Abstract
Aluminum hydroxide is an eye-catching and extensively researched adsorbent for fluoride removal and its defluoridation performance is closely related to the preparation method and crystalline phase. In this research, the defluoridation performances of aluminum hydroxides with different crystalline phases are compared and evaluated in terms of fluoride removal capacity, sensitivity to pH values and residual Al contents after defluoridation. It is found that the defluoridation performance of different aluminum hydroxides follows the order of boehmite > bayerite > gibbsite. The fluoride adsorption on aluminum hydroxides follows the pseudo-second-order kinetic model and Langmuir isotherm model, and the maximum defluoridation capacities of boehmite, bayerite and gibbsite are 42.08, 2.97 and 2.74 mg m−2, respectively. The pH values and FTIR analyses reveal that the ligand exchange between fluoride and surface hydroxyl groups is the fluoride removal mechanism. Different aluminum hydroxides have different surface hydroxyl group densities, which results in the different defluoridation capacities. This work provides a new idea to prepare aluminum hydroxide with outstanding defluoridation performance.
HIGHLIGHTS
The defluoridation performance of boehmite, bayerite and gibbsite is compared.
Boehmite exhibits better defluoridation performance than bayerite and gibbsite.
Fluoride removal mechanism was revealed through pH value and FTIR analyses.
Adsorption capacity of aluminum hydroxide depends on surface hydroxyl group density.
Graphical Abstract
INTRODUCTION
Fluorine is one of the necessary trace elements for human health, and appropriate fluorine can prevent dental caries and maintain the stability of calcium and phosphorus metabolism (Hegde et al. 2020). However, both insufficient and excessive fluorine will cause harmful effects on human health. The World Health Organization (WHO) has set 1.5 mg L−1 as the guideline of fluoride content in drinking water (Gai & Deng 2021). When the fluoride concentration is lower than 0.5 mg L−1, the incidence rate of dental caries increases. Long-term drinking of high fluoride water (>1.5 mg L−1) can cause a series of fluorosis based diseases, e.g. dental fluorosis, skeletal fluorosis, joint stiffness, paralysis, thyroid dysfunction and kidney dysfunction (He et al. 2020). Tens of million people worldwide are suffering from various degrees of fluorosis due to long-term intake of high fluoride drinking water (Alkurdi et al. 2019). Therefore, how to remove the excess fluoride from water is an urgent issue worldwide.
Adsorption is a well-studied and used defluoridation technology due to its simple design, convenient operation and low cost. During the past decades, a variety of adsorbents have been developed, such as natural and modified minerals (García-Sánchez et al. 2016), biomass adsorbents (Zhang & Huang 2019), metal oxides and metal hydroxides (Kameda et al. 2015), carbon-based adsorbents (Xie et al. 2019) and so on. Among these adsorbents, aluminum oxides and hydroxides have the advantages of simple preparation, low cost and strong affinity for fluoride, making them eye catching and extensively researched adsorbents for fluoride removal. Activated alumina is widely used for defluoridation in vast rural areas, but its defluoridation capacity is as low as 1.45 mg g−1 (Ghorai & Pant 2004). Some researchers found that the defluoridation performance of alumina could be improved through changing the crystalline phases. For example, the defluoridation capacity of α-Al2O3 and γ-Al2O3 at 25 °C is 2.73 and 14.0 mg g−1, respectively (Valdivieso et al. 2006; Kumar et al. 2011). In addition, metal oxides including MgO, MnO2, CeO2 and La2O3, can be uses as activators to modify alumina and improve defluoridation performance (Dayananda et al. 2015; He et al. 2019). However, alumina adsorbent has high residual aluminum content after defluoridation, inhibiting its wide application (George et al. 2010).
