OXIDE SORBENT: SYNTHESIS, CHARACTERIZATION AND SORPTION BEHAVIOR TO FLUORIDE AND PHOSPHATE IONS FROM AQUEOUS SOLUTION

. Excess quantities of fluoride and phosphate in water bodies can lead to fluorosis and eutrophication problems, respectively. In search of a promising adsorbent targeting these ions, Fe 3 O 4 /Al 2 O 3 /ZrO 2 ternary oxide was synthesized via co-precipitation method and characterized by X-ray diffraction (XRD), Fourier transform-infrared (FTIR) and Brunauer-Emmer-Teller (BET). Its specific surface area was found to be 205 m 2 /g. The effects of solution pH, adsorbent dose, contact time, agitation speed and initial fluoride and phosphate concentrations were also investigated and the optimum values were 4, 0.5 g, 12 h, 100 rpm and 20 mg/L, respectively, for fluoride and 5, 0.1 g, 8 h, 100 rpm and 10 mg/L, respectively, for phosphate. Fluoride and phosphate adsorptions fitted well with Freundlich and Langmuir isotherm models, respectively and their kinetic data correlated well with the pseudo-second order model. Desorbability study revealed that maximum desorption was achieved at pH 12. Thermodynamics study on the other hand showed that adsorption of fluoride was nonspontaneous and endothermic whereas that of phosphate was spontaneous and exothermic. Application on real water sample decreased the concentration of fluoride from 4.92 to 1.97 mg/L in ground water and phosphate from 1.7 to 0.35 mg/L lake water showing its potential as a promising adsorbent. structure peak was observed and it has a high specific surface area (205 m 2 /g) as indicated by XRD and BET study, respectively. The adsorption isotherms of fluoride better fitted to Freundlich model whereas that of phosphate better fitted to Langmuir model. The adsorption kinetics data indicated that the adsorption of both ions on the synthesized nanosorbent follows pseudo-second order reaction. The adsorption of fluoride was found to be non-spontaneous and endothermic whereas that of phosphate was spontaneous and exothermic. Results from this study demonstrated potential utility of tested Al 2 O 3 /Fe 3 O 4 /ZrO 2 as a viable technology for fluoride and phosphate removal from drinking water.


INTRODUCTION
Quality of water resources is deteriorating continuously due to geometrical growth of population, industrialization, civilization, domestic and agricultural activities and other geological and environmental changes [1,2]. In this regard, water pollution has become a serious issue in the present scenario, affecting all living creatures, household, recreation, fishing, transportation, and other commercial activities [3,4].
Thousands of organic, inorganic and biological pollutants have been reported as water contaminants. Some of them have serious side effects and toxicities with a few being lethal and carcinogenic [5,6]. These pollutants are very dangerous for all of us, aquatic conditions and the ecosystem of the earth as a whole. The United State National Primary Drinking Water Standards specify a maximum contaminant level (MCL) for a number of inorganic anions, including fluoride, nitrite, nitrate, chloride, phosphate and sulfate. Among such contaminants fluoride and phosphate are well known. Disposal of untreated waste containing fluoride and phosphate into water bodies may deteriorate water quality subsequently [7].
Contamination of fluoride in the ground and surface water could come either from natural geological sources including various minerals, such as fluorite, biotites, topazite and shale or from industries that uses fluoride containing compounds as raw materials, e.g., glass and ceramic production, semiconductor manufacturing, electroplating and aluminum smelters. In similar way industries such as fertilizer, semiconductor and phosphoric acid processing produce wastewater that contains high concentration of phosphate [8]. Excessive intake of these ions through food and synthesized at the laboratory scale and used as an adsorbent to remove fluoride and phosphate from aqueous solutions. The synthesis process followed co-precipitation method due to its advantages including simple and rapid preparation, easy control of particle size and composition, homogeneity in mixed precipitates and high specific surface of the products [17]. The adsorption kinetics, adsorption isotherms, the effect of coexisting ions, thermodynamic parameters, desorption and applicability to real water samples were investigated. Additionally, the influence of operating parameters such as solution pH, the effect of ionic strength, contact time, adsorbent dose, initial adsorbate concentration and agitation speed were determined.

