Groundwater for Sustainable Development 10 (2020) 100339

Contents lists available at ScienceDirect

Groundwater for Sustainable Development

journal homepage: http://www.elsevier.com/locate/gsd

Research paper

Synthesis of FeCl3-activated carbon derived from waste Citrus limetta peels

for removal of fluoride: An eco-friendly approach for the treatment of
groundwater and bio-waste collectively
Adil Siddique a, 1, Ashish Kumar Nayak b, 1, Jiwan Singh a, *
a
Department of Environmental Science, Babasaheb Bhimrao Ambedkar University Lucknow, Uttar Pradesh, 226025, India
b
Department of Civil Engineering, Indian Institute of Technology Kharagpur, West Bengal, 721302, India

A R T I C L E I N F O A B S T R A C T

Keywords: In this study, the iron-impregnated activated carbons derived from waste Citrus limetta peels (AC-CLPs) has been
Citrus limetta peels employed for the first time to remove the fluoride ions from water. The adsorbents were prepared by activating
Defluoridation the selected biomass with FeCl3 and subsequent carbonization at two different temperatures 250 � C and 500 � C,
FeCl3 activation
and the synthesized carbon materials were designated as AC-CLP250 and AC-CLP500, respectively. The synthe­
Activated carbon
sized activated carbons were characterized by particle size distribution, point of zero charge (pHZPC), scanning
Adsorption kinetics and isotherms
electron microscopy (SEM), energy dispersive X-ray (EDX), Fourier transform infrared spectroscopy (FTIR) and
X-ray diffraction (XRD) analyses. The influence of different operating parameters including adsorbent dose,
initial concentration of fluoride, contact time, solution pH and temperature on the sequestration of the fluoride
ions were studied. For both carbon materials, the optimal conditions were observed as pH of 6.6, adsorbent
dosage of 1.0 g L 1 and contact time of 240 min. The Langmuir and Freundlich isotherms were analyzed to
understand the adsorption process on the surface of the adsorbents. Langmuir isotherm model agreed with the
equilibrium data to a greater extent, indicating monolayer adsorption of fluoride ions onto AC-CLPs. The fluoride
uptake capacities were found to be 4.926 and 9.709 mg g 1 by the Langmuir isotherm for AC-CLP250 and AC-
CLP500, respectively. It was found that the adsorption process follows the pseudo-second-order kinetics, sug­
gesting that the defluoridation process is chemisorption. Also, the boundary layer effect and intra-particle
diffusion models were employed to scrutinize the rate-limiting step. Thermodynamic analysis revealed that
the adsorption process was feasible, spontaneous, and endothermic in nature. The removal efficacy of defluor­
idation process was increased with increase in temperature from 25 to 45 � C. These findings provide insights to
explore further the adsorption process for removal of fluoride ions using FeCl3-activated carbons derived from
other agricultural wastes.

1. Introduction involved in discharging of fluoride in surface water and groundwater.

These industries are semiconductor manufacturing factories, pharma­
Fluoride is a very well known toxic contaminate existing in surface ceutical companies, beryllium extraction plants, aluminum smelters,
water and groundwater. It is naturally found in groundwater and is also fertilizer manufacturing, and mining industries (Paudyal et al., 2011).
released into the groundwater from mineral resources such as fluorite The contamination of fluoride and its related diseases has been re­
and weathering of rocks (e.g., shale, basalt, granite). Leaching of fluo­ ported in many countries including Algeria, Benin, Cameroon, Egypt,
ride in groundwater may be possible if groundwater makes contact Ethiopia, Ghana, Ivory Cost, Kenya, Libya, Malawi, Morocco, Nigeria,
regularly with ores, minerals, and rocks; consequently, high fluoride Rwanda, Sierra Leone, South Africa, Sudan, Tanzania, Togo, Tunisia,
concentration may yield in groundwater (Jacks et al., 2005; Vithanage Uganda, Zimbabwe, China, India, Indonesia, Iran, Pakistan, Saudi Ara­
and Bhattacharya, 2015a; Kumar et al., 2016; Raghav and Kumar, 2018; bia, South Korea, Turkey, Yemen, Canada, Mexico, USA, Argentina,
Ali et al., 2019; Kimambo et al., 2019). Various industries are also Brazil, Estonia, Sweden, UK (England and Wales), and Norway (Yadav

* Corresponding author. Department of Environmental Science, Babasaheb Ambedkar University, Lucknow, 226025, India.
E-mail address: [email protected] (J. Singh).
1
Both authors contributed equally in this work and should be regarded as joint first authors.

https://doi.org/10.1016/j.gsd.2020.100339
Received 5 December 2018; Received in revised form 21 January 2020; Accepted 21 January 2020
Available online 25 January 2020
2352-801X/© 2020 Elsevier B.V. All rights reserved.
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

et al., 2013; Kimambo et al., 2019). The problems related to fluoride and value-added sorbents will contribute to the remediation of fluoride-
contamination are more severe in various states of India like Andhra laden wastewater. The peels of C. limetta have high pectin content with
Pradesh, Punjab, Haryana, Rajasthan, Gujarat, Uttar Pradesh, Tamil the abundant presence of carboxyl groups, which is an excellent antidote
Nadu, Karnataka, and Maharashtra (Meenakshi et al., 2008; Arlappa for controlling cholesterol and blood sugar (Baker, 1994). Also,
et al., 2013; Ali et al., 2018, 2019). The concentration of fluoride in C. limetta peels have been found to be excellent adsorbent for heavy
groundwater in a few parts of India is reported to be more than 30 mg metals removal (Schiewer and Balaria, 2009; Suryavanshi and Shukla,
L 1 (Sharma et al., 2011). If the amount of fluoride in drinking water is 2010; Njikam and Schiewer, 2012). However, segregation of spent
more than 1.5 mg L 1, then it causes endemic dental and skeletal fluo­ sorbing materials from liquid solution is a nontrivial task due to their
rosis, and crippling (Vithanage and Bhattacharya, 2015b). This was low specific gravity and tendency to agglomerate. These issues have
attributed to the fact that fluoride has a strong attraction with calcium widely restricted its application in batch and column mode. In this re­
ions occurring in the internal part of teeth and bones (Getachew et al., gard, magnetic separation is a convenient technique in which adsorbent
2015). particles can be separated from a suspension by using an external
Due to the rapid growth and industrial demand in the global popu­ magnetic field (Zhao et al., 2010). This separation technique has
lation has increased the use of groundwater, resulting in the availability recently gained prominence in water treatment processes and is now
of poor water quality and reduction in the groundwater level. Inorganic recognized as a potential technique to address the above issues.
pollutants such as fluoride, heavy metals, etc. have a significant role in In this work, a locally sourced waste biomass, i.e., C. limetta peels
the contamination of groundwater. Fluoride is such a contaminant were processed into fine particles and then treated with FeCl3⋅6H2O
which holds an extreme threat to human beings. It has been assessed that followed by carbonization at two different temperatures 250 � C and 500
approximately 100 million people suffer from a high concentration of �
C, intending to develop novel magnetic bio-waste activated carbons (i.
fluoride worldwide (Essadki et al., 2009). The ingestion of high fluoride e., AC-CLPs). However, to the best of our knowledge, this kind of
contaminated drinking water results in numerous types of diseases such treatment practice has not been invented yet to adsorb fluoride ions
as bone and skeletal fluorosis (Jagtap et al., 2012). Since, the primary from aqueous media. Our preliminary investigation indicated that there
source of fluoride consumption is drinking water, hence, Indian Stan­ is a strong affinity between fluoride and activated carbons derived from
dard (IS 10500) has imposed the permissible value of 1 mg L 1 of C. limetta peels. Thus, this work is devoted to studying in depth the
fluoride in drinking water (Bureau of Indian Standards BIS, 2012) which potentiality of the studied AC-CLPs towards the defluoridation process
is lower than 1.5 mg L 1 as suggested by WHO (World Health Organi­ with respect to different physico-chemical parameters such as adsorbent
zation WHO, 2006). Therefore, fluorosis mitigation has become an dose, initial fluoride concentration, pH, contact time and temperature.
essential area of research and continues to attract more attention. Different models of adsorption kinetic such as pseudo-first-order,
Different techniques have been applied to remove the excess fluoride pseudo-second-order, intra-particle diffusion, and Boyd models were
from polluted water before discharge to the environment. These tech­ employed to reveal the mechanism of fluoride removal by the adsor­
niques include coagulation-precipitation, adsorption, ion-exchange, bents as mentioned earlier. Isotherm models and thermodynamic pa­
reverse osmosis, nano-filtration, electro-dialysis, and electro- rameters were also analyzed by fitting the experimental data.
coagulation (Choong et al., 2007; Jagtap et al., 2012; Vithanage and
Bhattacharya, 2015a; Thakur and Mondal, 2017; Maity et al., 2018; 2. Materials and methods
Kimambo et al., 2019). Among all these techniques, the adsorption
process is a highly encouraging method which typically uses natural or 2.1. Chemicals
synthetic solid adsorbents. Due to its favorable economics, high energy
efficiency, ease of operation, and reuse of the adsorbent used after its All the chemicals used during the research work were of analytical
regeneration, this technique is highly acceptable. grade and used without further purification. This includes anhydrous
Several adsorptive media (non-biomass) have also been used to sodium chloride (NaF), ferric chloride hexahydrate (FeCl3⋅6H2O), so­
remove fluoride from drinking water, these includes activated alumina dium hydroxide (NaOH) and hydrochloric acid (HCl). Ionic strength
(Ghorai and Pant, 2005), kaolinite and bentonite (Srimurali et al., 1998; adjuster (ISA) solution (order code-FL-ISA1S, batch no. 0218) purchased
Meenakshi et al., 2008), hydrated ferric oxide (Dey et al., 2004), acti­ from the Fisher Scientific (a part of Thermo Fisher Scientific Inc.) is also
vated bauxite (Das et al., 2005), light-weight concrete materials (Oguz, used in our research.
2007), and many natural as well as synthetic media (Fan et al., 2003;
Mohapatra et al., 2009). Activated carbon (AC) is reflected to be a 2.2. Preparation of adsorbent
flexible adsorbing material due to its highly microporous structure, large
surface area, and excellent surface reactivity and ample availability of Citrus limetta (sweet lime) peels were procured from a local juice
functional groups. ACs have been applied widely for fluoride removal vendor located nearby areas of BBAU campus, Lucknow (Uttar Pradesh),
from waters by many researchers (Abe et al., 2004; Daifullah et al., India. It was washed several times with distilled water to remove the
2007; Tomar et al., 2014; Mariappan et al., 2015). Apart from this, the impurities and dust particles and subsequently dried in an air oven at
interest is growing to employ the inexpensive and eco-friendly materials 105 � C until a constant weight was achieved. The dried bio-materials
(such as agricultural/food waste and their by-products) for the adsorp­ were then grounded using a domestic mixer grinder and screened out
tive removal of fluoride ions from aqueous effluents due to its ubiqui­ to obtain the size fraction of 10 μm. The resulting product was then
tousness, cost-effective and readily accessible features (Sarkar et al., mixed with 150 mL distilled water and FeCl3⋅6H2O to achieve
2006; Meenakshi et al., 2008). Conversely, the utilization of waste FeCl3⋅6H2O: C. limetta peel powder in the ratio of 1:3 (i.e., 10 g: 30 g).
bio-materials has the benefit of not entailing extreme care and mainte­ The mixture was mechanically stirred at room temperature and 120 rpm
nance as well as solving their disposal issues. However, this biomass is for 4 h, then it was kept overnight. The mixture was separated by
deliberated as waste and waste minimization by bio-waste is a sustain­ peristaltic pump filtration and dehydrated at 105 � C in a hot air oven for
able decision for inventing the efficient fluoride removal technology. 12 h. The completely dehydrated material was then subjected to thermal
Citrus limetta peels are naturally abundant pectinaceous waste bio- treatment in a muffle furnace (ICON Instruments Company, India) at
materials which are generated in large quantities from the fruit juice two different temperature, i.e., 250 � C and 500 � C with a retention time
shops. However, these waste peels have no economic value and are of 2 h. After carbonization, the produced adsorbents were washed with
considered as food/agricultural waste. Thus, in this research work, the double distilled water to remove residual chemical until a neutral pH
emphasis is on the production of adsorbent materials by using this food (6.0–7.0) was achieved. Finally, the samples were dried at 105 � C for 12
waste. This alteration of unwanted waste peels to valuable commodity h and stored in an air-tight container for further defluoridation

