-phase transformation. The FTIR and O 1s results point to changes in surface hydroxyl and metal–oxygen environments, while the appearance of F 1s and the variation in Ce 3d show that fluoride is bound to Ce-containing surface sites. The XRD patterns further suggest that adsorption is accompanied by changes in Fe/Ce-bearing crystalline domains, without providing direct evidence for a separate bulk fluoride phase. These observations support a mechanism in which fluoride is immobilized through hydroxyl exchange and coordination with Ce sites, with Fe- and Al-containing phases acting as part of the reactive red mud matrix.

4.

4

Open in a new tab

XPS spectra of Ce-modified red mud before and after fluoride adsorption (a) Survey; (b) O 1s; (c) Ce 3d; (d) F 1s; (e)Fe 2p; (f) Al 2p.

3.2. A Single-Factor Experimental Study on Fluoride Ion Adsorption

3.2.1. The Effect of Red Mud on F– Adsorption before and after Modification

As shown in the Figure , the pristine red mud (RM) exhibited a relatively low removal capacity toward F–, with an adsorption efficiency of only 28.69% and an equilibrium adsorption capacity of 13.49 mg·g–1, indicating that unmodified red mud has limited capability for fluoride removal. In contrast, the acid-washed, calcined, and cerium-modified material (Ce-RM) achieved a markedly enhanced F– removal efficiency of 97%, with an adsorption capacity reaching 55.91 mg·g–1, which was substantially higher than that of the pristine red mud. These results demonstrate that the modification process effectively improved the surface reactivity and pore structure of red mud, while introducing cerium-based active sites with strong affinity toward fluoride. Consequently, more accessible sites were provided for coordination exchange, surface complexation, and electrostatic adsorption of F–, thereby significantly overcoming the limitations of the original red mud, including its low adsorption capacity and insufficient selectivity.

5.

5

Open in a new tab

Comparison of adsorption performance before and after med mud modification.

3.2.2. The Effect of Adsorption Time

Figure shows that Ce-RM exhibited rapid fluoride uptake during the early stage of adsorption, with both removal efficiency and adsorption capacity increasing sharply within the first 10–30 min. The fluoride removal efficiency reached 81% at 10 min, with an adsorption capacity of approximately 49 mg·g–1, and increased to 95% at 20 min before approaching 98% at 30 min. At a contact time of 60 min, Ce-RM reached its maximum fluoride removal efficiency and adsorption capacity, suggesting that adsorption equilibrium was effectively established. From 60 to 180 min, the fluoride removal rate stayed above 98% and the adsorption capacity remained around 60 mg·g–1 with almost no further increase, indicating that most active sites on the adsorbent had already been occupied by F– and that the system had reached adsorption equilibrium. Overall, the adsorption reached equilibrium at 60 min; To ensure that equilibrium was fully attained in the subsequent adsorption experiments, 60 min was selected as the standard reaction time.

6.

6

Open in a new tab

Effect of adsorption time on adsorption efficiency and adsorption capacity.

3.2.3. The Effect of Ce-RM Dosage

The effect of adsorbent dosage on fluoride adsorption was studied at pH 5 with an initial fluoride concentration of 20 mg·L–1, a reaction time of 60 min, and a temperature of 298 K, and the results are shown in Figure .

7.

7

Open in a new tab

Effect of different Ce-RM dosages.

As the Ce-RM dosage increased from 0.2 to 1.0 g·L–1, fluoride removal first increased and then declined. The highest removal efficiency of 21.86% was obtained at 0.4 g·L–1, whereas further dosage increases reduced the removal efficiency to 9.68%. This trend can be attributed to the fact that, at relatively low dosages, increasing the amount of adsorbent provides more available active sites, thereby enhancing fluoride removal efficiency. However, at excessive dosages, adsorbent particles are prone to aggregation, which reduces the number of effectively accessible adsorption sites. Meanwhile, because the total amount of F– in solution is limited, the surplus adsorbent cannot be fully utilized; consequently, the removal efficiency no longer increases and may even decline slightly. In contrast, the equilibrium adsorption capacity (qe) decreases continuously with increasing adsorbent dosage, declining from 16.16 mg·g–1 at 0.2 g·L–1 to 2.3 mg·g–1 at 1.0 g·L–1. This decrease occurs because qe represents the amount of F– adsorbd per unit mass of adsorbent. As the adsorbent dosage increases, the limited amount of F– is distributed over a larger number of adsorbent particles, resulting in a lower fluoride loading per unit mass of adsorbent. Overall, the optimal Ce-RM dosage in this study was 0.4 g·L–1, at which a relatively high F– removal efficiency was achieved while maintaining efficient utilization of the adsorbent.