Synthesis procedure of the adsorbents
Al2O3/Fe3O4/ZrO2 ternary composite was prepared by co-precipitation of Fe:Al:Zr at a mole ratio of 70:25:5. In the first step, stoichiometrically calculated FeCl2.4H2O and FeCl3.6H2O (at 0.5 mole ratio of Fe 2+ /Fe 3+ ) were dissolved in 100 mL of 0.3 M HCl solution. Then, the solution was added drop-wise from separatory funnel into the solution of 120 mL of 3 M NaOH (pH 12) over a period of 2 h, under a vigorous stirring at 80 o C in N2 atmosphere to obtain Fe3O4 Ferro fluid. Following this, the magnetite-alumina-zirconia oxide composite was prepared by adding 100 mL of Al(NO3)3.9H2O and ZrOCl2.8H2O (obtained by dissolving stoichiometric amounts of both salts in 100 mL of deionized water separately) into the obtained Fe3O4 suspension. The pH of the mixtures was adjusted to 8.0 using 0.1 M NaOH and the mixture was magnetically stirred under N2 atmosphere for 1.5 h at 70 o C. Finally, the resulting black magnetic fluid was made to settle and washed with deionized water several times to remove impurities such as Cl -, NO3and excess OHions and then dried at 60 o C for 24 h to obtain the desired product [19].

Elemental analysis
In this particular experiment, 0.1 g of the as-synthesized powder was digested with a mixture of acids to perform elemental analysis of Al, Fe, and Zr which was finally performed by flame atomic absorption spectrophotometer [20].

Characterization of the as-synthesized powder
The crystalline phases and the crystallite sizes of the as-synthesized nanosorbent before and after adsorption of fluoride and phosphate were determined using X-ray diffraction (XRD) instrument equipped with CuKα radiation (λ = 0.15405 nm). The surface functional groups of the nanocomposite before and after adsorption were determined by FTIR. Similarly, the specific surface area was investigated by Brunauer-Emmett-Teller (BET) technique.

Batch adsorption studies
Batch experiments were conducted using 250 mL Erlenmeyer flasks under continuous mixing conditions with magnetic stirrer at room temperature. The effect of different parameters such as dose of adsorbent, agitation speed, pH, ionic strength, contact time, initial concentration, and the presence of foreign ions were studied by varying one parameter at a time and keeping the others constant for both analytes in separate experiments. The residual Fconcentration was measured immediately using a fluoride selective electrode after an equal volume of TISAB was added on each sample solution. Then the potential response was converted to concentration by using the relation: where CFis the free fluoride concentration, E is the potential, S is the slope of the calibration curve and the constant is its y intercept [21]. On the other hand, final concentrations of the phosphate ions were determined spectrophotometrically by the molybdenum blue method by monitoring the absorbance at 880 nm on UV-Vis spectrophotometer [22]. Finally, removal efficiency was calculated by using the relationship in equations 2 and 3: where Co and Ce are initial and equilibrium concentrations of F -/PO4 3-(mg/L), respectively, m is quantity of the nanocomposite (g) and V is the volume required for the reaction mixture (L) [23].

Determination of the effect of ionic strength and pH study
The effect of pH on the adsorption of fluoride and phosphate was investigated by varying initial solution pH in the range 2-11 (2, 4, 5, 6, 7, 9 and 11) [25].

The effect of concentration and coexisting ions
The effect of initial concentration was evaluated by varying it in the range of 10-140 (10, 20, 40, 60, 80, 100, 120, and 140) mg/L. On the other hand effect of coexisting ions on the adsorption of fluoride and phosphate ions on the as synthesized nanocomposite was also evaluated in the presences of 20 mg/L of SO4 -2 , NO3 -, HCO3 -2 and Cl -, respectively, and separately keeping other parameters (pH, adsorbent dose, contact time and agitation speed) constant at optimized values.

Adsorption isotherm, adsorption kinetics and thermodynamic study
Data for plotting isotherms were obtained by mixing initial concentrations ranging from 10-140 (10, 20, 40, 60, 80, 100, 120, and 140) mg/L and the optimized amount of adsorbent dose for both ions. Adsorption kinetics was studied by varying time ranging from 1, 4, 8, 12, 16 and 24 h and the nature and thermodynamic feasibility of the sorption process was determined by performing the adsorption process at 20, 30, 40, 50 and 60 o C. In all cases, the pre-optimized parameters were kept constant at their optimum values.

