Different bone char (BC) preparation methods affect the physical chemical properties and therefore the capacity to remove fluoride. Fluoride removal capacities of BC prepared at three controlled temperatures, with particle grain sizes of 250–500 μm, 500–1,000 μm and 1,000–1,800 μm were determined in column experiments with an initial fluoride concentration of 8.55 mg/L. The sorbent was calcined at 400°C, 500°C and 600°C. BC calcined at 400°C had better fluoride removal performance compared to those calcined at 500°C and 600°C due to decreased pore volume and surface area as the calcination temperature was increased. There was a reduced mass transfer effect to the adsorption sites in pores as the BC particle size was increased. The equilibrium adsorptions Langmuir and Freundlich isotherms were tested. For the Langmuir equilibrium adsorption isotherm, maximum monolayer coverage (Q) was determined to be 3.512 mg/g, and the value of the separation factor (r) obtained was 0.1394 indicating favorable adsorption as it lies between 0 and 1. The Freundlich governing equilibrium adsorption isotherm model gave a value of 1/n equal to 0.445. This indicates a favorable adsorption process, since the bond energies increase with surface density of the adsorbent.

INTRODUCTION

Application of the adsorption process using locally available adsorbents is one of the cheap methods of defluoridation of fluoridated water in Tanzania. However, the adsorbents have to be prepared properly. Calcination of animal bones produces an effective material for removing fluoride. Depending on the charring process, the bones may result in a product with low defluoridation capacity (Dahi 2000). The production process of bone char (BC) therefore is one of the most important steps for development of proper structural and physico-chemical properties of BC. Calcination temperature and duration influence the pore structure formation and the surface chemistry of BC (Bonilla-Petriciolet et al. 2013).

Good quality BC is an effective material for removing fluoride (Jacobsen 2007). Effective bone calcination ensures removal of organic matter that otherwise adds taste and color to water (Dahi 2000).

This paper reports on BC preparation method using cow bones obtained from abattoirs around Arusha in Tanzania. The objective of the research was to study the effects of temperature on the quality of BC developed and its adsorption performance.

MATERIALS AND METHODS

BC preparation

Cattle bones were collected from slaughter houses in the Arusha area. The collected bones were calcined under controlled temperature and rate of heating in a thermo scientific (Thermolyne) programmable furnace in a laboratory condition. Bones were calcined at temperatures of 400°C, 500°C and 600°C in a furnace. According to Posner (1987), temperatures higher than 600°C damage the apatite structure, while temperatures below 400°C result in charred bones, giving a bad taste and odor when used for treating water owing to incomplete volatilization of organic matter. The temperature of the electric furnace was raised from room temperature at a controlled rate of 8°C/min and maintained at a constant temperature for three hours. Smoke release was used as an indicator of removal of volatile matter. For example, at 400°C, smoke started to be emitted at 250°C and at 350°C dense smoke was emitted. Smoking stopped after one hour at 400°C, which means charring was continued for a sufficient time at 400°C to allow all volatile matter to be exhausted and an extra time to allow for necessary structural change without leading to sintering. After charring, grinding and sieving, the BC was washed with deionised water in order to remove the dust before packing into the column filter. In so doing smell, color and bad taste were controlled. The calcined bones were allowed to cool to room temperature before further treatment.

Calcined bones were milled using a laboratory milling machine, Retschmuhle, GmbH, type: SK1 Number 72307. Thereafter, ground BC was sieved and graded using electromagnetic vibration sieves set with diameters ranging between 100 μm and 1,800 μm in order to get different BC particle sizes according to Laboratory Test Sieve ISO 565, part no. 667924, Endecotts Ltd, London, UK. These sizes comply with wire mesh series standard ISO 3310-1:2000 and BS 410-1:2000.

