In the present study, the defluoridation capabilities and adsorption mechanisms of cryptocrystalline magnesite were evaluated. All experiments were done by batch procedure. Conditions assessed include time, dosage, concentration, pH and the effects of competing ions. Optimum defluoridation conditions were observed to be 20 g/L magnesite, 2:100 solid:liquid ratio, 20 min of agitation and 60 mg/L fluoride concentration. Adsorption of fluoride by magnesite was observed to be independent of pH. Cryptocrystalline magnesite showed >99% efficiency for fluoride removal. Adsorption kinetics fitted better to a pseudo-second order than a pseudo-first order thus confirming chemisorption. Adsorption data fitted better to a Langmuir than a Freundlich adsorption isotherm thus confirming monolayer adsorption. Cryptocrystalline magnesite successfully removed excess fluoride from aqueous solution to below Department of Water Affairs and Forestry water quality guidelines. As such, this material can be used for a point source defluoridation technique in rural areas and households in South Africa and other developing countries. Based on comparison studies, cryptocrystalline magnesite proved to have high adsorption capacity for fluoride removal and can be used as a substitute for conventional treatment methods.

INTRODUCTION

Excessive uptake of water rich in fluoride has led to serious health problems (Shen & Schäfer 2014). Depending on the concentration, fluoride can be beneficial or harmful to human health. Permitted levels of fluoride assist in bone and tooth formation. Fluoride also prevents the development of dental cavities. However, if it is consumed in excess, it can lead to the development of dental and skeletal fluorosis (Tor 2006; Tor et al. 2009; Zhu et al. 2009; Trikha & Sharma 2014; Masindi et al. 2014b; Zhang et al. 2014; Zhou et al. 2014; Gitari et al. 2015; Zhao et al. 2015). According to the Department of Water Affairs and Forestry (DWAF), recommended fluoride concentrations in drinking water are around 1 mg/L (Masindi et al. 2014b; Gitari et al. 2015).

Contamination of groundwater by fluoride may be due to anthropogenic activities or the geochemical environment. It enters the aquatic ecosystems naturally, through weathering of fluoride-bearing rock, such as fluorapatite, sellaite and cryolite. It enters the environment artificially, through use of fluoride-containing materials in manufacturing industries, fertilizers, leaching and cleaning processes. All these activities lead to accelerated fluoride pollution. To remedy this, fluoride need to be removed from drinking water prior to human exposure (Xiuru et al. 1998; Meenakshi & Maheshwari 2006; Wang et al. 2007, 2013; Kamble et al. 2009; Zhu et al. 2009; Karthikeyan et al. 2011; Miretzky & Cirelli 2011; Sujana & Anand 2011; Wu et al. 2011; Srivastav et al. 2013; Tomar et al. 2014; Masindi et al. 2014b; Zhou et al. 2014; Gitari et al. 2015; Yu et al. 2015; Zhao et al. 2015).

Recently, scientific communities have been searching for practical ways of removing fluoride from underground water. Several low-cost and point-based technologies have been developed but the cost factor, unsustainable treatment processes, inefficient treatment capabilities and generation of huge secondary sludge limit their application in the defluoridation process. An array of materials and technologies have been tested, and they entail ion exchange, adsorption, precipitation, reverse osmosis, distillation, biosorption, filtration, flocculation and coagulation (Choi & Chen 1979; Chaturvedi et al. 1988; Gaciri & Davies 1993; Czarnowski et al. 1996; Dorenbos 2000; Çengeloğlu et al. 2002; Fan et al. 2003; Fletcher et al. 2006; Daifullah et al. 2007; Bansiwal et al. 2009; Biswas et al. 2009; Zhu et al. 2009; Camacho et al. 2010; Chen et al. 2010a, b, 2011; Bhatnagar et al. 2011; Chang et al. 2011; Brunson & Sabatini 2014; Dayananda et al. 2014; Gwala et al. 2014).

Calcium-based materials and modified clay have been used for defluoridation (Tor 2006; Viswanathan & Meenakshi 2010; Sujana & Anand 2011; Masindi et al. 2014b). To date, no study has reported on use of cryptocrystalline minerals for defluoridation of contaminated water. The primary aim of this study was to explore the feasibility of using cryptocrystalline magnesite as an adsorbent for defluoridation of groundwater. A study by Masindi et al. (2014a) documented that Folovhodwe region, in Limpopo province, South Africa has large deposits of magnesite (18 MT) which have been prospected to be mined for the coming 20 years unless the demand increases.