Compared with alumina, aluminum hydroxide has a low residual Al content after defluoridation, making it have more potential than alumina. However, the fluoride removal capacity of commercial aluminum hydroxide is very poor. To enhance the defluoridation efficiency, some methods were developed to prepare aluminum hydroxide adsorbents (Du et al. 2014; Sun et al. 2016). Shimelis et al. (2006) prepared aluminum hydroxide by hydrolysis of aluminum sulfate, and found that the adsorption capacity was 23.7 mg g−1. Gai et al. (2015) synthesized aluminum hydroxide by Al-water reaction under ultrasonic field, and demonstrated the maximum defluoridation capacity of 6.78 mg g−1. Jia et al. (2015a, 2015b) synthesized bayerite/boehmite adsorbent with 3D feather-like structures through the hydrothermal method, and the adsorbent exhibited excellent adsorption capacity of 56.8 mg g−1. According to the above discussion, aluminum hydroxides prepared by different methods have different defluoridation performances, but the specific reason is unclear. In fact, aluminum hydroxides prepared by different methods have different microstructures and crystalline phases, which may be the reason for their different defluoridation performances. In this work, the impact of crystalline phases on defluoridation efficiency of aluminum hydroxides was systematically analyzed through model experiment, and the related physicochemical mechanism was revealed.
EXPERIMENTAL PROCEDURE
NaF with analytical reagent grade and aluminum hydroxides with three crystalline phases, namely boehmite, bayerite and gibbsite, were used as received in the present experiment. Stock solution with F− concentration of 1,000 mg L−1 was obtained through dissolving an appropriate amount of NaF in deionized water. The aqueous solution with different initial F− concentration was prepared by diluting the F− stock solution with deionized water. In each fluoride removal test, 1 g of aluminum hydroxide adsorbent was added into a PTFE beaker containing 250 ml of F− solution. The suspension was stirred at a speed of 500 rpm using a magnetic stirrer in order to improve the adsorption efficiency. All the adsorption tests were conducted at 25 ± 1 °C, which was controlled by thermostatic water bath. At a predetermined time, 10 ml of suspension was collected from a beaker using a syringe-driven filter with 0.22 μm polyethersulfone (PES) membrane. After the adsorption test, the adsorbent was recovered by filtering and drying and used for further analysis.
The pH value of F− solution before and after fluoride removal was determined using a pH meter (PHS-3E, Inesa Scientific Instrument Co., Shanghai, China). The crystalline phases and morphologies of aluminum hydroxides were analyzed using X-ray diffractometry (XRD, D8 Advance, Germany) and scanning electron microscopy (SEM, Sigma 500/VP, Germany), respectively. The functional groups and chemical bonds of adsorbents before and after fluoride removal were analyzed by Fourier transform infrared spectrometer (FTIR, Avatar 370, USA). Inductively coupled plasma atomic emission spectrometer (ICP-AES, ICAP 6300, USA) was used to analyze the residual Al concentration in solution after defluoridation.
RESULTS AND DISCUSSION
Characterization of aluminum hydroxides
Figure 1 gives the micrographs of aluminum hydroxides with different crystalline phases. It is found that all the aluminum hydroxide particles have irregular shapes, and their particle sizes are non-uniform. The average particle sizes of boehmite, bayerite and gibbsite adsorbents are 17.1, 2.5 and 4.1 μm, respectively.
SEM micrographs of as-received (a) boehmite, (b) bayerite and (c) gibbsite adsorbents.
SEM micrographs of as-received (a) boehmite, (b) bayerite and (c) gibbsite adsorbents.
Figure 2(a) presents the X-ray patterns of aluminum hydroxides with different crystalline phases. For all the aluminum hydroxide adsorbents, only one crystalline phase (boehmite, bayerite and gibbsite) is detected, which validates the high purity of aluminum hydroxides.
X-ray patterns (a) and FTIR spectra (b) of as-received boehmite, bayerite and gibbsite adsorbents.
X-ray patterns (a) and FTIR spectra (b) of as-received boehmite, bayerite and gibbsite adsorbents.
Figure 2(b) shows the FTIR spectra of aluminum hydroxides with different crystalline phases. Clearly, aluminum hydroxides with different crystalline phases have obviously different FTIR spectra. The boehmite adsorbent has the bands at 3,302, 3,087, 2,102, 1,631, 1,384, 1,151, 1,074, 742, 613 and 488 cm−1, which are in good agreement with the results of AlOOH in the literature (Wang et al. 2009; Zhang et al. 2016). The bands at 3,302, 3,087, 1,636, and 1,074 cm−1 ascribe to the stretching and bending vibrations of Al-OH, and the bands at 488, 613 and 742 cm−1 represent the vibration mode of AlO6. The bands at 2,102 and 1,384 cm−1 correspond to the surface hydroxyl groups (Xu et al. 2015). The FTIR spectrum of bayerite is consistent with the standard spectrum of bayerite (Musić et al. 1999). The peaks of 3,657, 3,553, 3,467 and 3,424 cm−1 belong to the stretching vibrations of hydroxyl groups. For gibbsite, the bands at 3,621, 3,523, 3,463 cm−1 result from the OH-stretching vibrations, the bands at 1,644, 1,023 and 798 cm−1 originate from the OH-bending vibrations, and the peaks of 666 and 516 cm−1 represent AlO6 (Alex et al. 2014).
Defluoridation performances of aluminum hydroxides
Figure 3(a) shows the fluoride removal performances of aluminum hydroxides with different crystalline phases. Obviously, the crystalline phase of aluminum hydroxide has significant influence on its defluoridation performance. When the initial F− concentration is 10 mg/L, the defluoridation efficiency of boehmite, bayerite and gibbsite is 80.7%, 30.3% and 16.7%, respectively, indicating that the defluoridation performances of aluminum hydroxides with different crystalline phases follow the order of boehmite > bayerite > gibbsite. For boehmite, it can remove >60% fluoride in the first 1.5 h, and reach adsorption equilibrium within 12 h. To further research the defluoridation performance of aluminum hydroxide, the effect of initial F− concentration on defluoridation efficiency was tested, as shown in Figure 3(b)–3(d). For boehmite, bayerite and gibbsite, the impact trend of initial F− concentration on defluoridation efficiency is consistent. When increasing the initial F− concentration, the defluoridation capacity increases, while the defluoridation efficiency decreases. This is reasonable, because the adsorbent cannot provide enough active sites for fluoride adsorption when increasing the initial F− concentration.
(a) Fluoride removal from aqueous solution using different aluminum hydroxide adsorbents, where the initial F− concentration is 10 mg L−1; and the effect of initial F− concentration on fluoride removal from aqueous solution using (b) boehmite, (c) bayerite and (d) gibbsite adsorbents.
(a) Fluoride removal from aqueous solution using different aluminum hydroxide adsorbents, where the initial F− concentration is 10 mg L−1; and the effect of initial F− concentration on fluoride removal from aqueous solution using (b) boehmite, (c) bayerite and (d) gibbsite adsorbents.
Figure 4 shows the effect of initial pH value on the defluoridation performances of different aluminum hydroxides. Clearly, the defluoridation performance is closely related to the initial pH value. For bayerite and gibbsite adsorbents, the defluoridation capacity remains stable at pH range of 6–9, while it decreases when increasing the pH value from 9 or decreasing the pH value from 6. When the pH value is below 6, some aluminum hydroxide will dissolve, resulting in the reduction of defluoridation capacity. When the pH value is above 9, competitive adsorption between F− and OH− occurs, which decreases the defluoridation capacity. For the F− solution with pH values of 3 and 11, the defluoridation capacities of bayerite and gibbsite are only 0.89 and 0.82 mg g−1, and 0.28 and 0.23 mg g−1, respectively. Compared with bayerite and gibbsite adsorbents, boehmite exhibits excellent defluoridation capacity at a wide pH range of 4 − 10, indicating that boehmite is less sensitive to pH value than bayerite and gibbsite.
Fluoride removal from F- solution with different pH values using different aluminum hydroxide adsorbents, where the initial F− concentration is 20 mg L−1 and the temperature is 25 °C.
Fluoride removal from F- solution with different pH values using different aluminum hydroxide adsorbents, where the initial F− concentration is 20 mg L−1 and the temperature is 25 °C.
Figure 5 shows the effect of adsorption temperature on the defluoridation capacities of different aluminum hydroxides. It was found that temperature has little effect on the defluoridation capacities of different aluminum hydroxides. The defluoridation capacity slightly decreases when increasing the adsorption temperature, indicating that fluoride adsorption on aluminum hydroxide is an exothermic process.
Fluoride removal from F− solution at different temperatures using different aluminum hydroxide adsorbents, where the initial F− concentration is 20 mg L−1.
Fluoride removal from F− solution at different temperatures using different aluminum hydroxide adsorbents, where the initial F− concentration is 20 mg L−1.
Adsorption kinetics
Figure 6 shows the pseudo-first-order and pseudo-second-order nonlinear kinetic plots for the fluoride adsorption on different aluminum hydroxides, and the corresponding parameters are given in Table 1. According to the correlation coefficient (R2), the adsorption data fits better with the pseudo-second-order model than the pseudo-first-order model, implying that the adsorption of fluoride on aluminum hydroxide follows second-order adsorption.
Kinetic parameters of pseudo-first-order and pseudo-second-order models obtained by the nonlinear regression approach for fluoride adsorption on aluminum hydroxides
Adsorbents . | qe(exp) (mg g−1) . | Pseudo-first-order . | Pseudo-second-order . | ||||
---|---|---|---|---|---|---|---|
qe (mg g−1) . | k1 (h−1) . | R2 . | qe (mg g−1) . | k2 (g mg−1 h−1) . | R2 . | ||
Gibbsite | 1.10 | 1.00 | 1.761 | 0.959 | 1.10 | 2.169 | 0.990 |
Bayerite | 1.77 | 1.67 | 2.294 | 0.986 | 1.82 | 1.726 | 0.987 |
Boehmite | 4.98 | 4.72 | 5.288 | 0.957 | 4.98 | 1.734 | 0.987 |
Adsorbents . | qe(exp) (mg g−1) . | Pseudo-first-order . | Pseudo-second-order . | ||||
---|---|---|---|---|---|---|---|
qe (mg g−1) . | k1 (h−1) . | R2 . | qe (mg g−1) . | k2 (g mg−1 h−1) . | R2 . | ||
Gibbsite | 1.10 | 1.00 | 1.761 | 0.959 | 1.10 | 2.169 | 0.990 |
Bayerite | 1.77 | 1.67 | 2.294 | 0.986 | 1.82 | 1.726 | 0.987 |
Boehmite | 4.98 | 4.72 | 5.288 | 0.957 | 4.98 | 1.734 | 0.987 |
Nonlinear pseudo-first-order kinetic fit (a) and nonlinear pseudo-second-order kinetic fit (b) for fluoride removal from aqueous solution using different aluminum hydroxide adsorbents, where the initial fluoride concentration is 40 mg L−1 and the temperature is 25 °C.
Nonlinear pseudo-first-order kinetic fit (a) and nonlinear pseudo-second-order kinetic fit (b) for fluoride removal from aqueous solution using different aluminum hydroxide adsorbents, where the initial fluoride concentration is 40 mg L−1 and the temperature is 25 °C.
Adsorption isotherm
Isotherm parameters of Langmuir and Freundlich models for fluoride adsorption by aluminum hydroxides
Adsorbents . | Langmuir . | Freundlich . | ||||
---|---|---|---|---|---|---|
qmax (mg g−1) . | kL (L mg−1) . | R2 . | kF . | n . | R2 . | |
Gibbsite | 1.67 | 0.0385 | 0.921 | 0.0964 | 1.497 | 0.956 |
Bayerite | 2.97 | 0.0506 | 0.950 | 0.2947 | 1.894 | 0.869 |
Boehmite | 5.05 | 0.3414 | 0.965 | 1.6286 | 2.799 | 0.945 |
Adsorbents . | Langmuir . | Freundlich . | ||||
---|---|---|---|---|---|---|
qmax (mg g−1) . | kL (L mg−1) . | R2 . | kF . | n . | R2 . | |
Gibbsite | 1.67 | 0.0385 | 0.921 | 0.0964 | 1.497 | 0.956 |
Bayerite | 2.97 | 0.0506 | 0.950 | 0.2947 | 1.894 | 0.869 |
Boehmite | 5.05 | 0.3414 | 0.965 | 1.6286 | 2.799 | 0.945 |
Langmuir (a) and Freundlich (b) adsorption isotherms for fluoride removal from aqueous solution using different aluminum hydroxide adsorbents.
Langmuir (a) and Freundlich (b) adsorption isotherms for fluoride removal from aqueous solution using different aluminum hydroxide adsorbents.
Residual Al after defluoridation
As is known, some aluminum can leach into water in the defluoridation process using Al-based materials as adsorbents. Al is a potential neurotoxin, and WHO has set 0.2 mg L−1 as the concentration limit of Al in drinking water (George et al. 2010). Long-term drinking water with excessive Al may enhance the risk for the development of Alzheimer's disease in humans (Rondeau et al. 2000). Therefore, it is necessary and meaningful to analyze the residual Al after defluoridation when evaluating the defluoridation performance of aluminum hydroxides. Figure 8 illustrates the residual Al concentration of aqueous solution after defluoridation using aluminum hydroxides with different crystalline phases. As can be seen, the residual Al concentrations after defluoridation by gibbsite, bayerite and boehmite are 0.009, 0.007 and 0.003 mg L−1, respectively, which are far below the WHO limit. This indicates that all three aluminum hydroxides can be used for fluoride removal.
Residual Al concentration of aqueous solution after fluoride removal using different adsorbents, where the initial F− concentration is 20 mg L−1.
Residual Al concentration of aqueous solution after fluoride removal using different adsorbents, where the initial F− concentration is 20 mg L−1.
Comparison of defluoridation performance
To comprehensively compare the defluoridation performance of aluminum hydroxides with different crystalline phases, Table 3 summarizes the fluoride removal capacities and residual Al contents of different aluminum hydroxides. As is known, the adsorption capacity of an adsorbent is closely related to the specific surface area. To get rid of the impact of surface area, the defluoridation capacity per unit area (qmax,s) is also calculated and given in Table 3. Boehmite, bayerite and gibbsite have the defluoridation capacity per unit area of 42.08, 2.97 and 2.74 mg m−2. Based on the defluoridation capacity and residual Al content, it can be inferred that the defluoridation performance of different aluminum hydroxides follows the order of boehmite > bayerite > gibbsite.
Defluoridation performances of aluminum hydroxides with different crystalline phases
Adsorbents . | Particle size (μm) . | Surface area (m2 g−1) . | Residual Al (mg L−1) . | qmax,ma (mg g−1) . | qmax,sa (mg m−2) . |
---|---|---|---|---|---|
Boehmite | 17.1 | 0.12 | 0.003 | 5.05 | 42.08 |
Bayerite | 2.5 | 1.00 | 0.007 | 2.97 | 2.97 |
Gibbsite | 4.1 | 0.61 | 0.009 | 1.67 | 2.74 |
Adsorbents . | Particle size (μm) . | Surface area (m2 g−1) . | Residual Al (mg L−1) . | qmax,ma (mg g−1) . | qmax,sa (mg m−2) . |
---|---|---|---|---|---|
Boehmite | 17.1 | 0.12 | 0.003 | 5.05 | 42.08 |
Bayerite | 2.5 | 1.00 | 0.007 | 2.97 | 2.97 |
Gibbsite | 4.1 | 0.61 | 0.009 | 1.67 | 2.74 |
aqmax,m and qmax,s represent the maximum defluoridation capacity per unit mass and area, respectively.
In order to further assess the fluoride removal performances of different aluminum hydroxides, the adsorption capacities of aluminum hydroxides from present study were compared with other Al-based adsorbents, as shown in Table 4. It can be seen that boehmite exhibits excellent defluoridation capacity and fast adsorption rate, indicating that boehmite has more potential than bayerite and gibbsite for fluoride removal.
Adsorption capacities of different Al-based adsorbents for fluoride removal from water
Adsorbents . | C0 (mg L−1) . | pH . | Contact time (h) . | qmax (mg g−1) . | References . |
---|---|---|---|---|---|
Al2O3 modified expanded graphite | 3–100 | 4.0 | 2.5 | 5.75 | Jin et al. (2015) |
Mesoporous CoAl2O4 | 5–50 | 7.0 | 5 | 14.80 | Zhao et al. (2015) |
Al(OH)3 modified magnetite | 1–30 | 7.8 | 24 | 1.51 | García-Sánchez et al. (2016) |
Bayerite/boehmite | 10–100 | 7.0 | 24 | 56.80 | Jia et al. (2015a, 2015b) |
Sulfate modified AlOOH | 20–100 | 6.0 | 24 | 40.0 | Du et al. (2014) |
Activated Al2O3 | 2.5–14 | 7 | 24 | 1.45 | Ghorai & Pant (2004) |
Al2O3 cement granule | 2.5–100 | 6.5–7 | 3 | 10.21 | Ayoob & Gupta (2008) |
α-Al2O3 | 3–10 | 5 | 24 | 2.73 | Valdivieso et al. (2006) |
γ-Al2O3 | 1–100 | 6.15 | 24 | 14.0 | Kumar et al. (2011) |
Al2O3 nanoparticles | 1–50 | 5–6 | 2 | 9.73 | Tangsir et al. (2016) |
Mesoporous Al2O3 | 20–250 | 6 | 24 | 14.26 | Lee et al. (2010) |
Gibbsite | 10–40 | 7.5 | 12 | 1.67 | This study |
Bayerite | 10–40 | 7.5 | 12 | 2.97 | This study |
Boehmite | 10–40 | 7.5 | 12 | 5.05 | This study |
Adsorbents . | C0 (mg L−1) . | pH . | Contact time (h) . | qmax (mg g−1) . | References . |
---|---|---|---|---|---|
Al2O3 modified expanded graphite | 3–100 | 4.0 | 2.5 | 5.75 | Jin et al. (2015) |
Mesoporous CoAl2O4 | 5–50 | 7.0 | 5 | 14.80 | Zhao et al. (2015) |
Al(OH)3 modified magnetite | 1–30 | 7.8 | 24 | 1.51 | García-Sánchez et al. (2016) |
Bayerite/boehmite | 10–100 | 7.0 | 24 | 56.80 | Jia et al. (2015a, 2015b) |
Sulfate modified AlOOH | 20–100 | 6.0 | 24 | 40.0 | Du et al. (2014) |
Activated Al2O3 | 2.5–14 | 7 | 24 | 1.45 | Ghorai & Pant (2004) |
Al2O3 cement granule | 2.5–100 | 6.5–7 | 3 | 10.21 | Ayoob & Gupta (2008) |
α-Al2O3 | 3–10 | 5 | 24 | 2.73 | Valdivieso et al. (2006) |
γ-Al2O3 | 1–100 | 6.15 | 24 | 14.0 | Kumar et al. (2011) |
Al2O3 nanoparticles | 1–50 | 5–6 | 2 | 9.73 | Tangsir et al. (2016) |
Mesoporous Al2O3 | 20–250 | 6 | 24 | 14.26 | Lee et al. (2010) |
Gibbsite | 10–40 | 7.5 | 12 | 1.67 | This study |
Bayerite | 10–40 | 7.5 | 12 | 2.97 | This study |
Boehmite | 10–40 | 7.5 | 12 | 5.05 | This study |
Physicochemical mechanism
To confirm the above fluoride removal mechanism, the pH values before and after fluoride removal by different aluminum hydroxides were measured and are shown in Table 5. For all the aluminum hydroxides, the pH values increase after fluoride removal. Furthermore, the higher the defluoridation efficiency, the larger the change in pH value (ΔpH) becomes. When defluoridation efficiency increases, more F− adsorbs on aluminum hydroxide, resulting in the release of more OH−. This confirms the above fluoride removal mechanism.
pH values of different solutions before and after fluoride removal by different aluminum hydroxides
Adsorbentsa . | pH (before) . | pH (after) . | ΔpH . |
---|---|---|---|
Gibbsite | 7.72 | 7.89 | 0.17 |
Bayerite | 7.72 | 8.01 | 0.29 |
Boehmite | 7.49 | 7.94 | 0.45 |
Adsorbentsa . | pH (before) . | pH (after) . | ΔpH . |
---|---|---|---|
Gibbsite | 7.72 | 7.89 | 0.17 |
Bayerite | 7.72 | 8.01 | 0.29 |
Boehmite | 7.49 | 7.94 | 0.45 |
aThe initial fluoride concentration of solution is 20 mg/L.
Figure 9 demonstrates the FTIR spectra of boehmite before and after adsorption. It can be seen that the peak of 1,384 cm−1 belonging to the surface hydroxyl groups obviously decreases after fluoride adsorption (Xu et al. 2015), meaning that surface hydroxyl groups are involved in the fluoride adsorption process. This validates that the exchange between surface hydroxyl groups and fluoride is the main mechanism for fluoride adsorption on aluminum hydroxide.
FTIR spectra of boehmite adsorbent before and after fluoride removal from aqueous solution with initial F− concentration of 40 mg L−1.
FTIR spectra of boehmite adsorbent before and after fluoride removal from aqueous solution with initial F− concentration of 40 mg L−1.
Based on the above analyses, it can be concluded that the defluoridation capacity of aluminum hydroxide greatly depends on its surface hydroxyl group density. In fact, the surface hydroxyl group density of aluminum hydroxide varies with the crystalline phase. The surface hydroxyl group density of boehmite, bayerite and gibbsite is 16.5, 12.0 and 8.0 sites nm−2, respectively (Sposito 1996; Lefèvre et al. 2002; Karamalidis & Dzombak 2010). This may be the reason why the defluoridation capacities of different aluminum hydroxides follow the order of boehmite > bayerite > gibbsite.
CONCLUSIONS
In this research, the defluoridation performances of aluminum hydroxides with different crystalline phases are compared and evaluated in terms of fluoride removal capacity, sensitivity to pH values and residual Al concentration after defluoridation. The results indicated that the defluoridation performances of different aluminum hydroxides follow the order of boehmite > bayerite > gibbsite. For boehmite, it can remove >60% fluoride in the first 1.5 h, and reach adsorption equilibrium within 12 h. The maximum defluoridation capacities of boehmite, bayerite and gibbsite based on Langmuir isotherm model are 42.08, 2.97 and 2.74 mg m−2. The pH values and FTIR analyses reveal that the fluoride removal is mainly attributed to the ligand exchange between fluoride and surface hydroxyl groups. Different aluminum hydroxides have different surface hydroxyl group density, which results in the different defluoridation capacity. This work provides a new idea for the preparation of aluminum hydroxide with excellent adsorption performance.
ACKNOWLEDGEMENTS
This work is supported by the National Natural Science Foundation of China (Grant Nos. 51872181 and 51801093), the Key scientific research projects of Henan Province Colleges and Universities (No. 22A480006), and the Youth Backbone Teacher of Henan Province (No. 2020GGJS197).
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.