Journal of Colloid and Interface Science 310 (2007) 411–424

www.elsevier.com/locate/jcis

Adsorptive accumulation of Cd(II), Co(II), Cu(II), Pb(II), and Ni(II)

from water on montmorillonite: Influence of acid activation
Krishna G. Bhattacharyya ∗ , Susmita Sen Gupta
Department of Chemistry, Gauhati University, Guwahati, Assam, India
Received 18 November 2006; accepted 26 January 2007
Available online 26 March 2007

Abstract
The present work investigates the influence of acid activation of montmorillonite on adsorption of Cd(II), Co(II), Cu(II), Ni(II), and Pb(II) from
aqueous medium and comparison of the adsorption capacities with those on parent montmorillonite. The clay–metal interactions were studied
under different conditions of pH, concentration of metal ions, amount of clay, interaction time, and temperature. The interactions were dependent
on pH and the uptake was controlled by the amount of clay and the initial concentration of the metal ions. The adsorption capacity of acid-activated
montmorillonite increases for all the metal ions. The interactions were adsorptive in nature and relatively fast and the rate processes more akin
to the second-order kinetics. The adsorption data fitted both Langmuir and Freundlich isotherms, indicating that strong forces were responsible
for the interactions at energetically nonuniform sites. The Langmuir monolayer capacity of the acid-activated montmorillonite is more than that
of the parent montmorillonite (Cd(II): 32.7 and 33.2 mg/g; Co(II): 28.6 and 29.7 mg/g; Cu(II): 31.8 and 32.3 mg/g; Pb(II): 33.0 and 34.0 mg/g;
and Ni(II): 28.4 and 29.5 mg/g for montmorillonite and acid-activated montmorillonite, respectively). The thermodynamics of the rate processes
showed the adsorption of Co(II), Pb(II), and Ni(II) to be exothermic, accompanied by decreases in entropy and Gibbs free energy, while the
adsorption of Cd(II) and Cu(II) was endothermic, with an increase in entropy and an appreciable decrease in Gibbs free energy. The results have
established the potential use for montmorillonite and its acid-activated form as adsorbents for Cd(II), Co(II), Cu(II), Ni(II), and Pb(II) ions from
aqueous media.
© 2007 Elsevier Inc. All rights reserved.

Keywords: Montmorillonite; Acid-activated montmorillonite; Adsorption kinetics; Adsorption isotherm; Temperature; Enthalpy; Metal ions