EXPERIMENTAL METHODS

Sampling and preparations of magnesite adsorbent

Raw magnesite rocks were collected prior to any processing at the mine from the Folovhodwe Magnesite Mine, Limpopo Province, South Africa (22°35′47.0″S; 30°25′33″E). Magnesite samples were milled to a fine powder using a Retsch RS 200 miller and passed through a 32-μm particle size sieve.

Physiochemical characterization of the adsorbent

Elemental characteristics of the magnesite samples were measured by X-ray fluorescence spectroscopy (XRF) using a Philips PW 1480 X-ray spectrometer. Mineralogical characteristics of the magnesite samples were ascertained using a Philips X-ray diffractometer with Cu-Kα radiation. Phase identification was performed by searching and matching obtained spectra with the powder diffraction file data base with the help of JCPDS (Joint Committee of Powder Diffraction Standards) files for inorganic compounds. Morphology and elemental composition of magnesite were investigated by energy dispersive X-ray spectrometry (EDS: focused ion beam, Auriga from Carl Zeiss) attached to a scanning electron microscope (SEM: JEOL JSM7500 microscope) (SEM-EDS).

Preparation of stock solution and working solution

The accuracy of the analysis was monitored by analysis of National Institute of Standards and Technology water standards. Simulated fluoride-rich water was prepared using sodium fluoride salt. Fluoride stock solution (1,000 mg/L) was prepared by dissolving 2.2 g NaF in 1,000 mL ultra-pure water. A working solution (10 mg/L) was prepared by diluting the stock solution 1:100 in ultra-pure water.

Optimization of defluoridation conditions

Effect of agitation time

Removal of F with contact time was evaluated by using 10 mg/L F solution and 1:100 solid:liquid (S:L) ratio. Nine 100 mL aliquots of the 10 mg/L F solution were pipetted into 250 mL bottles and magnesite was added at a dose of 10 g/L. The mixtures were agitated for 1, 5, 10, 15, 20, 30, 60, 180 and 360 min at 250 rpm using a Stuart reciprocating shaker. After equilibration, the mixture was filtered through a 0.45 μm pore size nitrate cellulose filter membrane. The samples were analyzed for F by a CRISON ion selective electrode.

Effect of magnesite dosage

Removal of F according to the magnesite dose was evaluated using 10 mg L F solution. One hundred millilitre aliquots of the 10 mg/L F solution were pipetted into eight 250 mL bottles and 0.1, 0.3, 0.5, 1, 2, 4, 8 and 10 g of magnesite, respectively, were added to the flasks. The mixtures were then agitated for 20 min at 250 rpm using a Stuart reciprocating shaker. After equilibration, the mixtures were filtered through a 0.45 μm pore size nitrate cellulose filter membrane. The samples were analyzed for F by a CRISON ion selective electrode.

Effect of ion concentration

Removal of F with concentration was evaluated at 20 min of contact time, 20 g/L of magnesite dosage, 2:100 S:L and 250 rpm mixing. Ten 100 mL samples at initial concentration of 2, 4, 6, 8, 15, 20, 30, 40, 50 and 60 mg/L F solution were pipetted into 250 mL bottles and 20 g/L of magnesite was added to each. The mixtures were then agitated for 60 min at 250 rpm using a Stuart reciprocating shaker. After equilibration, the mixtures were filtered through a 0.45 μm pore size nitrate cellulose filter membrane. The samples were analyzed for F by CRISON ion selective electrode.

Effect of supernatants pH

Removal of fluoride as a function of pH was evaluated at 20 min of contact time, 20 g/L of magnesite, 2/100 S:L and 60 mg/L F. One hundred millilitre aliquots of 60 mg/L F solution were pipetted into six 250 mL bottles. The pH of the solution was adjusted to 2, 4, 6, 8, 10 and 12 using 0.1 M of HCl and 0.1 M NaOH. Four grams of magnesite was added to each bottle. The mixture was then agitated for 60 min using a reciprocating shaker at 250 rpm. After equilibration, the mixtures were filtered through a 0.45 μm pore size nitrate cellulose filter membrane. The samples were analyzed for F by CRISON ion selective electrode.