3.2.4. The Effect of Different Solution pH Levels

Solution pH is a key factor governing F– adsorption by the Ce-RM adsorbent. To evaluate its influence on the reaction performance of the material, fluoride removal by Ce-RM was examined over an initial pH range of 1–7 at 298 K, with an adsorbent dosage of 0.4 g·L–1, an initial F– concentration of 20 mg·L–1, a reaction time of 60 min. The experimental results are shown in Figure a. When the solution pH was in the range of 1–3, the F– removal efficiency by Ce-RM remained above 97%, while the adsorption capacity was maintained at 55–58 mg·g–1, indicating that acidic conditions favor the adsorptive removal of F–. To further elucidate the mechanism by which pH affects the adsorption performance, the zeta potential of Ce-RM was measured under different pH conditions, as shown in Figure b. The point of zero charge (pHpzc) of Ce-RM was 5.60, indicating that the adsorbent surface was overall positively charged when the solution pH was below 5.60, whereas it gradually became negatively charged at pH values above 5.60. Combined with the zeta-potential results, these findings indicate that under strongly acidic conditions (pH 1–3), the Ce-RM surface carries abundant positive charges, which favors the attraction of negatively charged F–. Meanwhile, surface hydroxyl groups on metal oxides such as Fe, Al and Ce are readily protonated to form positively charged active sites­(≡MOH2+), thereby enhancing the electrostatic adsorption of F– and its ligand–exchange interactions with the material. , As the pH increased from 3 to approximately 5, the F– removal efficiency of Ce-RM decreased markedly, with a particularly pronounced decline at pH 5. This behavior is primarily attributable to the fact that pH 5 is close to the point of zero charge (pHpzc) of Ce-RM. Under this condition, the positive charge density on the adsorbent surface is markedly reduced, weakening its electrostatic attraction toward F–. Meanwhile, as the pH increases, the number of protonated hydroxyl sites (≡MOH2+) on the material surface gradually decreases, accompanied by a corresponding reduction in the active sites available for ligand exchange or surface complexation with F–. In addition, the OH– concentration in solution gradually increases, and OH– competes with F– for metal active sites such as Ce, Fe and Al, further inhibiting F– adsorption. Therefore, near pH 5, the combined effects of reduced positive surface charge, deprotonation of hydroxyl sites and competitive adsorption by OH– led to a marked decrease in the fluoride-removal efficiency of Ce-RM. When the pH further increased beyond the point of zero charge (pHpzc = 5.60), the Ce-RM surface became negatively charged, resulting in electrostatic repulsion with the likewise negatively charged F–. This repulsion hindered the migration and binding of F– to the adsorbent surface, thereby markedly reducing the fluoride-removal performance. Overall, acidic conditions are more favorable for F– adsorption by Ce-RM. Although superior adsorption performance was obtained at pH 1–2, controlling the solution pH at approximately 3 is more feasible for practical application, considering acid consumption, equipment corrosion and operating costs in real wastewater treatment.

8.

8

Open in a new tab

Effect of pH on the adsorption of F– by Ce-RM (a) and the zeta potential plot of Ce-RM (b).

3.2.5. The Effect of Coexisting Ions

In practical fluoride-bearing wastewater systems, a range of coexisting anions, including PO4 3–, SO4 2–, NO3 –, Cl– and HCO3 –, are typically present. These anions may compete with F– for adsorption sites and occupy surface-active sites on the adsorbent or altering the interfacial adsorption environment, thereby weakening the selective uptake of fluoride and ultimately reducing the overall fluoride-removal efficiency. As shown in Figure , the competitive adsorption experiments revealed distinct effects of different coexisting anions on fluoride removal. The competitive adsorption experiments showed that, in the absence of coexisting anions, Ce-RM achieved an F– adsorption efficiency of approximately 97%, indicating its intrinsically favorable defluoridation performance. After the introduction of coexisting anions at different concentrations, the F– adsorption efficiency varied to different extents, suggesting clear differences in the competitive effects of these anions during the adsorption process. Considering that bicarbonate (HCO3 –) is the predominant carbonate species under the experimental pH conditions, HCO3 – was selected to represent carbonate-type coexisting anions in this study, and a separate CO3 2– experimental group was therefore not included. As the HCO3 – concentration increased from 20 to 100 mg·L–1, the F– adsorption efficiency steadily decreased from about 70.50% to about 38.70%, showing a much larger decline than that observed for Cl–, SO4 2–, NO3 – and PO4 3–. This suggests that HCO3 – strongly inhibits fluoride removal. The inhibition may be related to the increase in solution pH, changes in the surface charge of the adsorbent, and stronger competition between F– and OH–/HCO3 – for adsorption sites, which weakens the binding of F– to surface metal-hydroxyl groups or metal active sites. Therefore, the presence of HCO3 – markedly reduces the fluoride removal performance of the material. PO4 3– also noticeably affected F– adsorption. As its concentration increased, the adsorption efficiency declined and reached about 64.62% at 100 mg L–1, likely because PO4 3– readily coordinates with metal sites such as Ce, Fe and Al, occupying part of the available adsorption sites.In contrast, Cl–, NO3 – and SO4 2– had relatively weak effects on F– adsorption. Over the concentration range of 20–100 mg·L–1, the adsorption efficiency remained at about 88–96%, indicating that Ce-RM maintained good adsorption stability in the presence of these common coexisting anions.Overall, the inhibitory effect of the different anions on F– adsorption roughly followed the order HCO3 – > PO4 3– > Cl–, NO3 – and SO4 2–. Therefore, the effects of bicarbonate and phosphate should be given particular attention when treating fluoride-containing wastewater.

9.

9

Open in a new tab

Effect of coexisting ions on F– adsorption.