Experimental set up

A defluoridating BC column unit was constructed using a transparent Plexiglas pipe with a diameter of 38 mm. The column was closed at the lower end with an inlet regulating valve and at the upper end with an outlet valve. Naturally fluoridated water with an initial concentration of 8.55 mg/L maintained an upward flow through the column at a rate of 2.2 cm3/s. The water flow rate was maintained through the Plexiglas pipe by a constant head reservoir. The Plexiglas pipe had a height of 370 mm and was packed with 250 g of BC. The diameters of BC particle sizes used were within 250–500 μm, 500–1,000 μm and 1,000–1,800 μm, each calcined at temperatures of 400°C, 500°C and 600°C. The BC with particles of diameter less than 250 μm clogged the flow in the experiment column and therefore it was not used in this study. The flow rate was adjusted by inlet valve and maintained at 2.2 cm3/s (Figure 1). The effluent was collected in a sampling tube. Samples were collected at a time interval of 2 minutes for the first 10 minutes, then at intervals of 5 minutes up to 30 minutes, 10 minutes up to 60 minutes, 30 minutes up to 180 minutes and finally 1 hour up to 6 hours. This was to take care of fast fluoride adsorption in the initial stages.
Figure 1

A schematic diagram of BC column experimental set up.

Figure 1

A schematic diagram of BC column experimental set up.

Fluoride concentration measurement

Standard analytical methods (APHA 1989; HACH 1997) were used to quantify the presence of fluoride ion concentration. A 5 ml sample solution was mixed with 5 ml of 1,2-cyclohexanedinitrilo-tetraacetic acid-total ionic strength adjustment buffer, and the fluoride concentrations were measured using an ion selective electrode technique as described by Thole et al. (2012).

BC morphology and pore structure

The scanning electron microphotograph (SEM) of the calcined BC was studied using TESCO SEM model HCITMSAAW11, 5 V, 500 mA, SC: 102535 to evaluate surface morphology. The surface area and pore volume of the BC were determined by the Brunauer–Emmet–Teller method using a computer-controlled nitrogen gas adsorption analyzer at 273.0 K in the Chemistry Department laboratory at the University of Dar es Salaam.

BC X-ray fluorescence analysis

The X-ray fluorescence (XRF) Bruker AXS S8 Tiger model with serial number 206548 was used to determine the chemical composition of the BC media charred at different temperatures. Results are presented in Table 1. Taking BC charred at 400°C as a base, the oxide composition was found to change for some of the minor components (SiO2, SO3, Al2O3, Fe2O3 and CuO) in BC charred at temperatures of 500°C and 600°C. Therefore, this is an indication of structural change in the developed BC as the charring temperature was changed.

Table 1

Result of XRF analysis showing major component of bones charred at different temperatures

  Bones charred at 400°C Bones charred at 500°C
 
Bones charred at 600°C
 
Compound Compound content (%) Compound content (%) % decrease or increase Compound content (%) % decrease or increase 
CaO 55.21 55.27 0.1 55.2 0.0 
P2O5 39.23 40.88 4.2 41.43 5.6 
Na21.5 1.44 −4.0 1.41 −6.0 
MgO 1.22 1.18 −3.3 1.21 −0.8 
SiO2 1.04 0.37 −64.4 0.13 −87.5 
SO3 0.83 0.31 −62.7 0.12 −85.5 
Al2O3 0.28 −100.0  −100.0 
Cl 0.18 0.17 −5.6 0.16 −11.1 
Fe2O3 0.15 0.07 −53.3 0.08 −46.7 
SrO 0.13 0.13 0.0 0.13 0.0 
K20.1 0.11 10.0 0.11 10.0 
CuO 0.09 0.03 −66.7  −100.0 
ZnO 0.02 0.02 0.0 0.03 50.0 
  Bones charred at 400°C Bones charred at 500°C
 