Effect of competing ions

Removal of F with competing ions was evaluated at 20 min of contact time, 20 g/L of magnesite dosage, 2/100 S:L and 250 rpm. One hundred millilitres of solution containing of fluoride, sulphate, nitrate, chloride and bromide, all at 60 mg/L, was mixed with 20 g/L of cryptocrystalline magnesite. The mixture was equilibrated for 20 min at 250 rpm using a Stuart reciprocating shaker. After equilibration, the mixtures were filtered through a 0.45 μm pore size nitrate cellulose filter membrane. The samples were analyzed for anions by CRISON ion selective electrode and ion chromatography.

Adsorption of F from borehole water at optimized conditions

The optimized conditions of 20 min of contact time, 20 g/L of magnesite dosage, 2/100 S:L, 25 °C and 250 rpm shaking were used for defluoridating groundwater. The borehole water came from Siloam, Nzhelele, Limpopo province, South Africa.

Modelling of analytical results

Adsorption kinetics

The pseudo-first order is a kinetic model described by the following equation 
formula
1
where qe (mg g−1) is adsorption capacity at equilibrium, qt (mg g−1) is the adsorption capacity at time t, and K (min−1) is the rate constant of pseudo-first order. The value of K1 can be obtained from the slope by plotting t vs. log (qeqt).
The pseudo-second order model is used when the applicability of the first order kinetics becomes untenable. The equation of the pseudo-second order model is 
formula
2

This equation is used to obtain K2, the second order rate from the plots t vs. t/qe.

Adsorption isotherms

The relationship between the amount of fluoride adsorbed and the fluoride concentration remaining in solution is described by an isotherm. The two most common isotherm types for describing this type of system are Langmuir and Freundlich adsorption isotherms. These models describe adsorption processes on a homogenous (monolayer) or heterogeneous (multilayer) surface respectively. The most important model of monolayer adsorption came from Langmuir. This isotherm is given as follows: 
formula
3
The constants Q0 and b are characteristics of the Langmuir equation and can be determined from a linearized form of Equation (4). The Langmuir isotherm is valid for monolayer sorption due to a surface with finite number of identical sites and can be expressed in the following linear form: 
formula
4
where Ce = equilibrium concentration (mg L−1), Qe =amount adsorbed at equilibrium (mg g−1), Qm =Langmuir constant related to adsorption capacity (mg g−1) and b = Langmuir constant related to energy of adsorption (L mg−1). A plot of Ce vs. Ce/Qe should be linear if the data is described by the Langmuir isotherm. The value of Qm is determined from the slope and the intercept of the plot. It is used to derive the maximum adsorption capacity and b is determined from the original equation and it represents the intensity of adsorption.
The Freundlich adsorption isotherm describes the heterogeneous surface energy by multilayer adsorption. The Freundlich isotherm can be formulated as follows: 
formula
5
The equation may be linearized by taking the logarithm of both sides of the equation and can be expressed in linear form as follow: 
formula
6
where Ce = equilibrium concentration (mg L−1), qe = amount adsorbed at equilibrium (mg g−1), K = partition coefficient (mg g−1) and n = intensity of adsorption. The linear plot of log Ce vs. log qe indicates if the data is described by Freundlich isotherm. The value of K implies that the energy of adsorption on a homogeneous surface is independent of surface coverage and n is an adsorption constant which reveals the rate at which adsorption is taking place. These two constants are determined from the slope and intercept of the plot of each isotherm. An error analysis is required in order to evaluate the fit of the adsorption isotherms to experimental data. In the present study, the linear coefficient of determination of (R2) was employed for the error analysis. The linear coefficient of determination is calculated by using the equation 
formula
7
Theoretically, the values of R2 range from 0 to 1. A R2 value of 1 shows that 100% of the variation of experimental data is explained by the regression equation. The coefficient of determination R2 was applied to determine the relationship between the experimental data and the isotherms in most studies.

RESULTS AND DISCUSSION

Elemental composition of magnesite by XRF

XRF analysis showed that cryptocrystalline magnesite is composed of MgO (92%), SiO2 (5%), CaO (2%), FeO3 (0.5%) and MnO (0.5%).

Thus magnesite from Folovhodwe is dominated by MgO as the main element. It also contains impurities of silicon, calcium, iron and manganese which might have been incorporated during deposition processes. The results corroborated previous findings (Masindi et al. 2014a).

