17

CHAPTER 2

REVIEW OF LITERATURE

2.1 GENERAL

Fluoride ion in water exhibits unique properties, as its concentration

in optimum dose in drinking water is advantageous to health and excess

concentration beyond the prescribed limits affects the health (Vekata Mohan

et al 1995). High fluoride concentration in the ground water and surface water

in many parts of the world is a problem, to be paid more attention. High

fluoride in drinking water was reported from different geographical regions.

The removal of the excess fluoride from waters and wastewaters is

important in terms of protection of public health and environment.

Defluoridation is the only practicable way to overcome the problem of

excessive fluoride in drinking water.

Several adsorbents were tried for the removal of fluoride from water

and wastewater and are categorized in the following section.

2.2 DEFLUORIDATION OF WATER USING NATURALLY

AVAILABLE WASTE MATERIAL

2.2.1 Corn Cobs

The adsorption of fluoride on corn cobs powder was investigated by

Parmar et al (2006). Powdered corn cobs did not show remarkable adsorption

but aluminum treated corn cobs had good adsorption capacity. The parameters

studied include the contact time, concentration, temperature and pH. Near

18

neutral pH was identified as the optimum condition of the medium, and

90 to 120 minutes was the best contact time for maximum fluoride adsorption.

The adsorption process was found to follow Freundlich isotherm. The study

showed that Ca-treated corn cobs powder was found to be more effective as

an adsorbent compared to Al-treated corn cobs powder. The rate of uptake of

fluoride was very high with both the materials. The fluoride containing corn cobs

powder can be dumped as a solid waste material in pits. The raw materials

employed for the preparation of the substrate were cheap and easily available.

2.2.2 Tamarind Seed

Murugan and Subramanian (2006) investigated tamarind seed,

a household waste from the kitchen, for the sorptive removal of fluoride from

synthetic aqueous solution as well as from field water supplies. Batch sorptive

defluoridation was conducted under variable experimental conditions such as

pH, agitation time, initial fluoride concentration, particle size and sorbent

dose. Maximum defluoridation was achieved at pH 7.0. Defluoridation

capacity decreased with increase in temperature and particle size and the

adsorption followed first order kinetics and Langmuir adsorption isotherm.

Desorption was carried out with 0.1N HCl and it was found to be 90 %. The

surface and sorption characteristics were analyzed using FTIR and SEM

techniques. For domestic and industrial applications, defluoridation with

100 % achievement and subsequent regeneration of adsorbent was performed

with a household water filter and fixed bed column respectively.

2.3 CARBON BASED ADSORBENTS FOR FLUORIDE

REMOVAL

2.3.1 Commercial Activated Carbon

Batch adsorption studies were undertaken to assess the suitability of

commercially available activated charcoal (Tembhurkar and Dongre 2006) to

19

remediate fluoride- contaminated water. The effects of some of the major

parameters of adsorption, viz. pH, and dose of adsorbent, rate of stirring,

contact time and initial adsorbate concentration on fluoride removal efficiency

were studied and optimized. Based on these studies, it was concluded that

activated charcoal could be fruitfully utilized for the removal of fluoride.

The uptake of fluoride ions was possible between pH of 2.0 and 8.0, however,

pH of 2 gave maximum fluoride removal since neutralization of OH- ions by

large number of H+ ions takes place at less pH values. The percentage of

fluoride removal was found to be a function of adsorbent dose and contact

time at a given initial solute concentration. The removal increased with time

and adsorbent dose. The study on defluoridation using activated carbon

revealed that the equilibrium data fitted better to Langmuir isotherm than

Frendulich isotherm.

Emmanuel et al (2008) prepared commercial activated carbon

(CAC) and indigenously prepared activated carbons (IPACs) from

Pithacelobium dulce, Ipomoea batatas and Peltophorum ferrugineum for the

removal of fluoride. The effects of various experimental parameters like pH,

dose of the adsorbent, adsorbate concentration and contact time have been

investigated using a batch adsorption technique. The results of the

experiments have shown that the percentage of fluoride removal has increased

with the increase in contact time and dose of adsorbent. On the contrary the

percentage of removal has decreased with the increase in initial concentration

of the standard fluoride solution. The adsorption process was found to be of

first order with the intra particle diffusion as one of the rate determining steps.

Among the adsorbents considered PLDC (pithacelobium dulce carbon)

possessed the highest or the maximum adsorption capacity. The adsorption

capacity and efficacy in the removal of fluoride were far greater than CAC.

20

2.3.2 Alum-impregnated Carbons

Most of the carbon from different carbonaceous sources showed

fluoride removal capacity after alum impregnation. High fluoride removal

capacities of various types of activated carbon have been reported

(Venkataraman 1960). Activated carbon prepared from cotton waste, coffee

waste, coconut waste, paddy husk, corn cob, neem leaf etc. was tried for

defluoridation. All these materials proved to be of academic interest only

(Bulusu et al 1979).

The activated carbon prepared by carbonization of coconut shell

(CSC) in the presence of sulphuric acid was used for the removal of fluoride

by Arulanantham et al (1992). The developed carbon after impregnation with

Al (III) ions when used in wet condition showed a capacity for fluoride

removal three times higher than the same material used after drying.