Bones charred at 600°C
 
Compound Compound content (%) Compound content (%) % decrease or increase Compound content (%) % decrease or increase 
CaO 55.21 55.27 0.1 55.2 0.0 
P2O5 39.23 40.88 4.2 41.43 5.6 
Na21.5 1.44 −4.0 1.41 −6.0 
MgO 1.22 1.18 −3.3 1.21 −0.8 
SiO2 1.04 0.37 −64.4 0.13 −87.5 
SO3 0.83 0.31 −62.7 0.12 −85.5 
Al2O3 0.28 −100.0  −100.0 
Cl 0.18 0.17 −5.6 0.16 −11.1 
Fe2O3 0.15 0.07 −53.3 0.08 −46.7 
SrO 0.13 0.13 0.0 0.13 0.0 
K20.1 0.11 10.0 0.11 10.0 
CuO 0.09 0.03 −66.7  −100.0 
ZnO 0.02 0.02 0.0 0.03 50.0 

Defluoridation capacity of BC

The defluoridation capacity of the prepared BC was determined using the following mass balance equation (Thole et al. 2012): 
formula
1
where Dc is the defluoridation capacity (mg/g), Ci is the initial fluoride concentration (mg/L), Cf is the residual fluoride concentration (mg/L), V is the volume of treated water (L), and m is the mass of BC media used (g).

The governing equilibrium adsorption isotherm

The governing equilibrium adsorption isotherm for the BC prepared under controlled conditions was studied. The relationship between the equilibrium amount of fluoride on the adsorbent and solute concentration in water is well described by an isotherm. The purpose of an adsorption isotherm is to relate the adsorbate concentration in the bulk solution to the amount adsorbed at the solid/water interface (Kishore et al. 2011). To model the process, a batch experiment was carried out. The equilibrium data were fitted to both Langmuir and Freundlich isotherm equations. The Langmuir adsorption isotherm is used to describe the formation of monolayer adsorbate on the outer surface of the adsorbent quantitatively at the equilibrium concentration whereby no further adsorption takes place, and hence representing the equilibrium distribution of the ions between the solid and liquid phase (Said & Machunda 2014).

Batch experiments for adsorption studies were carried out in the laboratory by adding a known amount of BC to four flasks containing 500 mL of fluoride water with an initial fluoride concentration of 21 mg/L. The mixture was stirred at the rate of 100 rpm using Phipps & Birds, PB-700TM Jartester Machine (Catalog No. 7790 – 752B, Serial No. 198070026 269900) until the equilibrium of residual fluoride concentration was attained. Table 2 gives the residual fluoride concentration in each flask at equilibrium.

Table 2

Residual fluoride values at equilibrium

Flask no. Mass of BC (g) Volume in flask (mL) Initial fluoride concentration (mg/L) Residual fluoride at equilibrium, Ce (mg/L) 
500 21 15 
500 21 3.5 
500 21 2.2 
10 500 21 1.3 
500 21 21 
Flask no. Mass of BC (g) Volume in flask (mL) Initial fluoride concentration (mg/L) Residual fluoride at equilibrium, Ce (mg/L) 
500 21 15 
500 21 3.5 
500 21 2.2 
10 500 21 1.3 
500 21 21 

The amount of solute adsorbed at equilibrium was determined using Equations (2) and (3) as tabulated in Table 3. 
formula
2
 
formula
3
where Ci is the initial fluoride concentration (mg/L), Ce is the fluoride concentration at the equilibrium (mg/L), m is the mass of BC used (g), V is the volume of fluoridated water used (mL) and qe is the equilibrium amount of fluoride adsorbed on BC (mg/g).
Table 3

Computation of equilibrium amount of fluoride adsorbed on BC

Flask no. Mass of adsorbate (g) qe (mg/g) 
1.75 
1.34 
10 0.99 
Flask no. Mass of adsorbate (g) qe (mg/g) 
1.75 
1.34 
10 0.99 
The Langmuir isotherm (Mutchimadilok et al. 2014) is given by 
formula
4
where qe is the equilibrium amount of fluoride adsorbed on BC (mg/g), Ce is the equilibrium fluoride concentration (mg/L), Q is the Langmuir isotherm constant related to capacity (mg/g), and b is the Langmuir isotherm constant related to energy (L/mg).
To determine the Langmuir adsorption parameters, the linear form transformed equation was used: 
formula
5
or 
formula
6

Using the values in Tables 2 and 3, the Langmuir equation parameters in Table 4 were calculated.