Mineralogical composition by X-ray diffraction

The mineralogical composition of cryptocrystalline magnesite is presented in Figure 1 and the quantitative results are shown in Table 1.

Table 1

Percentage composition (weight) of magnesite

Mineral phases Weight% 3 σ error 
Brucite 15.44 0.48 
Forsterite 3.11 0.22 
Periclase 81.45 0.48 
Mineral phases Weight% 3 σ error 
Brucite 15.44 0.48 
Forsterite 3.11 0.22 
Periclase 81.45 0.48 
Figure 1

Mineralogical composition of cryptocrystalline magnesite.

Figure 1

Mineralogical composition of cryptocrystalline magnesite.

As shown in Figure 1, cryptocrystalline magnesite contains periclase and brucite as the main crystalline minerals. It also contains impurities of forsterite and monticellite. The amorphous phases were not detected on the analysis. Amorphous phases were quantified by XRF, Fourier transform infrared spectroscopy and EDS.

Elemental composition by EDS and morphology by SEM

The elemental composition and morphology of cryptocrystalline magnesite is shown in Figure 2.

Figure 2

EDS (left) and SEM (right) analysis of cryptocrystalline magnesite.

Figure 2

EDS (left) and SEM (right) analysis of cryptocrystalline magnesite.

SEM-EDS results show that cryptocrystalline magnesite contains Mg, C, O, Si and Ca, respectively. This corroborates results reported by Sparks (2003) who state that theoretically, pure magnesite contains 47.8% MgO and 52.2% CO2. The sum of Mg, C and O in the material under study indicate that the material is a carbonate of amorphous nature since it was detected by the XRD. SEM reveals the morphology of cryptocrystalline magnesite to be characterized by spherical shapes. Granular and rough surfaces promote the interaction of the surface and aqueous solution (Sparks 1995; Selim & Sparks 2001; Artiola et al. 2004; Harrison 2006; Vicente et al. 2013).

Batch experiments

Effects of contact time

Figure 3 shows the variation of fluoride concentration with agitation time. Fluoride removal was evaluated at 10 g/L of cryptocrystalline magnesite, 10 mg/L of fluoride concentration, 25 °C ambient temperature and 1 g/100 mL S:L ratio. The shaking time was varied from 1 to 360 min.

Figure 3

Effect of contact time on defluoridation of borehole water.

Figure 3

Effect of contact time on defluoridation of borehole water.

Figure 3 shows that there was a gradual increase in the percentage removal of fluoride from aqueous solution with an increase in contact time. From the first minute of contact, the reaction kinetics were very rapid and then stabilized after 20 min (approximately 90% fluoride removal). After 20 min, no further significant removal of fluoride was observed, indicating that the reaction has reached an equilibrium. Therefore, 20 min was taken as the optimum contact time and was used for subsequent experiments. Chen et al. (2010b) reported that ceramic adsorbent takes 48 h to remove fluoride from wastewater. Biswas et al. (2010) noted that 1.5 h is required to remove fluoride from aqueous solution using synthetic iron, aluminium and chromium (III) ternary mixed oxide. This study managed to remove fluoride in the shortest time of 20 min as compared to selected conventional methods.

Effects of magnesite dosage

Figure 4 shows the variation of fluoride concentration with dose of magnesite added to borehole water. Fluoride removal was evaluated at 20 min of agitation, 10 mg/L fluoride concentration, and 25 °C ambient temperature. The doses of magnesite were varied from 0.1 to 10 g.

Figure 4

Effect of dosage on defluoridation of borehole water.

Figure 4

Effect of dosage on defluoridation of borehole water.

The percentage removal of fluoride from aqueous solution was observed to increase with an increase in dosage of magnesite. As dosage increases, more surfaces which are suitable for the uptake of fluoride become more available. From 1 to 2 g per sample, the fluoride uptake was observed to be gradually increasing until it reached a steady state at 2 g (% removal ≥100%). This is an indication that more surfaces were present to remove all the fluoride from the aqueous solution. After 2 g, no further reaction was observed hence indicating that 4 g of magnesite provides enough surfaces to remove approximately 10 mg/L of fluoride. At low adsorbent dosage, the fluoride adsorption rate is rapid since the active sites are easily available and at high adsorbent dosage, the adsorbate species find it increasingly difficult to access the adsorption sites and equilibrium is established. As such, it was noted that the optimum dosage suitable for defluoridation is 2 g/100 mL of magnesite. Therefore, 20 min of agitation and 20 g/L of magnesite are the optimum defluoridation conditions which will be used in subsequent experiments.