1. Introduction toms of Cu(II) poisoning, particularly after drinking beverages

The multifold increase in the use of heavy metals over the
past few decades has inevitably resulted in an increased flux
of metallic substances into the environment. Under favorable
pH and other conditions, metals go into solution in water, con-
taminating natural water bodies. Cd(II), Co(II), Cu(II), Pb(II),
and Ni(II) are among the toxic metals: Cd(II) is known to cause
lung insufficiency, bone lesions, and hypertension [1]; Co(II)
affects the respiratory system, the kidneys, and the gastroin-
testinal tract, and Co dermatitis and sensitization as a result
of dermal exposure have been well documented [2]; gastroin-
testinal distress is one of the reported health effects of Cu(II),
and vomiting, nausea and abdominal pain are common symp-
* Corresponding author. Fax: +91 361 2700311.
E-mail address: krishna2604@sify.com (K.G. Bhattacharyya).
0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2007.01.080
stored in copper or untinned brass containers [3]; low levels of
Pb(II) have been identified with anemia, as it causes injury to
the blood-forming system, while high levels cause severe dys-
function of the kidneys, the liver, and the central and peripheral
nervous system and high blood pressure [4]; the respiratory
tract and immune systems are sensitive targets of Ni(II) toxi-
city, as it causes chronic bronchitis, emphysema, and impaired
lung function, and the immune response to nickel is elicited as
allergic contact dermatitis in humans [5]. Most of these metals
have been known as carcinogens or co-carcinogens.
Considering the threat of these toxic metals to humans and
all other living beings, an appropriate solution must be found
either by removing them completely from water or by bring-
ing them below the maximum permissible level. A good num-
ber of technologies are available for removal of heavy metals
from solution, namely reverse osmosis, chemical precipitation,
412 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424
ion exchange, solvent extraction, etc. But adsorption has be-
come a preferred method for removal, recovery, and recycling
of toxic heavy metals from wastewater [6]. Different conven-
tional and nonconventional adsorbents have been reported for
removal of various metal ions, viz., red mud [7], activated car-
bon [8], tree fern [9], sewage sludge [10], sawdust [11], sil-
ica [12], bone char [13], rice husk [14], bagasse fly ash [15],
resin [16], polymetallic sea nodules [17], modified zeolite [18],
spirogyra bioadsorbent [19], etc.
The clay minerals in soils play the role of natural scavengers
in removing and accumulating contaminants in water passing
through the soil by ion exchange and adsorption. High spe-
cific surface area, chemical and mechanical stability, layered
structure, high cation exchange capacity (CEC), tendency to
hold water in the interlayer sites, and the presence of Brøn-
sted and Lewis acidity have made clays excellent adsorbent
materials [20,21]. A good number of studies describing the ad-
sorption of various metals onto natural and modified clays have
been reported. Mellah and Chegrouche [22] used natural ben-
tonite for uptake of zinc from aqueous solution and reported
the maximum Langmuir monolayer capacity as 52.91 mg/g
for an initial Zn(II) concentration of 300 mg/L at 293 K.
The bentonite–Zn(II) interactions were exothermic and were
favored by high initial Zn(II) concentration and decreased par-
ticle size of clay adsorbent. The removal of Cr(III), Cr(VI), and
Ag(I) by bentonite was reported by Khan et al. [23], where
the adsorption of Cr(III) and Ag(I) onto bentonite was found
to be exothermic and Cr(VI)–bentonite interactions were en-
dothermic. Sepiolite has been used as an adsorbent for Cd(II),
Cu(II), and Zn(II) [24]. Yuvaz et al. [25] used raw kaolin for
removing Mn(II), Co(II), Ni(II), and Cu(II) from aqueous so-
lution, where the adsorption affinity was in the order Cu(II) >
Ni(II) > Co(II) > Mn(II) and the monolayer adsorption capac-
ity was 10.78, 1.67, 0.92, and 0.45 mg/g, respectively, at 298 K.
Alvarez-Ayuso and Garcia-Sanchez [26] used natural and Na-
exchanged bentonites for the removal of Cr(III), Ni(II), Zn(II),
Cu(II), and Cd(II) from aqueous solution. The authors reported
that the Langmuir model was most suitable for explaining the
results and the maximum adsorption capacity was 49.8, 24.2,
23.1, 30.0, and 26.2 mg/g, respectively, for Cr(III), Ni(II),
Zn(II), Cu(II), and Cd(II). The process, favored by increasing
pH, was shown to be very rapid, with >95% uptake in 30 min.
Strawn et al. [27] reported the adsorption of Cu(II) by smectites
(montmorillonite and bedellite) as a function of varying ionic
strength and pH. The X-ray adsorption fine structure (XAFS)
and electron paramagnetic resonance (EPR) spectroscopic ex-
periments on Cu(II) adsorbed on clays revealed that Cu–Cu
linkages in the multinuclear complexes were 2.65 Å apart, hav-
ing a coordination number near one. Lin and Juang [28] tried
to modify the adsorptive properties of montmorillonite by in-
troducing sodium dodecyl sulfate before Cu(II) and Zn(II) ad-
sorption. The modified clay had a higher affinity toward the
metal ions and equilibrium was attained within 120 min. The
interactions were endothermic with pseudo-first-order kinetics,
yielding rate constants of 6.64 × 10−4 and 3.14 × 10−3 min−1
at 298 K for Zn(II) and Cu(II), respectively. Kaolinite and its
modified forms have been successfully used for adsorption of
Cr(VI) from water [29]. The adsorption was strongly dependent
on the pH of the medium, with Cr(VI) uptake increasing from
pH 1.0 to 7.0, after which the uptake decreases and the process
attains equilibrium within 240 min. The Langmuir monolayer
capacity of the clay adsorbents was in the range from 10.6 to
13.9 mg/g.
The present work has been undertaken to investigate the use
of montmorillonite and its acid-activated form for removal of
Cd(II), Co(II), Cu(II), Pb(II), and Ni(II) from aqueous solution
by adsorption under various environmental conditions in single-
batch systems.
2. Materials and method
2.1. Clay adsorbent
The parent clay, montmorillonite, SWy-2 (M1), was ob-
tained from the University of Missouri-Columbia, Source Clay
Minerals Repository, USA. The clay was modified by treating
it with 0.25 M H2 SO4 (E. Merck, Mumbai, India) to acid-
activated montmorillonite (M2) using the procedure of Espan-
taleon et al. [30]. A quantity of 20 g of montmorillonite was
refluxed with 200 ml of 0.25 M H2 SO4 for 3 h. The resulting
activated clay was centrifuged and washed with water several
times till it was free of SO2−
4 and dried at 383 K in an air oven
until constant weight was attained. Both the clays were calcined
at 773 K for 10 h before they were used as adsorbents.
2.2. XRD measurement
A Phillips Analytical X-ray spectrometer (PW 1710) using
CuKα radiation was used to characterize the adsorbents.
2.3. FTIR measurement
FTIR spectra of the adsorbents were taken with a Perkin–
Elmer FTIR Spectrometer (Model Spectrum RXI, range 4400–
440 cm−1 ).
2.4. Surface area
The surface areas of the clay adsorbents were estimated fol-
lowing Sears’ method [31]. A quantity of 0.5 g of an adsorbent
was acidified with 0.1 N HCl to pH 3–3.5. The volume was
made up to 50 ml with distilled water after addition of 10.0 g of
NaCl. The titration was carried out with standard 0.1 M NaOH
in a thermostatic bath at 298 ± 0.5 K to pH 4.0, and then to
pH 9.0. The volume, V , required to raise the pH from 4.0 to 9.0
was noted and the surface area was computed from the follow-
ing equation:
S (m2 /g) = 32V − 25. (1)
2.5. Cation exchange capacity
The copper bis-ethylenediamine complex method has been
used to estimate the CEC of the clays [32]. A quantity of 50 ml
 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424
of 1 M CuCl2 solution was mixed with 102 ml of 1 M ethyl-
enediamine solution to allow the formation of the [Cu(en)2 ]2+
complex. A slight excess of the amine ensures complete forma-
tion of the complex. The solution is diluted with water to 1 L to
give a 0.05 M solution of the complex. A quantity of 0.5 g of
dry adsorbent was mixed with 5 ml of the complex solution in
a 100-ml flask and diluted with distilled water to 25 ml and the
mixture was agitated for 30 min in a thermostatic water bath
shaker and centrifuged. The concentration of the complex re-
maining in the supernatant was determined by mixing 5 ml of
it with 5 ml of 0.1 M HCl to destroy the [Cu(en)2 ]2+ complex,
followed by adding 0.5 g KI per ml and then titrating iodomet-
rically with 0.02 M Na2 S2 O3 in the presence of starch as an
indicator. The CEC was calculated from the formula
CEC (meq/100 g) = MSV (x − y)/1000m,
where M is the molar mass of the complex, S the concentration
of the thio solution, V the volume (ml) of the complex taken
for iodometric titration, m the mass of adsorbent taken (g),
x the volume (ml) of thio required for blank titration (without
the adsorbent), and y the volume (ml) of thio required for the
titration (with the adsorbent).
2.6. Metal adsorbate
The adsorption experiments were conducted using water
spiked with Cd(II), Co(II), Cu(II), Pb(II), and Ni(II), sep-
arately. Stock solutions containing 1000 mg of metal ion
per liter were prepared by dissolving appropriate amounts of
Cd(NO3 )2 ·4H2 O (Qualigens, Mumbai, India), Co(NO3 )2 ·6H2 O
(E. Merck, Mumbai, India), CuSO4 ·5H2 O (Qualigens, Mum-
bai, India), Pb(NO3 )2 (Glaxo, Mumbai, India), and Ni(NO3 )2 ·
6H2 O (Qualigens, Mumbai, India) separately in double-distilled
water and was used to prepare the adsorbate solutions of the
required concentrations. The pH of the aqueous solution of
Cd(II), Co(II), Cu(II), Pb(II), and Ni(II) as prepared was 5.5,
5.8, 5.7, 5.7, and 5.7, respectively.
2.7. Adsorption experiments
Batch adsorption experiments were carried out in 100-ml
Erlenmeyer flasks by mixing together a constant amount of ad-
sorbent with a constant volume of the aqueous solution of metal
ions. The contents in the flasks were agitated by placing them
in a water bath thermostat for a known time interval. The mix-
ture was then centrifuged and the concentration of metal ions
remaining unadsorbed in the supernatant liquid was determined
by atomic absorption spectrometry (Varian SpectrAA 220 with
air–acetylene oxidizing flame). The pH of the adsorptive solu-
tion was adjusted by addition of 0.01 N NaOH or 0.01 N HNO3
as needed. The following conditions were maintained for dif-
ferent sets of experiments:
2.7.1. Effect of pH
Clay 2 g/L, metal ion concentration 50 mg/L, tempera-
ture 303 K, interaction time 240 min (Cd, Co), 360 min (Cu),
180 min (Pb, Ni), pH 1.0–10.0 (Cd, Co, Ni), 1.0–6.0 (Cu, Pb).
413
2.7.2. Kinetics
Clay 2 g/L, metal ion concentration 50 mg/L, temperature
303 K, interaction time 20, 40, 60, 90, 120, 180, 240, 300,
360 min, pH 5.5 (Cd), 5.8 (Co), 5.7 (Cu, Pb, Ni).
2.7.3. Effects of adsorbent amount
Clay 2, 3, 4, 5, 6 g/L, metal ion concentration 50 mg/L,
temperature 303 K, interaction time 240 min (Cd, Co), 360 min
(Cu), 180 min (Pb, Ni), pH 5.5 (Cd), 5.8 (Co), 5.7 (Cu, Pb, Ni).
2.7.4. Effects of adsorbate concentration
Clay 2 g/L, metal ion concentration 10, 20, 30, 40, 50 mg/L,
temperature 303 K, interaction time 240 min (Cd, Co), 360 min
(Cu), 180 min (Pb, Ni), pH 5.5 (Cd), 5.8 (Co), 5.7 (Cu, Pb, Ni).
(2) 2.7.5. Adsorption isotherm
Clay 2 g/L, metal ion concentration 10, 20, 30, 40, 50, 75,
100, 150, 200, 250 mg/L, temperature 303 K, interaction time
240 min (Cd, Co), 360 min (Cu), 180 min (Pb, Ni), pH 5.5 (Cd),
5.8 (Co), 5.7 (Cu, Pb, Ni).
2.7.6. Adsorption thermodynamics
Clay 2 g/L, metal ion concentration 10, 20, 30, 40, 50 mg/L,
temperature 303, 308, 313 K, interaction time 240 min (Cd,
Co), 360 min (Cu), 180 min (Pb, Ni), pH 5.5 (Cd), 5.8 (Co),
5.7 (Cu, Pb, Ni).
2.8. Theoretical basis
Two isotherm equations were used to find out the relation
between the equilibrium concentrations of the adsorbate in the
liquid phase and in the solid phase. These isotherms [33,34] are
as follows:
(a) Freundlich isotherm:
qe = Kf Cen , (3)
where Ce and qe are the equilibrium concentration of metal
ions in the liquid phase and in the solid phase, respectively,
Kf and n being Freundlich coefficients.
(b) Langmuir isotherm:
Ce /qe = 1/(bqm ) + (1/qm )Ce , (4)
where b and qm are Langmuir coefficients representing the
equilibrium constant for the adsorbate–adsorbent equilib-
rium and the monolayer capacity. The linear Freundlich and
Langmuir plots are obtained by plotting (i) log qe vs log Ce
and (ii) Ce /qe vs Ce respectively, from which the adsorp-
tion coefficients could be evaluated.
The adsorption kinetics was tested with respect to the fol-
lowing well-known models:
(a) Pseudo-first-order kinetics using the Lagergren equation
[35,36]:
ln(qe − qt ) = ln qe − k1 t, (5)
414 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424
where qe and qt are the values of amount adsorbed per unit
mass at equilibrium and at any time t, and k1 is the pseudo-
first-order adsorption rate constant. The values of k1 can be
obtained from the slope of the linear plot of log(qe − qt )
vs t .
(b) The second-order kinetics is applied when it is found that
ln qe is not equal to the intercept of the first-order plot as
obtained from Eq. (5) by using the equation [37]
 