3.3. Adsorption Kinetics and Isotherms

3.3.1. Adsorption Kinetics

To further elucidate the adsorption behavior of F– on Ce-RM, the experimental data were fitted using nonlinear pseudo-first-order and pseudo-second-order kinetic models, with the corresponding results presented in Figure and Table . As shown in Figure , Ce-RM exhibited a pronounced rapid adsorption behavior toward F–. During the initial stage of the reaction­(0–10 min), the adsorption capacity increased rapidly, then gradually approached equilibrium, and reached a stable state at around 60 min. This phenomenon may be ascribed to the abundance of vacant active sites on the adsorbent surface during the initial stage of adsorption. As the process progressed, progressive occupation and eventual saturation of these sites reduced the availability of binding sites, thereby slowing the adsorption rate. To further investigate the adsorption mechanism, the experimental data were fitted using pseudo-first-order and pseudo-second-order kinetic models. The kinetic fitting results revealed that the pseudo-first-order model exhibited a lower correlation coefficient (R 2 = 0.979) than the pseudo-second-order model (R 2 = 0.994). Moreover, the equilibrium adsorption capacity predicted by the pseudo-second-order model (q e = 61.312 mg·g–1) was in closer agreement with the experimentally determined value (q e = 59.875 mg·g–1), indicating that the pseudo-second-order kinetic model provided a more accurate description of F– adsorption by Ce-RM. The pseudo-second-order kinetic model is generally predicated on the assumption that adsorption is governed primarily by chemisorption, with the adsorption rate controlled by electron sharing or exchange between surface active sites and the adsorbate. Accordingly, F– removal by Ce-RM can be inferred to proceed predominantly through chemisorption, involving mechanisms such as Ce–F coordination, together with ion exchange or complexation between surface hydroxyl groups and F–. Furthermore, cerium incorporation markedly increased the density of surface active sites and enhanced the affinity of the material for F–, thereby promoting its adsorption performance. Taken together, F– adsorption by Ce-RM was governed predominantly by chemisorption, accompanied by rapid surface reactions, and its kinetic behavior was more accurately described by the pseudo-second-order model.

10.

10

Open in a new tab

Adsorption kinetics curve of Ce-RM.

1. Kinetic Model Parameters for Ce-RM.
pseudo-first-order Kinetic Model q e (mg·g–1) k 1 (min–1) R 2
q t = q e (1- e – k 1 t ) 57.884 0.297 0.979
Pseudo-Second-Order Kinetic Model qe (mg·g–1) k 2 (g·mg–1·min–1) R 2
q t = k 2 q e 2 t /(1+ kq e t ) 61.312 0.01 0.994
Open in a new tab

3.3.2. Adsorption Isotherm

The fitted curves in Figure a,b, together with the fitting parameters presented in Table , indicate that F– adsorption on Ce-RM over 298–313 K can be well described by both the Langmuir and Freundlich isotherm models, with the Freundlich model providing the better representation of the experimental data. At 298, 303, and 313 K, the Freundlich model yielded higher coefficients of determination (R 2 = 0.983, 0.976 and 0.967, respectively) than the Langmuir model (R 2 = 0.950, 0.933 and 0.924), indicating that F– adsorption on Ce-RM is governed by heterogeneous surface sites and is more consistent with multilayer adsorption. The Freundlich constant (KF) varied from 54.10 to 57.38 L·mg–1, reflecting the high adsorption capacity of Ce-RM. In addition, the 1/n values ranged between 0.178 and 0.183, remaining below unity at all tested temperatures, which indicates that the adsorption process was thermodynamically favorable. The Langmuir-derived maximum adsorption capacities (q max,cal = 110.70, 107.75, and 107.03 mg·g–1) were comparable to the corresponding experimental values (q max,exp = 121.46, 119.50, and 117.50 mg·g–1). This indicates that monolayer adsorption behavior still exists to some extent during the adsorption process. The slight decline in the experimental maximum adsorption capacity with increasing temperature from 298 to 313 K suggests the exothermic nature of the adsorption process. A comprehensive analysis indicates that the adsorption of F– by cerium-modified red mud following acid leaching and calcination primarily occurs at heterogeneous active sites on the surface, with multilayer adsorption being the dominant mechanism, accompanied by a certain degree of monolayer adsorption behavior.

11.

11

Open in a new tab

Adsorption isotherm models for Ce-RM:(a) Freundlich model and (b) Langmuir model.

2. Parameters of the Langmuir and Freundlich Isotherm Models for Ce-RM.
  langmuir isotherm freundlich isotherm
T(K) q max.exp, (mg·g–1) q max,cal (mg·g–1) K L (L·mg–1) R 2 K F (L·mg–1) 1/n R 2
298 K 121.46 110.70 2.555 0.950 54.10 0.183 0.983
303 K 119.50 107.75 2.095 0.933 55.09 0.181 0.976
313 K 117.50 107.03 1.709 0.924 57.38 0.178 0.967
Open in a new tab

In addition, the maximum equilibrium adsorption capacity (q max) of Ce-RM is compared with other reported adsorbents, and the comparative result is presented in Table . Overall, Ce-RM shows good fluoride adsorption performance. Its maximum adsorption capacity is higher than that of some conventional red-mud-based and composite adsorbents, although it remains lower than that of several high-performance Ce-based MOFs, Mn-modified red mud, and CeO2-loaded geopolymer materials. Because the reported studies used different pH values, dosages, temperatures, and contact times, adsorption capacity alone is not sufficient for a direct comparison. Considering raw-material cost, preparation requirements, resource utilization, and removal efficiency, Ce-RM remains a promising adsorbent for fluoride-containing wastewater.