2
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

experiments. The activated carbon materials obtained from two 2.4. Characterization of magnetic AC-CLPs
different calcination temperatures - 250 � C and 500 � C were labeled as
AC-CLP250 and AC-CLP500, respectively. The steps of adsorbent prepa­ Zeta nanosizer (Nano-ZS90) was used to analyze the particle size
ration are discussed in Fig. 1. distribution of AC-CLPs after dissolving particles in ethanol. The particle
size of the adsorbent was determined by particle size analyzer. To
2.3. Determination of fluoride concentration in solution determine the point of zero charge (pHZPC) of the adsorbent, the solution
pH was adjusted in a range of 2.0–10.0 (with intervals of ‘2’) with 0.1 M
Fluoride solution was prepared by dissolving 224.0 mg of NaF in HCl and 0.1 M NaOH using a multi-parameter water analyzer (SYS­
1000 mL of distilled water, and a series of diluted samples were then TONIC 371, India) (Singh et al., 2011). The surface morphology of the
prepared such that they have working concentrations of 5, 10, 15, 20 adsorbents was analyzed using SEM (scanning electron microscopy)
and 30 mg L 1. The residual fluoride concentration and pH of the analysis (JSM 6490-LV, JEOL, Japan). Elemental analysis was con­
aqueous solution were determined using ion-selective electrode method ducted using the energy-dispersive X-ray (EDX) analysis. Fourier trans­
according to the standard methods (APHA, 4500-F-C-97) by employing a form infrared (FTIR) spectra were recorded using FTIR
digital ion meter (Eutech Ion 6þ, Thermo Scientific). Spectrophotometer (NICOLETTM 6700, Thermo Scientific, USA) for the
identification of functional groups present on the adsorbent surface

Fig. 1. Steps for preparation of activated carbons from Citrus limetta waste peels.

3
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

using KBr as reference material, and the spectral range were recorded in ethanol and subsequently analyzed through Zeta nanosizer at room
the region 400 to 4000 cm 1. X-ray diffraction patterns of AC-CPs were temperature. Particle size data is plotted as shown in Fig. 2(a). AC-
obtained on a powder X-ray diffractometer (PW 3040/60, PANalytical, CLP250 peak depicts a higher intensity than AC-CLP500 as represented in
Netherlands) and the samples were scanned from 10 to 80� at the rate of the particle size scattering curve. The peak showing the size of particles
2� per min. in the solution was observed in the range of 325–1118 nm for AC-CLP250
and 249–1278 nm for AC-CLP500. More extensive range of particles were
2.5. Adsorption experiments for removal of fluoride by AC-CLPs found in AC-CLP500, it may be explained that at high temperature
biomass was first thermally decomposed into smaller particles, and then
All the adsorption studies were performed in batch mode at 25 � C by some particles were clumped into bigger particles after cooling.
varying the initial fluoride concentration from 5 to 30 mg L 1. In each The point of zero charge (pHZPC) of an adsorbent is an important
experiment, an accurately weighed amount of AC-CLPs was mixed to 25 representative that determines the electrical charge behavior of AC-CLPs
mL of fluoride solution with a known initial concentration taken in 150 surfaces. Fig. 2(b) reveals that the interpretation between pHinitial and
mL capacity of volumetric flask and stirred in a water bath shaker (LI- ΔpH (i.e., pHfinal - pHinitial) are interconnected at pH 2.17 and 4.61,
WBIS-20, Labard Instruchem Pvt. Ltd., India) at a speed of 120 rpm, which are represented as pHZPC values of the activated carbons AC-
allowing equilibrium to be achieved. At the end of the desired contact CLP250 and AC-CLP500, respectively. In case of pH < pHZPC, the adsor­
time, the volumetric flasks were removed from the shaker and filtered bent surface charge is positive whereas when pH > pHZPC, the surface
through Whatman 42 filter paper in order to segregate the solid adsor­ charge is negative. From the results observed in Fig. 2(b), both the
bent particles. The filtrate was then measured by using ion-selective activated carbons seem to have more acid-active groups than basic
electrode method to determine the residual fluoride concentrations. components on their surfaces. This might have occurred due to the
Adsorption study was conducted to define the basic parameters; contact minimization of acidic groups with the increase in carbonization tem­
time (30–300 min), solution pH (4–10), adsorbent dose (0.5–3.0 g L 1), perature, and accordingly, the loss of significant acidic groups such as
temperature (25–45 � C) and initial fluoride concentration (5–30 mg phenolic, carboxylic, or lactonic groups are also enhanced. Moreover,
L 1). The experiments were carried out at optimum pH (pH ~6.6) of the the thermal activation contributes to the liberation of acidic metal ox­
fluoride solution except for the effect of pH variation study. Data ob­ ides on the material surfaces (Tchomgui-Kamga et al., 2010). In contrast,
tained from the tests as mentioned earlier also assisted in investigating the number of basic groups increases with increase in the carbonization
the adsorption isotherm (i.e., Langmuir and Freundlich models) as well temperature. For most of the parts, the synthesized AC-CLPs are char­
as its kinetics mechanism through pseudo-first-order, pseudo-second- acterized as acidic (Fig. 2(b)). The pHZPC study also signifies that a
order, intra-particle diffusion, and Boyd models. Each test was per­ positive charge behavior in AC-CLPs surface will be favorable for the
formed in duplicates in order to diminish the error, and the average adsorptive removal of anionic pollutants when the pH values are lower
values were reported here. The defluoridation efficacy and equilibrium than the pHZPC. From the results of Fig. 7, we found that the pHZPC in­
adsorption capacity were calculated using the following equations: creases with increasing carbonization temperature.
� �
Removalð%Þ ¼
C0 Ce
� 100 (1) 3.1.2. SEM and EDX analyses
C0 Morphological analysis of AC-CLPs was carried out by SEM (Fig. 3).
The SEM images were aimed to distinguish the micrographs obtained
qe mg ​ g 1
�
¼
ðC0 Ce ÞV
(2) from the activation process with two different temperatures 250 � C (i.e.,
m AC-CLP250) and 500 � C (i.e., AC-CLP500). In Fig. 3(a and c), the SEM
images showed an external surface which is quite uneven and formed by
where C0 and Ce are the initial and equilibrium concentrations of fluo­
irregular shapes, which could not be considered as pores but provide
ride ions respectively (mg L 1), V is the volume of the solution (L) and m
channels to the interior of the material. Consequently, it is assumed to be
is the adsorbent mass (g).
saturated by the availability of pores of different sizes. During the
activation process, the chemical activator from the synthesized acti­
3. Result and discussion
vated carbon experiences the evaporation, which further creates the
framework of pores (Uçar et al., 2009). The formation of cavities also
3.1. Characterization of the adsorbents (AC-CLPs)
offered the pathway to the fluoride ions for accessing the pores of the
carbon surface. Moreover, the SEM images of AC-CLPs explore the
3.1.1. Particle size analysis and point of zero charge (pHZPC)
cuboidal/pyramidal shaped crystal structures due to the cluster
The particle size of AC-CLPs was determined by homogenizing with

Fig. 2. (a) Particle size analysis using Zeta nanosizer and (b) surface charge distribution of AC-CLPs.