t/qt = 1/ k2 qe2 + (1/qe ) · t, (6)
where k2 qe2 is described as the initial adsorption rate as
t → 0. The plot of t/qt vs t gives a straight line, which
allows computation of qe and k2 .
(c) When there is a possibility of diffusion of adsorbate species
into the interior of the pores of the adsorbent, the intraparti-
cle diffusion [38] rate constant (ki ) is obtained from the
equation:
qt = ki · t 0.5 . (7)
Intraparticle diffusion plays a significant role in controlling
the kinetics of the adsorption process, if the plots of qt vs
t 0.5 yield straight lines passing through the origin with the
slope giving the rate constant, ki .
(d) The liquid film diffusion model [39] given by
ln(1 − F ) = −kfd t (8)
is applicable when the flow of the reactant from the bulk
liquid to the surface of the adsorbent determines the rate
constant F and the fractional attainment of equilibrium
(qt /qe ) and kfd is the adsorption rate constant. A linear plot
of − ln(1 − F ) vs t with zero intercept suggests that the ki-
netics of the adsorption process is controlled by diffusion
through the liquid film surrounding the solid adsorbent.
The thermodynamic parameters for the adsorption process,
H (kJ/mol), S (J/(K mol)), and G (kJ/mol), are evalu-
ated using the equations [40]
G = −RT ln Kd , (9)
ln Kd = −G/RT = S/R − H /RT , (10)
where K d = the distribution coefficient of the adsorbate (= qe /
Ce in L/g), T = the temperature (K), and R = the universal
gas constant (8.314 × 10−3 kJ/(K mol)). The plot of ln Kd vs
1/T should be linear with the slope (−H /R) and the in-
tercept (S/R) giving the values of H and S. The Gibbs
energy change, G, is the fundamental criterion of spontane-
ity. Processes occur spontaneously at a given temperature if G
is a negative quantity. All these relations are valid when the en-
thalpy change remains constant in the range of temperatures
considered.
3. Results and discussion
3.1. Characterization of clay adsorbents
3.1.1. XRD study
Acid treatment and calcination introduce some changes into
the crystal structure of clay minerals. The intensity of the char-
acteristic XRD peaks of the clay minerals are reduced on acid-
ification and the peaks become widened [41]. This implies
changes in the regular pattern of the clay structure along with
partial destruction compared to the original. The dispersion and
amorphization of the acid-treated clay minerals are known to
give rise to an increase in the very-low-angle diffraction inten-
sities [42] and a reduction in the corresponding tip widths [43].
The XRDs of both the clay adsorbents are given in Fig. 1. The
following observations can be drawn from the XRD measure-
ments:
(i) For the calcined, acid-activated montmorillonite, the basal
spacing expanded from 4.44 to 4.47 Å (2θ = 19.98◦ ),
which was accompanied by a decrease in intensity from
43.06 to 29.09%.
(ii) The intensity of most of the XRD peaks of montmoril-
lonite decreased sharply on acid treatment, so that both
octahedral and tetrahedral sites might have been affected
drastically.
(iii) Acid activation of montmorillonite yielded two new peaks
at 22.91 Å (2θ = 3.85◦ ) and 12.49 Å (2θ = 7.06◦ ), which
were absent in the untreated montmorillonite. Appearance
of new peaks indicated the formation of expansible phases
and interlamellar expansion [44].
(iv) The tip width of the 19.98◦ peak (2θ ) in montmorillonite
decreased from 0.32 to 0.30 (acid-activated montmoril-
lonite).
The XRD measurements showed that activation with 0.25 M
H2 SO4 influenced the structural properties of montmorillonite.
3.1.2. FTIR study
XRD results have already demonstrated some significant
changes brought about by modification of montmorillonite
through acid treatment. The FTIR measurements have lent addi-
tional support where assignment of bands to different structural
features is done as per Farmer [45]. The FTIRs of the clays are
given in Fig. 2. This is summarized below:
(i) The number of OH-stretching bands is the same for both
montmorillonite (M1) and acid-activated montmorillonite
(M2) with very little deviation of the band positions.
However, all the OH-stretching bands have reduced in-
tensities after acid treatment (3512 cm−1 : 76.5 to 65.4%;
3554 cm−1 : 77.3 to 66.6%; 3620 cm−1 : 74.8 to 64.8%
with the band shifting to 3621 cm−1 , etc.).
(ii) The SiO-stretching band (1033 cm−1 ) for M2 occurs at the
same position as that of M1, but the intensity decreases
from 59.3 to 49.6%.
 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424
(a)
(b)
Fig. 1. XRD of (a) montmorillonite M1 and (b) acid-activated montmorillonite
M2.
(iii) The SiO-bending band at 464 cm−1 decreases in intensity
from 72.1% (M1) to 61.1% after acid activation. The band
at 420 cm−1 (71.4%) shifts to 428 cm−1 with a decrease
in intensity to 64.0%.
(iv) The band at 722 cm−1 due to the presence of quartz in
M1 did not change its position in M2; however, the band
showed a decrease in intensity from 69.8 (M1) to 60.5%
(M2).
Montmorillonite has two characteristic FTIR regions [46–
48], viz., (i) 3500–3750 cm−1 (due to the surface structural
415
(a)
(b)
Fig. 2. FTIR spectra of (a) montmorillonite M1 and (b) acid-activated montmo-
rillonite M2.
OH groups of layered aluminosilicates and adsorbed water) and
(ii) 400–1150 cm−1 (due to lattice vibrations). Ravichandran
and Sivasankar [48] have demonstrated that the IR bands of
montmorillonite and vermiculite activated with HCl did not ex-
hibit any significant change from those of the parent clays. The
results in the present work are similar, but although principal IR
frequencies show only minor shifts, the intensities of the bands
are modified significantly. This has indicated that acid treatment
brings about considerable interaction between H+ ions and the
clay surface. It is also to be inferred that the interaction is taking
place with those surface groups whose IR frequencies exhibit a
416 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424

maximum shift. The SiO-bending frequencies show maximum

shifts (∼8 cm−1 ), while the OH-stretching, OH-bending, and
SiO-stretching bands have not shifted much due to acid treat-
ment.