Table 4

Computed Langmuir equation parameters

Flask no. 1/Ce 1/qe 
0.06667 0.33333 
0.28571 0.57143 
0.45455 0.74468 
0.76923 1.01523 
Flask no. 1/Ce 1/qe 
0.06667 0.33333 
0.28571 0.57143 
0.45455 0.74468 
0.76923 1.01523 

The Freundlich isotherm model is assumed to be associated with heterogeneous sites (Sarkar et al. 2006) and is used for fitting results of sorption experiment. The equation is as follows (Jayapriya et al. 2011): 
formula
7
where qe is the amount of fluoride ions adsorbed per unit weight of adsorbent (mg/g), Kd and 1/n are the Freundlich constants and Ce is the equilibrium fluoride concentration (mg/L).

Kishore et al. (2011) described this equation as follows: if < 1, the bond energies increase with the surface density of the adsorbent, and if > 1 the bond energy decreases with the surface density of the adsorbent, and if = 1 all surface sites are equivalent related to sorption capacity and sorption intensity.

The values of Ce and qe from Tables 2 and 3 were used to compute Freundlich isotherm parameters as tabulated in Table 5. 
formula
8
Table 5

Computed Freundlich adsorption isotherm parameters

Flask no. Ce (mg/L) Log Ce qe (mg/g) Log qe 
15 1.17609 0.47712 
3.5 0.54407 1.75 0.24304 
2.2 0.34242 1.34 0.12803 
1.3 0.11394 0.99 −0.00656 
Flask no. Ce (mg/L) Log Ce qe (mg/g) Log qe 
15 1.17609 0.47712 
3.5 0.54407 1.75 0.24304 
2.2 0.34242 1.34 0.12803 
1.3 0.11394 0.99 −0.00656 

RESULTS AND DISCUSSION

Effect of particle size and temperature on fluoride removal

The effect of particle size on fluoride removal for BC charred at 400°C and 500°C are shown in Figure 2. Reduced mass transfer effects to the adsorption sites in pores and on the particle surface is obvious as the particle size increases. Fluoride adsorption with particle sizes above 1,000 μm indicates the process to be mass transfer limited. The results are similar for all BC calcination temperatures of 400°C, 500°C and 600°C. Studies by Mavura et al. (2004) and Thole et al. (2012) also gave similar trends. In this study the effect of particle size was also studied by batch method with an initial fluoride concentration of 10.2 mg/L. The findings showed that the smallest particle sizes, that is 0.1–0.15 mm, had better adsorption ability than the rest. But particle sizes below 0.25 mm were found to clog the column, therefore it was found not suitable for use in a column applications. It was hence decided to work on particle size (dp) in the range of 0.25 mm < dp < 1.8 mm. For which, the smallest particle size suitable for column defluoridation adsorption was recommended to be 0.25–0.5 mm.
Figure 2

Effect of particle size and calcination temperature on fluoride removal.

Figure 2

Effect of particle size and calcination temperature on fluoride removal.

The effect of calcination temperature on the BC adsorption capacity was observed by looking at the residual fluoride concentration as a function of time. BC calcined at 400°C could be used for a longer time than one calcined at 500°C before reaching a maximum residual fluoride concentration of either 1.5 mg/L or 4 mg/L as per the World Health Organization standard and Tanzania Bureau of Standards, respectively.

The decrease in adsorption removal performance as the calcination temperature was increased from 400°C to 500°C and 600°C can be explained by the results of Figure 3 which show decreased pore volume and surface area of the BC as the calcination temperature was increased. Increase in calcination temperature led to sintering, hence, a decrease in pore volume and surface area. The resultant effect was a decrease in mobility of solution in the column and a decrease adsorption capacity of the adsorbent. The sintering effect is observed in the results of the SEM in Figure 4, which shows that, at low temperature (400°C) the surface of the BC is rough compared to that at higher temperature (600°C), which is smoother.
Figure 3

Effect of calcination temperature on BC pore volume and surface area.

Figure 3

Effect of calcination temperature on BC pore volume and surface area.

Figure 4

SEM analysis of BC calcined at different temperatures.

Figure 4

SEM analysis of BC calcined at different temperatures.

BC removal capacity

The effect of BC calcination temperature on fluoride removal rate capacities is shown in Figure 5. This study was done using 250 g of the media with particle grain size of 250–500 μm. It can be seen that the BC calcined at 400°C had the best removal rate capacity compared with those calcined at 500°C and 600°C. Thus, the calcination temperature has an effect on the developed structural morphology and temperatures above 400°C result in loss of fluoride adsorption performance of the BC because of the sintering effect.
Figure 5

Effect of temperature on fluoride removal capacities with time.

Figure 5

Effect of temperature on fluoride removal capacities with time.

The effect of particle size on the fluoride removal capacity for BC calcined at different temperatures were determined using Equation (1), at fluoride removal breakthrough point of 1.5 mg/L as shown in Figure 6. Equation (1) indicates the amount of fluoride in milligrams removed by one gram of BC used with the initial fluoride concentration being 8.55 mg/L. The results show that, BC with a particle size of 250–500 μm calcined at 400°C had the best fluoride removal capacity. Similar findings have been reported by Mjengera (1988) and Hauge et al. (1994), who reported that fluoride removal by BC is a surface reaction process and that as the surface area is increased, the effectiveness of the BC in removing fluoride also increases. This is clearly shown by the higher removal capacity with the BC particle range of 250–500 μm compared with the other, larger particles. Figure 6 also shows a decrease in fluoride removal capacity as the calcination temperature was increased to 500°C and 600°C as a result of a decrease in pore volume and surface area.
Figure 6

Fluoride removal capacities of BC at 1.5 mg/L breakthrough point.

Figure 6

Fluoride removal capacities of BC at 1.5 mg/L breakthrough point.

BC equilibrium adsorption isotherm: determination of the Langmuir isotherm equation parameters

The Langmuir adsorption isotherm describes quantitatively the formation of a monolayer adsorbate on the outer surface of the adsorbent. The transformed linear Langmuir isotherm, Equation (6), was plotted using values in Table 4. A plot of 1/qe against 1/Ce (Equation (6)) should yield a straight line if the data fit the model. The results gave a linear relationship as shown in Figure 7(a), with R2 value of 0.994. The maximum monolayer coverage capacity, Q, and Langmuir isotherm constant related to energy, b, in Equation (6) were obtained from the intercept and slope, respectively. From this research work, the values of the constants Q and b calculated from the intercept and slope were 3.512 mg/g and 0.294 L/mg, respectively.
Figure 7

Langmuir and Freundlich isotherm curves for BC fluoride adsorption.

Figure 7

Langmuir and Freundlich isotherm curves for BC fluoride adsorption.

The essential features of the Langmuir isotherm are described by a dimensionless quantity, r, referred to as separation factor (Prasad et al. 2000), which indicates the isotherm's shape and the nature of the adsorption process as unfavorable, favorable, linear or irreversible. The dimensionless quantity is expressed as 
formula
9
where C0 is the initial fluoride concentration.

The feasibility criteria of the adsorption process were judged from the r-values as shown in Table 6 (Sarkar et al. 2006; Hameed et al. 2008).

Table 6

The feasibility criteria of r-values

Value Description 
r > 1 Unfavorable 
r = 0 Irreversible 
0 < r < 1 Favorable 
r = 1 Linear 
Value Description 
r > 1 Unfavorable 
r = 0 Irreversible 
0 < r < 1 Favorable 
r = 1 Linear 

The value of r obtained was 0.139, indicating that the adsorption process was favorable (0 < r < 1) for the adsorbent in the removal of fluoride ions, and the R2 value of 0.994 proves that the adsorption data fitted well to the Langmuir isotherm model.