3. Comparison of F– Adsorption by Mn-RM and Other Adsorbents.
adsorbent pH dosage(g·L–1) T(K) t(min) qmax(g·L–1) reference
Ce-RM 3 0.4 298 60 121.46 this study
MIL-96(RM) 5.0 0.5 293 720 82.65
RM@GMs 2.0 1.0 298 60 76.57
CeO2@SGMs 2.0 1.0 298 90 121.77
Ce/Mn@MOFs 3.0 0.2 293 60 129.20
Ce@MOFs 3.0 0.2 293 60 154.08
Al-MOF-PUF@BC 5.5 2.0 298 480 16.52
Mn-RM 2.0 0.4 298 60 141.83
Open in a new tab

3.4. Suitability of Ce-RM Adsorbents for Real-World Water Samples

The fluoride concentration in the AlF3 solution, determined using a fluoride ion-selective electrode, was 27.82 mg·L–1. As show in Figure , in the actual AlF3 wastewater system, Ce-RM achieved an F– adsorption efficiency of 95.38% and an adsorption capacity of 66.52 mg·g–1, demonstrating its strong applicability in complex real-water matrices. Ion chromatography in the Table showed that Cl–, NO3 – and SO4 2– were the predominant coexisting anions in the actual AlF3 wastewater, with concentrations of 74.41, 9.72, and 1399.16 mg·L–1, respectively, whereas PO4 3– was not detected. Based on the interference experiments, it was found that Cl–, NO3 – and SO4 2– had little effect on fluoride removal efficiency within the range of 20–100 mg·L–1, whereas PO4 3– exhibited a significant concentration-dependent inhibitory effect. Therefore, despite the high SO4 2– concentration in the actual AlF3 wastewater, Ce-RM maintained efficient fluoride removal, indicating that SO4 2–, Cl– and NO3 – exerted limited competitive effects on fluoride adsorption by Ce-RM. Meanwhile, the absence of PO4 3– in the AlF3 wastewater prevented its strong competitive occupation of Ce active sites, thereby favoring the retention of high fluoride-capture capacity by Ce-RM in the AlF3 wastewater system.

12.

12

Open in a new tab

Comparison of Ce-RM removal efficiency in simulated and actual wastewater.

4. Anion Concentration in AlF3 Wastewater.

number retention time (min) name concentration (mg·L–1)
1 4.39 Cl– 74.41
2 7.22 NO3 – 9.72
3 8.88 SO4 2– 1399.16
4 - PO4 3– -
Open in a new tab

3.5. Release of Metals from the Supernatant after Ce–RM Adsorption and Assessment of Discharge Safety

To evaluate the potential risk of metal release from Ce-RM during fluoride removal, the supernatant obtained after adsorption of a 20 mg·L–1 NaF solution by Ce-RM was analyzed by ICP–MS, and its environmental safety was assessed with reference to the Integrated Wastewater Discharge Standard of China (GB 8978–1996). The results in the Table showed that the concentrations of the regulated metal elements As, Cd, Cr, Cu, Hg, Pb and Zn in the supernatant were 0.0007, 0.0001, 0.0856, 0.0110, 0.0008, 0.0009, and 0.0055 mg·L–1, respectively, all of which were below the corresponding discharge limits specified in GB 8978–1996. Among these elements, Cr exhibited the highest proportion relative to its regulatory limit, accounting for only 5.71%, while Cu and Hg accounted for 2.20% and 1.60%, respectively; the remaining regulated elements were all below 1% of their corresponding limits. In addition, although Al, Ce and Fe were detected in the supernatant, they are not included among the corresponding regulated metal parameters in GB 8978–1996. These results indicate that Ce-RM did not induce substantial release of regulated heavy metals during fluoride adsorption, demonstrating a low risk of secondary contamination and good environmental safety for application in aquatic systems.

5. Concentrations of Metal Elements in the Supernatant after Adsorption and Their Environmental Safety Assessment.

element concentration (mg·L–1) GB 8978–1996 limit values (mg·L–1) percentage of the limit (%) evaluation results
Al 6.104 not listed - items not included in the limit
As 0.0007 0.5 0.14 within limits
Cd 0.0001 0.1 0.10 within limits
Ce 0.121 not listed - items not included in the limit
Cr 0.0856 1.5 5.71 within limits
Cu 0.0110 0.5 2.20 within limits
Fe 0.1742 not listed - items not included in the limit
Hg 0.0008 0.05 1.60 within limits
Pb 0.0009 1.0 0.09 within limits
Zn 0.0055 2.0 0.28 within limits
Open in a new tab

3.6. Adsorption Mechanism

Based on the adsorption results and characterization analyses, fluoride removal by Ce-RM is mainly attributed to electrostatic attraction, surface hydroxyl exchange and coordination with Ce-containing sites (Figure ). These processes are not independent, but occur in a connected sequence from fluoride transport to surface binding.The rough and porous structure of Ce-RM provides favorable conditions for fluoride to reach the active surface. SEM shows that Ce-RM has a loose and rough morphology, and this structure remains largely stable after fluoride adsorption. The N2 adsorption–desorption results further indicate that acid activation, calcination and Ce loading improve the pore structure of red mud. As a result, F– can diffuse more easily to the surface and contact more active sites. The pore structure mainly promotes mass transfer and site accessibility, while the actual immobilization of fluoride depends on surface chemical reactions. Surface charge is the first factor affecting the approach of fluoride ions. Ce-RM shows high fluoride removal at pH 1–3, and its pHpzc is 5.60. When the solution pH is lower than pHpzc, the Ce-RM surface is positively charged. Meanwhile, surface metal hydroxyl groups can be protonated to form M–OH2+. This positively charged surface favors the attraction of F– and increases its local concentration near the adsorbent. Once F– reaches the surface, further binding occurs through reactions with hydroxylated metal sites.

