The Adsorptive Removal of Fluoride from Aqueous Solution by Modified Sludge: Optimization Using Response Surface Methodology

The sludge from the water supply plant was investigated to remove fluoride ions from the water. To improve the adsorption ability, the original sludge sample was treated with fuel oxidation, pyrolysis, hydrochloric acid, and sulphuric acid methods, and hydrochloric acid treatment improved the adsorption capacity of the sludge on the fluoride in water significantly, with a maximum adsorption capacity to 140 mg/kg. The adsorption experimental data was the well fitted pseudo-first-order model and the Langmuir isotherms model. SEM images and XRD patterns of the adsorbent were recorded to get a better insight into the adsorption process. The effect of three variables, hydrochloric acid treated sludge (HWS) dose, pH, and initial fluoride concentration were studied using a Box-Behnken statistical experimental design. The model of the adsorption and optimum conditions was investigated using the response surface methodology. The optimum removal efficiency of fluoride can reach 81.153% under the optimum condition: HWS dose of 14.10 g/L and pH value at 6.12. The effect of co-existing anions and the removal efficiency from the water were also studied. The results suggest that sludge from the water supply plant can be reused as a coagulant for the removal of fluoride from poor quality water.


Introduction
As a natural element, fluoride is universally present in varied water bodies, and it is considered beneficial up to 0.7 mg/L but detrimental if it exceeds 1.5 mg/L, which is the limit recommended by the World Health Organization (2004). Fluoride pollution has been a global environmental concern for decades and has caused great concerns due to its widespread nature and threat to human health. Human exposure to over 1.5 mg/L of fluoride often causes a lot of health problems through drinking water, such as brain damage, dental fluorosis and skeletal fluorosis, and thyroid disorder [1]. Fluoride-polluted surface water and groundwater is ubiquitous, especially in India, Africa, and the southwest of China. Fluoride has been found in groundwater in many parts of the globe, and it is present in at least 25 countries [2,3].
In the past decades, various techniques have been developed to remove excessive fluoride ions from river water, including coagulation-precipitation [4], membrane-based process [5], ion exchange [6], and adsorption [7,8]. In generally, adsorption has been regarded as one of the most important and widely used approaches in the defluoridation of wastewater [9]. Many adsorbents have been used for defluoridation, including activated alumina [10], carbonaceous materials [11], activated clay [12],

Experimental Process
Adsorption experiments were conducted by Organization for Economic Co-operation and Development (OECD). The adsorbent mixed with solution was placed in the flask with magnetic stirrer (Model 04803-02, Cole-Palmer-Instrument Company, Vernon Hills, IL, USA). The residual fluoride concentration was measured immediately after filtration using the microfiltration membrane. All the experiments were performed in triplicate, and the mean values were reported.

Reagent and Standard Solutions
To simulate natural water, fluoride stock solution was prepared by dissolving anhydrous sodium fluoride (99.0%, NaF, Kemiou Chemical Reagent Co. Ltd., Tianjin, China) with distilled water. Standards F − samples at a required concentration range were prepared using appropriate dilution of the stock solution.
The pH of the solution was adjusted by adding 0.1 mol/L HCl or NaOH solutions (Kemiou Chemical Reagent Co. Ltd., Tianjin, China). The pH values were periodically measured and readjusted until they were constant. The co-existing anions (SO 4 2− , NO 3 − , SiO 3 2− , PO 4 − ,) solution was prepared with corresponding sodium salts to the concentration of 5-10 mg/L.

Adsorption Kinetics and Isotherm
Adsorption experiment was carried out with four treated sludges at first. The fluoride removal efficiency of four treated sludges was investigated by using 2 g sludge and 48 mL of the 10 mg/L F − stock solution in 200 mL conical flask, respectively, to keep the constant initial fluoride concentration of 2.4 mg/L. In order to obtain an appropriate contact time between the HWS and fluoride ions, the samples were taken at time intervals of 2,4,8,16,20,24,30,36,40,48,60, and 70 min, then allowed to settle, and residual fluoride ion concentration was measured. The kinetics of fluoride adsorption on the HWS was determined under the initial fluoride of 2.4 mg/L and HWS dose at 12 g/L by using two different kinetic models, which are the pseudo-first-order model and pseudo-second-order model [22,23].
Isotherm experiments were conducted for the equilibration time of 70 mins by varying the fluoride concentration from 0.5 to 5 mg/L and the constant HWS dose at 15 g/L. The experimental data was fitted to Fruendlich and Langmuir isotherm models [24]. The removal efficiency (η) of fluoride ions and equilibrium sorption could be obtained by the Equations (1) and (2): where η (%) is the removal efficiency of fluoride ions, C 0 , C e are the initial and equilibrium concentrations of fluoride ions (mg/L), Q e is the equilibrium sorption (mg/g) at equilibrium, V (L) is the volume of the aqueous solution, and m is the mass (g) of adsorbent used in the experiments.