4
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

Fig. 3. SEM images depicting at different resolutions with an inset of EDX spectrum for (a) and (b) AC-CLP250 and (c) and (d) for AC-CLP500.

formation of Fe3O4 particles (Fig. 3(b and d)). were examined by FTIR spectra (Fig. 4(a)). The FTIR spectrum of AC-
The elemental compositions of prepared AC-CLPs at two different CLP250 reveals the presence of characteristic peaks at 3383, 2902, 1619,
temperatures were examined by EDX analysis. The quantitative mea­ 1442, 1249 and 582 cm 1, whereas in case of AC-CLP500 the predomi­
surements of all the elements present in the sorbent materials are nant peaks at 3422, 3032, 2876, 1618, 1441, 1217, 552 and 468 cm 1
compiled in Table 1. After the thermal activation process, it is under­ were observed. The broad absorption peaks at 3383 cm 1 (in case of AC-
standable that the residual chlorine contents in both the adsorbents (AC- CLP250) and 3422 cm 1 (in case of AC-CLP500) were attributed to the
CLPs) stimulated from the initial chloride salt (i.e., FeCl3). Furthermore, availability of surface groups such as phenol, carboxylic acids and its
it can be inferred that the weight percentage of chloride decreases from derivatives, as well as the presence of physically adsorbed water on the
21.74% to 13.97% with increase in the carbonization temperature from material (Singh et al., 2011). In Fig. 4(a), the peaks at 3011, 2902 and
250 � C to 500 � C. This chloride loss might be corroborated due to the 2876 cm 1 were attributed to the C–H stretching vibrations which
involvement of drying and carbonization steps, most likely as HCl. In further implies that minute quantity of some residual carbon of C. limetta
reality, we are unconvinced about the significance of residual chloride exists in AC-CLPs (Schiewer and Balaria, 2009). The peaks at 1618 and
for the removal of fluoride ions. However, more detailed studies are to 1619 cm 1 were assigned to the C꞊O stretching vibration of quinonic and
be carried out to reveal clearer insight. carboxylate groups (Mondal et al., 2019), and the peaks at 1000-1300
cm 1 corresponding to C–O stretching vibration of lactonic and phenolic
3.1.3. FTIR analysis groups present in the AC-CLP surfaces (Singh and Majumder, 2016).
The functional groups on the surface of AC-CLP250 and AC-CLP500 Next, Saha et al. (2013) reported that the peak at ~1440 cm 1 could be
attributed to symmetric bending of -CH3 groups. The peaks in the
400-600 cm 1 range could be attributed to Fe–O bond vibration of Fe3O4
Table 1
particles (Sun et al., 2006) and conversely, the characteristic absorption
Quantitative elemental distribution in weight % determined from EDX analysis
of AC-CLPs. peak around 630 cm 1 was not observed in the FTIR spectra of the
AC-CLP materials, which again confirmed the non-availability of
Element Weight percentage (%)
γ-Fe2O3 particles (Fu et al., 2008; Namduri and Nasrazadani, 2008).
AC-CLP250 AC-CLP500 Finally, it was perceived that the minor shift in the positions of the
Carbon 25.17 62.55 characteristics peaks after the fluoride adsorption indicate the plausible
Oxygen 12.65 12.65 interaction between the studied active carbons and fluoride ions through
Chlorine 21.74 13.97 -OH, C– – O, C–O and Fe–O functional groups.
Iron 21.55 2.46
Magnesium 0.56 –
Aluminum 8.78 – 3.1.4. XRD analysis
Potassium 0.55 0.30 Fig. 4(b) reveals the XRD patterns for the AC-CLP250 and AC-CLP500
Calcium 0.81 0.54 carbons. In both the cases for the adsorbent, the occurrence of board
Iron 21.55 9.09
peaks at about 24� was the confirmation of active carbons which have
Zirconium 1.75
Silicon – 0.90 amorphous structure. Consequently, it does not provide quantitative
Total 100.00 100.00 peaks to evaluate the crystallization mode of the AC-CLPs. The

5
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

Fig. 4. (a) FTIR spectra and (b) XRD diffractogram of synthesized AC-CLPs.

formation of the cubic spinels was corroborated from the XRD patterns 240 min. The equilibrium percentage removal in AC-CLP250 for fluoride
for magnetite by comparing with JCPDS file 89–43191. The diffraction ions is nearly 61.0%, while in case of AC-CLP500 it is equal to 73.2%.
peaks located at (220), (311), (400), (422), (511) and (440) are recog­ Consequently, AC-CLP500 exhibited higher fluoride removal percentage
nized the Fe3O4 magnetite (Shen et al., 2013). Furthermore, it can be than AC-CLP250 for identical optimized conditions. This is because of the
observed that iron oxide peaks became better delineated and possessed greater availability of internal pores in case of active carbon synthesized
higher intensities when the carbonization temperature is enhanced. at 500 � C, the as-prepared activated carbon was expected to be efficient
adsorbent for fluoride removal.

3.2. Preliminary batch adsorption study

3.2.1. Effect of adsorbent dose on fluoride removal efficacy
Fig. 6 (a and b) illustrate the influence of adsorbent dose on the
To investigate the adsorption potentiality of AC-CLP250 and AC-
percentage removal of fluoride ions by the activated carbons AC-CLP250
CLP500 for fluoride removal, several factors were analyzed. The factors
and AC-CLP500, respectively. The effect of adsorbent loading on the
include adsorbent dose, initial fluoride concentration, contact time,
removal of fluoride ions by AC-CLPs were carried out at four different
solution pH, and temperature. All adsorption experiments were per­
adsorbent dosages of 0.5, 1.0, 2.0 and 3.0 g L 1, while the experimental
formed at pH 6.6, excluding the effect of pH study. Simultaneously, the
conditions such as pH 6.6, temperature 25 � C, initial fluoride concen­
best value of the process parameter described above can be considered
tration 10 mg L 1 remain constant at varying time intervals of 30, 60,
as the optimized condition. The adsorption profile of fluoride removal
120, 180, 240 and 300 min. The removal percentage of fluoride by AC-
on the synthesized AC-CLPs is shown in Fig. 5. The experiment was
CLP250 increases from 20.4% to a maximum sorption rate of 65.0% with
carried out at 25 � C for 240 min by adding 1.0 g L 1 of adsorbent dosage
increasing the adsorbent dose from 0.5 to 1.0 g L 1, while in case of AC-
to 25 mL aqueous solution of fluoride ion having an initial concentration
CLP500 the removal efficacy is increased from 42.1% to 72.9% with
of 10 mg L 1. The adsorption of fluoride ions increased with the increase
increasing the adsorbent dose from 0.5 to 1.0 g L 1. The increase in the
of contact time for both the adsorbents. However, a saturation of
removal efficacy with an increase in adsorbent dose is due to the in­
adsorption rate for fluoride removal takes place when the contact time is
crease in adsorbent mass. Therefore, more number of active sites were
available for the defluoridation process. Next, the substantial decrease in
removal efficiency after the adsorbent dosage of 1.0 g L 1 might occur
due to an increase in particle aggregation experiencing from the
excessive sorbent mass. Consequently, it reduces the total availability of
sorption sites and also enhances the diffusional path length (Shukla
et al., 2002). Similarly defluoridation rate also increased with increase
in contact time up to 240 min, thereafter no significant removal was
observed by escalating the contact time (Fig. 6 (a and b)). It is because
initially a large number of vacant sorption sites were accessible, and
after optimum contact period, sorption rate slowed down due to plau­
sible monolayer formation on the external surface, pores of the sorbent
materials and pore diffusion onto the interior surface of the adsorbent.
The aforesaid observation attained through the adsorbate film layer due
to continuous shaking during the present sorption study (Yadav et al.,
2013). Therefore, the adsorbent dosage of 1.0 g L 1 in 240 min was
selected as optimum for further fluoride removal studies.

