71

Removal of Cr(VI) from Wastewater Using a Natural Nanoporous

Adsorbent: Experimental, Kinetic and Optimization Studies
Utkarsh Maheshwari and Suresh Gupta* Department of Chemical Engineering, Birla Institute of
Technology and Science, Pilani 333031, Rajasthan, India.

(Received date: 10 July 2014; Accepted date: 20 December 2014)

ABSTRACT: In the present work, a nanoporous adsorbent prepared from low-

cost neem bark is used for the removal of chromium (VI) ions from wastewater.
The adsorbent is characterized by scanning electron microscopy, energy-
dispersive spectroscopy, BET surface area, infrared analysis and X-ray diffraction
analysis. The effect of various parameters such as adsorbent dosage, pH, initial
concentration and contact time for the removal of chromium (VI) ions has been
studied. The data obtained have been tested with various isotherm and kinetic
models. Langmuir model is found to be the best suited isotherm model giving
maximum adsorption capacity as 26.95 mg g–1 for the removal of
chromium (VI) ions. Effluent from a plating industry was also used to evaluate
the performance of the developed adsorbent. Weber and Morris and Boyd models
show that both intra-particle and film diffusion are the controlling mechanism for
adsorption of chromium (VI) ions on the developed adsorbent. The mathematical
model is developed to optimize the adsorption capacity (q) in terms of the
influencing parameters with the help of differential evolution (DE) technique.

1. INTRODUCTION

A marked increase in population, industrialization and urbanization in the last few decades has
resulted in growing demand for water that will progressively lead to shortage of water. This will
directly or indirectly affect the environment. Chromium and its alloys have wider application in
industries such as metal finishing, dyes, pigments, inks, glass, ceramics, tanning, textile, wood
preserving and glues industries (Babu and Gupta 2008a; Gupta and Babu 2009a). The effluents
released from these industries are majorly contaminated with hexavalent chromium ions [Cr(VI)]
with concentrations ranging from 10 to 3000 mg l–1. Cr(VI) has been declared as a powerful
carcinogenic agent by the International Agency for Research on Cancer in 1982. The U.S.
Environmental Protection Agency approved 0.1 mg l–1 as the maximum allowable concentration
of Cr(VI) in drinking water. To maintain such a low level of Cr(VI) in the effluent streams, the
removal of Cr(VI) from industrial wastewater has become a research topic of great interest (Babu
and Gupta 2008b; Memon et al. 2009; Moreno-Piraján et al. 2011; Wan Ngah and Hanafiah 2008).
Methods such as chemical or electrochemical precipitation, ion exchange, solvent extraction
and membrane separation are available for Cr(VI) removal from wastewater. These methods are
either cost intensive or inefficient for large-scale handling of the effluent. Nonetheless, adsorption
has proven to be one of the effective methods for the removal of Cr(VI) on an industrial scale. In
general, diameter of various adsorbents is in the range of 10–6 m or less with significant porosity
value, leading to the large surface area available for adsorption. The most commonly used
adsorbent for Cr(VI) removal is commercially available activated carbon, which is very expensive
(Babel and Kurniawan 2003).
*Author to whom all correspondence should be addressed. E-mail: sureshg@pilani.bits-pilani.ac.in (S. Gupta).
72 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

Furthermore, nanotechnology was reported to be great importance in the wastewater treatment

process (Taha and Ibrahim 2014). An immense work has been carried out on the nanostructured
materials because of their unusual physical and chemical properties owing to their extremely
small size and larger surface area. Nanoporous materials can meet the requirements for a good
adsorbent due to the availability of large surface area for adsorption (Liao and Chen 2002).
Therefore, there is a need to develop a nanoporous low-cost adsorbent for the removal of Cr(VI).
Once the adsorption process is complete, the adsorbent gets saturated with Cr(VI). This
Cr(VI)-rich adsorbent is hazardous when released into the environment without any treatment. The
environmental constraints further emphasize the need to develop a suitable method for regeneration
of the saturated adsorbent. It is important to understand the mass-transfer mechanism of Cr(VI)
from the fluid phase to the nanoporous adsorbent. Different mathematical models are available in
the literature, which may be utilized to identify the mass-transfer controlling mechanism of metal
ion from the liquid phase to the nanoporous adsorbent (Kapur and Mondal 2013).
Under present conditions, there is a need for continuous improvement in technology for
optimization of the operations. To find the optimal values of independent parameters, an
appropriate mathematical model needs to be developed, which represents the response of the
system as a function of the independent parameters. These optimal values would be helpful in
designing the large-scale continuous adsorption system. Studies reported in the literature have not
attempted to develop a mathematical model and to find optimum parameter values for a given
adsorption system.
The present study aims for the preparation of a nanoporous adsorbent using the readily
available neem bark in the Pilani (Rajasthan) area. The neem bark is chemically and physically
treated to prepare the nanoporous activated neem bark (nANB) adsorbent. The prepared nANB is
characterized by various techniques such as scanning electron microscopy (SEM), energy-
dispersive spectroscopy (EDS), BET surface area, infrared (IR) and X-ray diffraction (XRD). The
performance of the developed adsorbent is evaluated by conducting batch experiments for the
removal of Cr(VI) from wastewater. The effects of various parameters are investigated. Various
adsorption isotherms, kinetics and mass transport models are applied to predict the rate and kinetic
parameters. The saturated nANB is regenerated by acid and base treatment. The performance of
developed adsorbent is also evaluated by using it in the treatment of actual plating industry
effluent for the removal of Cr(VI). A mathematical model is developed using experimental results
to predict the adsorption capacity as a function of all other affecting parameters. Further, the
modelled equation is optimized using the DE technique to obtain optimal parameter values for the
maximum adsorption capacity.

2. MATERIALS AND METHODS

2.1. Preparation of the Nanoporous Adsorbent

Neem bark is obtained from the trees in Birla Institute of Technology and Science (Pilani,
Rajasthan, India) for the preparation of the nanoporous adsorbent. The neem bark is repeatedly
washed with distilled water for removing dust and other soluble impurities. After drying it in a
shaded room for 24 hours, the dried neem bark is crushed into smaller pieces by passing it through
a roller crusher, jaw crusher and ball mill sequentially. The crushed material is then screened
through 12–14 BSS mesh screens. The particles retained on the 14-mesh screen [average particle
size (dp) = 1.20 mm] are used for the experiments. The retained particles are further activated by
treating them with concentrated sulphuric acid (H2SO4; 98% w/w) in a 1:1 weight ratio and later
Removal of Cr(VI) from Wastewater Using Neem Bark 73

subjected to a heat treatment in an oven at 70 °C for 24 hours. Finally, the particles are repeatedly
washed in distilled water for removing free acid and colour impurities. The neem bark obtained
after the activation process is referred as nANB throughout the study.

2.2. Characterization

Five techniques have been used for the characterization of the prepared adsorbent. The first
characterization technique used is SEM, which involves capturing various images by focusing a
very-high-energy beam of electrons on the surface of nanoporous materials. The next
characterization technique used is EDS, which uses the energy dispersivity to detect the specific
elements present in the test material. It provides the composition of the material in weight
percentage. The SEM and EDS analysis of nANB were carried out using field-emission gun-
scanning electron microscope (JSM-7600F). The available specific surface area is another
important parameter in adsorption, as adsorption is a surface phenomenon. Therefore, the specific
surface area of the nANB is determined using a BET surface analyzer (SMART SORB 92/93;
Smart Instruments, Thane, Maharashtra, India). The nANB is also characterized using the IR
spectrum from 400 to 4000 cm–1 to analyze the vibration mode of the functional groups available.
The last characterization method used was XRD for the phase analysis of the nANB developed. The
XRD characterization was carried out using the Rigaku MiniFlex II desktop X-ray diffractometer.

2.3. Batch Experiments

Batch experiments were performed with 25 ml of aqueous sample in 100-ml conical flask.
Approximately 2.8287 g of 99.9% potassium dichromate (K2Cr2O7) is dissolved in distilled water
and the solution is made up to 1000 ml for preparing 1000 mg l–1 stock solution of Cr(VI). The stock
solution is diluted to prepare different Cr(VI) concentration solutions ranging from 5 to 500 mg l–1.
These solutions are agitated at a speed of 150 rpm on a rotary shaker for a pre-determined period at
room temperature (35 ± 2 °C). The effect of contact time is studied with an initial Cr(VI)
concentration of 100 mg l–1, pH 2.70 and a constant adsorbent dosage of 6 g l–1. The effect of
adsorbent dosage is studied by varying the adsorbent amount from 4 to 28 g l–1 by keeping the initial
Cr(VI) concentration, pH and contact time as 100 mg l–1, 2.70 and 48 hours, respectively. The effect of
initial Cr(VI) concentration is studied by varying the Cr(VI) concentration from 60 to 200 mg l–1 by
maintaining adsorbent dosage, pH and contact time as 6 g l–1, 2.70 and 48 hours, respectively. The
effect of pH is also studied by varying the pH value of the solution from 2 to 12 by keeping the
initial Cr(VI) concentration, dosage of the adsorbent and contact time at 100 mg l–1, 6 g l–1 and
48 hours, respectively. All the experiments are performed two times and the averaged values
are taken.

2.4. Analysis of Chromium Ions

The concentration of total chromium ions left in the liquid phase is determined
spectrophotometrically using the atomic absorption spectroscope (AA-7000; Shimadzu) (Babu
and Gupta 2008a; Gupta and Babu 2009a; Memon et al. 2009). The amount of Cr(VI) adsorbed
by the nanoporous adsorbent is then calculated using the following equation:
(C0 – Ce )V
qe = (1)
W
74 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

where qe is the adsorption capacity [solid-phase concentration of Cr(VI) on the adsorbent; mg g–1],
C0 is the initial concentration of Cr(VI) (mg l–1), Ce is the concentration of chromium at
equilibrium (mg l–1), V is the initial volume of chromium solution (l) and W is the weight of the
adsorbent (g).

2.5. Regeneration

Previous studies have reported that the adsorption of Cr(VI) is highly favoured at low pH of 1
(Gupta and Babu 2009b; Kapur and Mondal 2013). Thus, the regeneration process is carried out
by treating the saturated nANB with 1 N NaOH for 24 hours. It is further treated with 1 N HCl
for the next 24 hours. After the treatment, the regenerated nANB is washed thoroughly with
distilled water to remove the traces of acid from the nANB surface (Gupta and Babu 2009a).

2.6. Study on Plating Industrial Effluent

The primary contaminant in the chrome plating industry effluent is Cr(VI). Therefore, in the
present study, plating industry effluent, which contains large amounts of Cr(VI), was used to
evaluate the performance of the prepared nANB for the removal of Cr(VI). The concentration of
Cr(VI) in the industrial sample was maintained as 100 mg l–1 by appropriate dilution. The diluted
sample is treated for a contact time of 48 hours with an adsorbent dosage of 6 g l–1.

3. RESULTS AND DISCUSSION

In the present study, fresh neem bark and nANB were used for the removal of Cr(VI) from
aqueous solution. The maximum adsorption capacity of fresh neem bark and nANB was found to
be 14.33 and 26.95 mg g–1, respectively, from equilibrium experiments at 35 °C with an adsorbent
dosage of 6 g l–1. The adsorption capacity of nANB for Cr(VI) removal was found to be almost
two times than the adsorption capacity of the fresh neem bark. Therefore, nANB is considered an
efficient adsorbent for further experimental studies.
The nANB adsorbent is characterized by SEM, EDS, surface area, IR and XRD at different
stages of the study. The effects of various influencing parameters such as the initial Cr(VI)
concentration, contact time, adsorbent dosage and pH are studied. The adsorption capacity of
nANB is compared with various adsorbents reported in literature (Table 1). The adsorption
capacity of nANB for Cr(VI) is found to be higher than many commercially available and low-
cost adsorbents. In the present work, the nanoporous adsorbent was prepared in granular form,
which makes it more applicable for continuous treatment of industrial effluent streams. The results
of this study indicate that the prepared nANB proves to be an efficient and effective adsorbent that
can be used for Cr(VI) removal from wastewater.

3.1. Characterization of the Adsorbent

In the present study, SEM is used to study the surface morphology of the developed nanoporous
adsorbent at a magnification of 50,000 ¥. The surface morphology and EDS analysis of the various
forms of adsorbent such as fresh adsorbent, used adsorbent, adsorbent used on industrial sample
and regenerated adsorbent are shown in the Figures 1–4, respectively.
Removal of Cr(VI) from Wastewater Using Neem Bark 75

TABLE 1. Maximum Adsorbent Capacity of Various Commercial and Low-Cost Adsorbents

Maximum adsorbent
S. No. Absorbent capacity, qm, (mg g–1) References
1 Nanoporous activated neem 26.95 Present study
bark (nANB)
2 Leaf mould 25.9 Sharma and Forster (1996)
3 Pine needles 21.50 Dakiky et al. (2002)
4 Coconut shell–based 20.0 Alaerts et al. (1989)
activated carbon
5 Sugar beet pulp 17.2 Sharma and Forster (1994)
6 Palm pressed fibres 15.0 Tan et al. (1993)
7 Maize cob 13.8 Sharma and Forster (1994)
8 Sugar cane bagasse 13.4 Sharma and Forster (1994)
9 Almond 10.0 Dakiky et al. (2002)
10 Maple saw dust 8.2 Yu et al. (2003)
11 Cactus 7.08 Dakiky et al. (2002)
12 Coal 6.78 Dakiky et al. (2002)
13 Biomass residual slurry 5.87 Namasivayam and Yamuna (1995)
14 Distillery sludge 5.7 Selvaraj et al. (2003)
15 Waste tea 1.55 Orhan et al. (1993)
16 Fe(III)/Cr(III) hydroxide 1.43 Namasivayam and
Ranganathan (1993)
17 Walnut shell 1.33 Orhan et al. (1993)
18 Soya cake 0.28 Daneshvar et al. (2002)

(a) (b)
Spectrum 1

0 1 2 3 4 5 6 7 8 9 10
Full scale 13721 cts cursor: 0.000 keV

Figure 1. (a) SEM and (b) EDS for fresh nANB.

SEM and EDS analysis of fresh nANB are shown in Figures 1(a and b), respectively. The SEM
images of Figures 1(a), 2(a), 3(a) and 4(a) show the morphology of the surface of the adsorbent
prepared. The presence of pores of different sizes is clearly visible on the surface of nANB
[Figure 1(a)]. After comparing the size of the pores with the scale of the figure, the presence of
10–100-nm pores on the surface of nANB had been confirmed. These results confirmed the
development of nanoporous adsorbent for the removal of Cr(VI). The EDS analysis [Figure 1(b)]
confirms that initially no Cr(VI) adheres to the surface of the adsorbent. The highest peak of
carbon (C), making approximately 55.35% of the total adsorbent by weight, may be due to the
carbonaceous material (neem bark) used to prepare the adsorbent.
76 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

(a) (b)
Spectrum 1

0 1 2 3 4 5 6 7 8 9 10
Full scale 9852 cts cursor: 0.000 keV

Figure 2. (a) SEM and (b) EDS for used nANB on pure Cr(VI) solution.

(a) (b)
Spectrum 1

0 1 2 3 4 5 6 7 8 9 10
Full scale 6495 cts cursor: 0.000 keV

Figure 3. (a) SEM and (b) EDS for used nANB on industrial effluent.

(a) (b)
Spectrum 1

0 1 2 3 4 5 6 7 8 9 10
Full scale 12387 cts cursor: 0.000 keV

Figure 4. (a) SEM and (b) EDS for regenerated nANB.

SEM and EDS analysis of the saturated nANB obtained after the treatment of aqueous Cr(VI)
solution and plating industry effluent are represented in Figures 2(a), 2(b), 3(a) and 3(b),
respectively. The EDS analysis [Figure 2(b) and 3(b)] shows the presence of Cr(VI) in the
saturated nANB obtained after the treatment of aqueous solution and industrial effluent,
respectively. The EDS analysis of saturated nANB [Figure 2(b) and 3(b)] suggests the presence of
10 wt% and 3.5 wt% of Cr(VI), respectively. The reduction in the percentage adsorption of Cr(VI)
for industrial effluent is due to the adsorption of other metals and impurities. The presence of
potassium (K+; 1 wt%) can be clearly seen in the saturated adsorbent [Figure 2(b)], which may be
due to the use of potassium dichromate as a preparatory agent.
Removal of Cr(VI) from Wastewater Using Neem Bark 77

100

80

60
%T

40

20

0
4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600
cm−1

Figure 5. IR spectra of nANB from 4000 to 400 cm–1 wavelength.

SEM image and EDS analysis of the regenerated nANB are shown in Figures 4(a and b),
respectively. Compared with the saturated nANB, there is a significant reduction in the amount of
Cr(VI) present on the surface of the regenerated nANB. Approximately 1.14% of Cr(VI) remains on
the regenerated adsorbent surface, which supports the removal of 91.4% of Cr(VI) from used nANB.
This indicates that the nANB can be regenerated and reused for further adsorption of Cr(VI).
Figures 1(a)–3(a) also reveal the important observation that the nanopores available in fresh
nANB are occupied by Cr(VI) after adsorption, which are again available after the regeneration
[Figure 4(a)]. The availability of the nano-sized pores may result in better adsorption capacity and
better removal of Cr(VI).
The surface area of the prepared nANB is analyzed using the BET surface area analyzer. The
measured value of specific surface area of nANB is found to be 36.64 m2 g–1, which is higher in
comparison with some of the other adsorbents such as tamarind seeds (Gupta and Babu 2009a),
tea factory waste (Malkoc and Nuhoglu 2007), Aspergillus niger (Namasivayam and Sureshkumar
2008), Zn(O) particles (Guo et al. 2012) and sawdust (Gupta and Babu 2009b). Higher surface
area of the adsorbent supports higher adsorption capacity (Hossain et al. 2010).
The IR spectrum of nANB is shown in Figure 5. The spectrum curve is not smooth due to the
composite and amorphous nature of the nANB. The peaks are found at 1030 and 1300, 1450 and
1640, 1540, 2350 and 2900 cm–1, which confirm the presence of C–O, C=O, C=C, O=C=O
and C–H stretching vibrations, respectively. The band appears at 3300 cm–1, which indicates the
presence of water molecules (O–H stretching mode of hydroxyl groups; Bakiler et al. 1999).
The peaks obtained for C–O, C=O and O=C=O indicate the presence of various acidic groups on
the surface of nANB (Depci et al. 2012).
The phase of the developed adsorbent is also studied using XRD with Cu-Ka radiation. The
wavelength considered is 1.54 Å and 2q ranges from 10° to 90° at a step size of 0.01°. The XRD
pattern for the nANB is shown in Figure 6. There is no peak identified in the spectrum (Figure 6),
and therefore, the particle size of developed adsorbent cannot be calculated using the Scherrer
formula. The distorted pattern depicts that there is no crystallinity in the nANB and that it purely
has only the amorphous nature, which is also confirmed by IR (Figure 5; Taha and Ibrahim 2014).

3.2. Batch Studies

The effects of various parameters such as contact time, adsorbent dosage, effect of pH and initial
adsorbate concentration on the removal of Cr(VI) from aqueous solution are studied (Figure 7).
The effect of contact time is an important parameter, which will be helpful to acquire the kinetic
78 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

200

150
Intensity

100

50

20 40 60 80
2Θ

Figure 6. Powder X-ray diffraction pattern for nANB.

Effect of contact time

25 Effect of mass of adsorbent
Effect of initial pH
Adsorption capacity, q (mg g−1)

Effect of initial concentration

20

15

10

0
0.0 0.2 0.4 0.6 0.8 1.0
Normalized values
(Contact time, Adsorbent dosage, Initial pH and Intial concentration)

Figure 7. Effect of contact time, adsorbent dosage, initial pH and initial concentration on the adsorption of Cr(VI) by nANB.

parameters and propose the rate kinetics of the Cr(VI) removal. It is observed that the adsorption
capacity rapidly increases with the increase in contact time up to 7 hours, and later it increases
gradually with further increase in contact time. This indicates that the adsorption is faster in the
initial period and it decreases later. This may be due to the availability of free active sites for
adsorption in the initial period, which decreases in the later part of the process as the active sites
get saturated with time.
The effect of adsorbent mass on Cr(VI) removal using nANB is studied to obtain the optimal
value of adsorbent dosage. It is found that the adsorption capacity decreases from 10.537 to
2.981 mg g–1 (Figure 7) while the percentage removal of Cr(VI) increases from 42% to 84% with
Removal of Cr(VI) from Wastewater Using Neem Bark 79

an increase in the adsorbent dosage from 4 to 28 g l–1. The decrease in adsorption capacity may
be due to the increase in the adsorbent amount, which leads to the increase in the number of active
sites available for the adsorption process. This also leads to an increase in percentage removal.
The effect of initial concentration is studied by varying Cr(VI) concentration from 60 to
200 mg l–1 and keeping other parameters constant (Figure 7). An increase in adsorption capacity
from 9.42 to 24.37 mg g–1 and a decrease in percentage removal from 94.21% to 73.11% occur
following an increase in the initial Cr(VI) concentration from 60 to 200 mg l–1. The increase in the
adsorption capacity may be due to the utilization of all the active sites available for adsorption at
higher initial Cr(VI) concentration. The decrease in the percent removal may be due to lack of
active sites available for adsorption. These active sites become saturated beyond a certain
concentration limit of adsorbate. Upon further increasing the Cr(VI) concentration, no adsorption
takes place and the Cr(IV) remains in the final solution.
The other significant parameter studied is the effect of pH on the removal of Cr(VI). The range
of pH considered in the study is 2–12. The adsorption capacity decreases from 16.26 to 1 mg g–1
with an increase in the pH value from 2 to 12 as shown in Figure 7. The chromium ion exists in
various forms at different solution pH [equation (2)]

+ +
2H 2H
2H + + 2HCrO –4 ←⎯⎯ → 2HCrO 4 ↔ 2H 2 O + Cr2 O 27 – ←⎯⎯ → 2CrO 3 + H 2 O (2)

It is well established in the literature that the dominating form of Cr(VI) at lower pH (pH ≈ 2)
is HCrO4– (Gupta and Babu 2009b; Kapur and Mondal 2013). The surface of the adsorbent is
positively charged, which is confirmed by the availability of acid sites using the IR spectrum. The
acid sites favour the adsorption of negatively charged HCrO4– at lower pH. With a rise in pH, there
is a decline in the adsorption capacity, which may be due to the transformation of HCrO4– into
other forms such as CrO42– and Cr2O72–. Furthermore, this can also be attributed to the dual
competition between the two different CrO42– and OH– forms available for adsorption, of which
the OH– form predominates (Babu and Gupta 2008a; Muthukumaran and Beulah 2011).

3.3. Mechanism of Cr(VI) Adsorption

The adsorbent used in this study was prepared from neem bark, which is activated by acid
treatment. According to literature sources, adsorption of Cr(VI) is more favourable for an acid-
treated adsorbent. The acid treatment makes available the acidic oxides groups on the surface of
nANB (Gupta and Babu 2009b). Because of the presence of these groups, the surface of the
adsorbent is positively charged. This is also supported by the fact that the nANB is a carbonaceous
substance and forms positively charged hydroxides (CxOH22+) on the surface when it comes in
contact with an aqueous solution [equation (3)].

C X O 2 + 2H 2 O ⇔ C X OH 22 + + 2OH – (3)

The IR spectrum (Figure 5) validates the presence of acidic groups on the adsorbent surface.
The solution pH and the pHzpc (zero point charge) value of the adsorbent can also be related to the
type and ionic state of the functional group present on the adsorbent surface. The pHzpc value of
nANB is found to be 2.80. It is a well accepted fact that if the pH of the solution is lower than the
value of pHzpc, the surface of the adsorbent is positively charged (Gupta and Babu 2009b; Kapur
and Mondal 2013). Studies have also shown that at a lower pH, the more stable form of Cr is
80 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

HCrO4–, which favours adsorption (Kapur and Mondal 2013). The suitability of nANB for the
adsorption of Cr(VI) may be due to the electrostatic attraction between the HCrO4– and the
positively charged surfaces of the nANB

C X OH 22 + + HCrO –4 ⇒ C X O 2 H 3CrO 3– (4)

Equations (3 and 4) are combined to obtain equation (5) as follows:

C X O 2 + 2H 2 O + HCrO 4– ⇔ C X O 2 H 3CrO 3– + 2OH – (5)

It is observed that the pH of the solution increases with the increase in Cr(VI) adsorption. From
equation (5), it can be seen that the adsorption of 1 mol of HCrO4– liberates 2 mol of hydroxyl
ions, which further increases the pH of the solution (Babu and Gupta 2008a; Gupta and Babu
2009c, 2010).

3.4. Use on an Industrial Sample

The applicability of nANB is also verified using the plating industry effluent containing a significant
amount of Cr(VI). There is 62.77% of Cr(VI) removal from the industrial sample, whereas it was
79.33% for aqueous solution containing 100 mg l–1 of initial concentration for an adsorbent dosage
of 18 g l–1. The adsorption capacity of nANB for Cr(VI) removal from plating industry sample was
found to be 3.49 mg g–1, whereas that from an aqueous solution was 4.53 mg g–1. The decrease in
adsorption capacity may be due to the presence of metal ions such as aluminium and nickel along
with other impurities such as chlorine in the industrial effluent, which may occupy the active sites
available for Cr(VI) adsorption. This is also supported by the EDS analysis of nANB saturated with
the industrial sample [Figure 3(b)]. The analysis supports the use of developed adsorbent for the
treatment of industrial wastewater containing Cr(VI) along with other ions and impurities.

4. THEORETICAL STUDIES
4.1. Adsorption Isotherms

In this study, a new adsorbent was developed which needs to establish an equilibrium relationship
between the liquid-phase and solid-phase concentrations of Cr(VI). This can be achieved with a
variety of isotherm models available in the literature (Gupta and Babu 2009a, b). In this study,
Langmuir, Freundlich and Temkin isotherm models are tested with experimentally obtained
equilibrium data. The different isotherm model results are compared using the experimental data
and shown in Figure 8. The parametric and R2 values for all the isotherms are reported in Table 2.
The best suited isotherm model from the equilibrium data would be helpful for designing the
fixed-bed adsorption column.
The R2 values indicate that the Langmuir isotherm model best describes the Cr(VI) adsorption on
nANB, which confirms the monolayer adsorption of Cr(VI) on nANB surface (Gupta and Babu
2009a). The maximum adsorption capacity obtained using the Langmuir isotherm is 26.95 mg g–1.
The experimental data also fit satisfactorily with the Temkin isotherm model, which assumes a linear
decrease in the heat of adsorption of all the molecules in the layer. This decrease may be due to the
Cr(VI)–Cr(VI) repulsion and homogeneous distribution of Cr(VI) on nANB (Hameed et al. 2008).
Removal of Cr(VI) from Wastewater Using Neem Bark 81

70

Equilibrium liquid-phase concentration

Experimental values
60 Langmuir isotherm
Freundlich isotherm
50 Temkin isotherm

C e (mg g−1)
40

30

20

10

8 10 12 14 16 18 20 22 24 26
Equilibrium solid-phase concentration q e (mg g−1)

Figure 8. Comparison of various isotherms with experimental data obtained for adsorption of Cr(VI) by nANB for an
adsorbent dosage of 6 g l–1, pH 2.70 and concentration of 60–200 mg l–1.

TABLE 2. Parametric and R2 Values for Various Adsorption Isotherm

Models Applied for Evaluating the Removal of Cr(VI) Using nANB
as the Adsorbent
Adsorption Isotherm
S. No. isotherm parameters R2
1 Langmuir Qm 26.95
B 0.144 0.995
Kf 6.896
2 Freundlich 0.961
n 3.083
AT 1.906
3 Temkin BT 5.219 0.989
bT 482.695

4.2. Adsorption Kinetics

Numerous rate kinetic models are available in the literature. Many studies reported application of
pseudo-first-order and second-order kinetic models for the adsorption of Cr(VI) (Gupta and Babu
2009a, b). In the present study, the pseudo-first-order and second-order kinetic models are applied
on obtained experimental data for the adsorption of Cr(VI) from aqueous solutions using nANB.
The model-predicted results are compared with experimental data (Figure 9). The kinetic
parameters along with R2 values for two kinetic models are presented in Table 3.
Upon comparing the values of qe and R2, it was found that the second-order kinetic model
provides a better account of the kinetics of Cr(VI) adsorption on nANB than the pseudo-first-
order model. The result suggests the chemical adsorption to be the premier rate-limiting step in
the process where the adsorbent surfaces got tied up by the metal ions, forming a firm covalent
bond. This probably occurs to maximize their co-ordination number with the surface. The kinetic
82 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

Solid-phase concentration (mg g−1)

12

10

4 Experimental values
Pseudo-first-order
Second order
2

0
0 10 20 30 40 50
Time (h)

Figure 9. Comparison of pseudo-first-order and second-order kinetic models obtained for adsorption of Cr(VI) by nANB
for adsorbent dosage of 6 g l–1, pH 2.70 and initial Cr(VI) concentration of 100 mg l–1.

TABLE 3. Kinetic Constants for Adsorption of Cr(VI) on nANB

R2

q e (mgg –1 ) K ad (min –1 )
Pseudo-first-order kinetics 0.966
9.88 0.0785
q e (mgg –1 ) K 2 (min –1 )
Second-order kinetics 0.997
13.37 0.0102

relationship is also important to predict the rate at which the adsorption is taking place, as this
will be helpful in designing the fixed-bed adsorption column. The mass-transfer coefficients and
the equilibrium capacity of the adsorbent can be calculated using the adsorption rate kinetic
constants. The parameters such as the shape of breakthrough curve and breakthrough time also
depend on the rate of adsorption (Babu and Gupta 2008a; Gupta and Babu 2009a, b; Kumar and
Kirthika 2009).

4.3. Mass-Transfer Mechanism

In the adsorption process, the movement of the adsorbate molecules from the fluid to solid phase
involves different mechanisms. Initially, the molecules of the adsorbate move from the bulk phase
of the liquid to the surface of the adsorbent, and this phenomenon is termed the film diffusion.
Further, by particle diffusion, molecules move from the surface to the interior of the particle. Here,
Weber and Morris (Kapur and Mondal 2013; Malash and El-Khaiary 2010; Weber and Morris
1963) and Boyd (Kapur and Mondal 2013; Malash and El-Khaiary 2010) models are used to study
the rate-limiting mechanism for Cr(VI) adsorption on nANB adsorbent.
Removal of Cr(VI) from Wastewater Using Neem Bark 83

12

10

qt (mg g−1) 6

0
0 1 2 3 4 5 6 7
1/2 1/2
t (min )

Figure 10. Weber and Morris plot for the Cr(VI) adsorption on nANB.

4.3.1. Weber and Morris Model

In the adsorption process, it might be possible that the mass-transfer rate of adsorbate molecules
is controlled by the intra-particle diffusion. Weber and Morris (1963) proposed the intra-particle
diffusion model proposed in equation (6)
q t = K id t1 2 (6)

where Kid (mg g–1 minute–1/2) represents the intra-particle diffusion rate constant.
The plot of qt versus t1/2 is shown in Figure 10. The model provides the value of Kid as
1.614 mg g–1 minute–1/2 at an initial Cr(VI) concentration of 100 mg l–1. It is suggested that the best-
fit line should pass through the origin for intra-particle diffusion, which is not the case here
(Figure 10). This confirms the fact that the intra-particle diffusion is not the only rate-controlling
step during the adsorption of Cr(VI) on nANB, and that there is also a significant contribution of
the film diffusion mechanism.

4.3.2. Boyd Model

The Boyd model is used for the film diffusion and is represented by equations (7 and 8) (Malash
and El-Khaiary 2010)
6
F = 1– exp(–Bt) (7)
π2
qt
F= (8)
qe

where the F is the fraction of solute adsorbed at time t, and Bt is a function of F. Equation (7) can
be rearranged in the linear form as follows:
Bt = [-0.4977 - ln(1 - F)] (9)
84 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

2.5

2.0

−0.4977-In(1-F)
1.5

1.0

0.5

0.0
0 500 1000 1500 2000 2500
t (min)

Figure 11. Boyd plot for the Cr(VI) adsorption on nANB.

The applicability of the Boyd model is confirmed by fitting the experimental data linearly for
[–0.4977 – ln(1 – F)] versus time. If the linear plot of the graph is not passing through the origin,
then it confirms that the process is film diffusion controlled.
It can be seen from Figure 11 that the line is not passing through the origin; however, the value of
the intercept is close to zero. This confirms that the process is controlled significantly by film diffusion
and partially by intra-particle diffusion. This is also confirmed by the Weber and Morris model.

5. MODELLING AND OPTIMIZATION

5.1. Formulation of Mathematical Model

Modelling is an extremely vital part of the process analysis, which helps in predicting the response
and behaviour of the system on the basis of dependent parameters. In this study, a second-order
polynomial [equation (10)] is used to predict the response of the system.

q pre = a 0 + ∑ a i X i + ∑ a ii X 2i (10)

where i represents the number of dependent variables varying from 1 to 4, qpre represents the
predicted adsorption capacity, ai and aii are constants and Xi represents the ith dependent variable.
The experimental data obtained for each parameter were normalized based on maximum
parameter value considered in all the experiments. The values were regressed using Microsoft
Excel 2007 (Microsoft, Redmond, WA, USA) for a second-order polynomial. The regressed
equation representing the adsorption capacity in terms of dependent variables is as follows:

q = -0.060375 + 0.73112929 ¥ X1 + 0.86501918 ¥ X2 + 0.81513482 ¥ X3 - 1.5179251 ¥ X4

+ 0.18959851 ¥ X21 - 1.3151705 ¥ X22 - 0.3902642 ¥ X23 + 0.81468689 ¥ X24 (11)

where X1, X2, X3 and X4 represent the normalized values of the initial concentration, pH, contact
time and adsorbent dosage, respectively. A graph (Figure 12) plotted between qpre (model prediction)
and qexp (experimental values) verifies the fit of the model. The R2 value obtained is 0.93, which
Removal of Cr(VI) from Wastewater Using Neem Bark 85

1 .2

qpre vs qexp
1 .0
Best-fitting linear fit

0 .8

qpre
0 .6

0 .4

0 .2

0 .0
0.0 0 .2 0 .4 0 .6 0.8 1.0
qexp

Figure 12. Comparison of the adsorption capacity qpre (predicted from model equation) with qexp (obtained by experiment).

indicates that the developed equation can be utilized to represent the adsorption capacity in terms of
other dependent parameters. Furthermore, the developed equation [equation (11)] is optimized using
DE to obtain the optimal parametric values for maximum adsorption capacity.

5.2. Optimization

In the last few decades, there has been a potential interest in optimization algorithms. There are a
number of well-recognized algorithms available for optimization, which include DE, genetic
algorithm (GA), genetic programming, evolutionary programming and evolution strategies. The
DE algorithm has more inbuilt advantages than GA. Unlike GA, which uses the binary coding for
expressing the range of parameters available in the problem, DE is an influential population-based
method. The DE algorithms are faster compared with the GAs. These algorithms are designed in
such a way that they facilitate easy location of the global optimum of the function along with an
improved accuracy (Babu 2007). A typical GE algorithm is shown in Figure 13.

5.2.1. Optimization Results

In the present study, the ranges of parameters selected are 60–200 mg l–1 for initial Cr(VI)
concentration, 4–28 g l–1 for adsorbent dosage, 0–48 hours for contact time and 2–12 for pH value
of the solution. The values of each parameter are normalized between 0 and 1. A code for
implementing the DE algorithm is developed using MATLAB 7. The convergence plots to obtain
optimized parametric values are shown in Figures 14(a–d).
The adsorption capacity is found to be maximum when the initial concentration, contact time,
pH and the adsorbent dosage are 200 mg l–1, 20 hours, 3.89 and 2.8 g l–1, respectively. The
adsorption capacity is more when the initial concentration and contact time are maximum, that is,
200 mg l–1 and 20 hours, respectively. Experimentally, it is found that the adsorption capacity
increases with the increase in the initial concentration and contact time, which supports the results
obtained from optimization study. The low values of pH and adsorbent mass are also supported
by the experimental results of individual parameters.
The optimization studies carried out in this study are helpful to obtain the optimum parametric
values without performing the extensive experimental studies. These values are further utilized to
design the continuous adsorption system.
86 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

Population size = NP

Target Xa Xb
Xc
vector
+
(Xa − Xb)
F
F(Xa − Xb)

Noisy
random +
vector Xc = Xc + F(Xa − Xb)

CR Xt Trial vector

Crossover

Cost of
Xi > Xt

Xt

Figure 13. Schematic diagram of DE (Babu 2007).

6. CONCLUSIONS

In the present study, a low-cost nanoporous adsorbent (nANB) is developed using neem bark (low-
cost material) and is used for the removal of Cr(VI) from aqueous solution and plating industry
wastewater. The SEM analysis supports the availability of nano-sized pores on the surface of
nANB. Adsorption capacity is found to increase with an increase in contact time and initial Cr(VI)
concentration. However, it decreases with an increase in the adsorbent dosage and initial pH of the
solution. The experimental equilibrium data are best fitted with the Langmuir isotherm model,
which supports the monolayer formation of Cr(VI) on nANB surface. The maximum adsorption
capacity of the adsorbate is obtained as 26.95 mg g–1, which is higher than many of the adsorbents
presently available. The kinetic data are well explained by the second-order kinetic model. The
adsorption of Cr(VI) on nANB is more favourable at low pH values. The form of chromium that
favours the adsorption process at lower pH is HCrO4–. The developed nANB is found to be efficient
for the treatment of plating industry effluent, which contains other ions and impurities along with
Cr(VI). It is found that the adsorption process of Cr(VI) removal using nANB is controlled by both
film and intra-particle diffusion mechanisms. The developed second-order polynomial equation is
found appropriate for expressing the adsorption capacity as a function of dependent parameters
such as initial Cr(VI) concentration, adsorbent dosage, contact time and initial pH of the solution.
The developed modelling equation is optimized using the DE technique and the optimal values of
initial Cr(VI) concentration, contact time, pH and the adsorbent dosage for maximum adsorption
capacity are obtained as 200 mg l–1, 20 hours, 3.89 and 2.8 g l–1, respectively.
Removal of Cr(VI) from Wastewater Using Neem Bark 87

All evalution values over parameter 1 All evalution values over parameter 2
200 200
Evaluation value

Evaluation value
150 150
100 100
50 50
0 0
−50 −50
−100 −100
−150 −150
0 0.2 0.4 0.6 0.8 1 0.2 0.4 0.6 0.8 1
next Parameter 1 prev. next Parameter 2

All evalution values over parameter 3 All evalution values over parameter 4
200 200

Evaluation value
Evaluation value

150 150
100 100
50 50
0 0
−50 −50
−100 −100
−150 −150
0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 1
prev. next Parameter 3 prev. next Parameter 4

Figure 14. Convergence plot for various parameters versus the evaluated value of q. (a) Parameter 1 (initial concentration),
(b) Parameter 2 (pH), (c) Parameter 3 (contact time) and (d) Parameter 4 (adsorbent dosage).

ACKNOWLEDGEMENTS

The authors thank the University Grants Commission (UGC), New Delhi, India, and the
Department of Science and Technology (DST), New Delhi, India, for financial assistance.

REFERENCES

Alaerts, G.J., Jitjaturant, V. and Kelderman, P. (1989) Water Sci. Technol. 21, 1701.
Babel, S. and Kurniawan, T.A. (2003) J. Hazard. Mater. 97, 219.
Babu, B. V. (2007) Process Plant Simulation, Oxford University Press, New Delhi, p. 272–277.
Babu, B.V. and Gupta, S. (2008b) J. Environ. Eng. Sci. 7, 553.
Babu, B.V. and Gupta, S. (2008a) Adsorption. 3, 85.
Bakiler, M., Maslov, I.V. and Akyüz, S. (1999) J. Mol. Struct. 476, 21.
Dakiky, M., Khamis, M., Manassra, A. and Mer’eb, M. (2002) Adv. Environ. Res. 6, 533.
Daneshvar, N., Salari, D. and Aber, S. (2002) J. Hazard. Mater. 94, 49.
Depci, T., Kul, A.R. and Önal, Y. (2012) Chem. Eng. J. 200–202, 224.
Guo, J., Li, Y., Dai, R. and Lan, Y. (2012) J. Hazard. Mater. 243, 265.
Gupta, S. and Babu, B.V. (2009a) J. Environ. Manage. 90, 3013.
Gupta, S. and Babu, B.V. (2009b) Chem. Eng. J. 150, 352.
Gupta, S. and Babu, B.V. (2009c) Bioresour. Technol. 100, 5633.
Gupta, S. and Babu, B.V. (2010) J. Water Resour. Prot. 2, 706.
88 U. Maheshwari and S. Gupta/Adsorption Science & Technology Vol. 33 No. 1 2015

Hameed, B.H., Mahmoud, D.K. and Ahmad, A.L. (2008) J. Hazard. Mater. 158, 65.
Hossain, M.A., Kumita, M. and Mori, S. (2010) Afr. J. Pure Appl. Chem. 4, 135.
Kapur, M. and Mondal, M.K. (2013) Chem. Eng. J. 218, 138.
Kumar, P.S. and Kirthika, K. (2009) J. Eng. Sci. Technol. 4, 351.
Liao, M.-H. and Chen, D.-H. (2002) J. Mater. Chem. 12, 3654.
Malash, G.F. and El-Khaiary, M.I. (2010) Chem. Eng. J. 163, 256.
Malkoc, E. and Nuhoglu, Y. (2007) Sep. Purif. Technol. 54, 291.
Memon, J.R.,Memon, S.Q., Bhanger, M.I., El-Turki, A., Hallam, K.R. and Allen, G.C. (2009) Colloids Surf.,
B. 70, 232.
Moreno-Piraján, J., Garcia-Cuello, V. and Giraldo, L. (2011) Adsorption. 17, 505.
Muthukumaran, K. and Beulah, S. (2011) Procedia Environ. Sci. 4, 266.
Namasivayam, C. and Ranganathan, K. (1993) Environ. Pollut. 82, 255.
Namasivayam, C. and Sureshkumar, M.V. (2008) Bioresour. Technol. 99, 2218.
Namasivayam, C. and Yamuna, R.T. (1995) Chemosphere. 30, 561.
Orhan, Y. and Büyükgüngör, H. (1993) Water Sci. Technol. 28, 247.
Selvaraj, K., Manonmani, S. and Pattabhi, S. (2003) Bioresour. Technol. 89, 207.
Sharma, D.C. and Forster, C.F. (1994) Bioresour. Technol. 47, 257.
Sharma, D.C. and Forster, C.F. (1996) Process Biochem. 31, 213.
Taha, M.R. and Ibrahim, A.H. (2014) J. Environ. Chem. Eng. 2, 1.
Tan, W.T., Ooi, S.T. and Lee, C.K. (1993) Environ. Technol. 14, 277.
Wan Ngah, W.S. and Hanafiah, M.A.K.M. (2008) Bioresour. Technol. 99, 3935.
Weber, W.J.J. and Morris, J.C. (1963) J. Sanit. Eng. Div. 89, 31.
Yu, L.J., Shukla, S.S., Dorris, K.L., Shukla, A. and Margrave, J.L. (2003) J. Hazard. Mater. 100, 53.