Batch Experiments of Variable Condition
The main factors affecting the adsorption of HWS were studied by RSM, including sludge doses (5, 10, 15 g/L), initial fluoride concentration (1, 3, 5 mg/L), and pH (4,7,10). RSM is one such statistical technique and is used for designing experiments, building models, evaluating the effects of several variables, and obtaining the optimum conditions for responses with a limited number of planned experiments [25,26].
Initially, the main experiments were conducted according to Box-Behnken design (BBD) to facilitate the modeling and optimization of the process (n = 17). Batch experiments for modeling and optimization were conducted according to design matrix presented in Table 1. For each experiment, 100 mL of solution with desired fluoride concentrations and HWS dose was mixed at pH = 4, 7, 10, for 70 min, respectively. Finally, the supernatant was drained, and fluoride concentration was determined. To testify the removal efficiency of the fluoride ions from water under optimal conditions, the adsorption experiment was studied with three water samples. W1 was collected from the effluent of a typical fluorite tailing pond, and W2 and W3 were collected from the effluent of two different glass processing factories. The concentration of three water samples are shown in Table 2.

Analysis
An ionic-activity meter (PXS-215, Shanghai instrument electric science instrument, Shanghai, China) equipped with combination fluoride-selective electrode (PF-1, Shanghai instrument electric science instrument, Shanghai, China) was employed for the measurement of fluoride ion concentration. The pH was measured with pH meter (DELTA320, Mettler Toldedo, Zurich, Switzerland). The compositions of sludge were analyzed using X-ray diffraction (XRD, Shimadzu, Kyoto, Japan), and the surface morphologies of sludge were examined using scanning electron microscope (SEM, JSM-6700F, FEI NanoPorts, Hillsboro, AL, USA). The surface area, pore area, pore volume, and average pore size determinations were carried out by N 2 adsorption isotherms using a Micrometrics ASAP 2020K surface area analyzer (Micromeritics Instrument Corp., Atlanta, GA, USA).

Effects of Four Treated Sludges when Removing Fluoride
The compositions of WS for the experiment are shown in Table 3. The WS contained large number of Al 2 O 3 and Fe 2 O 3 through the coagulation with the addition of polyaluminum chloride (PAC), aluminum sulfate (Al 2 (SO 4 ) 3 ), and ferric chloride (FeCl 3 ) in the process of producing water. SiO 2 was brought by the sludge in the settlement process . Compared to the WS, alum sludge coming from the alum plant [17] contained almost the same content of Al 2 O 3 and Fe 2 O 3 but less SiO 2 and more TiO 2 , which was decided by the raw materials and the production process. In order to improve the adsorption ability, the original sludge sample was treated with different methods. Meanwhile, fluorine removal experiments were carried out with the treated samples. The results shown in Figure 1 indicate that acid treatment can improve the ability of sludge to adsorb fluoride from water. Compared to other methods of treatment with original sludge, the sludge after acid treatment has the best efficiency, and HCl-acidified sludge is better than H 2 SO 4 . The fuel oxidation treatment can slightly improve the removal efficiency, while the pyrolysis displays a little suppression ( Figure 1).