Desorption and recyclability study
To investigate desorption of fluoride/phosphate from the exhausted nanocomposite, the optimized adsorbent dose was subjected to adsorption for an optimized time in a 50 mL of initial concentration of fluoride/phosphate ion solution. Then the exhausted adsorbent was conditioned (immersed) in 50 mL of 0.1 M NaOH solutions for desorption to take place by varying pH of the solution from 2-12. The recovery percentage was obtained from the following equations: where, Cd is desorbed concentration of fluoride/phosphate ion after the desorption process, Ce is equilibrium concentration and Co is initial concentration of fluoride and phosphate [26]. To investigate the extent of regeneration and reusability of the adsorbents, experiments were performed at pre optimized conditions and desorption was done by using 0.1M NaOH up to four cycles. Then the % adsorption of fluoride/phosphate on the adsorbent for each cycle was recorded [27].

Applicability on real water samples treatment
To assess the applicability of the method on real samples, ground water was taken from Adama Town on March 5, 2016 and lake water from Lake Adele on March 19, 2016 (Haramaya Woreda, Ethiopia) for fluoride and phosphate study, respectively. First, the initial concentration of fluoride and phosphate in these water samples were determined before loading the adsorbent. Then the synthesized nanosorbent was mixed with 50 mL of these water samples at the optimized condition to see its removal capacity.

Powder X-ray diffraction analysis
Adsorption reaction may lead to changes in the molecular and crystalline structure of the adsorbent and hence understanding of these properties and the resulting changes thereof would provide valuable information regarding adsorption reaction [28]. Hence XRD patterns of the adsorbent before and after treatment with fluoride and phosphate ions were studied and presented in Figure 1. As shown in Figure 1, the XRD pattern of Fe3O4/Al2O3/ZrO2 nanocomposite before and after adsorption exhibited no variation in the structure and this suggested that the fluoride and phosphate ions might diffuse into micropores and macropores without altering the structure of the adsorbent. A similar phenomenon was observed in the sorption study of phosphate by Fe-Al-Mn mixed oxide [29]. The diffraction pattern shows very weak intensities on a broad background suggesting the amorphous nature of the adsorbent. However, the presence of some low intensity peaks at 2θ values of 35.25, 43.42, 53.69, 57.46 and 63.33° could be attributed to a face centered cubic spinel structure of magnetite (Fe3O4) [19].

FTIR analysis
The FTIR spectra of Al2O3/Fe3O4/ZrO2 ternary oxide before and after adsorption of fluoride and phosphate are presented in Figure 2. The strong and broad band at 3600-3100 cm -1 region (O-H stretching vibration) may be assigned for the presence of hydroxyl of coordinated water molecules [30]. The band at 1633 cm -1 primarily attributed to the bending vibration of H-O-H (physisorbed water molecules) [29]. When comparing the spectrum of Al2O3/Fe3O4/ZrO2 ternary oxide before adsorption with the spectrum after adsorption, the peak at 1380 cm -1 (surface hydroxyl group, M-OH) before adsorption was weakened dramatically and this was accompanied by the formation of new peaks observed at 1305 cm -1 and 1260 cm -1 after adsorption of fluoride and phosphate, respectively. These results approved that fluoride and phosphate were successfully adsorbed on the surface of Al2O3/Fe3O4/ZrO2 ternary oxide [30]. The peak appearing at 1260 cm −1 in the spectrum after adsorption was attributed to the bending vibration of adsorbed phosphate. This indicates that replacement of surface hydroxyl groups (M-OH) by phosphate via the formation of inner sphere complex [29]. The spectral peak obtained at 1305 cm -1 may signify the adsorption of fluoride on the adsorbent possibly through electrostatic attraction. The bands at 589 cm -1 and 440 cm -1 may be assigned to M-O symmetrical stretching vibrations of the mixed metal oxides [31].

Surface area
The specific surface area of the as-synthesized adsorbent was determined by Brunauer-Emmett-Teller (BET) N2 adsorption and desorption isotherms. This study provided the specific surface area of Al2O3/Fe3O4/ZrO2 ternary oxide which was found to be 205 m 2 /g. The high specific surface area may have resulted from the presence of zirconium (ZrO2) oxides [32]. Therefore, Al2O3/Fe3O4/ZrO2 ternary oxide can be categorized as adsorbent with high adsorption capacity.