3.1.3. Surface area

The specific surface areas of montmorillonite and acid-acti-
vated montmorillonite were measured as 19.8 and 52.3 m2 /g,
respectively. Ravichandran and Sivasankar [48] reported a spe-
cific surface area of 19.0 m2 /g for montmorillonite, which on
treatment with HCl (0.1–0.7 M) increased to 188.3 m2 /g. Such
high values of specific surface area were not achieved in the
present work by treatment with 0.25 M H2 SO4 . The acid treat-
ment opens up the edges of the platelets and as a consequence,
the surface area and the pore diameter increase [43], which is
in conformity with the results obtained in this work. Kara et
al. [49] reported that an increase in the surface area of sepiolite (a)
upon acid activation followed by calcination could be attributed
to the removal of water molecules, both those formed during
acid activation and those inherently present as crystal water.

3.1.4. Cation exchange capacity (CEC)

The CEC of montmorillonite was measured as 153.0 meq/
100 g in agreement with the reported values [50]. On acid treat-
ment, the CEC increases to 341.0 meq/100 g.
Brønsted and Lewis acid sites are associated with the inter-
lamellar region and the edge sites of clays respectively. The ion-
exchange capacity of clay minerals is attributed to structural
defects, broken bonds, and structural hydroxyl transfers [51].
Treatment of the clay with 0.25 M H2 SO4 results in replace-
ment of a number of different cations with H+ ions and on
subsequent heating and calcination, dehydroxylation occurs,
leaving behind a number of Lewis sites. Much of the increase
in CEC in the present work is likely to be due to increase in
Lewis acidity, as the acid-treated clay was calcined at 773 K (b)
before CEC measurement. The CEC increase may be a com-
plex process involving these effects as well.

3.2. Adsorption of metal ions

3.2.1. Effects of pH
It is not possible to carry out adsorption experiments with
Cu(II) and Pb(II) at pH >6.0, or for Ni(II) and Co(II) at pH
>8.0, due to precipitation of the metals as the hydroxides,
which introduces uncertainty into the interpretation of the re-
sults. The precipitation was very rapid for Cu(II) and Pb(II) at
pH >6.0 and the experiments had to be terminated at pH 6.0.
Adsorption of Cd(II) could be studied over the pH range from
1.0 to 10.0 without any abrupt changes in adsorption. The gen-
eral trend in the extent of adsorption (%) as well as amount
(c)
adsorbed (qe ) with increase in pH is positive. Fig. 3 explains
the trend in the amount of metal ions adsorbed (qe ) with re-
Fig. 3. Influence of pH on the amount of metal ions adsorbed for adsorp-
spect to pH. The acid activation increases the number of sites tion of Cd(II), Co(II), Cu(II), Pb(II), and Ni(II) on montmorillonite, M1 and
responsible for adsorption of various metal species and at any acid-activated montmorillonite, M2: (a) Cu(II) and Pb(II), (b) Co(II) and Ni(II),
pH, the acid-activated montmorillonite has a higher adsorption (c) Cd(II) at equilibrium time (clay 2 g/L, initial metal concentration 50 mg/L,
capacity than the nonactivated clay. temperature 303 K).
 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424 417

Table 1
Effect of interaction time for adsorption of Cd(II), Co(II), Cu(II), Pb(II), and
Ni(II) on montmorillonite (M1) and acid-activated montmorillonite (M2) at the
natural pH (clay 2 g/L, initial metal ion concentration 50 mg/L, temperature
303 K)
Adsorbate Clay ad- Equilibrium Extent of adsorption qe
sorbent time (%) (mg/g)
(min) 40 min Equilibrium 40 min Equilibrium
Cd(II) M1 240 79.2 86.2 19.8 21.6
M2 84.8 92.2 21.2 23.0

Co(II) M1 240 50.4 63.1 12.6 15.8

M2 58.0 69.5 14.5 17.4

Cu(II) M1 360 74.9 83.9 18.7 21.0

M2 78.3 87.5 19.6 21.9

Pb(II) M1 180 77.9 86.9 19.5 21.7

M2 84.5 90.8 21.1 22.7

Ni(II) M1 180 50.9 62.8 12.7 15.7

M2 57.4 69.2 14.4 17.3

At very low pH, the number of H3 O+ ions exceeds that

of metal ions by several times and the metal ions can hardly (a) (b)
compete with H3 O+ ions for the binding sites on the clay adsor-
bents. With an increase in pH, the concentration of H3 O+ ions Fig. 4. Lagergren pseudo-first-order plots for adsorption of Cd(II), Co(II),
Cu(II), Ni(II), and Pb(II) on (a) montmorillonite M1 and (b) acid-activated
decreases and some of the sites become available to the metal
montmorillonite M2 at the natural pH of the metal solution (clay 2 g/L, ini-
ions. As the acidity decreases, more and more H3 O+ ions on tial metal ion concentration 50 mg/L, temperature 303 K).
the clay surface are replaced by metal ions and metal species
(such as Cd2+ , Cd(OH)+ , and Cd(OH)2 [52]; Co2+ , Co(OH)+ ,
Table 2
Co(OH)2 , Co(OH)3 , and Co(OH)4 [53]; Cu2+ , Cu(OH)+ , and The first-order and second-order rate coefficients for adsorption of Cd(II),
+ −
Cu2 (OH)2+ 2+
2 [54]; Pb , Pb(OH) , Pb(OH)2 , Pb(OH)3 , and Co(II), Cu(II), Pb(II), and Ni(II) on montmorillonite (M1) and acid-activated
2− 2+ +
Pb(OH)4 [55]; and Ni and NiOH [56]). This happens at a montmorillonite (M2) at the natural pH (clay 2 g/L, initial metal ion con-
centration 50 mg/L, temperature 303 K, units of k1 and k2 are min−1 and
comparatively lower pH (∼6.0) for Cu(II) [57] and Pb(II) [58],
g/(mg min), respectively)
while the adsorption could be safely carried out to pH ∼10.0
Metal Clay Pseudo-first order Pseudo-second order
for Cd(II) [59] and ∼8.0 for Ni(II) [60] and Co(II) [49].
ion adsorbent k1 × 102 r k2 × 102 r
3.2.2. Effects of interaction time and kinetics of adsorption Cd(II) M1 1.9 −0.74 3.1 +0.99
M2 2.4 −0.99 11.1 +1.00
For all the metal ions, adsorption is very fast initially and
the maximum uptake is recorded within 40 min, which slowly Co(II) M1 2.7 −0.92 4.7 +0.99
approaches equilibrium. Table 1 gives details of the increase in M2 2.6 −0.93 5.4 +0.99
the extent of adsorption (%) as well as the amount adsorbed Cu(II) M1 2.8 −0.98 15.4 +0.99
per unit mass (qe ) for all the metal ions at their respective equi- M2 3.1 −0.98 15.9 +0.99
libria. The initial uptake rate for all the metals is very high, as Pb(II) M1 2.8 −0.99 8.4 +0.99
a large number of adsorption sites are available for adsorption M2 4.0 −0.97 11.3 +0.99
at the onset of the process. As the sites are gradually filled up,
Ni(II) M1 3.1 −0.97 5.3 +0.99
adsorption becomes slow and the kinetics becomes more de- M2 3.2 −0.96 5.5 +0.99
pendent on the rate at which the adsorptive is transported from
the bulk phase to the actual adsorption sites [61].
The rates of adsorption have been tested first with the pseu- plots does not establish a first-order mechanism for the interac-
do-first-order mechanism of Lagergren by plotting log(qe − qt ) tions [62]. This is observed when qe values obtained from the
vs t (min) (Fig. 4). The regression coefficients (r ∼ −0.74 to Lagergren plots are compared with the experimental qe values
−0.99) and the first-order rate constants obtained from these (Table 3).
plots (values from 1.9 × 10−2 to 3.1 × 10−2 min−1 for mont- The first-order kinetics is therefore inappropriate to ex-
morillonite M1 and 2.4 × 10−2 to 4.0 × 10−2 min−1 for acid- plain the rate processes. Second-order kinetic plots (t/qt vs t )
activated montmorillonite M2) are given in Table 2. All the (Fig. 5) had better linearity (r ∼ +0.99, Table 2) and appeared
plots are linear, showing that the interactions might be follow- to give a better understanding of the interactions. Acid acti-
ing a first-order mechanism and that the rate is dependent on vation enhances the second-order rate constant for adsorption
the initial concentration. However, linearity of the Lagergren of the metal ions for all the cases (values from 3.1 × 10−2
418 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424