13.

13

Open in a new tab

Diagram of the adsorption mechanism.

The involvement of surface hydroxyl groups is supported by both FTIR and O 1s XPS. After fluoride adsorption, the hydroxyl and metal–oxygen vibration regions in the FTIR spectra change, and the M–OH component in the O 1s spectrum decreases from 61.08% to 42.93%. These changes indicate that surface hydroxyl groups take part in the adsorption process. Since O–H vibration can also be affected by hydrogen-bond changes, FTIR alone cannot fully determine the binding mode. The F 1s peak at approximately 686.85 eV further confirms that fluoride is retained on the material surface. These results suggest that F– partly replaces surface –OH groups and forms surface metal–fluoride species. Among the surface sites, Ce-containing species are the main active centers for fluoride fixation. The Ce 3d spectra show that Ce3+ and Ce4+ coexist on the Ce-RM surface. After fluoride adsorption, the Ce 3d envelope and the Ce3+/Ce4+ distribution change, indicating that the local chemical environment of Ce is affected by F–. This change supports the interaction between fluoride and Ce sites, most likely through Ce–F coordination. The comparison between RM and Ce-RM also confirms the role of Ce: pristine red mud shows limited fluoride removal, whereas Ce-RM exhibits much higher removal efficiency and adsorption capacity under the same conditions. Therefore, Ce loading introduces stronger fluoride-binding sites onto the red mud surface.

The original Fe- and Al-containing phases in red mud also contribute to the adsorption environment. XRD shows that hematite, aluminosilicate and Ce-containing oxide/silicate phases coexist in Ce-RM, and the Fe/Ce-related crystalline phases change after fluoride adsorption. The shift of Al 2p to higher binding energy suggests that Al-related sites are also affected by fluoride uptake. Fe 2p shows oxide/oxyhydroxide features with satellite peaks, which is consistent with the Fe-containing phases detected by XRD. Because Fe 2p contains overlapping multiplet and satellite structures, it is used here only for qualitative identification of Fe-containing phases, rather than for Fe2+/Fe3+ quantification or as direct evidence of Fe–F formation. The kinetic and isotherm results are consistent with the above surface-binding process. The pseudo-second-order model gives a better fit than the pseudo-first-order model, indicating that chemical interaction is involved in fluoride adsorption. The Freundlich model fits the equilibrium data better than the Langmuir model, suggesting that fluoride adsorption occurs on heterogeneous surface sites. This is consistent with the complex mineral composition of red mud and the coexistence of Ce-, Fe- and Al-related sites on the Ce-RM surface.

Therefore, fluoride removal by Ce-RM can be described as a stepwise surface process. Under acidic conditions, F– is first attracted to the positively charged surface. It then reacts with surface hydroxyl groups through hydroxyl exchange and local hydrogen-bond rearrangement. Finally, fluoride is mainly fixed by coordination with Ce-containing active sites, while Fe- and Al-containing phases provide additional surface environments. The improved pore structure helps expose these sites and facilitates fluoride transport. Overall, surface hydroxyl exchange and Ce–F coordination are the main pathways for fluoride removal, assisted by electrostatic attraction and the red mud matrix.

4. Conclusion

In this study, cerium was loaded onto red mud by an impregnation-calcination method to prepare a highly efficient adsorbent (Ce-RM) for the effective removal of fluoride from aqueous solution. Systematic batch adsorption experiments were carried out, and the results showed that the optimal adsorption performance was achieved for a fluoride solution with an initial concentration of 20 mg·L–1 at pH 3 and an adsorbent dosage of 0.4 g·L–1. The adsorption of fluoride by Ce-RM was well described by the pseudo-second-order kinetic model, indicating that the adsorption process was governed mainly by chemisorption. Meanwhile, the adsorption behavior could be well described by both the Langmuir and Freundlich isotherm models, with a better fit obtained using the Freundlich model, indicating that adsorption predominantly occurred on heterogeneous surface active sites and was mainly governed by multilayer adsorption, accompanied by a certain degree of monolayer adsorption. Although Ce-RM exhibited high fluoride removal efficiency and good environmental safety in both simulated fluoride-containing water and actual AlF3 wastewater systems, the regeneration and reusability of the adsorbent were not further evaluated in this study. Therefore, the structural stability, retention of active sites and long-term fluoride removal efficiency of Ce-RM during repeated adsorption–desorption cycles remain to be further verified.

Acknowledgments

This research was supported by Scientific Research Foundation for introduced Talents of Guangxi Minzu University (2021KJQD32), Guangxi Science and Technology Program (Guike AD23026195), Guangxi Key Research and Development Program (Guike AB25069424), Local Science and Technology Development Fund Projects Guided by the Central Government (Guike ZY24212039).