Freundlich isotherm

The Freundlich isotherm describes the adsorption characteristics for the heterogeneous surface. It is an empirical equation that encompasses the heterogeneity of sites and the exponential distribution of sites and their energies (Oladoja et al. 2008). Fitting of the computed Freundlich isotherm parameters in Table 5 using the linear form of the Freundlich isotherm, Equation (8), gave a correlation as shown in Figure 7(b). The constant, Kd, is an approximate indicator of adsorption capacity, while 1/n is a function of the strength of adsorption in the adsorption process (Dada et al. 2012). The plot of log qe against log Ce yielded a linear relationship as expected, with slope 1/n and the intercept log Kd of values 0.446 and −0.032, respectively.

Kishore et al. (2011) reported that, if 1/n < 1, bond energies between sorbent and sorbate increase with the surface density of the adsorbent; if 1/n > 1, bond energies decrease with the surface density of the adsorbent, and if 1/n = 1 all surface sites are equivalent. Dada et al. (2012) said that if the value of 1/n is below 1 it indicates a normal and favorable adsorption process. In this work, the plot of log(Ce) versus log(qe) was linear with R2 = 0.976, as shown in Figure 7(b), with 1/n = 0.445 confirming the applicability of the Freundlich isotherm model for the removal of fluoride ions by the BC calcined at 400°C.

The isotherm parameters evaluated from BC calcined at a temperature of 400°C for Langmuir and Freundlich equations are summarized in Table 7.

Table 7

Isotherm parameters

  Langmuir
 
Freundlich
 
Calcination temperature (°C) Q (mg/g) b (L/mg) 1/n Kd (L/g) 
400 3.512 0.294 0.139 0.446 0.929 
  Langmuir
 
Freundlich
 
Calcination temperature (°C) Q (mg/g) b (L/mg) 1/n Kd (L/g) 
400 3.512 0.294 0.139 0.446 0.929 

CONCLUSION

The SEM of the calcined BC showed an increased sintering effect as the calcination temperature was increased from 400°C to 500°C and 600°C. This decreases the adsorption capacity of the adsorbent and hence reduces its adsorption performance as calcination temperature was increased.

Investigation of equilibrium adsorption carried out indicated that for the Langmuir adsorption model, the separation factor or equilibrium parameter (r) value calculated was greater than 0 and less than 1, indicating that the Langmuir adsorption isotherm is favorable, while for the Freundlich model the value of 1/n indicates the strength of adsorption is less than 1, indicating a normal and favorable adsorption process. From this result, it can be concluded that the process of adsorption fitted the Langmuir model (R2 = 0.994) better than the Freundlich model (R2 = 0.976).

In this work, therefore, favorable preparation conditions for calcination of BC and particle sizing for fluoride adsorption were: calcination temperature of 400°C and a particle sizes range of 250–500 μm for the removal of fluoride from naturally fluoridated water. BC calcined at temperatures of 500°C and 600°C appeared to have less pore volume and reduced surface area. Particle sizes below 250 μm led to column clogging, while particles above 1,000 μm led to limited mass transfer.

ACKNOWLEDGEMENTS

The author acknowledges the financial support received from the Royal Society of London through Leverhulme Royal Society Africa Award Scheme. The sponsors did not have any involvement in the research design, execution and data analysis nor in the preparation and submission of this paper.

REFERENCES

REFERENCES
APHA
1989
Standard Methods for the Examination of Water and Wastewater
,
17th edn
.
American Public Health Association
,
Washington, DC
.
Bonilla-Petriciolet
A.
Rojas-Mayorga
C. K.
Aguayo-villarreal
I. A.
Hernández-Montoya
V.
Moreno-Virgen
M. R.
Tovar-Gómez
R.
Montes-Morán
M. A.
2013
Optimization of pyrolysis conditions and adsorption properties of bone char for fluoride removal from water
.
Journal of Analytical and Applied Pyrolysis
104
,
10
18
.
Dada
A. O.
Olalekan
A. P.
Olatunya
A. M.
Dada
O.
2012
Langmuir, Freundlich, Temkin and Dubinin-Radushkevich isotherms studies of equilibrium sorption of Zn2+ unto phosphoric acid modified rice husk
.
Journal of Applied Chemistry
3
,
38
45
.
Dahi
E.
2000
The state of the arts of small community defluoridation of drinking water
. In:
3rd International workshop on fluorosis and defluoridation of water
,
Chiang Mai
,
Thailand
, pp.
141
170
.
HACH
1997
Water analysis handbook
.
Loveland
,
Colorado
.
Hauge
S.
Osterberg
R.
Bjorvatn
K.
Selvig
K.
1994
Defluoridation of Drinking Water by the use of Pottery: Effect of Firing Temperature
.
Unpublished report
,
Artadveien, Bergen
,
Norway
.
Jacobsen
P.
2007
CDN'S Experiences in Producing Bone Char. Catholic Diocese of Nakuru report
,
Nakuru
,
Kenya
.
Jayapriya
G.
Ramya
R.
Rathinam
X.
Sudha
P. N.
2011
Equilibrium and kinetic studies of fluoride adsorption by chitin/cellulose composite
.
Arch. Appl. Sci. Res.
3
,
415
423
.
Kishore
M.
Hanumantharao
Y.
Ravindhranath
K.
2011
Preparation and development of adsorbent carbon from acacia farnesiana for defluoridation
.
Int. J. Plant, Animal Env. Sci.
1
(
3
),
209
223
.
Mavura
W. J.
Mwanyika
F. T.
Wrensford
G.
2004
Construction and optimization of a cartridge filter for removing fluoride in water
.
Bull. Chem. Soc. Ethiop.
18
,
1
6
.
Mjengera
H.
1988
Excess fluoride in potable water in Tanzania and defluoridation technology with emphasize on the use of polyaluminium chloride and magnesite
.
Unpublished report
,
University of Tampere
,
Finland
.
Mutchimadilok
Y.
Smittakorn
S.
Mongkolnchai-arunya
S.
Durnford
D.
2014
Defluoridation with locally produced Thai bone char
.
Adv. Env. Chem.
2014, article ID 483609, 9 pp
.
Oladoja
N. A.
Aboluwoye
C. O.
Oladimeji
Y. B.
2008
Kinetics and isotherm studies on methylene blue adsorption onto ground palm kernel coat
.
Turkish J. Eng. Environ. Sci.
32
,
303
312
.
Posner
A. S.
1987
Bone mineral and mineralization process
. In:
Bone and Mineral Research
,
Vol. 5, Ch. 2
(
Peck
W. A.
, ed.).
Elsevier Science
,
The Netherlands
.
Prasad
M.
Amritphale
S. S.
Saxena
S.
Chandra
N.
2000
Kinetics and isotherms for aqueous lead adsorption by natural minerals
.
Ind. Eng. Chem. Res.
39
,
3034
.
Said
M.
Machunda
R. L.
2014
Defluoridation of water supplies using coconut shells activated carbon: batch studies
.
International Journal of Science and Research (IJSR)
3
,
2327
2731
.
Sarkar
M.
Banerjee
A.
Pratim Pramanick
P.
2006
Kinetics and mechanism of fluoride removal using laterite
.
Ind. Eng. Chem. Res.
45
,
5920
5927
.
Thole
B.
Mtalo
F.
Masamba
W.
2012
Effect of particle size on loading capacity and water quality in water defluoridation with 200°C calcined bauxite, gypsum, magnesite and their composite filter
.
African Journal of Pure and Applied Chemistry
6
,
26
34
.