(a) (b)

(a) (b)
Fig. 6. Intraparticle diffusion plots for adsorption of Cd(II), Co(II), Cu(II),
Ni(II), and Pb(II) on (a) montmorillonite M1 and (b) acid-activated montmoril-
Fig. 5. Second-order plots for adsorption of Cd(II), Co(II), Cu(II), Ni(II), and lonite M2 at the natural pH of the metal solution (clay 2 g/L, initial metal ion
Pb(II) on (a) montmorillonite M1 and (b) acid-activated montmorillonite M2 at concentration 50 mg/L, temperature 303 K).
the natural pH of the metal solution (clay 2 g/L, initial metal ion concentration
50 mg/L, temperature 303 K).
Table 4
The intraparticle and liquid film diffusion rate coefficients for adsorption of
Table 3 Cd(II), Co(II), Cu(II), Pb(II), and Ni(II) on montmorillonite (M1) and acid-
Experimental and computed qe values from Lagergren and second-order plots activated montmorillonite (M2) at the natural pH (clay 2 g/L, initial metal
for adsorption of Cd(II), Co(II), Cu(II), Pb(II), and Ni(II) on montmorillonite ion concentration 50 mg/L, temperature 303 K, units of ki and kfd are mg/
(M1) and acid-activated montmorillonite (M2) at the natural pH (clay 2 g/L, (g min0.5 ) and min−1 , respectively)
initial metal ion concentration 50 mg/L, temperature 303 K) Metal Clay Intraparticle diffusion Liquid film diffusion
Metal Clay qe (mg/g) ion adsorbent ki × 10 Intercept r kfd × 102 Intercept r
ion adsorbents Experi- Lagergren Second Deviation (%)
Cd(II) M1 2.1 +18.5 +0.88 1.9 +1.8 +0.99
mental order Lagergren Second order M2 2.7 +19.4 +0.90 2.4 +1.6 +0.99
Cd(II) M1 21.6 3.5 18.1 −83.8 −16.2 Co(II) M1 4.3 +9.8 +0.96 2.6 +0.4 +0.92
M2 23.0 4.7 23.4 −79.5 +1.7 M2 4.0 +11.9 +0.96 1.8 +1.0 +0.99
Co(II) M1 15.8 11.9 16.6 −24.6 +5.0 Cu(II) M1 2.1 +17.9 +0.81 2.8 +1.6 +0.98
M2 17.4 10.2 18.1 −41.2 +4.0 M2 2.1 +18.8 +0.80 3.1 +0.2 +0.98
Cu(II) M1 21.0 4.1 21.3 −77.1 +1.2 Pb(II) M1 3.3 +17.3 +0.93 2.8 +1.1 +0.99
M2 21.9 4.4 22.2 −80.4 −1.3 M2 2.5 +19.4 +0.91 3.8 +0.9 +0.98
Pb(II) M1 21.7 7.4 22.3 −65.8 +2.7 Ni(II) M1 4.2 +10.1 +0.95 3.0 +0.3 +0.97
M2 22.7 10.2 23.2 −55.0 +2.2 M2 4.2 +11.6 +0.95 2.5 +0.7 +0.97
Ni(II) M1 15.7 12.7 16.5 −19.1 +5.0
M2 17.3 12.9 18.1 −25.4 +4.6
The plots (qt vs t 0.5 ) for intraparticle diffusion similarly
yield linear curves (Fig. 6, r ∼ +0.80 to +0.96) and the val-
to 15.4 × 10−2 g/(mg min) for montmorillonite M1 and 5.4 × ues of the rate constant, ki , are given in Table 4. Significantly,
10−2 to 15.9 × 10−2 g/(mg min) for acid-activated montmo- the plots do not have a zero intercept (range: +9.8 to +18.5
rillonite M2). The experimental qe values and those obtained for montmorillonite M1 and +11.6 to +19.4 for acid-activated
from the slopes of the second-order plots now show much bet- montmorillonite M2), indicating that the diffusion of metal
ter agreement (Table 3). Some deviations still exist that might species into the pores is not the dominating factor controlling
be due to the actual process not being in conformity with simple the mechanism of the process. Similar results are obtained with
first-order or second-order kinetics. the liquid film diffusion model, involving diffusion from the
 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424 419

creasing adsorbent amount provides greater surface area or ad-

(a) (b)
Fig. 7. Liquid film diffusion plots for adsorption of Cd(II), Co(II), Cu(II), Ni(II),
and Pb(II) on (a) montmorillonite M1 and (b) acid-activated montmorillonite
M2 at the natural pH of the metal solution (clay 2 g/L, initial metal ion con-
centration 50 mg/L, temperature 303 K).
bulk liquid phase to the surface of the adsorbent. The plots of
− ln(1 − F ) vs t are linear (Fig. 7, r ∼ +0.92 to +0.99) with
intercepts of +0.3 to +1.8 for montmorillonite M1 and +0.2
to +1.6 for acid-activated montmorillonite M2 (Table 4). The
model requires the curves to pass through the origin and al-
though the intercepts are not zero, their small values signify
some role played by diffusion from the liquid phase to the ad-
sorbent surface. The film diffusion rate coefficients are given in
Table 4.
3.2.3. Effects of adsorbent amount
In all the cases, the extent of adsorption (%) increases
rapidly with increase in the adsorbent amount. This is to be
expected because, for a fixed initial solute concentration, in-
sorption sites [63]. However, the amount adsorbed per unit mass
(qe ) decreases for both the clay adsorbents. A large amount of
adsorbent effectively reduces the unsaturation of the adsorption
sites and, correspondingly, the number of such sites per unit
mass comes down, resulting in comparatively less adsorption at
higher adsorbent amount.
For example, the amount of Cd(II) adsorbed per unit mass of
clay decreased from 21.6 to 8.2 mg/g and 23.0 to 8.3 mg/g for
montmorillonite and acid-activated montmorillonite, respec-
tively, as the clay amount was changed from 2.0 to 6.0 g/L.
The corresponding extent of adsorption increased from 86.2 to
98.5% and from 92.2 to 99.9% in the same order. Similar trends
were observed in the cases of clay–Co(II), clay–Cu(II), clay–
Pb(II), and clay–Ni(II) systems. The variation in the extent of
adsorption (%) and the amount adsorbed per unit mass (qe ) for
all clay–metal systems is given in Table 5. Similar observations
have been reported earlier (Pb(II) on sawdust [64], Cd(II) on
river bed sediment [65], etc.).
3.2.4. Effects of initial metal ion concentration
The number of metal ions adsorbed per unit mass (qe ) of
clay increases gradually with increase in metal concentration,
whereas the extent of adsorption (%) decreases with increasing
metal ion loading (Table 6). For example, by changing the ini-
tial concentration of Cd(II) from 10 to 50 mg/L, qe increases
from 4.7 to 21.6 mg/g and 5.0 to 23.0 mg/g for montmoril-
lonite (M1) and acid-activated montmorillonite (M2), respec-
tively, whereas the extent of adsorption (%) decreases from 94.9
to 86.2% and from 99.7 to 92.2%, respectively, for M1 and M2.
Similar trends have been found in all other cases.
At low initial metal ion concentration, the ratio of the num-
ber of metal ions to the number of available adsorption sites
is small and consequently the adsorption is independent of the
initial concentration, but as the concentration of metal ions in-
creases, the situation changes and the competition for adsorp-
tion sites becomes fierce [66]. At high concentration of metal,
a unit mass of the adsorbent is exposed to a larger number of
metal ions and a progressively higher number of metal ions are
taken up with the gradual filling up of the appropriate bind-