The authors declare no competing financial interest.

References

  1. Du H., Fang L., Shi X., Li T., Sui Z., Zhong J., Li K.. Efficient defluoridation of high-fluoride water using rare earth-based adsorbents: adsorption performance, mechanism, and kinetics. Sci. Rep. 2025;15(1):42709. doi: 10.1038/s41598-025-26780-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Rashid U. S., Das T. K., Sakthivel T. S., Seal S., Bezbaruah A. N.. GO-CeO2 nanohybrid for ultra-rapid fluoride removal from drinking water. Sci. Total Environ. 2021;793:148547. doi: 10.1016/j.scitotenv.2021.148547. [DOI] [PubMed] [Google Scholar]
  3. National Environmental Protection Administration Integrated Wastewater Discharge Standard (GB 8978–1996); China Environmental Science Press, 1996. [Google Scholar]
  4. Ullah M., Shalchian H., Younas T., Vegliò F., Ippolito N. M.. Sustainable wastewater treatment in lithium-ion battery recycling: Processes, challenges, and innovations. Process Saf. Environ. Prot. 2025;202:107710. doi: 10.1016/j.psep.2025.107710. [DOI] [Google Scholar]
  5. Miyashita T., Yasuda K., Uda T.. Removal of phosphorus and fluorine from wastewater containing PF6–via accelerated decomposition by Al3+ and chemical precipitation for hydrometallurgical recycling of lithium-ion batteries. Environ. Sci.:Water Res. Technol. 2024;10(5):1245–1255. doi: 10.1039/D3EW00854A. [DOI] [Google Scholar]
  6. Ezzeddine A., Bedoui A., Hannachi A., Bensalah N.. Removal of fluoride from aluminum fluoride manufacturing wastewater by precipitation and adsorption processes. Desalin. Water Treat. 2015;54(8):2280–2292. doi: 10.1080/19443994.2014.899515. [DOI] [Google Scholar]
  7. Zhou L., Zhong Y., Cao X., Zhao H., Liu R.. Treatment of Highly Concentrated Fluoride Wastewater from the Semiconductor Industry by CaCl2 based on the Adjustment of ζ-Potential. ACS ES&T Water. 2023;3(12):4074–4081. doi: 10.1021/acsestwater.3c00463. [DOI] [Google Scholar]
  8. Wei X., Xu N., Xu A., Zhang X.. Molecular identification guided process design for deep removal of fluoride from electroplating effluent. Environ. Sci.:Adv. 2022;1(4):584–593. doi: 10.1039/D2VA00177B. [DOI] [Google Scholar]
  9. Qiu Y., Ren L.-F., Shao J., Xia L., Zhao Y.. An integrated separation technology for high fluoride-containing wastewater treatment: Fluoride removal, membrane fouling behavior and control. J. Cleaner Prod. 2022;349:131225. doi: 10.1016/j.jclepro.2022.131225. [DOI] [Google Scholar]
  10. Singh S., German M., Chaudhari S., Sengupta A. K.. Fluoride removal from groundwater using Zirconium Impregnated Anion Exchange Resin. J. Environ. Manage. 2020;263:110415. doi: 10.1016/j.jenvman.2020.110415. [DOI] [PubMed] [Google Scholar]
  11. Wang X., Li N., Li J., Feng J., Ma Z., Xu Y., Sun Y., Xu D., Wang J., Gao X.. et al. Fluoride removal from secondary effluent of the graphite industry using electrodialysis: Optimization with response surface methodology. Front. Environ. Sci. Eng. 2019;13(4):51. doi: 10.1007/s11783-019-1132-5. [DOI] [Google Scholar]
  12. Huang H., Liu J., Zhang P., Zhang D., Gao F.. Investigation on the simultaneous removal of fluoride, ammonia nitrogen and phosphate from semiconductor wastewater using chemical precipitation. Chem. Eng. J. 2017;307:696–706. doi: 10.1016/j.cej.2016.08.134. [DOI] [Google Scholar]
  13. Qiu Y., Ren L.-F., Xia L., Shao J., Zhao Y., Van der Bruggen B.. Investigation of fluoride and silica removal from semiconductor wastewaters with a clean coagulation-ultrafiltration process. Chem. Eng. J. 2022;438:135562. doi: 10.1016/j.cej.2022.135562. [DOI] [Google Scholar]
  14. Oduro A. A., Seaneye W. S., Miezah K., Anku W. W., Agorku E. S.. Defluorination of borehole water using activated clay and bauxite-laden polyvinyl alcohol hybrid membranes. RSC Appl. Polym. 2025;3(6):1613–1628. doi: 10.1039/D5LP00188A. [DOI] [Google Scholar]
  15. Huang L., Sheng L., Wan K., Wang M., Zhang H., Yan J., Liu Y., Alhassan S. I., Chen Y., Arulmani S. R. B.. The diversified mechanism of adsorption and electro-adsorption technologies by using Ti3AlC2 for removing fluoride. Desalination. 2024;578:117449. doi: 10.1016/j.desal.2024.117449. [DOI] [Google Scholar]