Effects of ion concentration

Figure 5 shows the variation of percentage removal of fluoride from aqueous media with varying fluoride concentration. Fluoride removal was evaluated for 20 min contact time, 20 g/L of dosage, 2/100 S:L ratio and 25 °C ambient temperature. Concentrations were varied from 2 to 60 mg/L.

Figure 5

Effect of concentration on defluoridation of borehole water.

Figure 5

Effect of concentration on defluoridation of borehole water.

Figure 5 shows a decrease in percentage removal of fluoride with an increase in fluoride concentration. The general observation is that when the concentration was increased the percentage removal decreased. However, from 2 to 10 mg/L, there was a slight increase in the adsorption of fluoride because initially the adsorption rate may be higher due to an increase in the number of vacant sites available resulting in an increased concentration gradient between the sorbate in the solution and that of the sorbent surface. With time, the concentration gradient is reduced owing to the fluoride adsorption onto the vacant sites leading to the decreased adsorption during the later stages. Over the concentration range 2–15 mg/L magnesite managed to remove approximately 99% of fluoride from an aqueous medium. Over the concentration range 15–60 mg/L, it was capable of removing >95% of fluoride from the solution. This shows that magnesite has strong affinity for fluoride. From the obtained results, it can be concluded that 20 min of contact time, 20 g/L of dosage, and 60 mg/L of fluoride solution are the optimum conditions for defluoridation of borehole water.

Effect of supernatant pH

Figures 6 and 7 shows the variation of pH and percentage removal of fluoride from aqueous media with varying pH ranges. Fluoride removal was evaluated at 20 min contact time, 20 g/L of dosage, 60 mg/L F, 2/100 S:L ratio and 25 °C ambient temperature. The pH varied from 2 to 12.

Figure 6

Variation of initial and final pH on defluoridation of borehole water.

Figure 6

Variation of initial and final pH on defluoridation of borehole water.

Figure 7

Effect of pH on defluoridation of borehole water.

Figure 7

Effect of pH on defluoridation of borehole water.

Figure 6 shows an increase in pH after the interaction of magnesite and fluoride-rich water. Final pH for all solutions ranged from 10 to 11. Removal of fluoride was >95.0% at all pH ranges except pH 2 which was slightly lower than 94.5%. The percentage removal did not vary significantly at various initial pH suggesting the removal of fluoride using cryptocrystalline magnesite is independent of pH of media. Furthermore, it was observed that adsorption of fluoride using magnesite is independent of initial pH (Figure 7).

Effects of competing ions

The effect of competing ions on fluoride removal is shown in Figure 8 (20 min contact time, 20 g/L dosage, 60 mg/L, 2:100 S:L ratio and 25 °C ambient temperature).

Figure 8

Effect of competing ions on defluoridation of borehole water (60 min contact time, 20 g/L dosage, 60 mg/L, 2/100 S:L ratios and 25 °C ambient temperature).

Figure 8

Effect of competing ions on defluoridation of borehole water (60 min contact time, 20 g/L dosage, 60 mg/L, 2/100 S:L ratios and 25 °C ambient temperature).

The effects of co-existing nitrate, sulphates, chlorite, and bromide anions on fluoride removal was examined (Figure 8). In the presence of other anions, removal of fluoride by cryptocrystalline magnesite was greater than 99%. Sulphate was also observed to be adsorbed onto magnesite matrices hence showing that the anionic species has affinity to magnesite. As shown in Figure 8, the competing anions have no effects on the removal of fluoride from aqueous solution.

Possible fluoride removal mechanism

The mechanism of fluoride sorption on magnesite surfaces can be given by the following expression: 
formula
8
 
formula
9
It is expected that similar process occur on the presence study using cryptocrystalline magnesite.

Adsorption kinetics

The pseudo-first and second order adsorption kinetics is shown in Figure 9.

Figure 9

Pseudo-first order (a), and second order kinetics (b) (20 g/L of the magnesite, 10 mg L−1 of fluoride, 250 rpm and at ambient temperature). Time: 1–360 min.

Figure 9

Pseudo-first order (a), and second order kinetics (b) (20 g/L of the magnesite, 10 mg L−1 of fluoride, 250 rpm and at ambient temperature). Time: 1–360 min.

To evaluate the mechanisms of adsorption by magnesite, pseudo-first order and second order kinetic models were applied (Figures 9(a) and 9(b)). This was done in an attempt to gain insight on the mechanism and steps controlling removal of fluoride. The kinetics data showed better correlation with the pseudo-second order model because of the high values of correlation coefficients based on the linear regression (R2 = 1). This confirms chemisorption, since the rate limiting step is a chemical sorption.

Adsorption isotherms

The Langmuir and Freundlich adsorption isotherm is shown by Figure 10.

Figure 10

Langmuir (a), and Freundlich (b) adsorption isotherm (20 g/L of the magnesite, 20 min of equilibration and 250 rpm). Concentration was varied from 2 to 60 mg L−1. The Langmuir and Freundlich adsorption isotherms are shown in Table 2.

Figure 10

Langmuir (a), and Freundlich (b) adsorption isotherm (20 g/L of the magnesite, 20 min of equilibration and 250 rpm). Concentration was varied from 2 to 60 mg L−1. The Langmuir and Freundlich adsorption isotherms are shown in Table 2.

Table 2

Langmuir and Freundlich adsorption isotherm

  Langmuir adsorption isotherm
 
Freundlich adsorption isotherm
 
Species R2 Qm b RL R2 k n 
F 0.99 9.2 5.9 0.003 0.90 5.0 2.4 
  Langmuir adsorption isotherm
 
Freundlich adsorption isotherm
 
Species R2 Qm b RL R2 k n 
F 0.99 9.2 5.9 0.003 0.90 5.0 2.4 

The R2 values for the linear form of the Langmuir and Freundlich isotherms are 0.99 and 0.90, respectively (Figures 8(a) and 8(b)). According to R2 values, the Langmuir isotherm best represents the equilibrium adsorption of fluoride onto cryptocrystalline magnesite; it therefore means that the adsorption process occurred on a heterogeneous surface energy by multilayer adsorption. The RL between 0 and 1 means the adsorption is favourable (Kamble et al. 2009). As such, adsorption of fluoride onto magnesite matrices is favorable.

Removal of fluoride under optimized conditions

Table 3 shows the removal efficiency of fluoride from aqueous media using magnesite under optimized and natural conditions. The fluoride-rich groundwater had a slightly alkaline pH of 9.2. Magnesite was observed to remove fluoride from groundwater to below DWAF water quality guidelines. This shows that magnesite is an effective material that can be used for removal of fluoride in wastewaters or groundwater.

Table 3

Physiochemical conditions of borehole water before and after defluoridation

Parameter Untreated water Treated water 
pH 9.2 10.26 
EC (μS cm−1100 45.3 
TDS (mg L−166.4 185.3 
F (mg L−110 0.01 
Bromide 11 
Sulphate 47 0.01 
Nitrate 65 64 
Chloride 140 141 
Parameter Untreated water Treated water 
pH 9.2 10.26 
EC (μS cm−1100 45.3 
TDS (mg L−166.4 185.3 
F (mg L−110 0.01 
Bromide 11 
Sulphate 47 0.01 
Nitrate 65 64 
Chloride 140 141 

EC, electrical conductivity; TDS, total dissolved solids.

Table 4 shows the different adsorption capacities of various adsorbents of fluoride reported in the literature. Even though the adsorption capacities were obtained at different pHs and temperatures, they offer a useful criterion to compare the different adsorption capacities. From the table, it is clear that magnesite has a higher adsorption capacity than the other adsorbents. Therefore magnesite can be effectively applied for defluoridation of groundwater because of its ability to adsorb fluoride.