Table 5
Effect of clay amount on extent of adsorption (%) and amount of metal ion adsorbed per unit mass of clays (qe , mg/g) for adsorption of Cd(II), Co(II), Cu(II),
Pb(II), and Ni(II) on montmorillonite (M1) and acid-activated montmorillonite (M2) at the natural pH (initial metal concentration 50 mg/L, temperature 303 K) at
clay–metal equilibrium
Clay Clay amount Cd(II) Co(II) Cu(II) Pb(II) Ni(II)
adsorbent (g/L) % qe % qe % qe % qe % qe
M1 2 86.2 21.6 63.1 15.8 83.9 21.0 86.9 21.7 62.8 15.7
3 91.0 15.2 68.4 11.4 87.8 14.6 91.5 15.3 67.9 11.3
4 95.2 11.9 73.5 9.2 91.4 11.4 95.4 11.9 72.4 9.1
5 98.2 9.8 77.8 7.8 94.7 9.5 98.9 9.9 76.2 7.6
6 98.5 8.2 78.7 6.6 97.7 8.1 99.1 8.3 78.3 6.5

M2 2 92.2 23.0 69.5 17.4 87.5 21.9 90.8 22.7 69.2 17.3
3 95.2 15.9 74.7 12.5 91.4 15.2 94.0 15.7 74.5 12.4
4 97.7 12.2 79.6 10.0 94.9 11.8 97.2 12.3 79.3 9.9
5 99.6 10.0 83.8 8.4 98.0 9.8 99.2 9.9 83.4 8.3
6 99.9 8.3 84.8 7.1 99.3 8.3 99.9 8.3 85.5 7.1
420 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424

Table 6
Effect of initial concentration of adsorbates on extent of adsorption (%) and amount of metal ion adsorbed per unit mass of clays (qe ) for adsorption of Cd(II),
Co(II), Cu(II), Pb(II), and Ni(II) on montmorillonite (M1) and acid-activated montmorillonite (M2) at the natural pH (clay 2 g/L, temperature 303 K) at clay–metal
equilibrium
Clay Metal ions Cd(II) Co(II) Cu(II) Pb(II) Ni(II)
adsorbent (mg/L) % qe % qe % qe % qe % qe
M1 10 94.9 4.7 84.6 4.2 96.6 4.8 95.3 4.8 84.3 4.2
20 92.2 9.2 78.1 7.8 92.5 9.3 92.3 9.2 77.8 7.8
30 89.5 13.4 71.9 10.8 88.9 13.3 90.4 13.6 71.6 10.7
40 87.9 17.6 67.0 13.4 89.0 17.2 88.7 17.7 66.7 13.3
50 86.2 21.6 63.1 15.8 83.9 21.0 86.9 21.7 62.8 15.7
M2 10 99.7 5.0 91.8 4.6 99.9 5.0 99.9 5.0 91.5 4.6
20 98.1 9.8 85.0 8.5 96.1 9.6 98.1 9.8 84.7 8.5
30 95.4 14.3 78.7 11.8 92.5 13.9 95.8 14.4 78.4 11.7
40 93.9 18.8 73.5 14.7 89.6 17.9 92.0 18.4 73.2 14.6
50 92.2 23.0 69.5 17.4 87.5 21.9 90.8 22.7 69.2 17.3

(a) (b)
Fig. 8. Isotherm plots (qe vs Ce ) for adsorption of Cd(II), Co(II), Cu(II), Ni(II),
and Pb(II) on (a) montmorillonite M1 and (b) acid-activated montmorillonite
M2 at the natural pH of the metal solution (clay 2 g/L, initial metal ion con-
centration 10, 20, 30, 40, 50, 75, 100, 150, 200, 250 mg/L, temperature 303 K)
at clay–metal equilibrium.
(a) (b)
Fig. 9. Freundlich isotherm for Cd(II), Co(II), Cu(II), Ni(II), and Pb(II) on (a)
montmorillonite M1 and (b) acid-activated montmorillonite M2 at the natural
pH of the metal solution (clay 2 g/L, initial metal ion concentration 10, 20,
30, 40, 50, 75, 100, 150, 200, 250 mg/L, temperature 303 K) at clay–metal
equilibrium.

lonite and acid-activated montmorillonite have lower uptake of

ing sites. This gives rise to an increase in qe , although the
percent adsorption comes down. Similar results have been re-
ported earlier for metal ion adsorption on different adsorbents
(e.g., Cd(II) on activated carbon from coconut coir pith [67];
Cu(II) on maple sawdust [61]; Ni(II) and Cd(II) on bagasse fly
ash [68], etc.).
3.2.5. Adsorption isotherm
For initial metal ion concentrations of 10, 20, 30, 40, 50,
75, 100, 150, 200, and 250 mg/L, the plots of qe vs Ce for
metal–clay interactions for a fixed amount of adsorbent (2 g/L)
at 303 K are presented in Fig. 8. It is seen that both montmoril-
Ni(II) and Co(II) than of Cd(II), Cu(II), and Pb(II).
The empirical Freundlich isotherm yields linear plots (r ∼
+0.93 to +0.98, Fig. 9) and the values of the coefficients, n
(all values <1.0) and Kf (4.5 to 9.2 mg1−1/n L1/n /g for mont-
morillonite and 5.9 to 12.9 mg1−1/n L1/n /g for acid-activated
montmorillonite, Table 7) indicate that the clays have a good
potential to be used as adsorbents for the metal ions Cd(II),
Co(II), Cu(II), Ni(II), and Pb(II). It is to be noted that the
Freundlich isotherm applies to adsorption on nonspecific and
heterogeneous sites on solid surfaces, but no definite mecha-
nism could be arrived at. The isotherm is valid for weak van
der Waals type adsorption as well as for strong chemisorption.
 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424 421