  16. Zhao X., Zhang J., Dai Z., Lei Y., Liu X., Liu G.. Simple preparation and efficient fluoride removal of La anchored Zr-based metal–organic framework adsorbent. J. Environ. Chem. Eng. 2022;10(6):108807. doi: 10.1016/j.jece.2022.108807. [DOI] [Google Scholar]
  17. Yu T., Chen Y., Zhang Y., Tan X., Xie T., Shao B., Huang X.. Novel reusable sulfate-type zirconium alginate ion-exchanger for fluoride removal. Chin. Chem. Lett. 2021;32(11):3410–3415. doi: 10.1016/j.cclet.2021.04.057. [DOI] [Google Scholar]
  18. Yu Y.-L., Chen C.-Y., Dhanasinghe C., Verpoort F., Surampalli R. Y., Chen S.-C., Kao C.-M.. Development of modified MgO/biochar composite for chemical adsorption enhancement to cleanup fluoride-contaminated groundwater. J. Environ. Manage. 2024;370:123016. doi: 10.1016/j.jenvman.2024.123016. [DOI] [PubMed] [Google Scholar]
  19. Yang C., Shen C., Zhang N., Zhang X., Zhao L., Zheng J.. Removing Fluoride from Water by Nanostructured Magnesia-Impregnated Activated Carbon. Colloids Interfaces. 2025;9(2):22. doi: 10.3390/colloids9020022. [DOI] [Google Scholar]
  20. Guo S., Zheng F., Xu J., Jiang J., Cui Z., Wu C., Lin Y., Sun Q., Zheng Y., Sa B.. Enhanced fluoride removal from drinking water by activated carbon supported Ce–Al oxides: performance and mechanism. RSC Adv. 2025;15(18):14363–14374. doi: 10.1039/D5RA00397K. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Wei Y., Wang L., Li H., Yan W., Feng J.. Synergistic Fluoride Adsorption by Composite Adsorbents Synthesized From Different Types of MaterialsA Review. Front. Chem. 2022;10:900660. doi: 10.3389/fchem.2022.900660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Mohan D., Goyal H., Kushwaha R., Markandeya, Shukla S. P., Srivastava V.. Fluoride removal using activated red mud. Environ. Qual. Manag. 2024;33(4):559–566. doi: 10.1002/tqem.22112. [DOI] [Google Scholar]
  23. Wang J., Zhang H., Wang N., Mo H., Yang Z., Dong Y., Liu Q., Huang X., Han B.. Synthesis of Polymer Sodium Alginate–Red Mud Adsorbent and Its Application in the Removal of Low-Concentration Fluoride. Polymers. 2025;17(6):826. doi: 10.3390/polym17060826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Khairul M. A., Zanganeh J., Moghtaderi B.. The composition, recycling and utilisation of Bayer red mud. Resour., Conserv. Recycl. 2019;141:483–498. doi: 10.1016/j.resconrec.2018.11.006. [DOI] [Google Scholar]
  25. Niu A., Lin C.. Trends in research on characterization, treatment and valorization of hazardous red mud: A systematic review. J. Environ. Manage. 2024;351:119660. doi: 10.1016/j.jenvman.2023.119660. [DOI] [PubMed] [Google Scholar]
  26. Wang D., Luo L., Chen Y., Chen S., Qiu X., Liu M.. Fluoride removal from aqueous solution using Ce-modified red mud as adsorbent. Mater. Today Sustain. 2023;24:100502. doi: 10.1016/j.mtsust.2023.100502. [DOI] [Google Scholar]
  27. Pavan C., Santalucia R., Escolano-Casado G., Ugliengo P., Mino L., Turci F.. Physico-Chemical Approaches to Investigate Surface Hydroxyls as Determinants of Molecular Initiating Events in Oxide Particle Toxicity. Int. J. Mol. Sci. 2023;24(14):11482. doi: 10.3390/ijms241411482. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ellerbrock R., Stein M., Schaller J.. Comparing amorphous silica, short-range-ordered silicates and silicic acid species by FTIR. Sci. Rep. 2022;12(1):11708. doi: 10.1038/s41598-022-15882-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Castaldi P., Silvetti M., Santona L., Enzo S., Melis P.. XRD, FTIR, and thermal analysis of bauxite ore-processing waste (red mud) exchanged with heavy metals. Clays Clay Miner. 2024;56(4):461–469. doi: 10.1346/ccmn.2008.0560407. [DOI] [Google Scholar]
  30. Changes in the structure of red mud and black nickel mud after activation. Acta Montan. Slovaca 2024, 29 (v29/i1), 227–236. [Google Scholar]
  31. Hansen P. E., Spanget-Larsen J.. NMR and IR Investigations of Strong Intramolecular Hydrogen Bonds. Molecules. 2017;22(4):552. doi: 10.3390/molecules22040552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Liang W., Couperthwaite S. J., Kaur G., Yan C., Johnstone D. W., Millar G. J.. Effect of strong acids on red mud structural and fluoride adsorption properties. J. Colloid Interface Sci. 2014;423:158–165. doi: 10.1016/j.jcis.2014.02.019. [DOI] [PubMed] [Google Scholar]