Table 4

Comparison of different adsorption capacities (mg/g) of different adsorbents for fluoride

Adsorbent Adsorption capacity Source 
Cryptocrystalline magnesite 9.2 Present study 
Alum-bent 5.7 Masindi et al. (2014b)  
Al-oxide origin alumina Kamble et al. (2010)  
Mg2+ bentonite 2.3 Thakre et al. (2010)  
Polymer/bio-polymer composites 15 Karthikeyan et al. (2011)  
Activated alumina 1.1 Maliyekkal et al. (2006)  
Lanthanide impregnated silica gel 3.8 Zhou et al. (2004)  
Fe3+ modified bentonite clay 2.9 Gitari et al. (2015)  
Adsorbent Adsorption capacity Source 
Cryptocrystalline magnesite 9.2 Present study 
Alum-bent 5.7 Masindi et al. (2014b)  
Al-oxide origin alumina Kamble et al. (2010)  
Mg2+ bentonite 2.3 Thakre et al. (2010)  
Polymer/bio-polymer composites 15 Karthikeyan et al. (2011)  
Activated alumina 1.1 Maliyekkal et al. (2006)  
Lanthanide impregnated silica gel 3.8 Zhou et al. (2004)  
Fe3+ modified bentonite clay 2.9 Gitari et al. (2015)  

CONCLUSIONS AND RECOMMENDATIONS

This study arrived at the following conclusions:

  1. Adsorption of fluoride by magnesite is independent of pH.

  2. Maximum fluoride removal was found at 20 min of contact time, 20 g/L of dosage, 60 mg/L of fluoride concentration, 2/100 S:L and 25 °C ambient temperature.

  3. Kinetic studies revealed that the fluoride removal followed a pseudo-second order than pseudo-first order hence confirming chemisorption as the rate-limiting step.

  4. The adsorption data fitted better to Langmuir than Freundlich adsorption isotherms hence confirming monolayer adsorption

  5. To this end, it can be concluded that, effectiveness, availability and low cost of cryptocrystalline magnesite makes this material a potential candidate for defluoridation of groundwater.

  6. Cryptocrystalline magnesite managed to remove fluoride to below DWAF drinking water quality guidelines. As such, this technology can be used as a point source defluoridation technique in rural areas and households in South Africa and other developing countries.

ACKNOWLEDGEMENTS

The authors wish to convey their sincere acknowledgement to the council of scientific and industrial research (CSIR), SASOL-INZALO, ESKOM-TESP, National Research Foundation (NRF), and Department of Science and Technology (DST) and University of Venda research and publication committee for funding this project.

REFERENCES

REFERENCES
Artiola
J. F.
Pepper
I. L.
Brusseau
M. L.
2004
Environmental Monitoring and Characterization
.
Elsevier Academic Press
,
Burlington, MA
,
USA
.
Bansiwal
A.
Thakre
D.
Labhshetwar
N.
Meshram
S.
Rayalu
S.
2009
Fluoride removal using lanthanum incorporated chitosan beads
.
Colloid. Surf. B.
74
,
216
224
.
Bhatnagar
A.
Kumar
E.
Sillanpää
M.
2011
Fluoride removal from water by adsorption – A review
.
Chem. Eng. J.
171
,
811
840
.
Camacho
L. M.
Torres
A.
Saha
D.
Deng
S.
2010
Adsorption equilibrium and kinetics of fluoride on sol–gel-derived activated alumina adsorbents
.
J. Colloid Interf. Sci.
349
,
307
313
.
Çengeloğlu
Y.
Kır
E.
Ersöz
M.
2002
Removal of fluoride from aqueous solution by using red mud
.
Sep. Purif. Technol.
28
,
81
86
.
Chaturvedi
A. K.
Pathak
K. C.
Singh
V. N.
1988
Fluoride removal from water by adsorption on china clay
.
Appl. Clay Sci.
3
,
337
346
.
Chen
N.
Zhang
Z.
Feng
C.
Sugiura
N.
Li
M.
Chen
R.
2010a
Fluoride removal from water by granular ceramic adsorption
.
Colloid Interf. Sci.
348
,
579
584
.
Chen
N.
Zhang
Z.
Feng
C.
Li
M.
Zhu
D.
Chen
R.
Sugiura
N.
2010b
An excellent fluoride sorption behavior of ceramic adsorbent
.
J. Hazard. Mater.
183
,
460
465
.
Choi
W. W.
Chen
K. Y.
1979
Removal of fluoride from waters by adsorption
.
J. Am. Water Works Ass.
71
,
562
570
.
Czarnowski
W.
Wrześniowska
K.
Krechniak
J.
1996
Fluoride in drinking water and human urine in Northern and Central Poland
.
Sci. Total Environ.
191
,
177
184
.
Dayananda
D.
Sarva
V. R.
Prasad
S. V.
Arunachalam
J.
Ghosh
N. N.
2014
Preparation of CaO loaded mesoporous Al2O3: Efficient adsorbent for fluoride removal from water
.
Chem. Eng. J.
248
,
430
439
.
Fan
X.
Parker
D. J.
Smith
M. D.
2003
Adsorption kinetics of fluoride on low cost materials
.
Water Res.
37
,
4929
4937
.
Fletcher
H. R.
Smith
D. W.
Pivonka
P.
2006
Modeling the sorption of fluoride onto alumina
.
J. Environ. Eng.
132
,
229
246
.
Gitari
W. M.
Ngulube
T.
Masindi
V.
Gumbo
J. R.
2013
Defluoridation of groundwater using Fe(III)-modified bentonite clay: optimization of adsorption conditions
.
Desalination Water Treat.
53
,
1578
1590
.
Harrison
R. M.
2006
An Introduction to Pollution Science
.
RSC Publishing
,
Cambridge
,
UK
.
Kamble
S. P.
Dixit
P.
Rayalu
S. S.
Labhsetwar
N. K.
2009
Defluoridation of drinking water using chemically modified bentonite clay
.
Desalination
249
,
687
693
.
Kamble
S. P.
Deshpande
G.
Barve
P. P.
Rayalu
S.
Labhsetwar
N. K.
Malyshew
A.
Kulkarni
B. D.
2010
Adsorption of fluoride from aqueous solution by alumina of alkoxide nature: batch and continuous operation
.
Desalination
264
,
15
23
.
Meenakshi
R. C.
2006
Fluoride in drinking water and its removal
.
J. Hazard. Mater.
137
,
456
463
.
Selim
H. M. E.
Sparks
D. L.
2001
Physical and Chemical Processes of Water and Solute Transport/Retention in Soils: Proceedings of a Symposium Sponsored by Divisions S-1 and S-2 of the Soil Science Society of America in Baltimore, MD, 18 to 22 October 1998
.
Soil Science Society of America
,
Madison, WI
,
USA
.
Sparks
D. L.
1995
Environmental Soil Chemistry
.
Academic Press
,
Amsterdam
.
Sparks
D. L.
2003
Environmental Soil Chemistry
.
Academic Press
,
Amsterdam
.
Srivastav
A. L.
Singh
P. K.
Srivastava
V.
Sharma
Y. C.
2013
Application of a new adsorbent for fluoride removal from aqueous solutions
.
J. Hazard. Mater.
263
,
342
352
.
Thakre
D.
Rayalu
S.
Kawade
R.
Meshram
S.
Subrt
J.
Labhsetwar
N.
2010
Magnesium incorporated bentonite clay for defluoridation of drinking water
.
J. Hazard. Mater.
180
,
122
130
.
Tor
A.
Danaoglu
N.
Arslan
G.
Cengeloglu
Y.
2009
Removal of fluoride from water by using granular red mud: Batch and column studies
.
J. Hazard. Mater.
164
,
271
278
.
Vicente
M. A.
Gil
A.
Bergaya
F.
2013
Chapter 10.5, Pillared clays and clay minerals
. (
Faïza
B.
Gerhard
L.
, eds.)
Developments in Clay Science
,
Elsevier, Amsterdam
, pp.
523
557
.
Viswanathan
N.
Meenakshi
S.
2010
Enriched fluoride sorption using alumina/chitosan composite
.
J. Hazard. Mater.
178
,
226
232
.
Wang
J.
Xu
W.
Chen
L.
Jia
Y.
Wang
L.
Huang
X. -J.
Liu
J.
2013
Excellent fluoride removal performance by CeO2–ZrO2 nanocages in water environment
.
Chem. Eng. J.
231
,
198
205
.
Xiuru
Y.
Kuanxiu
S.
Jianping
W.
Liuchang
H.
Zhaohui
Y.
1998
Preparation of CeO2–TiO2/SiO2 and its removal properties for fluoride ion
.
J. Rare Earths
16
,
279
280
.
Zhu
P.
Wang
H.
Sun
B.
Deng
P.
Hou
S.
Yu
Y.
2009
Adsorption of fluoride from aqueous solution by magnesia-amended silicon dioxide granules
.
J Chem. Technol. Biotechnol.
84
,
1449
1455
.