3.2.2. Effect of initial fluoride ion concentration and contact time

The combined influence of initial fluoride ion concentration and
Fig. 5. Profile of the defluoridation ability of activated carbons derived at 250 contact time for the maximum removal efficacy of fluoride was studied.
and 500 � C. The adsorbent dosage, solution pH, temperature and shaking speed were

6
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

Fig. 6. Effect of adsorptive parameters on fluoride removal using AC-CLPs: (a) and (b) corresponds to influences of adsorbent dose on removal efficacy by AC-CLP250
and AC-CLP500, respectively (conditions: initial fluoride concentration 10 mg L 1; contact time 0–300 min; volume of fluoride solution 25 mL; temperature 25 � C),
and (c) and (d) reflects to the initial concentration effects on the fluoride removal using AC-CLP250 and AC-CLP500, respectively (conditions: initial fluoride con­
centration 5–30 mg L 1; contact time 0–300 min; volume of fluoride solution 25 mL; temperature 25 � C).

fixed at 1.0 g L 1, 6.6, 25 � C and 120 rpm, respectively. The percentage removal on chitin based adsorbent. Hence, the optimum equilibrium
removal of fluoride was quantified by choosing an initial fluoride ion contact time for subsequent adsorption studies was considered to be 240
concentration from 5 mg L 1 to 30 mg L 1 with varying the contact time min.
from 30 min to 300 min. The percentage of defluoridation rate decreased
with an increase in initial fluoride ion concentration (Fig. 6 (c and d)). 3.2.3. Effect of pH on fluoride removal efficacy
This is due to the fact that the number of active sites on the adsorbent The influence of the solution pH on the defluoridation process is one
surface is not adequate to accommodate the fluoride ions under higher of the important parameter which relates with the active surface sites
concentration of fluoride. In contrast, at lower concentration of fluoride, (protonated or deprotonated) of the adsorbent material as well as the
the relative proportion of surface active sites to total fluoride is much degree of ionization of the adsorbate in the solution (Iriel et al., 2018).
higher, and consequently, fluoride ions can interact with the active sites Fig. 7 shows the effect of pH on the removal percentage of fluoride by
present on the adsorbent surface. The maximum removal efficacy of AC-CLPs. The fluoride adsorption was studied in the pH range of
fluoride is found to be 94.7% and 94.8% by employing the adsorbents 4.0–10.0 by keeping all the optimized parameters (i.e., fluoride ion
AC-CLP250 and AC-CLP500, respectively with an initial concentration of concentration 10 mg L 1, contact times of 240 min, AC-CLPs dose 1.0 g
5 mg L 1, whereas the percentage removal diminished up to 5.0% and L 1, and temperature 25 � C) constant. As seen in Fig. 7, the fluoride
33.3% in case of 30 mg L 1 of initial fluoride concentration, respec­ removal efficiency of AC-CLPs increased significantly by increasing the
tively. Similar observation has been reported for fluoride removal using pH from 4.0 to 6.6, and then decreased when the pH was further
neem charcoal adsorbent (Chakrabarty and Sarma, 2012). Therefore, increased up to 10.0. The maximum fluoride removal efficacy was found
AC-CLP250 and AC-CLP500 are recognized to be effective adsorbents for to be 64.5% and 73.2% at natural pH (~6.6) for AC-CLP250 and
fluoride removal under aqueous environment. AC-CLP500 respectively. Moreover, the decrease in fluoride removal was
Simultaneously, Fig. 6 (c and d) depict the effect of contact time on observed at highly alkaline (pH ~10.0) and acidic (pH ~4.0) conditions,
fluoride removal efficacy of AC-CLPs. It was observed that fluoride which are of no practical significance in real-field engineering applica­
removal gradually enhances with an increase in contact time and finally tions. From the experimental data, the pHZPC values for AC-CLP250 and
reaches the equilibrium at 240 min (i.e., 4 h), after that no significant AC-CLP500 are 2.17 and 4.61, respectively (Fig. 2(b)). The behavior of
change in fluoride removal was observed. The overall defluoridation fluoride adsorption onto AC-CLPs can be described by considering the
process was carried out in three different stages as explored in the point of zero charge (pHZPC) of the adsorbent. When the pH is lower than
aforesaid figures. The initial stage revealed the high sorption uptake of pHZPC, the adsorbent surface was positively charged and when it is
fluoride ions onto the AC-CLPs. The second stage referred the gradual higher than pHZPC, the adsorbent surface is negatively charged. Thus,
uptake of fluoride ions illustrating the utilization of all active sites over the defluoridation efficiency reduced for both the activated carbons
the surface of prepared adsorbents. The third stage signified the satu­ when pH > pHZPC. This is due to electrostatic repulsion between fluoride
ration stage in which the adsorption uptake was relatively constant. ions and negatively charged adsorbent surface, and competition be­
Similar results were reported earlier by Thakre et al. (2015) for fluoride tween fluoride ions and excess of hydroxyl anions. But, high removal

7
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

binding forces on the adsorbent surface at higher temperature (Ma et al.,

2011). According to Elhalil et al. (2016), it can cause a swelling effect
within the internal structure of AC-CLPs as well as increase the mobility
of fluoride ions, in order to penetrate the fluoride ions further.

3.3. Modeling of adsorption isotherms

Adsorption isotherm models were accompanied in order to achieve

the adsorption behavior and adsorption capacity for removing a unit
mass of fluoride by different AC-CLPs. Although a variety of isotherm
models are available in order to explore the mechanism of the adsorp­
tion process. Nevertheless, the Langmuir and Freundlich isotherm
models are the most commonly employed for surface sorption studies
(Hameed and Ahmad, 2009). In the present work, Langmuir and
Freundlich isotherm models were used to elucidate the relationship
between the solid-phase concentration (i.e., qe, mg g 1) and the
liquid-phase concentration (i.e., Ce, mg L 1).
In this regard, AC-CLPs were contacted by increasing the fluoride
Fig. 7. Effect of initial pH on fluoride removal by the AC-CLP sorbents per­ concentrations from 5 mg L 1 to 30 mg L 1 at optimized conditions of
formed in batch study (conditions: initial fluoride concentration 10 mg L 1; the adsorbent dose of 1.0 g L 1, pH of 6.6 and temperature of 25 � C.
contact time 240 min; adsorbent dose 1.0 g L 1; volume of fluoride solution 25
mL, temperature 25 � C). 3.3.1. Langmuir isotherm
The adsorption capacities of the AC-CLP250 and AC-CLP500 corre­
efficiency of fluoride at pH < pHZPC is due to electrostatic attraction sponding to monolayer adsorption have been estimated by the Langmuir
(Chen et al., 2011). The reduction in fluoride removal under acidic model. This model is indicative of saturation of binding sites of the
conditions (4.0–6.0) was not only due to the formation of weak hydro­ adsorbent with negligible interaction among the adsorbed species.
fluoric acid, but also for the loss of iron oxide from AC-CLPs (Asgari Saturated monolayer isotherm can be represented in Eq. (3) (Langmuir,
et al., 2012). Hence, the optimum pH value has been found to be 6.6, and 1918):
this pH value was used for the following research. Similar pH profiles
Ce Ce 1
have been reported in the adsorption of fluoride onto tea waste sup­ ¼ þ (3)
qe qmax KL qmax
ported by hydrous aluminium oxide nanoparticles (Cai et al., 2015).
where qe is the amount of fluoride ions adsorbed per mass of adsorbent
3.2.4. Effect of temperature on fluoride removal efficacy (mg g 1), Ce is the equilibrium concentration of fluoride ions (mg L 1),
The effect of temperature on the adsorption of fluoride ions by the qmax and KL are the Langmuir constants related to the adsorption ca­
synthesized adsorbents AC-CLPs were explored at three different tem­ pacity (mg g 1) and energy of adsorption (L g 1), respectively. The value
peratures (25, 35 and 45 � C), whereas other equilibrium conditions like of qmax and KL can be obtained using a linear plot in between Cqee versus
initial fluoride concentration 10 mg L 1, contact time 240 min, pH 6.6
Ce . To corroborate whether the defluoridation process was favorable or
and adsorbent dose 1.0 g L 1 were kept constant, hence the defluor­
not, a dimensionless constant namely separation factor was employed.
idation efficiency is revealed in Fig. 8 (a and b). For the activated car­
Separation factor (RL) according to the literature of Hameed and Ahmad
bons derived at 250 and 500 � C, the observation shows that the
(2009) is expressed in Eq. (4),
percentage removal of fluoride increases with increase in temperature.
The maximum fluoride adsorption of 61.4% and 73.1% were observed at 1
RL ¼ (4)
25 � C for the activated carbons AC-CLP250 and AC-CLP500, respectively, 1 þ KL C0
which further enhanced up to 90.7% and 91.3% at 45 � C. The increasing
adsorption trend at high temperature indicates that the defluoridation where C0 is the initial fluoride concentration (mg L 1). In Eq. (4), the
method is an endothermic process, which may be due to the improving values of RL can be employed to interpret the adsorption type specified
as favorable (0 < RL < 1), unfavorable (RL > 1), linear (RL ¼ 1) and

Fig. 8. Effect of temperature on adsorption of fluoride onto AC-CLP250 (a) and AC-CLP500 (b) (conditions: initial fluoride concentration 10 mg L 1; contact time 240
min; adsorbent dose 1.0 g L 1; volume of fluoride solution 25 mL, solution pH 6.6).