The studies clearly showed that the coconut shell after alum impregnation was

effective for the removal of fluoride from dilute aqueous solutions.

The optimum pH for fluoride removal was wide 5.0-8.0 for CSC as against

6.0-7.0 for alumina. The regeneration of CSC with aluminum sulphate

solution was found to be simple and problems associated with alumina were

not encountered. The CSC could be prepared from waste coconut shell

material with good bulk density comparable to that of coal based commercial

activated carbons. In addition this carbon had good attritional characteristics.

2.3.3 Graphite

Batch adsorption system using various grades of graphite as

adsorbents was investigated (Karthikeyan and Elango 2008) to remove

fluoride ions from aqueous solutions. The system variables studied include

initial concentration of the sorbate, agitation time, adsorbent dose, pH, co-ions

and temperature. Lower range of pH and high temperature ranges were found

21

as the optimum conditions for maximum fluoride adsorption by the

adsorbents. The result gained from this study was well described by the

Langmuir and Freundlich isotherms. On the basis of the kinetic and XRD

studies a mechanism in which surface adsorption as well as intra-particle

diffusion as rate limiting step has been proposed for the physisorption of

fluoride ions onto graphite.

2.3.4 Waste Carbon Slurry

Waste carbon slurry obtained from fuel oil based generators of

fertilizer industry was investigated for adsorption of fluoride by Gupta et al

(2007). They investigated the effect of contact time, pH, temperature and

adsorbent dose on the extent of adsorption by carbon slurry. The contact time

and pH for maximum fluoride uptake were found to be 1 h and 7.58

respectively. The experimental isotherm data were found to follow Langmuir

equation more closely. From kinetic analysis, the adsorption was found

to follow second-order mechanism. The performance of adsorbent was

excellent as it removed fluoride even at low concentration, i.e. < 2 mg L-1

.

The practical utility of the column was tested by drawing breakthrough curves

for fluoride adsorption from wastewater using carbon slurry, which showed

an implemental breakthrough capacity of 4.155 mg L-1

.

2.3.5 Carbon Nanotubes

Exploring the application of carbon nanotubes to adsorbing fluoride,

a team led by Li et al (2003a) prepared aligned carbon nanotubes (ACNT), by

the decomposition of xylene, catalysed by ferrocene. The authors found this

material to adsorb 4.5 mg g-1

fluoride from 15 mg L-1

fluoride at

pH 7. The adsorption isotherms generated under identical conditions for

activated carbon, -Al2O3, a typical soil and carbon nanotubes showed that the

order of adsorption was: carbon nanotubes > soil > -Al2O3 > activated carbon.

22

2.3.6 Carbonaceous Materials From Pyrolysis of Sewage Sludge

Mendoza et al (2011) investigated the removal of fluoride ions by

a carbonaceous material obtained from pyrolysis of sewage sludge. Effect of

contact time, pH, and fluoride concentration were evaluated. Equilibrium was

reached after 18 h of contact time and the maximum removal obtained at

pH 7.06 ± 0.08, which corresponds to the zero charge point of the adsorbent.

The maximum removal of 82.2 ± 0.5% of fluoride was found with 0.4 g L-1

and with 20 g L-1

.The kinetic data of the process could be fitted to the pseudo

second order and the intraparticle mass transfer diffusion models. The study

on defluoridation using carbonaceous material revealed that the equilibrium

data fitted to both Langmuir and Frendulich isotherms.

2.4 ION EXCHANGE RESINS AS ADSORBENTS FOR

REMOVAL OF FLUORIDE

Fluoride ion can be removed from water supplied with a strongly

basic anion-exchange resin containing quarternary ammonium fuctional

groups. The removal takes place according to the following reaction:

Matrix – N R3Cl-

+ F-

Matrix – N R3F- + Cl

-(2.1)

The fluoride ions replace the chloride ions of the resin. This process

continues until all the sites on the resin are occupied. The resin is then back

washed with water that is supersaturated with dissolved sodium chloride salt.

New chloride ions then replace the fluoride ions leading to recharge of the

resin and starting the process again. The driving force for the replacement of

chloride ions from the resin is the stronger electronegativity of the fluoride

ions.

23

Sundaram and Meenakshi (2009) developed organic-inorganic

hybrid type ion exchangers for fluoride removal. Synthesized hybrids were

characterized using FTIR studies. Batch adsorption studies were performed as

a function of contact time, pH and influence of other interfering anions.

All the ion exchangers possessed appreciable defluoridation capacity. The

values of defluoridation capacity of polyacrylamide Al (III) phosphate

(Al-Ex), polyacrylamide Ce(IV) phosphate (Ce-Ex), polyacrylamide Zr (IV)

phosphate(Zr-Ex) were found to be 2144, 2290 and 2166 mg F-

kg-1

,

respectively. Ce-Ex had slightly higher defluoridation capacity than Al-Ex

and Zr-Ex. The defluoridation capacity of sorbents was significantly

influenced by the pH of the medium and was altered in the presence of

bicarbonate ions. The adsorption followed Freundlich isotherm. The nature of

adsorption of all the exchangers studied was spontaneous and endothermic.

The hybrid type sorbents had more scope for field applicability because of

their usable form.

The defluoridation capacity of a chelating resin, namely

Indion FR 10 (IND), and Ceralite IRA 400 (CER), an anion-exchange resin,

were compared under various equilibrating conditions for the identification of

selective sorbent (Meenakshi and Viswanathan 2007). The results showed that

chelating resin is more selective than an anion-exchange resin for fluoride

removal. The surface morphology of resins before and after fluoride sorption

was observed using scanning electron microscopy (SEM). Fourier transform

infrared spectroscopy (FTIR) was used for the determination of functional

groups responsible for fluoride sorption. The sorption process was found to be

controlled by pseudo-second-order. The results of analysis of both the resins

clearly established that the sorption process was spontaneous and exothermic

and the equilibrium data agreed well with both Freundlich and Langmuir

isotherms. The pseudo-second-order kinetic reaction model was found to be

the best correlation of the data for fluoride removal on IND. It was found

24

to be an effective sorbent for fluoride removal than CER resin because of its

specificity towards fluoride. It could be concluded that though fluoride is an

anion it cannot be effectively exchanged using anion-exchange resins, as

its concentration is lower than the other anions present in water. Hence,

as an alternative, cation exchange / chelating type resins may be employed for

selective sorption of fluoride, which removes fluoride by an adsorption /

precipitation mechanism.

2.5 MINERAL BASED ADSORBENTS FOR FLUORIDE

REMOVAL

2.5.1 Low Cost Natural Minerals

Fan et al (2003) investigated the adsorption kinetics and adsorption

capacity of low cost materials at a low initial fluoride concentration.

The experiments were carried out at a neutral pH, and radioisotope18

F rather

than19

F was used since18

F can be rapidly measured by measuring the

radioactivity with a resolution of 1 X 10-13

mg or 0.01 µCi. The tested

materials are hydroxyapatite, fluorospar, calcite, quartz and quartz activated

by ferric ions. Their adsorption capacities followed the order:

Hydroxyapatite > Fluorospar > Quartz activated using ferric ions >

Calcite > Quartz

The uptake of fluoride on hydroxyapatite was an ion-exchange

process and followed the pseudo-first-and second-order equations. Calcite

has been seen as a good adsorbent in fluoride removal and has been patented.

However, the data suggested that its adsorption capacity was only better than

quartz.

Various low-cost materials like kaolinite, bentonite, charfines,

lignite and nirmali seeds were investigated (Srimurali et al 1998) to assess

25

their capacity for removal of fluorides from water by batch adsorption studies.

Studies were also conducted to determine optimum operating system

parameters; such as contact time, pH, dose and size of the adsorbent. From the

experimental investigation, the order of removal of fluoride from the test

solution using low-cost materials was found to be bentonite > charfines >

Kaolinite > lignite > nirmali seeds. At optimum system conditions, charfines

and bentonite exhibited highest removal capacity (around 40%). The study

indicated that removal of fluoride from water depended on the contact time,

pH, and dose of the adsorbent. Removal of fluoride increased with time and

approached more or less constant values. It decreased with the increasing pH

of the test solution. Removal of fluoride increased with the decreasing size of

the sorbent and with increasing sorbent dose. The study also revealed that

chemical pre-treatment of the sorbents (charfines) was not effective in

enhancing the fluoride removal capacity.

2.5.2 Synthetic Hydroxyapatite

Sundaram et al (2008) have reported the advantages of nano-

hydroxyapatite (n-HAp), a cost effective sorbent for fluoride removal, n-HAp

possessed a maximum deflouridation capacity (DC) of 1845 mg F- kg

-1 which

was comparable with that of activated alumina, a defluoridation agent

commonly used in the indigenous defluoridation technology. A new

mechanism of fluoride removal by n-HAp was proposed in which it was

established that this material removes fluoride by both ion-exchange and

adsorption process. There was no significant influence of other co-anions like

chloride, nitrate and sulphate on the DC of n-Hap except bi-carbonate ions.

The adsorption pattern followed both Langmuir and Freundlich isotherms,

but better fitted to Langmuir isotherm model. The rate of reaction followed

pseudo-second-order kinetics.

26

2.5.3 Granular Red Mud

Red mud is a very fine material (particle size of which is generally

below 75µ) and high specific surface area (10 m2

gm-1

) which is produced

during the Bayer process for alumina production (Hind et al 1999). It is the

insoluble product of bauxite digestion with sodium hydroxide at elevated

temperature and pressure. It is mainly composed of iron oxides and has

a variety of elements and mineralogical phases.

The removal of fluoride from aqueous solution using the original

and activated red mud forms has been studied by many researchers

(Lopez et al 1998). The fluoride adsorption capacity of activated form has

been found to be same as original form. The adsorption was highly dependent

on pH. The possibility of removal of fluoride ion by using red mud was

explained on the basis of the chemical nature and specific interaction with

metal oxide surfaces (Yunus et al 2002).

The removal of fluoride from water using granular red mud (GRM)

was investigated by Tor et al (2009) using batch and column adsorption

techniques. The main conclusions were the high capability of removing

fluoride at low concentrations, satisfactory adsorption capacity in batch and

column adsorption and fine reversibility to be regenerated rapidly for four

cycles indicated that GRM could be used in fluoride adsorption.

Batch experiments indicated that the time to attain equilibrium was 6 h and

adsorption followed the pseudo-second-order kinetic model. Maximum

removal of fluoride was achieved at a pH of 4.7. The adsorption of fluoride by

GRM in batch systems could be described by the Freundlich isotherm, and the

adsorption capacity was 0.851 mg g-1

. Higher fluoride sorption capacity

was obtained using column experiments than using batch experiments.

27

2.5.4 Activated Titanium Rich Bauxite

The potential of thermally activated titanium rich bauxite (TRB) for

adsorptive removal of excess fluoride from drinking water was examined by

Das et al (2005). Adsorption with respect to variation of pH, adsorbent dose,

initial fluoride concentration, presence of interfering ions and heat treatment

were investigated by batch equilibrium experiments. The optimum

temperature of thermal activation for maximum adsorption capacity was

found in between 300 and 450º C. Although uptake of fluoride was dependent

on contact time, adsorbent dose, pH, and concentration of adsorbate the

optimum pH for maximum uptake was found in between 5.5 and 6.5. The

maximum adsorption capacity was found to be 3.8 mg g-1

at pH 6 with

adsorbent dose and initial fluoride concentration of 1 g L-1

and 10 mg L-1

respectively. Adsorption of fluoride was fairly rapid in first 10-15 min and

thereafter increased slowly to reach the equilibrium in about 1.0-1.5 h.

The adsorption followed first–order kinetics and the data fitted reasonably

well to Langmuir and Freundlich isotherm models. The adsorption of fluoride

was not greatly affected by the presence of common interfering ions

indicating selectively of the material towards fluoride adsorption. Competitive

effect of interfering ions could be minimized by proper selection of operating

pH and fluoride could be removed to the desired level (<1.5 mg L-1

) from

contaminated water using appropriate dose of activated bauxite.

2.5.5 Plaster of Paris

Batch sorption system using plaster of Paris as an adsorbent was

investigated by Gopal and Elango (2007) to remove fluoride ions from

aqueous solutions. Initial concentration of the sorbate, agitation time,

adsorbent dose, pH, co-ions and temperature were discussed. Wide range of

pH and low temperature ranges were found as the optimum conditions for

maximum fluoride adsorption by the adsorbent. The results gained from this

28

study were extremely well described by the theoretical Freundlich and

Langmuir isotherm. The higher enthalpy change for the adsorption process

and XRD studies indicated that adsorption occurred through chemisorptions.

2.5.6 Clay Based Adsorbents

Fluoride removal using China clay was studied by researchers

(Chaturvedi et al 1988). Low fluoride concentration, high temperature and

acidic pH were factors favouring the adsorption of fluoride. It was concluded

that the alumina constituent of the China clay was responsible for fluoride

adsorption.

Ramdeni et al (2010) developed two types of natural clays for

removal of fluoride in Sahara region. Water supply for people in the Sahara

region was mainly assured by poor quality ground water which had excessive

minerals, hardness and high concentration of fluoride. This lead to many teeth

and bone diseases such as fluorosis. To eliminate the excess fluoride from the

El Oued Souf City water supply located in the South East of the Algerian

Sahara by retention process onto montmorillonite clay using potentiometric

method two types of natural clays were tested. The first one contained a

higher percentage of calcium (AC) and the second one without calcium

(ANC). These adsorbents were activated chemically and thermally with

temperatures ranging between 200 and 500ºC. Montmorillonite without

calcium ANC-Na+ was more efficient for deflouridation of Saharan water

than montmorillonite with higher percentage of calcium ANC-Na+. They

concluded that ANC-Na+ could be a promising alternative sorbent for

defluoridation of water.

2.5.7 Lateritic Ores

Nickel laterites and chromite mine overburden usually contain high

content of iron and small amounts of alumina, chromium, cobalt, nickel,

manganese. Due to their high iron content in the form of goethite, some

29

studies have been reported for fluoride adsorption (Sarkar et al 2006).

Recently Sujana et al (2009b) have compared the fluoride uptake capacity of

various goethite containing geo materials of India. Effect of various

experimental parameters such as time, temperature, pH, adsorbent and

adsorbate concentration has been reported, the kinetic, isothermic and

thermodynamic parameters were evaluated. Groundwater samples were also

tested for fluoride removal.

2.5.8 Magnesium Oxide

The application of Magnesium oxide for defluoridation is not a new

one (Venkateswarlu and Rao 1953, 1954). The mechanism of removal of

fluoride ions from water by magnesium oxide is as follows: Addition

of magnesium oxide to fluoride-bearing water results in the hydration of

magnesium oxide to magnesium hydroxide.

MgO + H2O Mg (OH) 2 (2.2)

The magnesium hydroxide formed in the above reaction combines

with fluoride ions to form practically insoluble magnesium fluoride as:

2NaF + Mg (OH) 2 MgF2 + 2NaOH (2.3)

Precipitation of fluoride ions as insoluble magnesium fluoride

lowers the fluoride ion concentration in water. It has been found that for

a given mass of magnesium oxide, the amount of fluoride retained increases

with concentration of fluoride ions in the spiked water samples. Further, at a

given solution concentration, the amount of fluoride retained by magnesium

oxide decreases in conjunction with calcium oxide (lime).

2.5.9 Lanthanum Oxide

Rare earth mineral based adsorbent viz. lanthanum oxide was

investigated by Rao and Karthikeyan (2011) for defluoridation. Results of

batch experiments indicated about 90 % removal in 30 min from a 4 mg L-1

30

synthetic fluoride solution. The effects of various parameters such as contact

time, pH, initial concentration and sorbent dose on sorbent efficiency were

investigated. Isothermal equilibrium sorption data suggested that sorption

reaction followed the BET isotherm model involving multilayer sorption.

Sorption capacity ranges from 0.5 to 2.5 mg g-1

depending upon the initial

concentration and higher sorption capacities were accomplished at low pH

values. Adsorbent showed negligible desorption of fluoride in distilled water.

Alum was most effective regenerant than HCl and NaOH. Hence, 1 % alum

solution was used for regeneration. Results of cyclic regeneration with alum

indicated that the sorbent could be regenerated for ten cycles without

significant loss of sorption capacity. Up flow column studies demonstrated

the engineering application of lanthanum oxide for removal of fluoride in

continuous flow column studies.

2.5.10 Hematite Modified with Aluminium Hydroxide

Rios and Hernandez (2012) investigated the modification effects of

hematite with aluminium hydroxide on the removal of fluoride ion from water

using batch experiments. The authors concluded that the modified hematite is

an efficient adsorbent, compared to unmodified hematite, for the removal of

fluoride ion. The optimum pH range for maximum adsorption was between

2.34 and 6.26. The maximum adsorption capacity was 116.75 mg g-1

.

The kinetic sorption processes followed Elovich model and the isotherm

conformed to Freundlich and Langmuir models.

2.6 REMOVAL OF EXCESS FLUORIDE FROM WATER USING

ALUMINIUM COMPOUND BASED ADSORBENTS

2.6.1 Waste Residue from Alum Manufacturing Process

The ability of waste residue, generated from alum manufacturing

process, to remove fluoride ion from water has been investigated (Nigusse

et al 2007). The alum sludge a waste material from alum manufacture,

31

containing different metal oxides with a heterogeneous surface, has shown

a superior adsorption capability for fluoride ion. Series of batch adsorption

experiments were carried out to assess parameters that influence the

adsorption process. The factors investigated include the effect of contact time,

adsorbent dose, thermal pretreatment of the adsorbent, neutralization of the

adsorbent, initial fluoride concentration, pH of the solution and effect of

co-existing anions. The results revealed that low cost, locally available

industrial waste material generated from aluminium sulphate manufacturing

process was promising material to remove excess fluoride from water.

Adsorption of fluoride was fairly rapid in first 5 min and thereafter increased

slowly to reach the equilibrium in about 1h. Based on the results obtained

it could be concluded that about 85% of fluoride was removed within the first

5 min at an optimum adsorbent dose of 16 g L-1

in pH range of less than 8 for

initial fluoride concentration of 10 mg L-1

. The adsorption followed

second–order kinetics. The adsorbent's fluoride removal efficiency was

affected significantly with carbonate ion concentrations and little or no effect

by other anions such as phosphates, chlorides, sulphates and nitrates.

2.6.2 Amorphous Fe / Al Mixed Hydroxides

Sujana et al (2009) have reported that the effectiveness of

amorphous iron and aluminum mixed hydroxides in removing fluoride from

aqueous solutions. A series of mixed Fe / Al samples were prepared at room

temperature by co-precipitating Fe and Al mixed salt solutions at pH 7.5.

The compositions (Fe : Al molar ratio) of the oxides were varied as 1:0, 3:1,

2:1, 1:1 and 0:1. Batch adsorption studies for fluoride removal on these

materials showed that the adsorption capacities of the materials were highly

influenced by solution pH, temperature and initial fluoride concentration. The

rate of adsorption was fast and equilibrium was attained within 2 h. The

adsorption followed first-order kinetics. The sample with molar ratio 1 has

shown maximum adsorption capacity of 91.7 mg g-1

. The optimum pH range

32

for fluoride adsorption was found to be 4.0 - 5.0 for the samples 1:0, 3:1 and

2:1, whereas it was in the range of 4.0 - 7.5 for 1:1 and 0:1. The equilibrium

data fitted to both Langmuir and Freundlich isotherm models and showed

high adsorption capacities.

2.7 BONE CHARCOAL AS ADSORBENT FOR FLUORIDE

REMOVAL

The uptake nature of fluoride on bone surfaces is an early method

of investigation of defluoridation of water supplies in 1937 itself (Smith et al

1937). The carbonate radical of the apatite comprising bone

Ca (PO4)6.CaCO3 was exchanged by fluoride ion, form insoluble fluorapatite

on the bone surface (Benefield et al 1982).

Ca (PO4)6.CaCO3 + 2F-

Ca (PO4)6.CaF + CO32-

(2.4)

This process of treatment was further improved by carbonizing bone

at temperature greater than 1100ºC, commonly known as bone char. Bone

char possesses better ion exchange property for fluoride ion as compared

to unprocessed bone. The column filled with bone char was regenerated by

washing it with alkali solution.

Batch adsorption studies were conducted to determine the effects of

parameters such as initial solute concentration and adsorbent dose on fluoride

adsorption by fish bone charcoal (Bhargava et al 2008). The fluoride removal

was dependent on both the dose of adsorbent and the contact time, at any given

initial solute concentration. For any dose of the adsorbent, the fluoride removal

increased with increasing contact times. The rate of increase of fluoride removal

was more significant up to 240 min contact time, while the rate significantly

decreased beyond 240 min till an equilibrium condition was attained in

33

about 540 min. A model was developed to predict the equilibrium fluoride

concentration for any given initial fluoride concentration and the adsorbent dose.

Bone charcoal obtained from thermal activation of bones of goat

was investigated by Bandyopadhyay et al (2009) for effective remediation of

fluoride contaminated water. The crushed dried bones were thermally

activated and sieved to separate the material into discrete size ranges

(2.36 mm to 1.18 mm) used as adsorbent for the purpose. Bone char had good

adsorption capacity for fluoride and as such this adsorbent showed excellent

removal of fluoride from water. The adsorption of fluoride on the surface of

the adsorbent was found to depend on pH of the solution as well as the

concentration of the adsorbate. The adsorption of fluoride at neutral pH

was found to be higher than those at acidic or alkaline range. It was found that

with the increase in initial fluoride concentration up to certain limit, the

percentage removal of fluoride increased with the adsorbent dose of 50 g L-1

,

but the percentage removal of fluoride decreased at same adsorbent dose,

when initial fluoride concentration exceeded beyond the limit. Experimental

equilibrium adsorption data obtained through batch study fitted reasonably

well to both Langmuir and Freundlich isotherm models.

2.8 VARIOUS FORMS OF ALUMINA FOR FLUORIDE

REMOVAL

2.8.1 Gamma Alumina

Aluminium compounds are known to have high potential for

removal of fluoride from water. Gamma alumina, a purest form of alumina

was investigated by Rao and Karthikeyan (2008) to assess its sorptive

removal capacity of fluoride from water employing a synthetic fluoride

solution of 4 mg L-1

. The adsorbent exhibited rapid and high uptake of

fluoride under experimental conditions. The effects of various parameters like

34

contact time, pH, initial fluoride concentration and sorbent dose were

investigated. The fluoride uptake by gamma alumina initially increased with

increase in pH from 3.0 to 4.0 and the removal increased from 84 % to 98 %

and remained fairly constant up to pH 7.0. The results indicated the usefulness

of the sorbent over a wide range of pH that was normally encountered in real

field conditions. The presence of anions, in general, had a negative influence

on sorptive uptake of fluoride by gamma alumina. The order of effect of

anions on uptake of fluoride by gamma alumina in general was carbonates >

bicarbonates > sulphates > chlorides. Results of cyclic regeneration with alum

indicated the potential usefulness of the sorbent for nearly 10 cycles without

significant loss of sorption capacity.

2.8.2 Mesoporous Alumina

Lee et al (2010) prepared two different kinds of Mesoporous

Alumina (MA) samples using aluminium tri-sec-butoxide in the presence of

either cetyltrimethyl ammonium bromide (MA-1) or stearic acid (MA-2) as

structure directing agent, and tested for adsorptive removal of fluoride in

water. Both materials contained a worm-hole–like mesopore structure,

but exhibited different textural properties. The effectiveness of these materials

for removal of fluoride ions in aqueous solution was evaluated in batch

adsorption experiments by measuring adsorption capacities and kinetic

parameters. Measured equilibrium adsorption data were fitted to the Langmuir

model, and kinetics data were fitted to a pseudo-second-order. Mesoporous

alumina demonstrated superior adsorption performances to gamma alumina in

both sorption capacity and initial sorption rates.

2.8.3 Activated Alumina

Activated alumina is regarded as excellent material for fluoride

removal. However, pH and alkalinity were the factors which affect

35

the sorption capacity. The exhausted material could be regenerated by various

treatments. The effect of other ions present in drinking water, like chlorides,

sulphates and carbonates, over the defluoridation efficiency of activated

alumina was minimum,even though the presence of bicarbonate ions showed

considerable influence in the process of defluoridation.

It is a granular, highly porous material consisting essentially of

aluminum trihydrate. It is widely used as a commercial desiccant and in many

gas drying processes. The crystal structure of alumina contains cation lattice

discontinuities giving rise to localized areas of positive charge. This makes

alumina attract various anionic species. Alumina has a high preference for

fluoride compared to other anionic species, and hence is an attractive

adsorbent. It also does not shrink, swell, soften nor disintegrate when

immersed in water. The activated alumina was proposed for the first time for

defluoridation of water for domestic use in the 1930s. Since then, the use of

activated alumina has become a popular defluoridation method. The

maximum adsorption capacity of activated alumina for fluoride was found

to be 3.6 mg F g-1

of alumina (Shrivastava and Vani 2009).

2.8.4 Alum-Impregnated Activated Alumina

The alum-impregnated activated alumina (AIAA) for removal of

fluoride from water through adsorption has been investigated by Tripathy et al

(2006). All the experiments were carried out by batch mode. The effect of

various parameters viz. contact time, pH effect (pH 2-8), adsorbent dose

(0.5-16 g L-1

), and initial fluoride concentration (1-35 mg L-1

) has been

investigated to determine the adsorption capacity of AIAA. The removal of

fluoride increased with increase in pH up to 6.5 then decreased with the

increasing pH. The optimum pH was found to be 6.5, which is suitable for the

potable purpose. Kinetic study showed that removal of fluoride was found

to be very rapid during the initial period, i.e. most of the fluoride was

36

removed during 10-60 min and reached a maximum of 92 % at 3 h.

Alum–impregnated activated alumina could remove fluoride effectively

(up to 0.2 mg L-1

) from water containing 20 mg L-1

fluoride.

2.9 TRI-CALCIUM PHOSPHATE AS ADSORBENT FOR

FLUORIDE REMOVAL

He et al (1995) compared Hydroxyapatite (HAP) and Tricalcium

phosphate (TCP) for fluoride removal. Hydroxyapatite is a natural material

that is strong in fixing fluoride. An X-ray diffraction analysis proved that the

densitometric tracings of tricalcium phosphates (TCP) are similar to those of

HAP. The study examined and compared the efficiency of defluoridation

from drinking water by TCP under various pH level, temperature and contact

time conditions. The defluoridation mechanism has been discussed in detail.

The results showed that there was a close negative correlation between

the defluoridation efficiency of TCP and the pH levels of raw water, positive

correlation between defluoridation efficiency and both the temperature and

the contact time, suggested that the defluoridation mechanism of TCP

could be a complex chemical reaction.

Static defluoridation of high fluoride (10-12 ppm) water by sixteen

different combinations of tricalcium phosphate (TCP), bone char (BC),

hydroxyapatite (HAP), and related substances has been investigated by

He and Cao (1996). The defluoridation efficiency of relatively insoluble

calcium phosphates was, TCP (87.0 %) > HAP (68.0 %) > BC (66.4 %),

when they were used singly in a batch procedure for 24 h under routine

conditions. When optimal amounts of free phosphate were added together

with BC or HAP to high-fluoride water, the removal of fluoride reached

95 %. The bone char (BC)-mono calcium phosphate (MCP) system seemed

to be the best combination for removal of fluoride from drinking water. With

300 mg BC plus 23 mg MCP per 100 mL of water, fluoride was reduced over

37

24 h from 10.4 mg L-1

to 0.6 mg L-1

by co- precipitation in the pH range

6.5-8.5. Addition of calcium did not improve the defluoridation efficiency of

calcium phosphate.

Observations of the gel structure of tri-calcium phosphate during the

process of manufacture suggested the possibility of its use as an adsorbent for

fluorides (Adler et al 1938). Test towers have been used to prove the

effectiveness of granular tri-calcium phosphate for the removal of fluorides

from natural fluoride – bearing water as well as from synthetic waters.

The effect of particle size on capacity as well as the effect of fluoride

concentration, pH and total hardness-fluoride ratio was recorded.

Capacity tests showed tri-calcium phosphate of -20 to +40 mesh sizes to have

approximately twice the efficiency of activated alumina reported by Fink and

Lindsay and investigated by Swope and Hess (1937).

2.10 CHEMICALLY MODIFIED ION EXCHANGER RESINS

FOR FLUORIDE REMOVAL

Indion FR 10 resin had sulphonic acid functional group (H+

form)

possessed appreciable defluoridation capacity and its capacity has been

enhanced by chemical modification into Na+

and Al3+

forms by loading

respective metal ions in H+ forms of resin (Viswanathan and Meenakshi,

2009). The defluoridation capacity of Na+ and Al

3+forms were found to be

445 and 478 mg F-kg

-1, respectively, whereas the defluoridation capacity of

H+ form was 265 mg F

-kg

-1 at 10 mg L

-1 initial fluoride concentration.

The defluoridation capacity of these sorbents was found to be independent of

pH of the medium and unaltered in the presence of co-anions present in the

medium. Aluminium (III) form had higher defluoridation capacity among the

sorbents studied. The mechanism of fluoride removal by the sorbents

is mainly controlled by chemisorptions. The sorption process followed

Freundlich, Langmuir and Redlich-Peterson isotherms. The values of

38

thermodynamic parameters indicated that the fluoride removal was

spontaneous and endothermic in nature. The rate of reaction of all the forms

was controlled by pseudo-second-order and particle diffusion kinetic models.

Field trial results indicated that these sorbents can be effectively used

to remove the fluoride from water. The best eluent for the registration of the

sorbents was identified as 0.1 M HCl.

Indion FR 10 is a commercially available ion exchange resin with

sulphonic acid functionality named as H+ form had appreciable defluoridation

capacity (DC). It has been chemically modified to La3+

, Fe3+

, Ce3+

, and Zr4+

forms by incorporating respective metal ions into the resin in order to know

their fluoride selectivity by measuring the DC of the respective resin

( Viswanathan and Meenakshi, 2008). The maximum DC of these chemically

modified ion exchange resins suggested their higher selectivity towards

fluoride than H+ form which had the DC of only 275 mg F

-kg

-1 at 11 mg L

-1

initial fluoride concentration. It was concluded that all modified resins

possessed higher DC which in turn indicated their affinity to fluoride than the

original resin. The DC of these sorbents was not influenced by pH of the

medium and was slightly influenced in the presence of co-anions except

bicarbonate. The sorption process followed Langmuir isotherm. The kinetics

of the modified resins followed pseudo-second-order.

Solangi et al (2010) described a convenient method for the

modification of Amberlite XAD-4 resin by introducing thio–urea (ATU)

binding sites onto the aromatic rings. The modified (ATU) resin has been

employed for the quantitative sorption of fluoride ions in batch as well as

column experiments. The parameters (i.e. pH, contact time, etc.) were

optimized and desorption of fluoride ions was fulfilled by using 0.01M HCl

solution. It has been noticed that the modified resin had high efficiency for the

removal of fluoride from water at a wide range of pH mainly at pH 7. The

39

resin can be regenerated several times with 0.01M HCl and may be used as an

ion exchange material in filters for the removal of fluoride from drinking

water. The study was extended to evaluate the efficacy of the resin towards

the real samples of drinking water from the Thar Desert of Pakistan with high

fluoride content.

Amberlite XAD-4 has been modified by introducing amino group

onto the aromatic ring for its application in fluoride remediation (Solangi et al

2009). The characteristics of the modified resin were studied by FTIR and

elemental analysis techniques. It has been observed that the modified resin

was efficient for the removal of fluoride ion from aqueous solution at various

pHs particularly at 9 pH. It has also been found that the resin was effective

even in the presence of other anions such as Br-, NO2

-, NO3

-, HCO3

-, SO4

2-

ions. It was found to be a suitable adsorbent and could be applied for the

removal of fluoride ion from the drinking water of Thar Desert.

2.11 AIM AND SCOPE OF THE WORK

Activated alumina is regarded as an excellent and widely used

adsorbent for defluoridation of drinking water in municipal supplies due to its

surface area, adsorption effect, highly porous structure and high degree of

surface reactivity. The presence of common anions presents in drinking water,

like chlorides, sulphates, and carbonates except bicarbonate have little effect

on the defluoridation efficiency of activated alumina. The maximum

adsorption capacity of activated alumina for fluoride was found to be

3.6 mg F g-1

of alumina (Shrivastava and Vani 2009).

Tri-calcium phosphate is another adsorbent that has gel structure

and better characteristics than that of activated alumina. Irrespective of the

pH of the influent water the pH of the treated water will move towards the

neutral value. In addition tri-calcium phosphate is found to have

40

approximately twice the efficiency of activated alumina for fluoride. But

the adsorbent is sparingly used in water treatment for fluoride removal from

drinking water. Both the adsorbents show increased adsorption capacity in the

powder form as the surface area and consequently more number of active sites

are exposed. However, the adsorbents in the powdered form are not suitable

for continuous column operations as they produce pressure drop.

It was decided to agglomerate the above powdered adsorbents into

granular form so that the adsorbents could be profitably used in column

operations without loss in their capacity for fluoride removal. A suitable

method could be agglomeration using of a cheap, readily available and

non-toxic polymer. Polyvinyl acetate could be a better choice as it is soluble

in simple organic solvents but insoluble in aqueous medium.

Apart from the preparation of granular adsorbents from powdered

activated alumina and tri-calcium phosphate by agglomeration it was also

decided to prepare a superior adsorbent for fluoride removal. The better

choice would be to chemically modify readily available granular ion exchange

polymer resin so that its physical characteristics could be retained,

simultaneously improving its capacity, pH tolerance and selectivity towards

fluoride in the presence of common anions.

Objectives:

1. There are no reports available with respect to agglomeration

of adsorbents using polyvinyl acetate.

2. The present study is aimed at preparing agglomerated granular

activated alumina and tri-calcium phosphate and evaluating

the same in fluoride removal from aqueous solution.

41

3. Batch adsorption studies are proposed to optimize

equilibration time, pH conditions and adsorbents dose for

maximum removal of fluoride. Adsorption isotherm studies

are used to characterize the equilibrium between the amount

of adsorbate that accumulated on the adsorbent and the

concentration of the adsorbate and to calculate adsorption

capacity of the adsorbents.

4. The kinetic models namely pseudo-first order and pseudo-

second order equations will be used to test the experimental

data to examine the adsorption kinetics.

5. Column studies will be carried out in teflon column of 70 cm

length and 1.5 cm dia. Optimum flow rate, bed height for

maximum adsorption will be found out. Effect of common

anions on the adsorption capacity of the adsorbents will be

studied.

6. Regeneration studies are aimed to be done to find suitable

regenerant and evaluate the concentration of the regenerant

used for regeneration and number of cycles of operation the

adsorbents could be subjected to.

7. SEM images of the adsorbent will be recorded to examine the

morphology before and after adsorption. FT-IR spectroscopy

analysis will be done on plain and chemically modified resin

to analyse the functional groups that were originally present in

the plain resin and the groups that have been replaced or

newly introduced after chemical modification.

8. The performance of the three adsorbents will be compared to

find out best among the three for applications in defluoridation

techniques.