Table 7
Freundlich and Langmuir coefficients for adsorption of Cd(II), Co(II), Cu(II),
Pb(II), and Ni(II) on montmorillonite (M1) and acid-activated montmorillonite
(M2) at the natural pH (clay 2 g/L, initial metal concentration 10, 20, 30, 40,
50, 75, 100, 150, 200, 250 mg/L, temperature 303 K, units of Kf , qm and b are
mg 1−1/n L1/n /g, mg/g and L/g, respectively) at clay–metal equilibrium
Metal Clay Freundlich coefficients Langmuir coefficients
ion adsorbent Kf n r qm b r
Cd(II) M1 8.6 0.3 +0.93 32.7 191.7 +0.99
M2 12.9 0.2 +0.97 33.2 225.4 +0.99

Co(II) M1 4.6 0.4 +0.97 28.6 72.1 +0.99

M2 6.0 0.3 +0.98 29.7 100.7 +0.99

Cu(II) M1 9.2 0.3 +0.97 31.8 192.8 +0.99

M2 12.4 0.2 +0.98 32.3 278.0 +0.99

Pb(II) M1 8.9 0.3 +0.93 33.0 206.6 +0.99

M2 11.3 0.3 +0.97 34.0 320.3 +0.99

Ni(II) M1 4.5 0.4 +0.97 28.4 71.5 +0.99

M2 5.9 0.3 +0.98 29.5 100.2 +0.99

The Kf values of acid-activated montmorillonite are higher than (a) (b)

those of the nonactivated montmorillonite.
The Langmuir isotherm is specific to strong monolayer
chemisorption, which might be the case in the present work
since the Langmuir plots have good linearity (Fig. 10, r ∼
+0.99) (Table 7). The Langmuir equilibrium coefficient, b, de-
termines the direction in which the adsorbate–adsorbent equi-
librium
clay(solid phase) + metal(II)(aqueous phase) = clay–metal(II)
is shifted. Large values of b ensure that the equilibrium is pre-
dominantly driven toward the right, leading to the formation of
the adsorbate–adsorbent complex.
From the values of Langmuir monolayer capacity, qm , it is
observed that acid activation did not have much influence on the
values. This suggests that treatment with acid does not increase
the number of adsorption sites to a large extent, although the
treatment influences the strength of the existing sites as revealed
by the equilibrium coefficient (b) data. Langmuir monolayer
capacity, qm , is considerable for both the clays and therefore the
clays can take up large amounts of the metal ions. The qm values
for Cd(II), Co(II), Cu(II), Pb(II), and Ni(II) on montmorillonite
(M1) are 32.7, 28.6, 31.8, 33.0, and 28.4 mg/g and those for
acid-activated montmorillonite (M2) are 33.2, 29.7, 32.3, 34.0,
and 29.5 mg/g in the same order of metal ions.
Similar ranges of adsorption coefficients for metal ions on
various adsorbents are reported in the literature. Adsorption of
Cd(II) on perlite follows the Freundlich isotherm model with
an adsorption capacity of 0.64 mg/g [59]. Yuvuz et al. [25]
have reported a Langmuir monolayer capacity, qm , of 0.919
and 1.470 mg/g at 298 and 313 K, respectively, for Co(II) ad-
sorption on Turkish kaolinite. Langmuir adsorption capacity for
Cu(II) adsorption on natural clinoptilolite and synthetic zeo-
lites has been shown to be 5.91 and 50.5 mg/g, respectively, by
Alvarez-Ayuso et al. [69]. The uptake of Cu(II) on electric fur-
nace slag in the temperature range from 293 to 313 K has Fre-
undlich adsorption capacity Kf = 0.44 to 1.09 mg1−1/n L1/n /g
Fig. 10. Langmuir isotherm for adsorption of Cd(II), Co(II), Cu(II), Ni(II), and
Pb(II) on (a) montmorillonite M1 and (b) acid-activated montmorillonite M2 at
the natural pH of the metal solution (clay 2 g/L, initial metal ion concentra-
tion 10, 20, 30, 40, 50, 75, 100, 150, 200, 250 mg/L, temperature 303 K) at
clay–metal equilibrium.
and Langmuir monolayer capacity qm = 32.68 to 39.22 mg/g,
whereas the uptake of Pb(II) on the electric furnace slag in the
same temperature range had Freundlich and Langmuir mono-
layer capacity in the range from 0.57 to 1.82 mg1−1/n L1/n /g
and 33.78 to 37.04 mg/g, respectively [70]. Lopez et al. [71] re-
ported that the Langmuir adsorption capacity for Ni(II) on red
mud was found as 10.95 mg/g.
3.2.6. Adsorption thermodynamics
Adsorption of metal ions on clay adsorbents shows different
behavior with respect to increase in temperature (Table 8).
Cu(II) adsorption follows an endothermic path. The extent
of adsorption (%) and the amount adsorbed per unit mass of
the adsorbents (qe ) increase appreciably in the temperature
range from 303 to 313 K. Cu(II) uptake changes from 83.9%
(21.0 mg/g) to 89.3% (22.3 mg/g), and 87.5% (21.9 mg/g)
to 92.7% (23.2 mg/g) for montmorillonite (M1) and acid-
activated montmorillonite (M2), respectively. Thus, adsorption
of Cu(II) onto clays has to overcome a small activation energy
barrier and increasing energy supply makes it easier for Cu(II)
to adsorb onto the clay surface. Such activated adsorption fol-
lowing an endothermic path has also been reported earlier
[72,73]. The effect of increasing temperature is quite different
in the case of Cd(II), where increase of temperature from 303
to 308 K is found to enhance the amount of Cd(II) adsorbed per
unit mass of the clays (as well as extent of adsorption, %), but
the trend is reversed between 308 and 313 K. For example, in
the case of montmorillonite, the adsorption of Cd(II) increases
from 21.6 mg/g (86.2%) to 22.8 mg/g (91.1%) on increasing
the temperature from 303 to 308 K, but the adsorption decreases
422 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424

Table 8
Effect of temperature on extent of adsorption (%) and amount of metal ion adsorbed per unit mass of clays (qe , mg/g) for adsorption of Cd(II), Co(II), Cu(II),
Pb(II), and Ni(II) on montmorillonite (M1) and acid-activated montmorillonite (M2) at the natural pH (clay 2 g/L, initial concentration of metal ions 50 mg/L) at
clay–metal equilibrium
Clay Temperature Cd(II) Co(II) Cu(II) Pb(II) Ni(II)
adsorbent (K) % qe % qe % qe % qe % qe
M1 303 86.2 21.6 63.1 15.8 83.9 21.0 86.9 21.7 62.8 15.7
308 91.1 22.8 62.0 15.5 86.9 21.7 79.0 19.8 56.8 14.2
313 89.7 22.4 59.8 15.0 89.3 22.3 74.7 18.7 51.3 12.8
M2 303 92.2 23.0 69.5 17.4 87.5 21.9 90.8 22.7 69.2 17.3
308 93.1 23.3 68.3 17.1 89.9 22.5 84.3 21.1 61.5 15.4
313 92.6 23.2 66.1 16.5 92.7 23.2 79.1 19.8 55.5 13.9