8
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

irreversible (RL ¼ 0) (Nayak and Pal, 2017). 1998) were employed to simulate the experimental data. The kinetic
study was carried out by varying the contact time from 30 min to 300
3.3.2. Freundlich isotherm min during the defluoridation process. The adsorbent dosage, solution
In the stated equation, the Freundlich isotherm model described the pH, and temperature were fixed at 1.0 g L 1, 6.6 and 25 � C, respectively.
multilayer adsorption behavior of adsorbate molecules on the surface of The validity of the best-fit kinetic model was tested against the qe,exp, qe,
2
the adsorbent (Freundlich, 1906). This model assumes that the surface cal, and R values. Equation (6) illustrates the pseudo-first-order kinetic
of the adsorbent has a different binding energy spectrum. The linear model.
form of the Freundlich isotherm equation can be expressed by Eq. (5):
k1 t
logðqe qt Þ ¼ log qe (6)
1 2:303
logqe ¼ logKF þ logCe (5)
n
where k1 represents the reaction rate constant (min 1) for the pseudo-
1
� �
1 1=n first-order model, qe is the adsorption capacity in the equilibrium (mg
where KF mg ​ g L mg and n are the Freundlich isotherm
constants, and KF is representing the adsorption capacity per unit mass g 1), and qt is the amount of fluoride ions adsorbed at time t (min). The
of adsorbent. The 1/n value reveals the heterogeneity factor (when the values of k1 and qe,cal were evaluated from the slopes and intercepts of
value is greater than 1 it signifies the favorable adsorption of adsorbate the linear plots of ln (qe - qt) versus t (figure not given), respectively and
onto the adsorbent surface). are showed in Table 4. Table 4 illustrates a disgraceful agreement be­
The Langmuir and Freundlich constants with correlation coefficient tween qe,exp and qe,cal values for the fluoride removal. The smaller R2
(R2) values are highlighted in Table 2 for both the adsorbents and the values for the pseudo-first-order kinetic model indicates the in­
linear plots are observed in Fig. 9 (a and b), respectively. The qe,exp was compatibility of the pseudo-first-order model for fluoride adsorption
compared with the qe,cal obtained from the two isotherm models at 30 onto the AC-CLPs.
mg L 1 of fluoride concentration. It is noticed that the values obtained The pseudo-second-order kinetic model was further selected based
from Langmuir and Freundlich adsorption isotherms have a lower per­ on the equilibrium of the adsorption capacity, as expressed in Eq. (7).
centage of deviations. Most importantly, in the case of Langmuir t 1 t
isotherm, the R2 values are close to unity as compared to the Freundlich ¼ þ (7)
qt k1 q2e qe
isotherm. Hence, this describes that the adsorptive behavior of fluoride
onto AC-CLPs obeyed the Langmuir isotherm. It signifies the monolayer where k2 denotes the reaction rate constant (g mg 1 min 1) for the
coverage of adsorbate over a homogeneous adsorbent surface. Further­ pseudo-second-order kinetic model. The values of k2 and qe,cal were
more, the values of n are much higher than unity, which also corroborate calculated from the intercepts and slopes of the plots of t/qt versus t
the favorable adsorption process. Similarly, to present the observation of (Fig. 11(a and b)) respectively and are highlighted in Table 4. Table 4
the adsorption of fluoride onto chemically modified palm kernel shell explores that the correlation coefficient (i.e., R2) values of adsorption of
was better explained by the Langmuir model against Freundlich fluoride ions for the pseudo-second-order model was higher than that of
isotherm (Bashir et al., 2015). Fig. 10 reveals the RL values are between the pseudo-first-order model. Furthermore, the adsorption capacity
0 and 1 for both the synthesized adsorbents suggesting favorable values (i.e., qe,cal) calculated by the pseudo-second-order model also
adsorption conditions. By increasing the initial fluoride concentration, possessed a good agreement between experimental values (i.e., qe,exp),
the RL values diminish progressively due to the superior favorability at as compared with the pseudo-first-order model. This corroborated that
high fluoride concentration. the adsorption process of fluoride onto the AC-CLPs could be considered
Table 3 illustrates the comparison of the maximum adsorption ca­ as a pseudo-second-order model process, which implies that the fluoride
pacity of fluoride by various biomass related low-cost adsorbents with uptake process was chemisorption mechanisms.
proposed adsorbents employed in this study. Compared to other adsor­ Moreover, the mechanism of the defluoridation process can be
bents, the performance of AC-CLPs as reported in this study signifies its described by the intra-particle diffusion model (Weber and Morris,
noteworthy real-field applications. In this study, the adsorbents syn­ 1963), which can be explored by Eq. (8):
thesized from waste C. limetta peels explore a remarkable adsorption
capacity than many other adsorbents procured till date. qt ¼ kid t0:5 þ C (8)

where kid represents the intra-particle diffusion constant (mg g 1

3.4. Adsorption kinetics and its mechanism min 0.5) and the intercept (C) reveals the boundary layer diffusion ef­
fects or external film resistance.
In order to examine the rate of adsorption for fluoride onto the AC- Generally, the entire adsorption process can be classified into four
CLPs, two different kinetic models, i.e., pseudo-first-order model segments: (i) the adsorbate ions move from the bulk solution to the
(Lagergren, 1898) and pseudo-second-order model (Ho and McKay, adsorbent surface by bulk diffusion phenomenon; (ii) film diffusion of
the adsorbate ions through the boundary layer in order to reach the
Table 2 external surface of the adsorbent; (iii) intra-particle diffusion of the
Adsorption isotherm parameters for the removal of fluoride ions using AC- adsorbate ions into an adsorption site by pore diffusion mechanism, and
CLP250 and AC-CLP500.
(iv) adsorption takes place onto the active sites of the adsorbent surface,
Isotherm Parameters Values while the energy will substantiate on the binding process either due to
AC-CLP250 AC-CLP500 physical or chemical mechanisms (Michalak et al., 2013). Additionally,
it is also mandatory to understand the step which governs the overall
Langmuir qe,exp (mg g 1) 5.000 10.000
qe,cal (mg g 1) 4.906 9.434 removal rate in the defluoridation process by interpreting the experi­
% deviation 1.88 5.66 mental data. The first and fourth steps were recognized to be rapid; thus,
qmax (mg g 1) 4.926 9.709 it cannot be suggested as the rate-limiting step for the defluoridation
KL (L mg 1) 8.120 1.144
process. The adsorption rate will be monitored by external boundary
R2 0.9964 0.9776
Freundlich qe,cal (mg g 1) 5.101 9.953
layer diffusion or by intra-particle diffusion or by both. To illustrate the
% deviation 2.02 0.47 effect of intra-particle diffusion, the plots of qt versus t0.5 are interpreted.
KF (mg1
g) (L mg 1)1/n 4.395 5.875 If the plot is linear and passing through the origin, then it signifies that
n 22.727 6.452 the entire adsorption process is governed by intra-particle diffusion,
R2 0.1675 0.9439
whereas if it does not pass through the origin irrespective of the linear

9
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

Fig. 9. Linear fits of the (a) Langmuir isotherms and (b) Freundlich isotherms for fluoride adsorption on the AC-CLPs (conditions: contact time 240 min; initial pH
6.6; adsorbent dosage 1.0 g L 1; initial fluoride concentration 5–30 mg L 1; volume of fluoride solution 25 mL).

Table 3
Comparison of the adsorption capacities of various biomass based adsorbents for
the defluoridation process.
Adsorbent Adsorbent References
capacity (mg
g 1)

Eichhornia crassipes biomass carbon 0.52 Sinha et al. (2003)

prepared at 300 � C
Eichhornia crassipes biomass carbon 1.54 Sinha et al. (2003)
prepared at 600 � C
Activated carbon derived from 0.23 Karthikeyan and Ilango
Moringa indica (2007)
KMnO4-modified activated carbon 15.90 Daifullah et al. (2007)
derived from rice straw
Scandinavia spruce wood modified 7.92 Tchomgui-Kamga et al.
with aluminum and iron oxides (2010)
and carbonized at 500 � C
Scandinavia spruce wood modified 13.64 Tchomgui-Kamga et al.
with aluminum and iron oxides (2010)
and carbonized at 650 � C
Scandinavia spruce wood modified 5.67 Tchomgui-Kamga et al.
Fig. 10. Effect of initial fluoride concentration on the separation factor (RL).
with aluminum and iron oxides (2010)
and carbonized at 900 � C
relationship, then it explores that the intra-particle diffusion is not only Thermally activated carbon derived 4.617 Alagumuthu et al.
from Cynodon dactylon (2011)
the rate-controlling step but other mechanisms are also included (Nayak
Thermally activatedcarbon derived 7.014 Veeraputhiran and
and Pal, 2018). Fig. 11 (c and d) show that the plots of qt versus t0.5 for from Phyllanthus emblica Alagumuthu (2011)
AC-CLP250 and AC-CLP500 respectively, were multi-linear which in­ Ammonium carbonate impregnated 22.33 Sivasankar et al. (2012)
dicates that more than one rate-controlling mechanism affected the activated carbon of Tamarindus
fluoride uptake onto the AC-CLPs. Based on these plots, the defluor­ indica fruit shells
Activated bagasse carbon 1.15 Yadav et al. (2013)
idation processes of the four steps are encompassed by three phases,
Acid treated Moringa oleifera leaves 1.1441 Dan and Chattree
suggesting that the intra-particle diffusion is not only the rate-limiting (2018)
step. The initial phase of the plot explored an external mass transfer, Alkali treated Moringa oleifera leaves 0.8340 Dan and Chattree
whereas the second phase reveals a gradual adsorption is mainly due to (2018)
Chemically modified palm kernel 2.35 Bashir et al. (2015)
the intra-particle diffusion. The last phase is the equilibrium stage in
shells
which the intra-particle diffusion slows down due to lower availability Citrus limetta peels-based 4.926 Present study
of the fluoride concentration in the aqueous solution. Moreover, due to activated carbon prepared at
the double scenario of intra-particle diffusion (consisting of both film 250 � C (i.e. AC-CLP250)
and intra-particle diffusion), Boyd’s model was employed to understand Citrus limetta peels-based 9.709 Present study
activated carbon prepared at
whether the boundary layer film diffusion or the intra-particle diffusion
500 � C (i.e. AC-CLP500)
controls the adsorption process. This model is expressed in Eq. (9) (Boyd
et al., 1947):
� �X ∞ � � qt
6 1 � (10)
F¼1 exp m2 Bt (9) F¼
qe
π2 m¼1 m2
Since it is impossible to evaluate the values of Bt for each fraction
where F is the fraction of fluoride ions adsorbed at any time t (min), and adsorbed from Eq. (9) by considering the Fourier transform and inte­
Bt is mathematical fraction of F. The parameter F can be obtained by the gration. Reichenberg (1953) postulated the following expression:
following expression:
Bt ¼ 0:4977 lnð1 FÞ (11)

10
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

Table 4
Kinetic parameters obtained through pseudo-first-order and pseudo-second-order models for the adsorption of fluoride onto AC-CLPs at initial concentration of 5–30
mg L 1.
1 1
Fluoride concentration (mg L ) qe,exp (mg g ) Pseudo-first-order model Pseudo-second-order model

qe,cal (mg g 1) % deviation k1 (min 1) R2 qe,cal (mg g 1

) % deviation k2 (g mg 1
min 1
) R2

AC-CLP250
5 4.63 3.77 18.57 0.120 0.8814 4.65 0.43 0.372 0.9999
10 5.70 3.22 43.51 0.021 0.8086 5.80 1.75 0.031 0.9990
15 3.12 2.69 13.78 0.013 0.9628 3.37 8.01 0.011 0.9816
20 3.49 1.09 68.77 0.016 0.2724 3.56 2.01 0.050 0.9975
30 1.63 1.05 35.58 0.012 0.7721 1.62 0.61 0.033 0.9822
AC-CLP500
5 4.74 0.73 84.60 0.018 0.4677 4.74 0.00 0.209 0.9999
10 7.32 5.29 27.73 0.008 0.8379 7.69 5.05 0.004 0.9629
15 7.51 5.51 26.63 0.021 0.9306 7.74 3.06 0.014 0.9975
4
20 8.15 19.58 140.25 0.044 0.8238 12.33 51.29 6.014 � 10 0.5842
4
30 10.07 10.24 1.69 0.010 0.9060 14.21 41.11 5.682 � 10 0.6835

Fig. 11. Pseudo-second-order model for (a) AC-CLP250 and (b) AC-CLP500, and intra-particle diffusion model for (c) AC-CLP250 and (d) AC-CLP500.

The Bt values at different time intervals can be evaluated using Eq. origin, which corroborates that the rate-determining step of the
(11) for various solute fractions. The calculated Bt values were inter­ defluoridation process was revealed with the combined effects of both
preted against time t. The Boyd plots are revealed in Fig. 12 (a and b) for boundary layer film diffusion and intra-particle pore diffusion.
the adsorption of fluoride ions onto AC-CLP250 and AC-CLP500, respec­
tively. These Boyd plots were employed to distinguish between the 3.5. Study of adsorption thermodynamics
controlled mechanisms of the defluoridation process by either boundary
layer film diffusion or intra-particle diffusion. If the plots of Bt versus Thermodynamics parameters is responsible for comprehensive in­
time t are linear and passing through the origin, then the actual rate- formation about the spontaneity and heat change of the adsorption
controlling step in the adsorption process is the intra-particle pore process. Thermodynamics parameters such as Gibbs free energy change
diffusion. Otherwise they are governed by external film diffusion (ΔG, kJ mol 1), enthalpy change (ΔH, kJ mol 1) and entropy change
(Nethaji et al., 2013). From the above-mentioned figures, it was (ΔS, J mol 1 K 1) were calculated using the following equation (Mah­
observed that the plots were non-linear and does not pass through the moodi et al., 2011):

11
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

Fig. 12. Boyd plots for fluoride adsorption onto the adsorbent (a) AC-CLP250 and (b) AC-CLP500, and (c) plot of ln KC versus 1/T for the determination of ther­
modynamic parameters (conditions: contact time 240 min; initial pH 6.6; adsorbent dosage 1.0 g L 1; initial fluoride concentration 10 mg L 1; volume of fluoride
solution 25 mL).

Kc ¼
qe
(12) of ΔH value lies in the range of 2.1–20.9 and 80–200 kJ mol 1 for
Ce physisorption and chemisorption, respectively (Yu et al., 2004). For the
present study, the ΔH value was found to be 72.581 and 64.833 kJ mol 1
ΔG ¼ RT lnðKc Þ (13) for the studied adsorbents of AC-CLP250 and AC-CLP500, respectively.
This concludes that the adsorption process followed the
ΔS ΔH
lnKc ¼ (14) physico-chemical mechanism. Hence, the experimental data reveals that
R RT
the adsorption of fluoride onto AC-CLPs was favored and blocked at high
where Kc is the distribution constant, qe (mg g 1) is the amount of temperatures. The positive values of ΔS indicate the increased
fluoride adsorbed on the AC-CLPs at equilibrium, Ce (mg L 1) is the randomness at solid-solution interface occurring in the internal struc­
equilibrium concentration in a solution, and R (8.314 J mol 1 K 1) is the ture during the adsorption of fluoride onto AC-CLP250 and AC-CLP500.
universal gas constant and T (K) is the absolute temperature. The
adsorption experiments were conducted at three different temperatures 4. Conclusion
of 25, 35 and 45 � C, and it was revealed that the removal percentage of
fluoride increased with increasing temperature suggesting that the In this study, Citrus limetta waste peels were employed to synthesize
adsorption was more feasible at higher temperature. ΔH and ΔS values the activated carbons after chemical activation by impregnating with
were calculated from the slope and intercept of Van’t Hoff plot of ln Kc FeCl3. The adsorbents obtained from two different carbonization tem­
versus 1/T (Fig. 12(c)) and obtained thermodynamic parameters were peratures 250 � C and 500 � C were assigned as AC-CLP250 and AC-CLP500,
presented in Table 5. The negative values of ΔG indicate the thermo­ respectively and the potentiality of the activated carbons were examined
dynamically viable and spontaneous nature of the adsorption process by for the adsorptive removal of fluoride ions from aqueous solution. By
AC-CLP250 and AC-CLP500 (Borah et al., 2015). The positive value of ΔH increasing the initial fluoride concentration from 5 to 30 mg L 1, the
displays the endothermic nature of adsorption process. The magnitude removal efficiency decreased from 94.7% to 5.0% for AC-CLP250 and
94.8%–33.3% for AC-CLP500. Both the activated carbons were equally
effective, but AC-CLP250 did not corroborate the significant magnetic
Table 5 property compared to AC-CLP500. Next, the adsorption studies were
Thermodynamics parameters for fluoride adsorption onto AC-CLP250 and AC-
performed as a function of adsorbent dosage, initial concentration of
CLP500.
fluoride, contact time, solution pH and temperature. For both the ad­
1
Adsorbent Temperature ΔG (kJ ΔH (kJ ΔS (J mol R2 sorbents, the removal efficiency was maximum at pH 6.6 (almost
(K) mol 1) mol 1) K 1)
neutral) and decreases significantly on either side of pH 6.6, which
AC- 298 0.698 72.581 246.219 0.9892 would be appropriate for real-field purification system. The adsorption
CLP250 308 3.455
kinetics was found to follow the pseudo-second-order model, whereas
318 5.610
AC- 298 1.319 64.833 222.333 0.9837 the adsorption isotherm followed the Langmuir isotherm with high R2
CLP500 values (AC-CLP250 ¼ 0.9964 and AC-CLP500 ¼ 0.9776). The maximum

12
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

monolayer capacity obtained using the Langmuir isotherm for AC- Das, N., Pattanaik, P., Das, R., 2005. Defluoridation of drinking water using activated
titanium rich bauxite. J. Colloid Interface Sci. 292, 1–10.
CLP250 and AC-CLP500 were 4.926 and 9.709 mg g 1 under the optimum
Dey, S., Goswami, S., Ghosh, U.C., 2004. Hydrous ferric oxide (HFO): a scavenger of
conditions of adsorbent dosage 1.0 g L 1, contact time 240 min and fluoride from contaminated water. Water Air Soil Pollut. 158, 311–323.
temperature 25 � C with fluoride concentration ranging from 5 to 30 mg Elhalil, A., Qourzal, S., Mahjoubi, F.Z., Elmoubarki, R., Farnane, M., Tounsadi, H.,
L 1. The rate-determining step of the defluoridation process was Sadiq, M., Abdennouri, M., Barka, N., 2016. Defluoridation of groundwater by
calcined Mg/Al layered double hydroxide. Emerg. Contam. 2, 42–48.
corroborated by both external film layer diffusion and intra-particle Essadki, A.H., Gourich, B., Vial, C., Delmas, H., Bennajah, M., 2009. Defluoridation of
diffusion mechanisms. Thermodynamic parameters were computed for drinking water by electrocoagulation/electroflotation in a stirred tank reactor with a
the removal of fluoride and their values evoked that the defluoridation comparative performance to an external-loop airlift reactor. J. Hazard Mater. 168,
1325–1333.
process was endothermic and spontaneous in the studied temperature Fan, X., Parker, D.J., Smith, M.D., 2003. Adsorption kinetics of fluoride on low cost
range. Finally, it could be concluded that AC-CLP can be effectively materials. Water Res. 37, 4929–4937.
utilized for the adsorptive removal of fluoride ions from the aqueous Freundlich, H.M.F., 1906. Over the adsorption in solution. J. Phys. Chem. 57, 385–471.
Fu, R., Wang, W., Han, R., Chen, K., 2008. Preparation and characterization of γ- Fe2O3/
environment. ZnO composite particles. Mater. Lett. 62, 4066–4068.
Getachew, T., Hussen, A., Rao, V.M., 2015. Defluoridation of water by activated carbon
prepared from banana (Musa paradisiaca) peel and coffee (Coffea arabica) husk. Int.
Acknowledgement J. Environ. Sci. Technol. 12, 1857–1866.
Ghorai, S., Pant, K.K., 2005. Equilibrium, kinetics and breakthrough studies for
Facilities provided by Babasaheb Bhimrao Ambedkar University, adsorption of fluoride on activated alumina. Separ. Purif. Technol. 42, 265–271.
Hameed, B.H., Ahmad, A.A., 2009. Batch adsorption of methylene blue from aqueous
Lucknow (India) and financial support provided by Science and Engi­
solution by garlic peel, an agricultural waste biomass. J. Hazard Mater. 164,
neering Research Board (SERB)-Department of Science and Technology 870–875.
(DST), Government of India under the Early Career Research Award Ho, Y.S., McKay, G., 1998. Sorption of dye from aqueous solution by peat. Chem. Eng. J.
scheme [No. ECR/2016/001924] are greatly acknowledged. 70, 115–124.
Iriel, A., Bruneel, S.P., Schenone, N., Cirelli, A.F., 2018. The removal of fluoride from
aqueous solution by a lateritic soil adsorption: kinetic and equilibrium studies.
Appendix A. Supplementary data Ecotoxicol. Environ. Saf. 149, 166–172.
Jacks, G., Bhattacharya, P., Chaudhary, V., Singh, K.P., 2005. Controls on the genesis of
some high-fluoride groundwaters in India. Appl. Geochem. 20, 221–228.
Supplementary data to this article can be found online at https://doi. Jagtap, S., Yenkie, M.K., Labhsetwar, N., Rayalu, S., 2012. Fluoride in drinking water and
org/10.1016/j.gsd.2020.100339. defluoridation of water. Chem. Rev. 112, 2454–2466.
Karthikeyan, G., Ilango, S.S., 2007. Fluoride sorption using Moringa indica-based
activated carbon. Iran. J. Environ. Health Sci. Eng. 4, 21–28.
References Kimambo, V., Bhattacharya, P., Mtalo, F., Mtamba, J., Ahmad, A., 2019. Fluoride
occurrence in groundwater systems at global scale and status of defluoridation - state
of the art. Groundwater Sustain. Dev. 9, 100223.
Abe, I., Iwasaki, S., Tokimoto, T., Kawasaki, N., Nakamura, T., Tanada, S., 2004.
Kumar, M., Das, N., Goswami, R., Sarma, K.P., Bhattacharya, P., Ramanathan, A., 2016.
Adsorption of fluoride ions onto carbonaceous materials. J. Colloid Interface Sci.
Coupling fractionation and batch desorption to understand arsenic and fluoride co-
275, 35–39.
contamination in the aquifer system. Chemosphere 164, 657–667.
Alagumuthu, G., Veeraputhiran, V., Venkataraman, R., 2011. Fluoride sorption using
Lagergren, S., 1898. About the theory of so-called adsorption of soluble substances.
Cynodon dactylon-based activated carbon. Hem. Ind. 65, 23–35.
K. Sven. Vetenskapsakad. Handl. 24, 1–39.
Ali, S., Fakhri, Y., Golbini, M., Thakur, S.K., Alinejad, A., Parseh, I., Shekhar, S.,
Langmuir, I., 1918. The adsorption of gases on plane surfaces of glass, mica and
Bhattacharya, P., 2019. Concentration of fluoride in groundwater of India: a
platinum. J. Am. Chem. Soc. 40, 1361–1403.
systematic review, meta-analysis and risk assessment. Groundwater Sustain. Dev. 9,
Ma, W., Zhao, N., Yang, G., Tian, L., Wang, R., 2011. Removal of fluoride ions from
100224.
aqueous solution by the calcination product of Mg-Al-Fe hydrotalcite-like
Ali, S., Shekhar, S., Bhattacharya, P., Verma, G., Chandrasekhar, T., Chandrashekhar, A.
compound. Desalination 268, 20–26.
K., 2018. Elevated fluoride in groundwater of Siwani Block, Western Haryana, India:
Mahmoodi, N.M., Hayati, B., Arami, M., Lan, C., 2011. Adsorption of textile dyes on pine
a potential concern for sustainable water supplies for drinking and irrigation.
cone from colored wastewater: kinetic, equilibrium and thermodynamic studies.
Groundwater Sustain. Dev. 7, 410–420.
Desalination 268, 117–125.
Arlappa, N., Aatif Qureshi, I., Srinivas, R., 2013. Fluorosis in India: an overview. Int. J.
Maity, J.P., Hsu, C.-M., Lin, T.-J., Lee, W.-C., Bhattacharya, P., Bundschuh, J., Chen, C.-
Res. Dev. Health 1, 97–102.
Y., 2018. Removal of fluoride from water through bacterial-surfactin mediated novel
Asgari, G., Roshani, B., Ghanizadeh, G., 2012. The investigation of kinetic and isotherm
hydroxyapatite nanoparticle and its efficiency assessment: adsorption isotherm,
of fluoride adsorption onto functionalize pumice stone. J. Hazard Mater. 217–218,
adsorption kinetic and adsorption thermodynamics. Environ. Nanotechnol. Monit.
123–132.
Manag. 9, 18–28.
Baker, R.A., 1994. Potential dietary benefits of citrus pectin and fiber. Food Technol. 48,
Mariappan, R., Vairamuthu, R., Ganapathy, A., 2015. Use of chemically activated cotton
133–139.
nut shell carbon for the removal of fluoride contaminated drinking water: kinetics
Bashir, M.T., Salmiaton, A., Nourouzi, M.M., Azni, I., Harun, R., 2015. Fluoride removal
evaluation. Chin. J. Chem. Eng. 23, 710–721.
by chemical modification of palm kernel shell-based adsorbent: a novel agricultural
Meenakshi, S., Sundaram, C.S., Sukumar, R., 2008. Enhanced fluoride sorption by
waste utilization approach. Asian J. Microbiol. Biotechnol. Environ. Sci. 17,
mechanochemically activated kaolinites. J. Hazard Mater. 153, 164–172.
533–542.
Michalak, I., Chojnacka, K., Witek-Krowiak, A., 2013. State of the art for the biosorption
Borah, L., Goswami, M., Phukan, P., 2015. Adsorption of methylene blue and eosin
process-a review. Appl. Biochem. Biotechnol. 170, 1389–1416.
yellow using porous carbon prepared from tea waste: adsorption equilibrium,
Mohapatra, M., Anand, S., Mishra, B.K., Giles, D.E., Singh, P., 2009. Review of fluoride
kinetics and thermodynamics study. J. Environ. Chem. Eng. 3, 1018–1028.
removal from drinking water. J. Environ. Manag. 91, 67–77.
Boyd, G.E., Adamson, A.W., Myers Jr., L.S., 1947. The exchange adsorption of ions from
Mondal, N.K., Basu, S., Sen, K., Debnath, P., 2019. Potentiality of mosambi (Citrus
aqueous solutions by organic zeolites. II: kinetics. J. Am. Chem. Soc. 69, 2836–2848.
limetta) peel dust toward removal of Cr(VI) from aqueous solution: an optimization
Bureau of Indian Standards (BIS), 2012. Drinking Water Specification IS: 10500 (New
study. Appl. Water Sci. 9, 116.
Delhi).
Namduri, H., Nasrazadani, S., 2008. Quantitative analysis of iron oxides using Fourier
Cai, H., Chen, G., Peng, C., Xu, L., Zhu, X., Zhang, Z., Dong, Y., Shang, G., Ke, F., Gao, H.,
transform infrared spectrophotometry. Corrosion Sci. 50, 2493–2497.
Wan, X., 2015. Enhanced removal of fluoride by tea waste supported hydrous
Nayak, A.K., Pal, A., 2017. Green and efficient biosorptive removal of methylene blue by
aluminium oxide nanoparticles: anionic polyacrylamide mediated aluminium
Abelmoschus esculentus seed: process optimization and multi-variate modeling.
assembly and adsorption mechanism. RSC Adv. 5, 29266–29275.
J. Environ. Manag. 200, 145–159.
Chakrabarty, S., Sarma, H.P., 2012. Defluoridation of contaminated drinking water using
Nayak, A.K., Pal, A., 2018. Rapid and high-performance adsorptive removal of hazardous
neem charcoal adsorbent: kinetics and equilibrium studies. Int. J. Chem. Tech. Res.
acridine orange from aqueous environment using Abelmoschus esculentus seed
4, 511–516.
powder: single- and multi-parameter optimization studies. J. Environ. Manag. 217,
Chen, R., Xue, Q., Zhang, Z., Sugiura, N., Yang, Y., Li, M., Chen, N., Ying, Z., Lei, Z.,
573–591.
2011. Development of long-life-cycle tablet ceramic adsorbent for geosmin removal
Nethaji, S., Sivasamy, A., Mandal, A.B., 2013. Adsorption isotherms, kinetics and
from water solution. Appl. Surf. Sci. 257, 2091–2096.
mechanism for the adsorption of cationic and anionic dyes onto carbonaceous
Choong, T.S.Y., Chuah, T.G., Robiah, Y., Koay, F.L.G., Azni, I., 2007. Arsenic toxicity,
particles prepared from Juglans regia shell biomass. Int. J. Environ. Sci. Technol. 10,
health hazards and removal techniques from water: an overview. Desalination 217,
231–242.
139–166.
Njikam, E., Schiewer, S., 2012. Optimization and kinetic modeling of cadmium
Daifullah, A.A.M., Yakout, S.M., Elreefy, S.A., 2007. Adsorption of fluoride in aqueous
desorption from citrus peels: a process for biosorbent regeneration. J. Hazard Mater.
solutions using KMnO4-modified activated carbon derived from steam pyrolysis of
213–214, 242–248.
rice straw. J. Hazard Mater. 147, 633–643.
Dan, S., Chattree, A., 2018. Sorption of fluoride using chemically modified Moringa
oleifera leaves. Appl. Water Sci. 8, 76.

13
A. Siddique et al. Groundwater for Sustainable Development 10 (2020) 100339

Oguz, E., 2007. Equilibrium isotherms and kinetics studies for the sorption of fluoride on Sun, J., Zhou, S., Hou, P., Yang, Y., Weng, J., Li, X., Li, M., 2006. Synthesis and
light weight concrete materials. Colloid. Surface. Physicochem. Eng. Aspect. 295, characterization of biocompatible Fe3O4 nanoparticles. J. Biomed. Mater. Res. A 80,
258–263. 333–341.
Paudyal, H., Pangeni, B., Inoue, K., Kawakita, H., Ohto, K., Harada, H., Alam, S., 2011. Suryavanshi, U., Shukla, S.R., 2010. Adsorption of Pb2þ by alkali-treated Citrus limetta
Adsorptive removal of fluoride from aqueous solution using orange waste loaded peels. Ind. Eng. Chem. Res. 49, 11682–11688.
with multi-valent metal ions. J. Hazard Mater. 192, 676–682. Tchomgui-Kamga, E., Alonzo, V., Nanseu-Njiki, C.P., Audebrand, N., Ngameni, E.,
Raghav, S., Kumar, D., 2018. Adsorption equilibrium, kinetics, and thermodynamic Darchen, A., 2010. Preparation and characterization of charcoals that contain
studies of fluoride adsorbed by tetrametallic oxide adsorbent. J. Chem. Eng. Data 63, dispersed aluminum oxide as adsorbents for removal of fluoride from drinking
1682–1697. water. Carbon 48, 333–343.
Reichenberg, D., 1953. Properties of ion-exchange resins in relation to their structure. III. Thakre, D., Dixit, P., Waghmare, S., Manwar, N., Labhsetwar, N., Rayalu, S.S., 2015.
kinetics of exchange. J. Am. Chem. Soc. 75, 589–597. Synthesis optimization and fluoride uptake properties of high capacity composite
Saha, R., Mukherjee, K., Saha, I., Ghosh, A., Ghosh, S.K., Saha, B., 2013. Removal of adsorbent for defluoridation of drinking water. Environ. Prog. Sustain. Energy 34,
hexavalent chromium from water by adsorption on mosambi (Citrus limetta) peel. 1576–1585.
Res. Chem. Intermed. 39, 2245–2257. Thakur, L.S., Mondal, P., 2017. Simultaneous arsenic and fluoride removal from
Sarkar, M., Banerjee, A., Pramanick, P.P., Sarkar, A.R., 2006. Use of laterite for the synthetic and real groundwater by electrocoagulation process: parametric and cost
removal of fluoride from contaminated drinking water. J. Colloid Interface Sci. 302, evaluation. J. Environ. Manag. 190, 102–112.
432–441. Tomar, V., Prasad, S., Kumar, D., 2014. Adsorptive removal of fluoride from aqueous
Schiewer, S., Balaria, A., 2009. Biosorption of Pb2þ by original and protonated citrus media using Citrus limonum (lemon) leaf. Microchem. J. 112, 97–103.
peels: equilibrium, kinetics, and mechanism. Chem. Eng. J. 146, 211–219. Uçar, S., Erdem, M., Tay, T., Karag€oz, S., 2009. Preparation and characterization of
Sharma, B.S., Agrawal, J., Gupta, A.K., 2011. Emerging challenge: fluoride activated carbon produced from pomegranate seeds by ZnCl2 activation. Appl. Surf.
contamination in groundwater in Agra district, Uttar Pradesh. Asian J. Exp. Biol. Sci. Sci. 255, 8890–8896.
2, 131–134. Veeraputhiran, V., Alagumuthu, G., 2011. Sorption equilibrium of fluoride onto
Shen, P., Jiang, W., Wang, F., Chen, M., Ma, P., Li, F., 2013. Preparation and Phyllanthus emblica activated carbon. Int. J. Res. Chem. Environ. 1, 42–47.
characterization of Fe3O4@TiO2 shell on polystyrene beads. J. Polym. Res. 20, 252. Vithanage, M., Bhattacharya, P., 2015a. Fluoride in the environment: sources,
Shukla, A., Zhang, Y.-H., Dubey, P., Margrave, J.L., Shukla, S.S., 2002. The role of distribution and defluoridation. Environ. Chem. Lett. 13, 131–147.
sawdust in the removal of unwanted materials from water. J. Hazard Mater. 95, Vithanage, M., Bhattacharya, P., 2015b. Fluoride in drinking water: health effects and
137–152. remediation. In: Lichtfouse, E., Schwarzbauer, J., Robert, D. (Eds.), Environmental
Singh, J., Mishra, N.S., Uma, Banerjee, S., Sharma, Y.C., 2011. Comparative studies of Chemistry for a Sustainable World: Volume 5: CO2 Sequestration, Biofuels and
physical characteristics of raw and modified sawdust for their use as adsorbents for Depollution. Springer International Publishing, Switzerland, pp. 105–151.
removal of acid dye. BioResources 6, 2732–2743. Weber, W.J., Morris, J.C., 1963. Kinetics of adsorption on carbon from solution. J. Sanit.
Singh, T.P., Majumder, C.B., 2016. Fluoride removal from sewage water using Citrus Eng. Div. Am. Soc. Civ. Eng. 89, 31–60.
limetta peel as biosorbent. Int. J. Pharm. Pharmaceut. Sci. 8, 86–92. World Health Organization (WHO), 2006. Chemical fact sheets: fluoride. In: Guidelines
Sinha, S., Pandey, K., Mohan, D., Singh, K.P., 2003. Removal of fluoride from aqueous for Drinking Water Quality (Electronic Source): Incorporation First Addendum,
solutions by Eichhornia crassipes biomass and its carbonized form. Ind. Eng. Chem. fourth ed. Recommendations, Geneva.
Res. 42, 6911–6918. Yadav, A.K., Abbassi, R., Gupta, A., Dadashzadeh, M., 2013. Removal of fluoride from
Sivasankar, V., Rajkumar, S., Murugesh, S., Darchen, A., 2012. Tamarind (Tamarindus aqueous solution and groundwater by wheat straw, sawdust and activated bagasse
indica) fruit shell carbon: a calcium-rich promising adsorbent for fluoride removal carbon of sugarcane. Ecol. Eng. 52, 211–218.
from groundwater. J. Hazard Mater. 225–226, 164–172. Yu, Y., Zhuang, Y.-Y., Wang, Z.-H., Qiu, M.-Q., 2004. Adsorption of water-soluble dyes
Srimurali, M., Pragathi, A., Karthikeyan, J., 1998. A study on removal of fluorides from onto modified resin. Chemosphere 54, 425–430.
drinking water by adsorption onto low-cost materials. Environ. Pollut. 99, 285–289. Zhao, X., Wang, J., Wu, F., Wang, T., Cai, Y., Shi, Y., Jiang, G., 2010. Removal of fluoride
from aqueous media by Fe3O4@Al(OH)3 magnetic nanoparticles. J. Hazard Mater.
173, 102–109.

14