Equilibrium and Kinetics of Adsorption
Adsorption kinetics, demonstrating the solute uptake rate, is one of the most important characters that represent the adsorption efficiency of the HWS. Two possible models of kinetics were used to fit the experimental results ( Figure 2a): (1) pseudo-first-order model and (2) pseudo-secondorder model [27,28].
As shown in Table 4, the pseudo-first-order model has a better fit with the experimental data with the higher squared correlation coefficients (R 2 = 0.9894). The adsorption reaction is fast during the initial 40 min, and equilibrium was reached around 70 min. This result was in agreement with the equilibrium time found for alum sludge and quartz, which are reported in other studies [17,24].
The analysis of Langmuir and Freundlich isotherm models [29] were presented in Figure 2b and Table 5. The RL values in Langmuir model are favorable (0 < RL< 1) [30]. The higher regression correlation coefficient (0.9959) was observed for Langmuir model, indicating that the Langmuir model was the most suitable for describing the adsorption equilibrium, meaning the formation of fluoride ion at the outer surface of the HWS was monolayer coverage [22].

Equilibrium and Kinetics of Adsorption
Adsorption kinetics, demonstrating the solute uptake rate, is one of the most important characters that represent the adsorption efficiency of the HWS. Two possible models of kinetics were used to fit the experimental results ( Figure 2a): (1) pseudo-first-order model and (2) pseudo-second-order model [27,28].
As shown in Table 4, the pseudo-first-order model has a better fit with the experimental data with the higher squared correlation coefficients (R 2 = 0.9894). The adsorption reaction is fast during the initial 40 min, and equilibrium was reached around 70 min. This result was in agreement with the equilibrium time found for alum sludge and quartz, which are reported in other studies [17,24].
The analysis of Langmuir and Freundlich isotherm models [29] were presented in Figure 2b and Table 5. The R L values in Langmuir model are favorable (0 < R L < 1) [30]. The higher regression correlation coefficient (0.9959) was observed for Langmuir model, indicating that the Langmuir model was the most suitable for describing the adsorption equilibrium, meaning the formation of fluoride ion at the outer surface of the HWS was monolayer coverage [22].

Equilibrium and Kinetics of Adsorption
Adsorption kinetics, demonstrating the solute uptake rate, is one of the most important characters that represent the adsorption efficiency of the HWS. Two possible models of kinetics were used to fit the experimental results ( Figure 2a): (1) pseudo-first-order model and (2) pseudo-secondorder model [27,28].
As shown in Table 4, the pseudo-first-order model has a better fit with the experimental data with the higher squared correlation coefficients (R 2 = 0.9894). The adsorption reaction is fast during the initial 40 min, and equilibrium was reached around 70 min. This result was in agreement with the equilibrium time found for alum sludge and quartz, which are reported in other studies [17,24].
The analysis of Langmuir and Freundlich isotherm models [29] were presented in Figure 2b and Table 5. The RL values in Langmuir model are favorable (0 < RL< 1) [30]. The higher regression correlation coefficient (0.9959) was observed for Langmuir model, indicating that the Langmuir model was the most suitable for describing the adsorption equilibrium, meaning the formation of fluoride ion at the outer surface of the HWS was monolayer coverage [22].    Note: Q e,exp refers to actual equilibrium adsorption capacity; Q e,cal refers to the fitted theoretical equilibrium adsorption capacity; K 1 , K 2 is the rate constant of the kinetics models. Table 5. Adsorption isotherm constants for fluoride adsorption onto HWS.

Surface Area Analysis
To further explore the surface area of the untreated and hydrochloric acid water sludge samples, the BET adsorption experiment was employed. Adsorption-desorption N 2 isotherms are shown in Figure 3, and the surface area and pore size data are listed in Table 6. N 2 adsorption-desorption loops were not closed, because when the experiments were below 0.15 Pa pressure, irreversible adsorption occurred, and the adsorbed N 2 could not be desorbed. This indicated that there was a strong adsorption potential in the micropores of the HWS, while the WS sample does not exist in this case, and Table 6 shows that the average pore size of HWS is 3.23694 nm, which is relatively smaller than the WS of 8.88669 nm, which also revealed the that there is no adsorption potential in the micropores of WS. In addition, the surface area and pore volume of HWS improved 6.82 and 7.93 times, respectively, compared to the untreated WS, so the adsorption capacity of HWS is higher than WS is reasonable.

Surface Area Analysis
To further explore the surface area of the untreated and hydrochloric acid water sludge samples, the BET adsorption experiment was employed. Adsorption-desorption N2 isotherms are shown in Figure 3, and the surface area and pore size data are listed in Table 6. N2 adsorption-desorption loops were not closed, because when the experiments were below 0.15 Pa pressure, irreversible adsorption occurred, and the adsorbed N2 could not be desorbed. This indicated that there was a strong adsorption potential in the micropores of the HWS, while the WS sample does not exist in this case, and Table 6 shows that the average pore size of HWS is 3.23694 nm, which is relatively smaller than the WS of 8.88669 nm, which also revealed the that there is no adsorption potential in the micropores of WS. In addition, the surface area and pore volume of HWS improved 6.82 and 7.93 times, respectively, compared to the untreated WS, so the adsorption capacity of HWS is higher than WS is reasonable.

SEM and XRD Studies
The SEM images of HWS before and after absorption are shown in Figure 4. There is no evident change observed after the adsorption. The XRD results ( Figure 5) also revealed that only peaks intensity changed, while the diffractograms are similar, and no important changes in the structure of the adsorbent after adsorption are observed, indicating that the process was mainly physical adsorption-dominated. The identified compounds were K1.2Al4Si8O2(OH)2·4H2O, TiFeCl3, SiO2, Al2SiO5, (Mg,Fe)2SiO4, Fe0.4Mg0.76SiO3, and (Mg,Al,Fe)6(Si,Al)4O10(OH)8.

SEM and XRD Studies
The SEM images of HWS before and after absorption are shown in Figure 4. There is no evident change observed after the adsorption. The XRD results ( Figure 5) also revealed that only peaks intensity changed, while the diffractograms are similar, and no important changes in the structure of the adsorbent after adsorption are observed, indicating that the process was mainly physical adsorption-dominated. The identified compounds were K 1.

Effects of Variable Conditions on the Adsorption of Fluoride
The effects of variable conditions on fluoride removal were studied using RSM. As mentioned earlier, RSM based on BBD was employed to investigate the effects of three independent variables, HWS dose, pH, and initial concentration on the adsorption of fluoride by HWS. The BBD factorial design along with five replicates at central points is presented in Table 1. Design Expert 8.0.6 software (Stat-Ease Corporation, Minneapolis, MN, USA) was used for experimental design and analysis. Experimental data were fitted to a second-order polynomial model [31]: in which Y is the predicted response (Removal efficiency in %) used as dependent variable, xi and xj are the in dependent variables, b0 is the constant coefficient, bi is the coefficient that determines the influence of variable i in the response, bij is the coefficient that determines the effect of interaction between variables i and j, bii is the parameter that determines the shape of the curve, and k is the number of variables studied [25,32]. RSM model and its validation based on the experimental results are presented in Table 2. Based on the experimental data, regression models using a second-order polynomial were represented by Eq.4, which was developed, after which statistically insignificant coefficients (p-value greater than 0.1) were excluded from the analysis.
The analysis of variance (ANOVA) for the proposed model and corresponding p-values and F-values for assessing the significance of the regression coefficients are presented in Table 7. A p-value of model less than 0.05 implies that the proposed model well predicts the experimental results at 5% confidence interval [21]. A large p-value for lack of fit (>0.05) is preferred, as it measures the

Effects of Variable Conditions on the Adsorption of Fluoride
The effects of variable conditions on fluoride removal were studied using RSM. As mentioned earlier, RSM based on BBD was employed to investigate the effects of three independent variables, HWS dose, pH, and initial concentration on the adsorption of fluoride by HWS. The BBD factorial design along with five replicates at central points is presented in Table 1. Design Expert 8.0.6 software (Stat-Ease Corporation, Minneapolis, MN, USA) was used for experimental design and analysis. Experimental data were fitted to a second-order polynomial model [31]: in which Y is the predicted response (Removal efficiency in %) used as dependent variable, xi and xj are the in dependent variables, b0 is the constant coefficient, bi is the coefficient that determines the influence of variable i in the response, bij is the coefficient that determines the effect of interaction between variables i and j, bii is the parameter that determines the shape of the curve, and k is the number of variables studied [25,32]. RSM model and its validation based on the experimental results are presented in Table 2. Based on the experimental data, regression models using a second-order polynomial were represented by Eq.4, which was developed, after which statistically insignificant coefficients (p-value greater than 0.1) were excluded from the analysis.
The analysis of variance (ANOVA) for the proposed model and corresponding p-values and F-values for assessing the significance of the regression coefficients are presented in Table 7. A p-value of model less than 0.05 implies that the proposed model well predicts the experimental results

Effects of Variable Conditions on the Adsorption of Fluoride
The effects of variable conditions on fluoride removal were studied using RSM. As mentioned earlier, RSM based on BBD was employed to investigate the effects of three independent variables, HWS dose, pH, and initial concentration on the adsorption of fluoride by HWS. The BBD factorial design along with five replicates at central points is presented in Table 1. Design Expert 8.0.6 software (Stat-Ease Corporation, Minneapolis, MN, USA) was used for experimental design and analysis. Experimental data were fitted to a second-order polynomial model [31]: in which Y is the predicted response (Removal efficiency in %) used as dependent variable, x i and x j are the in dependent variables, b 0 is the constant coefficient, b i is the coefficient that determines the influence of variable i in the response, b ij is the coefficient that determines the effect of interaction between variables i and j, b ii is the parameter that determines the shape of the curve, and k is the number of variables studied [25,32]. RSM model and its validation based on the experimental results are presented in Table 2. Based on the experimental data, regression models using a second-order polynomial were represented by Equation (4), which was developed, after which statistically insignificant coefficients (p-value greater than 0.1) were excluded from the analysis.
The analysis of variance (ANOVA) for the proposed model and corresponding p-values and F-values for assessing the significance of the regression coefficients are presented in Table 7. A p-value of model less than 0.05 implies that the proposed model well predicts the experimental results at 5% confidence interval [21]. A large p-value for lack of fit (>0.05) is preferred, as it measures the model failure in representing data points in the experimental domain [33]. In this case, the p-value of lack of fit is 0.1792, implying that lack of fit of the model is insignificant. Adequate precision (AP) is the ratio of the predicted responses from the design points to their average standard deviation, which, for a good model fit, its desired value is 4 or more [21]. The ratio of 48.388 implies that the model is acceptable. The overall prediction performance of the model is described by coefficient of determination (R 2 ). A high R 2 value, close to 1, is desirable to ensure a satisfactory adjustment of the model to the experimental data [26]. The value of R 2 = 0.9957 and a reasonable agreement with R 2 adjusted is necessary [25,34]. In the present models, the values of R 2 adjusted = 0.9901 was close to R 2 , indicating high significance of the model.  A, B, C refer to the linear entries (initial concentration, pH, HWS dose, respectively) of the analysis of variance and A 2 , B 2 , C 2 are its quadric entries; AB, AC, BC refer to the interaction item (initial concentration-pH, initial concentration-HWS dose, pH-HWS dose, respectively, df refers to the degree of freedom).
A plot of the residuals was also used to assess the adequacy of the model. The residual plots of the models are presented in Figure 6a. The residuals are normally distributed if the points on the plot follow a straight line [31]. As Figure 6a illustrates, the assumption of normality is satisfied for the models [26]. Figure 6b presents the observed and predicted values. The statistical significances of the models are evident from Figure 6a, as observed and predicted values fit each other well. The statistical model can be used to predict the removal efficiency in this experiment in the range above.
A plot of the residuals was also used to assess the adequacy of the model. The residual plots of the models are presented in Figure 6a. The residuals are normally distributed if the points on the plot follow a straight line [31]. As Figure 6a illustrates, the assumption of normality is satisfied for the models [26]. Figure 6b presents the observed and predicted values. The statistical significances of the models are evident from Figure 6a, as observed and predicted values fit each other well. The statistical model can be used to predict the removal efficiency in this experiment in the range above.  Response surface graphs presented in Figure 7 demonstrate the effects of variables and their interactive effects on the removal of fluoride. These plots are generated as a function of two variables at the same time, keeping the third variable at a centre level. As shown in (a) and (b), the increase of initial fluoride concentration leads to the decrease of final efficiency. As shown in (c), the removal efficiency increased as the HWS dose increased under the same pH. The increase in fluoride adsorption was possibly attributed to the increase in availability of F − due to the presence of a greater number of active sites [35,36]. Considering initial concentration = 3 mg·L −1 , the WHO standard for permissible limit of fluoride in water (≤1.5 mg·L −1 ), which would be fulfilled with HWS (≥6.17 g·L −1 ) in the neutral condition of the present work, can be less at pH = 6. However, the effect of removal efficiency of fluoride was more prominent by initial concentration compared to HWS dose and pH from (a) and (b), meaning initial concentration of fluoride has an adverse effect on its removal. Samarghandi reported a similar result in adsorption of fluoride [31]. At low initial concentration, most of fluoride will interact with the binding sites of the adsorbent, resulting in higher removal percentage. On the other hand, at high initial concentration, only some of the ions will combine with the finite available sites for binding [37]. Response surface graphs presented in Figure 7 demonstrate the effects of variables and their interactive effects on the removal of fluoride. These plots are generated as a function of two variables at the same time, keeping the third variable at a centre level. As shown in (a) and (b), the increase of initial fluoride concentration leads to the decrease of final efficiency. As shown in (c), the removal efficiency increased as the HWS dose increased under the same pH. The increase in fluoride adsorption was possibly attributed to the increase in availability of F − due to the presence of a greater number of active sites [35,36]. Considering initial concentration = 3 mg·L −1 , the WHO standard for permissible limit of fluoride in water (≤1.5 mg·L −1 ), which would be fulfilled with HWS (≥6.17 g·L −1 ) in the neutral condition of the present work，can be less at pH = 6. However, the effect of removal efficiency of fluoride was more prominent by initial concentration compared to HWS dose and pH from (a) and (b), meaning initial concentration of fluoride has an adverse effect on its removal. Samarghandi reported a similar result in adsorption of fluoride [31]. At low initial concentration, most of fluoride will interact with the binding sites of the adsorbent, resulting in higher removal percentage. On the other hand, at high initial concentration, only some of the ions will combine with the finite available sites for binding [37]. Usually, pH has been seen as an important factor influencing adsorption of the crystalline form to the adsorbent. It has been reported that, in case of zeolite and activated alumina, the pH of zero charge (pHpzc) may vary from 5.5 to about 8.3 [24], and the optimum pH for maximum adsorption is between 5 and 7 [31,38]. The influence of the initial pH on the removal efficiency of this study is shown in Figure 7b and 7c. The percentage of fluoride removal remains nearly constant within the pH range of 4-7. Further increase in the pH of the solution slightly decreases the removal efficiency. The fluoride uptake capacity of this media is not affected in the pH range less than or equal to 7, Usually, pH has been seen as an important factor influencing adsorption of the crystalline form to the adsorbent. It has been reported that, in case of zeolite and activated alumina, the pH of zero charge (pHpzc) may vary from 5.5 to about 8.3 [24], and the optimum pH for maximum adsorption is between 5 and 7 [31,38]. The influence of the initial pH on the removal efficiency of this study is shown in Figure 7b,c. The percentage of fluoride removal remains nearly constant within the pH range of 4-7. Further increase in the pH of the solution slightly decreases the removal efficiency. The fluoride uptake capacity of this media is not affected in the pH range less than or equal to 7, possibly due to the presence of positively charged and neutral sites at the surface of the adsorbent [24]. The decline at pH > 7 may be due to the competition between OH − and F − [39]. This is in agreement with fluoride removal studies on activated alumina by other researchers.
The numerical simulation optimum conditions for removal efficiency of fluoride using HWS were carried by RSM with the help of the desirability function. In this study, the desirability function approach was employed for optimization using Design Expert, which provides several possible options including minimum, maximum, target, within the range, none (only for response), and equal to (factors only) for choosing a desired goal for each variable and response [39]. The average values of confirmation tests in triplicates and predicted by the model under optimum conditions are presented in Table 7. As shown in Table 8, the removal efficiency of confirmation tests was closed to the prediction. The optimum removal efficiency of fluoride can reach 81.153% under the optimum condition: HWS dose of 14.10 g/L and pH value at 6.12. Meanwhile, the lower initial fluoride concentration is better.

Effect of Co-Existing Anions
As we all know, water contains other anions such as sulfate, phosphate, nitrate, and silicate in addition to fluoride. The results for the removal efficiency of HWS with concentration of each anion are shown in Figure 8. Some studies showed that the presence of other co-existing ions in water had an effect on fluoride removal [17,24,40,41]. It was observed that SO 4 2− , PO 4 3− , SiO 4 4− , and NO 3− ions showed a negative effect on removal of fluoride. The fluoride adsorption efficiency of the adsorbent decreased from about 80% to 14% in case of silicate and 29% in phosphate, and although the sulfate and nitrate showed slight influence on fluoride removal, the removal efficiency still decreased from 80% to 65% for nitrate and 46% for sulfate. This can be due to the competition between the anions and fluoride [41]. The competition ability of four anions followed the order SiO 4  options including minimum, maximum, target, within the range, none (only for response), and equal to (factors only) for choosing a desired goal for each variable and response [39]. The average values of confirmation tests in triplicates and predicted by the model under optimum conditions are presented in Table 7. As shown in Table 8, the removal efficiency of confirmation tests was closed to the prediction. The optimum removal efficiency of fluoride can reach 81.153% under the optimum condition: HWS dose of 14.10 g/L and pH value at 6.12. Meanwhile, the lower initial fluoride concentration is better.

Effect of Co-Existing Anions
As we all know, water contains other anions such as sulfate, phosphate, nitrate, and silicate in addition to fluoride. The results for the removal efficiency of HWS with concentration of each anion are shown in Figure 8. Some studies showed that the presence of other co-existing ions in water had an effect on fluoride removal [17,24,40,41]. It was observed that SO4 2− , PO4 3− , SiO4 4− , and NO 3− ions showed a negative effect on removal of fluoride. The fluoride adsorption efficiency of the adsorbent decreased from about 80% to 14% in case of silicate and 29% in phosphate, and although the sulfate and nitrate showed slight influence on fluoride removal, the removal efficiency still decreased from 80% to 65% for nitrate and 46% for sulfate. This can be due to the competition between the anions and fluoride [41]. The competition ability of four anions followed the order SiO4 4− > PO4 3− > SO4 2− > NO3 − ; the results are in good agreement with similar work done by others [24] for activated alumina.

The Adsorption Efficiency from Water
The adsorption experiment results with wastewater samples are shown in Table 9. To examine the optimal removal efficiency, the 14.1g/L of adsorbent was first added. Results showed that the removal efficiencies were 79.21% for W1, 79.84% for W2, and 80.67% for W3, respectively, which was

The Adsorption Efficiency from Water
The adsorption experiment results with wastewater samples are shown in Table 9. To examine the optimal removal efficiency, the 14.1g/L of adsorbent was first added. Results showed that the removal efficiencies were 79.21% for W1, 79.84% for W2, and 80.67% for W3, respectively, which was in agreement with the results obtained from RSM analysis, i.e., the removal efficiency for the lower initial fluoride concentration is better, for example, and the fluoride concentration of W3 was 8.6 mg/L, relating to a highest removal efficiency of 80.67%. However, once the optimal removal efficiency was acquired with such high adsorbent dose, the adsorption capacity inevitably decreased and was significantly lower than the adsorbents of other studies. In order to make the results of this study comparable to the others, the adsorbent dose was drastically reduced to 2 g/L to acquire a satisfy adsorption capacity. As a result, the harvested maximum adsorption capacity was 2.03 mg/g appearing at adsorption processes in W1 sample, which is lower than composite adsorbent, desugared reed root, and lignite from other literatures, while higher than pumice, modified montmorillonite, modified hematite, and modified zeolite (Table 10). Of course, its greatly decreased removal efficiency (only 20.10%) was foreseeable.

Conclusions
This work has shown that WS can be considered as a promising material for removing fluoride from poor quality water. The conclusions drawn from this study are given below: (1) Acid treatment and high temperature ranges found as the conditions for better fluoride sorption and hydrochloric acid treatment will gain the best efficiency for removing fluoride for WS. (2) A model of adsorption has been proposed for the adsorption of F − onto HWS. The results gained from this study were well described by the theoretical Langmuir isotherms. The values of the equilibrium parameter and RL indicated that the F − /treated HWS system was favorable. Kinetic studies reveal that the adsorption is first order. Thermodynamic parameters were calculated, indicating that the adsorption was spontaneous.
(3) RSM based on BBD was employed to investigate the effects of the three independent variables, namely, HWS dose, pH, and initial concentration on the adsorption of fluoride by HWS. The optimum removal efficiency of fluoride from wastewater using HWS was observed with the help of RSM, and it can reach 81.153% under the optimum condition: HWS dose of 14.10 g/L, at pH = 6.12. (4) In addition, the results showed that presence of other co-exiting ions, such as SO 4 2