  33. Moreno de la Calle A. ´., Vizcaíno A. J., Carrero A., Calles J. A., Megía P. J.. Study of Mesostructured CeO2 Synthesis via Nanocasting Using SBA-15 as a Template: Influence of the Cerium Precursor. Int. J. Mol. Sci. 2024;25(23):13016. doi: 10.3390/ijms252313016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Easton C. D., Morgan D. J.. Critical examination of the use of x-ray photoelectron spectroscopy (XPS) O 1s to characterize oxygen vacancies in catalytic materials and beyond. J. Vac. Sci. Technol. A. 2025;43(5):053205. doi: 10.1116/6.0004686. [DOI] [Google Scholar]
  35. Bêche E., Charvin P., Perarnau D., Abanades S., Flamant G.. Ce 3d XPS investigation of cerium oxides and mixed cerium oxide (CexTiyOz) Surf. Interface Anal. 2008;40(3–4):264–267. doi: 10.1002/sia.2686. [DOI] [Google Scholar]
  36. Wu X., Zhang Y., Dou X., Zhao B., Yang M.. Fluoride adsorption on an Fe–Al–Ce trimetal hydrous oxide: Characterization of adsorption sites and adsorbed fluorine complex species. Chem. Eng. J. 2013;223:364–370. doi: 10.1016/j.cej.2013.03.027. [DOI] [Google Scholar]
  37. He Y., Zhang L., An X., Wan G., Zhu W., Luo Y.. Enhanced fluoride removal from water by rare earth (La and Ce) modified alumina: Adsorption isotherms, kinetics, thermodynamics and mechanism. Sci. Total Environ. 2019;688:184–198. doi: 10.1016/j.scitotenv.2019.06.175. [DOI] [PubMed] [Google Scholar]
  38. Biesinger M. C., Payne B. P., Grosvenor A. P., Lau L. W. M., Gerson A. R., Smart R. S. C.. Resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Cr, Mn, Fe, Co and Ni. Appl. Surf. Sci. 2011;257(7):2717–2730. doi: 10.1016/j.apsusc.2010.10.051. [DOI] [Google Scholar]
  39. Mukhopadhyay K., Ghosh A., Das S. K., Show B., Sasikumar P., Chand Ghosh U.. Synthesis and characterisation of cerium­(iv)-incorporated hydrous iron­(iii) oxide as an adsorbent for fluoride removal from water. RSC Adv. 2017;7(42):26037–26051. doi: 10.1039/C7RA00265C. [DOI] [Google Scholar]
  40. Dhillon A., Kumar Sharma T., Soni S. K., Kumar D.. Fluoride adsorption on a cubical ceria nanoadsorbent: function of surface properties. RSC Adv. 2016;6(92):89198–89209. doi: 10.1039/C6RA16962G. [DOI] [Google Scholar]
  41. Mohan D., Pittman C. U.. Arsenic removal from water/wastewater using adsorbentsA critical review. J. Hazard. Mater. 2007;142(1–2):1–53. doi: 10.1016/j.jhazmat.2007.01.006. [DOI] [PubMed] [Google Scholar]
  42. Wang X., Zhu H., Sun T., Dai H.. Synthesis of a Matériaux Institut Lavoisier metal-organic framework 96 (MIL-96­(RM)) using red mud and its application to defluorination of water. Mater. Today Commun. 2020;25:101401. doi: 10.1016/j.mtcomm.2020.101401. [DOI] [Google Scholar]
  43. Yi M., Wang K., Wei H., Wei D., Wei X., Wei B., Shao L., Fujita T., Cui X.. Efficient preparation of red mud-based geopolymer microspheres (RM@GMs) and adsorption of fluoride ions in wastewater. J. Hazard. Mater. 2023;442:130027. doi: 10.1016/j.jhazmat.2022.130027. [DOI] [PubMed] [Google Scholar]
  44. Wang Z., Gu X., Zhang X., Wang X., Zhang J., Liu Y., Tan X., Zhao Y., Kang D., Guo W.. et al. New easily recycled carrier based polyurethane foam by loading Al-MOF and biochar for selective removal of fluoride ion from aqueous solutions. Sci. Total Environ. 2023;901:166312. doi: 10.1016/j.scitotenv.2023.166312. [DOI] [PubMed] [Google Scholar]
  45. Jian S., Shi F., Zhu Y., Jiang W., Liu Y., Yang W., Hu J., Jiang S.. Ce-MOFs and Ce/Mn bimetallic MOFs for efficient fluoride removal and their defluoridation mechanism. Colloids Surf., A. 2024;701:134938. doi: 10.1016/j.colsurfa.2024.134938. [DOI] [Google Scholar]
  46. Luo C., Liu L., Ren S., Lan Y., Xu C., Wang J., Yuan C., Li Y., Pan D., Huang J.. et al. Development of Fluorophilic Magnetic Adsorbents with Ultrahigh Adsorption Capacity from Red Mud: Defluorination Behavior and Mechanisms. ACS Omega. 2025;10(11):11397–11410. doi: 10.1021/acsomega.4c11404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Chen J., Yang R., Zhang Z., Wu D.. Removal of fluoride from water using aluminum hydroxide-loaded zeolite synthesized from coal fly ash. J. Hazard. Mater. 2022;421:126817. doi: 10.1016/j.jhazmat.2021.126817. [DOI] [PubMed] [Google Scholar]
  48. Foo K. Y., Hameed B. H.. Insights into the modeling of adsorption isotherm systems. Chem. Eng. J. 2010;156(1):2–10. doi: 10.1016/j.cej.2009.09.013. [DOI] [Google Scholar]

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES