Fluoride removal from water using high-activity aluminum hydroxide prepared by the ultrasonic method

Abstract

Different Al(OH)₃ powders were used as adsorbents for fluoride removal from water. The results showed that the defluoridation performance of ultrasonically prepared Al(OH)₃ (UAH) is much better than that of commercially available Al(OH)₃ and is comparable to that of activated alumina, because the ultrasonic waves effectively break the agglomerates in the suspension so that the UAH particles are fine and have a beneficial phase constituent. Furthermore, the residual aluminum concentration in aqueous solution after defluoridation by Al(OH)₃ was found to be one order of magnitude lower than that obtained with activated alumina which is below the World Health Organization (WHO) guideline for aluminum (0.2 mg L⁻¹) in drinking water. The defluoridation dynamics and mechanism for UAH are discussed in detail.

---

1. Introduction

Fluoride is an essential mineral for the growth of teeth and bones in mammals. Fluoride at a relatively low level (0.5–1.5 mg L⁻¹) in drinking water is beneficial to human health, as it promotes the calcification of dental enamel and the maintenance of healthy bones. However, an excessive intake of fluoride for a long period can result in dental fluorosis, skeletal fluorosis, osteoporosis and brittle bones.¹ Therefore, it is necessary to maintain the fluoride concentration in drinking water at an acceptable level. The WHO has set 1.5 mg L⁻¹ of fluoride as the upper limit in drinking water.²

High fluoride concentration in drinking water comes from two different channels: natural sources and anthropogenic discharge. Fluorides are released into the environment naturally through weathering and the dissolution of rock minerals, leading to a high fluoride concentration in groundwater in some areas. On the other hand, the discharge of wastewater containing fluorides from various industries, e.g. mining, semiconductor fabricating, electroplating, rubber and fertilizer manufacturing etc., is another cause for fluoride enrichment in groundwater. Most rural areas and some urban areas in developing countries use groundwater as the drinking water. As per a conservative estimate, more than 200 million people worldwide are at risk of different forms of “fluorosis” due to excess fluoride in drinking water, especially in Africa, Mexico, China, India, Pakistan, and Thailand.³

Fluoride removal from water has long been an engineering challenge, much effort has been put into developing effective defluoridation technologies. In order to reduce fluoride concentration to an acceptable level in drinking water, different methods have been developed such as precipitation, adsorption, ion exchange, membrane separation and electrodialysis, and reverse osmosis.^4,5 Among the various defluoridation techniques, adsorption is an environmentally friendly and economically viable method due to its flexibility and simplicity of design, relative ease of operation and low cost.

During the past decades, many natural mineral and biopolymer adsorbents have been used to remove fluoride, e.g. alum sludge, hydroxyapatite, aluminum hydroxide coated rice husk ash, surface modified pumice, modified natural siderite, zeolitic tuff, pseudoboehmite and chitosan shell.^6–13 Although natural mineral and biopolymer adsorbents are low in cost, they have low adsorption capacity and require a complex modification procedure to improve their defluoridation performance. In order to enhance the adsorption capacity of adsorbents, some nanometer metal oxides and hydroxides with high surface areas such as nano alumina,^14–16 carbon nanotube supported alumina,¹⁷ manganese-oxide-coated alumina,^18,19 nano boehmite or bayemite,^20–22 double or trimetal hydroxides,^23–27 Fe₃O₄/Al₂O₃, CeO₂/ZrO₂, MnO/Al₂O₃ nanoparticles,^28–30 aluminum sulfate/graphene hydrogel, MgO/MgCO₃, and alumina modified graphite,^31–33 etc. have been developed in recent years.

In general, the criteria for selecting an adsorbent mainly include the adsorption capacity, cost, production technology and water quality after fluoride removal. Among the various adsorbents available, activated alumina has been commercialized and widely used in many countries due to its high binding affinity with F⁻ ions and its cost-effectiveness. However, the main disadvantage of activated alumina is its high residual aluminum concentration in aqueous solution after defluoridation. George et al.³⁴ found that the residual aluminum concentration was up to 1.5 mg L⁻¹ when the initial fluoride concentration was 10 mg L⁻¹ and the activated alumina dose was 4 g L⁻¹, which is far higher than the WHO guideline of 0.2 mg L⁻¹ of aluminum in drinking water. A high concentration of aluminum in drinking water could cause Alzheimer’s disease.³⁵ For the aforementioned nanometer adsorbents, there is not much data relating to the residual aluminum concentration in water after defluoridation. In this work, it was found that a low residual aluminum concentration in drinking water was reached using aluminum hydroxides to remove fluoride, making aluminum hydroxides promising alternatives to activated alumina.

However, the commercial aluminum hydroxides produced by the Bayer process using bauxite have a poor fluoride adsorption capacity.³⁶ In order to improve the defluoridation performance of aluminum hydroxides, Shimelis et al.³⁷ prepared aluminum hydroxide adsorbent through hydrolysis of aluminum sulfate. Jia et al.²² synthesized a feather like bayerite/boehmite adsorbent by a facile one-pot hydrothermal method. However, these methods use alkaline solution and the amount of alkaline solution requires precise control. In a previous work,³⁸ a high-activity Al(OH)₃ was prepared by the reaction of Al with water using an ultrasonic procedure. In this work, the ultrasonically prepared Al(OH)₃ was used as the adsorbent for fluoride removal from water, and it was found that its defluoridation performance is comparable to that of activated alumina.

2. Experimental procedure

Al powder with an average size of 2.25 μm (99.9% purity, Henan Yuan Yang Aluminum Industry Co., China) and two kinds of commercial Al(OH)₃ powders purchased from Tianjin (99.4% purity, Tianjin Damao Chemical Reagent Factory, Tianjin, China, referred to as T-CAH hereafter) and Japan (99.99% purity, 2.5 μm, High Purity Chemical Co., Tokyo, Japan, referred to as J-CAH) were used in the present research. In order to study the defluoridation performance, NaF, NaCl, Na₂SO₄, NaNO₃, Na₂HPO₄, NaOH and HCl (analytical reagent grade, Sinopharm Chemical Reagent Co., Ltd, Shanghai, China) were used in the experiment.

In addition to the commercial Al(OH)₃ powders, two other Al(OH)₃ suspensions and powders were prepared in this work. One was prepared by reacting pure Al powder with water in a closed glass reactor in vacuum (the initial pressure is 7.4 kPa) at 40 °C to form Al(OH)₃ suspension (S-VAH), and then filtering by a filter paper and drying at 60 °C to form Al(OH)₃ powder (VAH).³⁹ Another was prepared by putting pure Al powder into a beaker with deionized water, then ultrasonically treating in an ultrasonic vessel (40 kHz, 100 W) at 40 °C for a time period (∼2 h) to form Al(OH)₃ suspension (S-UAH), and finally filtering by a filter paper and drying at 60 °C to form Al(OH)₃ powder (UAH).³⁸

The fluoride stock solution (1000 mg L⁻¹) was prepared by dissolving an appropriate amount of NaF in deionized water. The fluoride solution used for the adsorption experiments was prepared by diluting the stock solution to a setting concentration using deionized water. Adsorption tests were carried out in 250 mL of fluoride solution, and 1 g of Al(OH)₃ adsorbent (4 g L⁻¹) was used in each test. A magnetic agitation bar was used at a speed of ∼500 rpm to stir the mixture of adsorbent and fluoride solution. At a pre-setting time, 10 mL of sample was taken from the adsorption solution and filtrated through a filter with a 0.22 μm polyethersulfone (PES) resin membrane. The residual F⁻ concentration in the filtrate was measured using an ion chromatograph (IC, Model no. MIC-II, Metrohm Co., Switzerland). The F⁻ removal ratio α can be calculated using the following equation:

where C₀ and C_t are the initial and residual F⁻ concentration at time t, respectively. Moreover, the residual aluminum concentration in filtrate was measured using an inductively coupled plasma atomic emission spectrometer (ICP-AES, Model no. ICAP 6300, Thermo Fisher Co., USA). All the adsorption tests were conducted three times and the average values were reported.

A zeta potential analyzer (type: ZETASIZER 3000HSA, Malvern Instruments Ltd, UK) was used to measure the zeta (ζ) potential of the Al(OH)₃ suspensions prepared by the ultrasonic procedure (S-UAH). A pH meter (Model no. MP512, Shanghai Sanxin Instrument Co., Shanghai, China) was used to measure the pH value of the different solutions. X-ray diffractometry (XRD, Model no. D/max-2200, Rigaku Co., Japan) was used to analyze the phases in different Al(OH)₃ powders. Scanning electron microscopy (SEM, Model no. JSM-6700F, JEOL Co., Japan) was used to observe the morphologies of Al and different Al(OH)₃ powders.

3. Results and discussion

3.1. Characterization of aluminium hydroxides

Fig. 1 shows the morphologies of as-received pure Al and three kinds of Al(OH)₃ powders. It can be seen that the Al particles are spherical, their surfaces are dense and smooth (Fig. 1(a)). The surface morphologies of the three kinds of Al(OH)₃ powders differ. The particles of J-CAH are large and have a columnar structure (Fig. 1(b)). However, UAH has a flower-like structure and its particles are fine (Fig. 1(d)). VAH also has a flower-like structure, but some of its particles are large and dense. From the morphology analysis, it seems that the surface area of Al(OH)₃ powders is in the order of UAH > VAH > J-CAH.

Fig. 1 SEM micrographs of pure Al powder with an average particle size of 2.25 μm (a), commercial Al(OH)₃ powder purchased from Japan (J-CAH) (b), Al(OH)₃ powders prepared by the reaction of 2.25 μm Al powder with water in vacuum (VAH) (c) and using an ultrasonic procedure (UAH) (d), respectively.

Fig. 2 shows the X-ray diffraction patterns of four kinds of Al(OH)₃ powders. It can be seen that T-CAH and J-CAH have a phase of gibbsite and bayerite, respectively. VAH has a phase of bayerite, but UAH has a phase composition of bayerite and boehmite. Furthermore, the diffraction peaks of UAH are wider than those of other Al(OH)₃ powders (see the red arrows), so the grain sizes in UAH are smaller than those in other Al(OH)₃ powders, implying that the surface area of UAH is the largest of the powders studied, which is consistent with the morphology observation presented in Fig. 1.

Fig. 2 X-ray diffraction patterns of different Al(OH)₃ powders: (a) T-CAH, (b) J-CAH, (c) VAH and (d) UAH.

3.2. Fluoride removal performance

Fig. 3(a) shows fluoride removal from aqueous solution with an initial F⁻ concentration of 20 mg L⁻¹ using different Al(OH)₃ powders or suspensions at 25 °C. It can be seen that the preparation methods have a significant impact on the defluoridation performance of Al(OH)₃. The commercial Al(OH)₃ powders have a poor F⁻ adsorption capacity, the final fluoride removal by T-CAH and J-CAH is only 9.5% and 27.3%, respectively. The Al(OH)₃ powders and suspensions prepared by the reaction of Al with water in vacuum or using an ultrasonic procedure have a better defluoridation performance than the commercial Al(OH)₃ powders. For the Al(OH)₃ prepared by the reaction of Al with water, its powders (UAH and VAH) have a quicker defluoridation behavior than its suspensions (S-UAH and S-VAH), but the final fluoride removal by UAH and VAH is lower than that by S-UAH and S-VAH. Among these Al(OH)₃ powders and suspensions, UAH has the quickest defluoridation performance and it removes ∼75% of fluoride within 5 h, which corresponds to a defluoridation capacity of ∼3.8 mg g⁻¹ UAH. This indicates that the defluoridation capacity of UAH is comparable, under similar conditions, to that of the activated alumina (∼2.4 mg g⁻¹ activated alumina) discussed in ref. 14.

Fig. 3 Fluoride removal from aqueous solution at 25 °C (a) with an initial F⁻ concentration of 20 mg L⁻¹ using different Al(OH)₃ powders or suspensions and (b) with different initial F⁻ concentrations using UAH.

Fig. 3(b) shows fluoride removal from aqueous solution with different initial F⁻ concentration using UAH at 25 °C. It can be seen that the initial F⁻ concentration has a significant effect on the fluoride removal ratio, which decreases when the initial F⁻ concentration is increased. When the initial F⁻ concentration is 10 mg L⁻¹, more than 90% of fluoride is removed within 3 h and the residual F⁻ concentration is ∼1.0 mg L⁻¹, which is within the guideline value of WHO. This implies that UAH is an effective and viable adsorbent for defluoridation if the F⁻ concentration in groundwater is below 10 mg L⁻¹.

The effect of initial pH value in aqueous solution on the fluoride removal using UAH was investigated, as shown in Fig. 4(a), where fluoride solutions with different initial pH values were obtained by adding a suitable amount of HCl or NaOH. It can be seen that the final fluoride removal is closely related to the initial pH value. When the initial pH value is between 5 and 9, it has almost no impact on the fluoride removal. When the initial pH value decreases from 5 to 3, the fluoride removal increases. When the initial pH value is >9, fluoride removal decreases with an increase in the pH value.

Fig. 4 (a) Effect of pH value on fluoride removal from aqueous solution using UAH at 25 °C (initial F⁻ concentration = 20 mg L⁻¹, contact time = 24 h) and (b) zeta potential curve for S-UAH.

In fact, the F⁻ adsorption onto Al(OH)₃ or AOOH is closely related to its surface chemical characteristics, changing the pH value leads to a change in its zeta potential. Fig. 4(b) shows the zeta potential curve of S-UAH, it can be seen that the isoelectric point of S-UAH is at ∼10.3. When the pH value is below this isoelectric point, the zeta potential of Al(OH)₃ suspension is positive, i.e. the Al(OH)₃ particles have a positive surface charge.⁴⁰ In this case, there is an electrostatic attraction between the Al(OH)₃ surfaces and F⁻ ions, so decreasing the pH value from 5 to 3 promotes the F⁻ adsorption onto UAH. However, when the pH value is above the isoelectric point, the Al(OH)₃ particles have a negative surface charge and there is an electrostatic repulsion between the Al(OH)₃ surfaces and F⁻ ions, so increasing the pH value inhibits the F⁻ adsorption when the pH value is >9 (Fig. 4(a)). Moreover, there is a competition between the OH⁻ and F⁻ ions for the adsorption sites on Al(OH)₃ surfaces at alkaline pH range.⁴⁰

In addition to fluoride, drinking water often contains other anions such as Cl⁻, SO₄²⁻, NO₃⁻ and HPO₄²⁻, etc.,⁴¹ and the co-existing anions in water may influence the defluoridation performance of UAH. Fig. 5(a) shows the fluoride removal from a solution containing 50 mg L⁻¹ of different co-existing anions separately using UAH at 25 °C. It can be seen that Cl⁻ and NO₃⁻ ions almost have no effect on the fluoride removal. However, SO₄²⁻ and HPO₄²⁻ ions inhibit the F⁻ adsorption onto UAH, this is more marked for HPO₄²⁻ ions, which decrease the fluoride removal from ∼80% to just 24.4%. Fig. 5(b) shows the residual co-existing anions in aqueous solution after defluoridation by UAH. It can be seen that the concentrations of Cl⁻ and NO₃⁻ do not change before and after defluoridation, but the concentrations of SO₄²⁻ and HPO₄²⁻decrease to 47.0 and 4.4 mg L⁻¹, respectively, after defluoridation. This indicates that SO₄²⁻ and HPO₄²⁻ ions are also adsorbed onto UAH surfaces, and there is competition between F⁻ and, SO₄²⁻and HPO₄²⁻ ions for the adsorption sites on the UAH surfaces.²⁰

Fig. 5 (a) Effect of co-existing anions on fluoride removal from aqueous solution using UAH at 25 °C and (b) residual co-existing anions in aqueous solution after defluoridation (initial co-existing anion concentration = 50 mg L⁻¹, initial F⁻ concentration = 20 mg L⁻¹, contact time = 24 h).

Fig. 6 shows fluoride removal from solutions with different initial F⁻ concentration using UAH at different temperatures. It can be seen that the defluoridation performance of UAH decreases with an increase in the temperature. When the temperature increases from 25 °C to 55 °C, the fluoride removal decreases from 92.6% to 82.5% and from 68.1% to 45.4% for initial F⁻ concentrations of 10 and 40 mg L⁻¹, respectively, implying that F⁻ adsorption onto UAH is an exothermic process.⁴²

Fig. 6 Effect of temperature on fluoride removal from aqueous solution using UAH, where the initial F⁻ concentrations in (a) and (b) are 10 and 40 mg L⁻¹, respectively.

Our adsorption dynamics analyses indicated that F⁻ adsorption onto UAH is pseudo-second-order adsorption. The adsorption isotherm showed that F⁻ adsorption onto UAH is a chemical adsorption process. Thermodynamic study confirmed that F⁻ adsorption onto UAH is an exothermic process, which is favorable and spontaneous in nature (the details are given in the ESI material†).

3.3. Residual aluminum

Aluminum will dissolve into water from an adsorbent surface when Al-based adsorbent is used for water treatment. Fig. 7 shows the residual aluminum in aqueous solution after defluoridation using different Al(OH)₃ powders or suspensions at 25 °C, the results obtained for activated alumina under similar conditions are presented for comparison.³⁴ Compared with the activated alumina, the residual aluminum concentrations in aqueous solution after defluoridation using different Al(OH)₃ powders or suspensions are low, and are below the WHO guideline of 0.2 mg L⁻¹ of aluminum in drinking water. Moreover, the residual aluminum concentration in aqueous solution after defluoridation by S-UAH is 0.17 mg L⁻¹, which is higher than those obtained after defluoridation using other Al(OH)₃ powders. This is because some of the Al(OH)₃ particles in S-UAH are probably amorphous, and the solubility of amorphous Al(OH)₃ in water is higher than that of crystalline Al(OH)₃.⁴³

Fig. 7 Residual Al in aqueous solution at 25 °C after defluoridation using different adsorbents (initial F⁻ concentration = 20 mg L⁻¹, contact time = 24 h). The result of activated alumina (AA) is added for comparison³⁴ and the red line represents the WHO guideline for aluminum concentration in drinking water.

Fig. 8–10 show the effect of pH value, initial F⁻ concentration and temperature on the residual aluminum concentration in aqueous solution after defluoridation by UAH. When the pH value is between 5.5 and 8.5, the pH value has a small effect on the residual aluminum concentration in aqueous solution, but it is in the range of 0.040–0.055 mg L⁻¹ (Fig. 8), which is below the WHO guideline. In fact, the pH value of drinking water is about 6.5–8.5, so UAH is suitable for fluoride removal in drinking water. Fig. 9 indicates that the initial F⁻ concentration has a significant impact on the residual aluminum concentration and it increases with an increase in the F⁻ concentration. When the initial F⁻ concentration is >40 mg L⁻¹, the residual aluminum concentration exceeds the WHO guideline. The possible cause is that at higher initial F⁻ concentration, Al(OH)₃ interacts with F⁻ ions to form more AlF_x species. As AlF_x has a definite solubility in water, more residual aluminum appears in the solution.⁴⁴ Fig. 10 indicates that the residual aluminum concentration increases with an increase in the temperature, because the solubility of Al(OH)₃ in water increases with an increase in the temperature.

Fig. 8 Residual Al in aqueous solutions with different initial pH values at 25 °C after defluoridation using UAH (initial F⁻ concentration = 20 mg L⁻¹, contact time = 24 h).

Fig. 9 Residual Al in aqueous solution with different initial F⁻ concentration at 35 °C after defluoridation using UAH (contact time = 24 h).

Fig. 10 Residual Al in aqueous solution at different temperatures after defluoridation using UAH (initial F⁻ concentration = 10 mg L⁻¹, contact time = 24 h).

3.4. Desorption

An adsorbent is economically effective if the adsorbent can be regenerated and reused. In order to study the desorption capacity of the adsorbent, used UAH was prepared by filtering the aqueous solution with an initial F⁻ concentration of 20 mg L⁻¹ after defluoridation for 24 h at 25 °C and then drying the filtered powder. Desorption tests were then carried out on the used UAH, in each test 0.5 g of the used UAH was added to 100 mL of NaOH solution and then stirred with a magnetic bar at a speed of 500 rpm for 10 h. The desorption capacity of the used UAH was obtained by measuring the F⁻ concentration in the desorption solution. Fig. 11 shows the effect of the pH value of aqueous solutions on the F⁻ desorption from the used UAH at 25 °C. It can be seen that few F⁻ ions were released from the used UAH at a pH value < 11. However, the F⁻ desorption increased abruptly when the pH value was >11. At a pH value of 14, the used UAH released 93.1% of F⁻ ions within 10 h, indicating that after defluoridation the UAH could be regenerated using NaOH solution.²⁰

Fig. 11 Effect of pH value in aqueous solution on desorption of F⁻ ions from the used UAH at 25 °C, where the desorption time is 10 h and the used UAH adsorbed fluoride in an aqueous solution with an initial F⁻ concentration of 20 mg L⁻¹ at 25 °C for 24 h.

3.5. Physicochemical mechanisms

It is known that there is a layer of hydroxyl groups on Al(OH)₃ particle surfaces. When Al(OH)₃ particles are placed in an aqueous solution containing F⁻ ions, the Al(OH)₃ particles interact with the F⁻ ions by a surface complexation reaction, which is responsible for the F⁻ adsorption. To a certain degree, a surface hydroxyl group can be viewed as a Lewis base, which has an oxygen atom as a donor that can coordinate with protons or Al ions. The Al ions are Lewis acids, which can exchange the hydroxyl groups for other coordinating anions.⁴⁵ Thus, the specific adsorption of F⁻ ions onto Al(OH)₃ particle surfaces can be described as ligand exchange, where an hydroxyl group is exchanged for a F⁻ ion. According to the ligand exchange model, such an adsorption is a simple exchange of an hydroxyl group for a F⁻ ion

F⁻ + ≡AlOH ↔ ≡AlF + OH⁻ (2)

where ≡AlOH is the aluminol group.⁴⁶ According to eqn (2), the Al(OH)₃ will release a OH⁻ ion after adsorbing a F⁻ ion, resulting in an increase of pH value.

In order to validate the adsorption mechanism proposed above, Table 1 gives the pH values of the solutions with different F⁻ concentrations before and after defluoridation by different Al(OH)₃ at 25 °C. It can be seen that the pH value increases after defluoridation by Al(OH)₃. The change (ΔpH) in pH value for UAH is higher than for other Al(OH)₃ adsorbents, the ΔpH increases with an increase in the initial F⁻ concentration and is proportional to the amount of F⁻ removal (Fig. 3). This confirms the above proposed mechanism, because more OH⁻ ions are released when more F⁻ ions adsorb onto Al(OH)₃ particle surfaces. As the increase in reaction by-product inhibits the reversible reaction, the F⁻ removal ratio decreases with increasing F⁻ concentration (Fig. 3(b)).

Table 1 pH values of the solution before and after defluoridation using different adsorbents at 25 °C

Adsorbent	pH (before defluoridation)	pH (after defluoridation)	ΔpH
T-CAH, 20 mg L^−1	7.18	7.36	0.18
J-CAH, 20 mg L^−1	7.22	7.51	0.29
S-VAH, 20 mg L^−1	7.37	7.76	0.39
S-UAH, 20 mg L^−1	7.53	8.27	0.74
VAH, 20 mg L^−1	7.38	7.62	0.24
UAH, 10 mg L^−1	6.82	6.98	0.16
UAH, 20 mg L^−1	7.03	7.51	0.48
UAH, 30 mg L^−1	7.35	8.05	0.70
UAH, 40 mg L^−1	7.46	8.53	1.07

---

From the above analyses, it is clear that F⁻ ion adsorption onto Al(OH)₃ particles is closely related to their surface hydroxyl groups. The surface hydroxyl group density and surface area of Al(OH)₃ are two key factors affecting its F⁻ removal efficiency. In fact, different aluminum oxides and hydroxides have different surface hydroxyl group density. For example, the surface hydroxyl group density of α-Al₂O₃, boehmite and gibbsite is 6, 16.5 and 12 groups per square nanometer, respectively.⁴⁶ This may be one reason why T-CAH and J-CAH have different F⁻ removal performance, because they have different Al(OH)₃ phases. Moreover, the good adsorption kinetics of UAH probably also results from its high surface hydroxyl group density, because UAH has a phase composition of bayerite and boehmite (Fig. 2), and boehmite has a high surface hydroxyl group density as mentioned above.

It is well known that the sonochemical effect of ultrasound in a liquid–solid system mainly arises from acoustic cavitation.⁴⁷ During cavitation, bubble collapse produces intense heating and high pressure at the interfacial region around cavitation bubbles, which promote the formation of boehmite during the UAH preparation process, because boehmite forms at a higher temperature than bayerite.⁴⁸ As aforementioned, the existence of boehmite in UAH is favorable for the removal of F⁻. In addition, ultrasonic cavitation can create microjets and shock waves, which effectively break agglomerates and increase the Al(OH)₃ and AlOOH surface area, leading to a higher F⁻ removal ratio of UAH than VAH, T-CAH and J-CAH (Fig. 3(a)).

The above analyses indicated that the ultrasonic procedure is an important reason for the high F⁻ removal of UAH. In order to determine whether the ultrasonic procedure can improve the F⁻ removal of J-CAH and VAH, J-CAH and S-VAH were ultrasonically treated for 1 h and then used in F⁻ adsorption tests. Fig. 12 and 13 show F⁻ removal from aqueous solution with an initial F⁻ concentration of 20 mg L⁻¹ at 25 °C using J-CAH, VAH, S-VAH and those after the ultrasonic treatment. It can be seen that the ultrasonic treatment has a negligible impact on the removal of F⁻by J-CAH (Fig. 12). However, the ultrasonic treatment greatly improves the removal of F⁻by VAH and S-VAH (Fig. 13).

Fig. 12 Fluoride removal from aqueous solution with an initial F⁻ concentration of 20 mg L⁻¹ at 25 °C using commercial Al(OH)₃ powder (J-CAH) and those after the ultrasonic treatment, where “S-J-CAH, US 1 h” is the Al(OH)₃ suspension prepared by adding J-CAH into deionized water and then ultrasonically treating for 1 h, and “J-CAH, US 1 h” is the Al(OH)₃ powder obtained through filtering and drying the “S-J-CAH, US 1 h”. For comparison, the defluoridation curves of S-UAH and UAH are added.

Fig. 13 Fluoride removal from aqueous solution with an initial F⁻ concentration of 20 mg L⁻¹ at 25 °C using (a) S-VAH and (b) VAH and those after ultrasonic treatment, where “S-VAH, US 1 h” is the Al(OH)₃ suspension prepared by ultrasonically treating S-VAH for 1 h and “VAH, US 1 h” is the Al(OH)₃ powder obtained by filtering and drying the “S-VAH, US 1 h”. For comparison, the defluoridation curves of S-UAH and UAH are added.

In fact, there are many Al(OH)₃ soft agglomerates in S-VAH due to its lack of breaking mechanism. After the ultrasonic treatment, the soft agglomerates in S-VAH are broken, which increases the Al(OH)₃ surface area, leading to an increase in its F⁻ removal. However, the agglomerates in J-CAH are hard due to the sintering and grain growth during drying, coalescence between particles occurs and the bonding strength between the particles is strong.³⁹ Therefore, it is not easy to break these hard agglomerates, and the surface area of J-CAH almost has no change after the ultrasonic treatment. This is why the ultrasonic treatment has a negligible impact on the F⁻ removal by J-CAH.

In fact, the Al–water reaction is a promising hydrogen-generation technology for portable fuel cell application and Al(OH)₃ is the byproduct of Al–water reaction.^49–51 Therefore, the Al–water reaction byproduct can be used to remove fluoride in drinking water through ultrasonic treatment, which provides an ideal and economic way to dispose of the Al–water reaction byproduct.

4. Conclusions

In this work, a high-activity Al(OH)₃ powder prepared by the reaction of Al with water using the ultrasonic procedure (UAH) was used as the adsorbent for the removal of fluoride from water. It was found that the defluoridation performance of UAH is much better than that of commercial Al(OH)₃ powders and is comparable to that of activated alumina. The residual aluminum concentration in aqueous solution after defluoridation by UAH is much lower than that obtained after defluoridation by activated alumina, making UAH a promising alternative to activated alumina. The mechanism analyses revealed that ultrasonic cavitation effectively breaks the agglomerates in Al(OH)₃ suspension and promotes the formation of a boehmite phase with a high surface hydroxyl group density, leading to the high fluoride removal capacity of UAH. The present work provides a new way of disposing of Al(OH)₃ produced by the Al–water reaction used for hydrogen generation.

Acknowledgements

We would like to thank the Innovation Program of Shanghai Municipal Education Commission (13ZZ079) and the “085 project” of Shanghai Municipal Education Commission for their financial support.

References

1. Meenakshi and R. C. Maheshwari, J. Hazard. Mater., 2006, 137, 456–463.
2. A. Bhatnagar, E. Kumar and M. Sillanpää, Chem. Eng. J., 2011, 171, 811–840.
3. S. Ayoob, A. K. Gupta and V. T. Bhat, Crit. Rev. Environ. Sci. Technol., 2008, 38, 401–470.
4. M. Mohapatra, S. Anand, B. K. Mishra, D. E. Giles and P. Singh, J. Environ. Manage., 2009, 91, 67–77.
5. E. Kumar, A. Bhatnagar, W. Hogland, M. Marques and M. Sillanpää, Chem. Eng. J., 2014, 241, 443–456.
6. M. G. Sujana, R. S. Thakur and S. B. Rao, J. Colloid Interface Sci., 1998, 206, 94–101.
7. X. Fan, D. J. Parker and M. D. Smith, Water Res., 2003, 37, 4929–4937.
8. V. Ganvir and K. Das, J. Hazard. Mater., 2011, 185, 1287–1294.
9. A. Salifu, B. Petrusevski, K. Ghebremichael, L. Modestus, R. Buamah, C. Aubry and G. L. Amy, Chem. Eng. J., 2013, 228, 63–74.
10. M. N. Sepehr, V. Sivasankar, M. Zarrabi and M. S. Kumar, Chem. Eng. J., 2013, 228, 192–204.
11. Y. Shan and H. M. Guo, Chem. Eng. J., 2013, 223, 183–191.
12. A. Teutli-Sequeira, M. Solache-Ríos, V. Martínez-Miranda and I. Linares-Hernández, J. Colloid Interface Sci., 2014, 418, 254–260.
13. Z. Wan, W. Chen, C. Liu, Y. Liu and C. L. Dong, J. Colloid Interface Sci., 2015, 443, 115–124.
14. S. Ghorai and K. K. Pant, Chem. Eng. J., 2004, 98, 165–173.
15. L. M. Camacho, A. Torres, D. Saha and S. G. Deng, J. Colloid Interface Sci., 2010, 349, 307–313.
16. E. Kumar, A. Bhatnagar, U. Kumar and M. Sillanpää, J. Hazard. Mater., 2011, 186, 1042–1049.
17. Y. H. Li, S. Wang, X. Zhang, J. Wei, C. Xu, Z. Luan, D. Wu and B. Wei, Environ. Technol., 2003, 24, 391–398.
18. S. M. Maliyekkal, A. K. Sharma and L. Philip, Water Res., 2006, 40, 3497–3506.
19. S. X. Teng, S. G. Wang, W. X. Gong, X. W. Liu and B. Y. Gao, J. Hazard. Mater., 2009, 168, 1004–1011.
20. S. G. Wang, Y. Ma, Y. J. Shi and W. X. Gong, J. Chem. Technol. Biotechnol., 2009, 84, 1043–1050.
21. R. P. Liu, W. X. Gong, H. C. Lan, Y. P. Gao, H. J. Liu and J. H. Qu, Chem. Eng. J., 2011, 175, 144–149.
22. Y. Jia, B. S. Zhu, Z. Jin, B. Sun, T. Luo, X. Y. Yu, L. T. Kong and J. H. Liu, J. Colloid Interface Sci., 2015, 440, 60–67.
23. J. B. Zhou, Y. Cheng, J. G. Yu and G. Liu, J. Mater. Chem., 2011, 21, 19353–19361.
24. Q. Chang, L. H. Zhu, Z. H. Luo, M. Lei, S. C. Zhang and H. Q. Tang, Ultrason. Sonochem., 2011, 18, 553–561.
25. B. Zhao, Y. Zhang, X. M. Dou, X. M. Wu and M. Yang, Chem. Eng. J., 2012, 185–186, 211–218.
26. X. M. Wu, Y. Zhang, X. M. Dou, B. Zhao and M. Yang, Chem. Eng. J., 2013, 223, 364–370.
27. T. Zhang, Q. R. Li, H. Y. Xiao, Z. Y. Mei, H. X. Lu and Y. M. Zhou, Appl. Clay Sci., 2013, 72, 117–123.
28. L. Y. Chai, Y. Y. Wang, N. Zhao, W. C. Yang and X. Y. You, Water Res., 2013, 47, 4040–4049.
29. J. Wang, W. H. Xu, L. Chen, Y. Jia, L. Wang, X. J. Huang and J. H. Liu, Chem. Eng. J., 2013, 231, 198–205.
30. S. Alemu, E. Mulugeta, F. Zewge and B. S. Chandravanshi, Environ. Technol., 2014, 35, 1893–1903.
31. Y. Q. Chen, Q. K. Zhang, L. B. Chen, H. Bai and L. Li, J. Mater. Chem. A, 2013, 1, 13101–13110.
32. K. S. Zhang, S. B. Wu, X. L. Wang, J. Y. He, B. Sun, Y. Jia, T. Luo, F. L. Meng, Z. Jin, D. Y. Lin, W. Shen, L. T. Kong and J. H. Liu, J. Colloid Interface Sci., 2015, 446, 194–202.
33. H. Y. Jin, Z. J. Ji, J. Yuan, J. Li, M. Liu, C. H. Xu, J. Dong, P. Hou and S. Hou, J. Alloys Compd., 2015, 620, 361–367.
34. S. George, P. Pandit and A. B. Gupta, Water Res., 2010, 44, 3055–3064.
35. V. Rondeau, D. Commenges, H. Jacqmin-Gadda and J. F. Dartigues, Am. J. Epidemiol., 2000, 152, 59–66.
36. F. Habashi, Hydrometallurgy, 2005, 79, 15–22.
37. B. Shimelis, F. Zewge and B. S. Chandravanshi, Bull. Chem. Soc. Ethiop., 2006, 20, 17–34.
38. Y. Yang, W. Z. Gai, Z. Y. Deng and J. G. Zhou, Int. J. Hydrogen Energy, 2014, 39, 18734–18742.
39. W. Z. Gai and Z. Y. Deng, Int. J. Hydrogen Energy, 2014, 39, 13491–13497.
40. Y. Ku and H. M. Chiou, Water, Air, Soil Pollut., 2002, 133, 349–360.
41. W. Z. Gai and Z. Y. Deng, J. Power Sources, 2014, 245, 721–729.
42. M. G. Sujana, G. Soma, N. Vasumathi and S. Anand, J. Fluorine Chem., 2009, 130, 749–754.
43. J. L. R. Bahena, A. R. Cabrera, A. L. Valdivieso and R. H. Urbina, Sep. Sci. Technol., 2002, 37, 1973–1987.
44. H. Farrah, J. Slavek and W. F. Pickering, Aust. J. Soil Res., 1987, 25, 55–69.
45. R. Kummert and W. Stumm, J. Colloid Interface Sci., 1980, 75, 373–385.
46. G. Sposito, Environmental chemistry of aluminum, CRS Press, 1996.
47. K. S. Suslick and G. J. Price, Annu. Rev. Mater. Sci., 1999, 29, 295–326.
48. W. Z. Gai, W. H. Liu, Z. Y. Deng and J. G. Zhou, Int. J. Hydrogen Energy, 2012, 37, 13132–13140.
49. Z. Y. Deng, J. M. F. Ferreira, Y. Tanaka and J. H. Ye, J. Am. Ceram. Soc., 2007, 90, 1521–1526.
50. Z. Y. Deng, J. M. F. Ferreira and Y. Sakka, J. Am. Ceram. Soc., 2008, 91, 3825–3934.
51. C. S. Fang, W. Z. Gai and Z. Y. Deng, J. Am. Ceram. Soc., 2014, 97, 44–47.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra14706a

---