Elemental analysis
Analysis of the synthesized nanocomposite by flame atomic absorption spectrometer indicated that the percentage compositions of Fe, Al and Zr in the as-synthesized nanocomposite were found to be 72.3%, 23.9% and 3.8%, respectively. These percentage compositions are close to the ratio of each metal ion (70:25:5) taken during the synthesis process.

pH of point of zero charge
The pH of point of zero charge of the Al2O3/Fe3O4/ZrO2 nanocomposite was found to be 5.71 as presented on Figure 3a. Thus, the adsorption of phosphate and fluoride on the adsorbent is favored at pH below 5.71.

The effect of ionic strength
Ionic strength is an important parameter especially when solute removal is affected by slight changes in the supporting electrolyte solution [33]. Table 1 shows the % removal and adsorption capacity of phosphate and fluoride in the presence of NaNO3. The result indicates that the percentage removal of fluoride onto Al2O3/Fe3O4/ZrO2 decreased as concentration of NaNO3 increased under the concentration range studied. High NaNO3 concentration indicates more NO3ions existing on the outer layer which would exert a strong repulsive force against the approaching Fion leading to a decrease in the adsorption capacity of Al2O3/Fe3O4/ZrO2. On the other hand, no significant change was observed on phosphate adsorption as concentration of NaNO3 increased from 0.001 to 0.1 M [33]. Therefore, from the results it can be inferred that outer sphere complexation could be formed when fluoride was adsorbed on Al2O3/Fe3O4/ZrO2 as removal capacity depends on concentration of NaNO3. On the other hand, PO4 3adsorption was through the formation of inner sphere complexes since concentration of NaNO3 has no significant change on its percentage removal.

Optimization of experimental parameters
Batch mode of adsorption was used to optimize experimental parameters, including solution pH, adsorbent dose, speed of agitation, contact time and initial fluoride and phosphate ions concentration. The optimum value for each parameter was found to be 4, 0.5 g, 100 rpm, 12 h and 20 mg/L for fluoride and 5, 0.1 g, 100 rpm, 8 h and 10 mg/L for phosphate.
The adsorption of fluoride onto the as-synthesized adsorbent increased as pH increased from 2 to 4 (90 to 92.93%) (Figure 3b). A similar trend was observed on phosphate adsorption up to pH 5 (increased from 72.75 to 96.63%). Thus, maximum adsorption for fluoride and phosphate was achieved at pH 4 and 5, respectively. The decrease in the sorption capacity at higher pH can be attributed to the competition for the active site by OHions and or the development of negative charge on the adsorbent surface, since both OHions and F -/PO4 3ions have the same charge [34,35].

Adsorption isotherm study
The relationship between the amount of substance adsorbed at constant temperature and its concentration in the equilibrium solution is called the adsorption isotherm [36]. The most widely accepted surface adsorption models for single-solute systems (Langmuir and Freundlich) were selected in this study [37]. The Freundlich isotherm in its linear form is represented by: Log qe = log Kf + (5) qe = Kf Ce 1/n (6) where, qe is the amount of solute adsorbed per unit weight of the sorbent (mg/g), Ce is the equilibrium concentration of solute in solution (mg/L), Kf is a measure of adsorption capacity and 1/n is the adsorption intensity. The Freundlich isotherm constants 1/n and Kf can be calculated from the slope and intercept of the plot logqe vs logCe. The values of 1/n lying between 0 and 1 and the n values lying between 1 and 10 indicate favorable conditions for adsorption. The intercept of the line, Kf, is roughly indicator of the adsorption capacity [38]. On the other hand, the Langmuir equation is useful for the estimation of maximum adsorption capacity corresponding to complete monolayer coverage and expressed by: The linearized form of equation 7 is given in equation 8: where, Qo is the amount of adsorbate at complete monolayer coverage (mg/g) and gives the maximum sorption capacity of sorbent, Ce is the equilibrium concentration of fluoride/phosphate (mg/L) and b (L/mg) is the Langmuir isotherm constant that relates to the energy of adsorption. Langmuir constants Qo and b can be calculated from the slope and intercept of the plot Ce/qe versus Ce, respectively. The plot Ce/qe vs Ce yields a straight line with a slope 1/Qo and an intercept 1/Qob. The values of all parameters of the two models are given in Table 2.  The value of RL indicates the type of isotherm: to be irreversible if (RL = 0), favorable if (0 < RL < 1), linear if (RL = 1), or unfavorable for (RL ˃ 1) [38]. Thus, the calculated values of RL for phosphate and fluoride adsorption onto Al-Fe-Zr ternary oxide were found to be between 0 and 1 indicating the favorability of the adsorption process. Langmuir's and Freundlich's curves were interpreted with respect to correlation coefficient R 2 , a statistical measure of how well the regression line approximates the real data point. An R 2 value of 1.0 indicates that the regression line perfectly fits the data. In this study, the R 2 values of the nano sized adsorbent in Freundlich's equation exhibits a higher value than Langmuir's (i.e. 0.9478 and 0.9091, respectively) in the case of fluoride, whereas phosphate adsorption indicates a higher R 2 value in Langmuir plot than Freundlich plot (0.9686 and 0.9465, respectively). Therefore, fluoride and phosphate adsorption onto the nanosized adsorbent best fits with Freundlich and Langmuir equation respectively. The Freundlich and Langmuir equation for the fluoride adsorption process has been developed as indicated in the equation (10) and (11), respectively. qe = 0.973Ce 0.679 (10) qe = 1.01361Ce/1+0.113Ce (11) Similarly, the Freundlich and Langmuir equations for the phosphate adsorption process are indicated in the equation (12) and (13), respectively. qe = 13.24Ce 0.427 (12) qe = 34.5Ce/1+1.035Ce (13)

Adsorption kinetics
The adsorption kinetics of the synthesized nanosorbent for the two analytes was studied by varying contact time from 1 to 24 h. In this study the Lagergren pseudo-first and pseudo-second order models were considered. The pseudo-first order adsorption kinetics model is given in where, qe and qt are the amount of fluoride/ phosphate adsorbed (mg/g) at equilibrium and at any time t (min), respectively. The adsorption rate constant, k1 was determined from the slope of the linear plot of log(qe -qt) versus t (i.e., k1 = -2.303 × slope) and qe is obtained from intercept (log qe = intercept). On the other hand, the integrated form at boundary conditions (t = 0 to t = t) and (qt = 0 to qt = qt) for the pseudo-second order adsorption kinetic model is expressed as: where, qt and qe are the amounts of adsorbed analyte at a time t and equilibrium (mg/g), respectively, k2 is the rate constant (g mg -1 min -1 ), t is the stirring time (min), k2 which is (slope 2 /intercept) can be determined from plotting t/qt against t based on the above equation and the value of qe is 1/slope. The Lagergren pseudo-first and pseudo-second order reactions were interpreted with respect to coefficient of determination, R 2 , and by comparing the calculated qe value with the corresponding experimental qe value. Thus, the result in Table 3 indicates that the adsorptions of fluoride and phosphate onto Al2O3/Fe3O4/ZrO2 follow second order reaction.

Adsorption thermodynamics study
The standard free energy change (ΔG), enthalpy change (ΔH) and entropy change (ΔS) are the main thermodynamic characteristics of any adsorption system in equilibrium. The fundamental thermodynamic equation that relates all the above three parameters is given in equation 17 [39].
where ΔG is in kJ/mol, ΔH is in kJ mol -1 , ΔS is in kJ mol -1 K -1 and T is in K. The free energy of adsorption, considering the adsorption equilibrium constant K is given by equation 18: ΔG = -RT lnKc (18) where, R is the universal gas constant (8.314 J mol -1 K -1 ) and Kc is the equilibrium constant obtained from qe/Ce. ΔH and ΔS can be determined from the plot of lnKc versus T -1 by computing their slope and intercept (i.e. ΔH = -slope × R and ΔS = intercept × R). From equations (17) and (18) a relationship between lnK and T can be derived and is indicated in equation 19. The calculated thermodynamic parameters for the adsorption process of fluorine and phosphate onto the adsorbent are as shown in Table 4.
ln Kc = ΔS/R -ΔH/TR (19) The positive values of ΔG in fluoride adsorption study confirms that the adsorption process is non spontaneous. The value of ΔH was positive, indicating that the adsorption reaction was endothermic. So, increasing temperature encourages fluoride ion adsorption. Enhancement of adsorption capacity of Al2O3/Fe3O4/ZrO2 at higher temperature may be attributed to the enlargement of the pore size and/or activation of the adsorbent surface and also increase in the mobility of fluoride ions. The positive value of ΔS reflects high degrees of disorderliness at the solid solution interface during the adsorption of fluoride onto Al2O3/Fe3O4/ZrO2. This may be due to the fact that the adsorbed water molecules which are displaced by the adsorbate species, gain more translational entropy than it is lost by the adsorbate molecules, thus allowing the prevalence of randomness in the system [28]. In case of the phosphate adsorption process, the negative values of ΔG confirm that the adsorption process is spontaneous, which becomes a less negative value with an increase in temperature, indicating that adsorption occurred at a lower temperature. The negative values of enthalpy confirm that the exothermic nature of the adsorptions process and  The effect of coexisting ions study In the present study we investigated the possibility of interference by HCO3 -, Cl − , NO3and SO4 2on fluoride and phosphate removal. The results in Figure 4 indicates that the percentage removal in the presence of these ions ranged from 72.65 -84% for fluoride and 95.5 -99.3% for phosphate. Fluoride adsorption is more affected in the presence of sulphate and phosphate due to their competion for the active sites as they are all negatvely charged [11]. Bicarbonate and nitrate have less effect on its adsorption whereas phosphate adsorption was more affected in the presence of nitrate and fluoride when compared with the other ions.

Desorption study
Ions desorbability can be defined as the ratio of the desorbed ions over the total adsorbed ions by the adsorbent. Therefore, the desorbability can be used to indicate the degree of fluoride/phosphate desorption from the adsorbent [40]. In this particular experiment, pH of the solution was varied from 2-12 (2, 4, 5, 6, 8, 10 and 12) to observe the effect of solution pH on desorption of fluoride and phosphate from the nanosized adsorbent. Desorption of fluoride increased from 3.31 to 77.7% while that of phosphate from 0.25 to 85% when pH of the solution was increased from 2 to 12. These results indicated that desorption was more favorable at higher pH for both ions. Up to pH 8 desorption of both ions took place slightly, then increased abruptly. When compared to fluoride, desorption of phosphate was more difficult at low pH. The maximum desorption was obtained at pH 12 (77.7 and 85.1%) for fluoride and phosphate, respectively, which is in agreement to the inner sphere adsorption mechanism of the two ions on the sorbent material [41].

Recyclability study
To make a cost effective and user-friendly process, the adsorbent should be regenerated, so as to reuse for further fluoride and phosphate adsorption. In this study the experiments were performed using optimized conditions and desorption was done by using 0.1 M NaOH up to four cycles. As presented in Table 5, the percentage removal of fluoride and phosphate was found to be promising up to the fourth cycle.

Applicability on real water samples treatment
To assess the applicability of the method for real samples, ground water was taken from Adama town for fluoride and lake water from Lake Adele (Haramaya Woreda) for phosphate adsorption study. The initial concentrations of fluoride and phosphate in these water samples were determined before loading the adsorbent. Then the water samples were mixed with the adsorbent by both batch and column methods and the result are presented in Table 6. In column method, the sample solution was passed through the column at a flow rate of 5 mL/min. Results of the batch method show that the fluoride content of the ground water was decreased to 1.97 mg/L from 4.92 mg/L and the phosphate content of the lake water decreased to 0.35 mg/L from 1.7 mg/L. The fluoride content was decreased to maximum permissible level after passing through the column packed with the adsorbent for three times (i.e., below 1.5 mg/L).

CONCLUSION
The findings of the present study indicate that fluoride and phosphate uptakes from the water by Al2O3/Fe3O4/ZrO2 were found to be dependent on the pH, contact time, initial fluoride and phosphate concentration, adsorbent dose and temperature. The results of the study also revealed that the synthesized nanosorbent has high removal capacity of phosphate than fluoride. Al2O3/Fe3O4/ZrO2 oxide is amorphous at the synthesized temperature since no strong crystal structure peak was observed and it has a high specific surface area (205 m 2 /g) as indicated by XRD and BET study, respectively. The adsorption isotherms of fluoride better fitted to Freundlich model whereas that of phosphate better fitted to Langmuir model. The adsorption kinetics data indicated that the adsorption of both ions on the synthesized nanosorbent follows pseudo-second order reaction. The adsorption of fluoride was found to be non-spontaneous and endothermic whereas that of phosphate was spontaneous and exothermic. Results from this study demonstrated potential utility of tested Al2O3/Fe3O4/ZrO2 as a viable technology for fluoride and phosphate removal from drinking water.