Table 9 now almost in the region of chemisorption enthalpies. The

Thermodynamic data for adsorption of Cd(II), Co(II), Cu(II), Pb(II), and Ni(II) trends are similar for exothermic adsorption of Ni(II) (mont-
on montmorillonite (M1) and acid-activated montmorillonite (M2) at the nat- morillonite and acid-activated montmorillonite: H −45.2 and
ural pH (clay 2 g/L, initial concentration of metal ions 50 mg/L) at clay–metal
equilibrium
−56.9 kJ/mol) and Co(II) (montmorillonite and acid-activated
montmorillonite: H −15.1 and −19.9 kJ/mol). The mag-
Metal Clay H S −G (kJ/mol)
ions adsorbent (kJ/mol) (J/(K mol))
nitude of the adsorption enthalpy, H , indicates moderately
303 K 308 K 313 K Mean
strong bonding between the metal ions and the clay minerals.
Cd(II) M1 +40.2 +147.5 44.6 45.4 46.1 45.4 The endothermic enthalpy change (H ) associated with
M2 +25.0 +107.3 32.5 33.0 33.6 33.0
Cu(II) adsorption decreases on acid activation (montmoril-
Co(II) M1 −15.1 −47.1 14.3 14.5 14.7 14.5 lonite and acid-activated montmorillonite: H +50.7 and
M2 −19.9 −59.3 17.9 18.2 18.5 18.2 +45.8 kJ/mol). Similar endothermic interaction has been ob-
Cu(II) M1 +50.7 +180.7 54.7 55.6 55.9 55.4 served in Cd(II) adsorption (montmorillonite and acid-activated
M2 +45.8 +334.5 101.3 102.8 104.6 102.9 montmorillonite: H +40.2 and +25.0).
Pb(II) M1 −75.5 −235.9 71.4 72.5 73.8 72.5 Decrease in entropy during Co(II), Pb(II), and Ni(II) adsorp-
M2 −83.4 −258.7 78.3 79.5 80.8 79.5 tion helps to stabilize the clay–metal adsorption complex (S
are in the range from −47.1 to −235.9 J/(K mol) for mont-
Ni(II) M1 −45.2 −146.4 44.4 45.1 45.8 45.1
M2 −56.9 −181.8 55.0 55.8 56.8 55.8 morillonite and −59.3 to −258.7 J/(K mol) for acid-activated
montmorillonite). S values for interactions of Cu(II) and
Cd(II) with montmorillonite are +180.7 and +147.5 J/(K mol)
to 22.4 mg/g (89.7%) on further increase of temperature to and those for acid-activated montmorillonite are +334.5 and
313 K. The process is controlled by the adsorbate–adsorbent +107.3 J/(K mol), respectively. The endothermic interactions
and adsorbate–adsorbate forces and it is clear from the results are driven by entropy increase, which illustrates increased ran-
that the first one becomes weak in comparison to the latter as domness at the solid–solution interface indicating strong affin-
the temperature increases. ity of the adsorbent for Cu(II) and Cd(II).
For Co(II), Ni(II), and Pb(II), increase in temperature results The clay–metal interactions take place spontaneously and
in a decrease in the extent of adsorption (%) and the amount are accompanied by decrease in Gibbs energy (Table 9).
of metal adsorbed per unit mass of clay (qe ), showing the in- G values vary from −14.5 to −72.5 kJ/mol for montmo-
teractions to be exothermic. For example, in the temperature rillonite and from −18.2 to −102.9 kJ/mol for acid-activated
range of 303 to 313 K, Co(II) uptake changes from 63.1% montmorillonite, respectively.
(15.8 mg/g) to 59.8% (15.0 mg/g) and 69.5% (17.4 mg/g) Thermodynamic data on metal adsorption on clays are
to 66.1% (16.5 mg/g) for montmorillonite and acid-activated scarce. Yuvaz et al. [25] reported the thermodynamic para-
montmorillonite, respectively. Thus, the adsorbate ions have a meters for adsorption of Cu(II), Ni(II), Co(II), and Mn(II) on
tendency to escape from the solid phase to the solution. More- Turkish kaolinite. H and S values were +39.52 kJ/mol
over, rise in temperature and excess energy supply promote and +11.7 J/(K mol) for Cu(II), +37.27 kJ/mol and +10.7 J/
desorption. Such results are not uncommon for adsorption of (K mol) for Ni(II), +21.52 kJ/mol and +5.4 J/(K mol) for
metal ions from the aqueous phase [74]. Co(II), and +36.73 kJ/mol and +10.1 J/(K mol) for Mn(II).
The thermodynamic adsorption parameters, H , S, and Echeverria et al. [75] have reported that H , S, and G for
G (Table 9), are computed from the plots of ln Kd vs 1/T . adsorption of Ni(II) onto illite have values of +16.8 kJ/mol,
In the case of clay–Pb(II) interaction, the mean adsorption +58 J/(mol K), and −1.04 kJ/mol, respectively. Lin and
enthalpy, H , is −75.5 kJ/mol for montmorillonite, which Juang [28] investigated the adsorption of Cu(II) and Zn(II)
changes to −83.4 kJ/mol after acid activation. The exothermic onto montmorillonite modified with sodium dodecyl sulfate.
enthalpy is higher after acid activation, with a clear indica- The values of H , S, and G were reported as 7.05 kJ/mol,
tion that Pb(II) ions are held more strongly by acid-activated 9.09 J/(K mol), and −9.66 kJ/mol, respectively, for Cu(II) and
montmorillonite. H for acid-activated montmorillonite is 7.39 kJ/mol, 6.39 J/(K mol), and −9.17 kJ/mol, respectively,
 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424
for Zn(II) at 298 K. These values are not much different from
the values obtained in this work.
4. Conclusions
Montmorillonite and its acid-activated forms are capable of
removing metal ions, namely Cd(II), Co(II), Cu(II), Ni(II), and
Pb(II), from aqueous solution. Acid activation enhances the ad-
sorption capacity compared to the untreated clay mineral due to
increased surface area and pore volume. Adsorption increases
with pH till the metal ions are precipitated out in alkaline solu-
tion.
The rate of uptake of the metal ions by the clays is very
high initially, followed by a low rate indicating the entry of
the metal ions into the interior of the adsorbent particles as
the interactions proceed. The processes are much more akin to
second-order kinetics. The uptake of metal ions depends appre-
ciably on the amount of clay. The amount adsorbed per unit
mass (qe ) and the net adsorption (%) show reverse trends with
increase in adsorbent amount.
Both Langmuir and Freundlich isotherms yield good fits
and the adsorption coefficients agree well with the condi-
tions supporting favorable adsorption. All the metal–clay in-
teractions are thermodynamically favorable. Clay–Pb(II), clay–
Ni(II), and clay–Co(II) interactions are exothermic in nature,
whereas clay–Cu(II) and clay–Cd(II) are endothermic. The
clay–metal interactions are spontaneous in all cases supported
by decrease in Gibbs energy.
Acknowledgment
One of the authors (S.S.G.) is grateful to the University
Grants Commission, New Delhi, for providing assistance for
this work under the FIP scheme.
References
[1] ATSDR, Toxicological profile for cadmium, Agency for Toxic Substances
and Disease Registry (http://www.atsdr.cdc.gov), U.S. Department of
Health and Human Services, Atlanta, GA, 1999.
[2] ATSDR, Draft toxicological profile for cobalt, Agency for Toxic Sub-
stances and Disease Registry (http://www.atsdr.cdc.gov), U.S. Department
of Health and Human Services, Atlanta, GA, 2001.
[3] ATSDR, Draft toxicological profile for copper, Agency for Toxic Sub-
stances and Disease Registry (http://www.atsdr.cdc.gov), U.S. Department
of Health and Human Services, Atlanta, GA, 2002.
[4] ATSDR, Toxicological profile for lead, Agency for Toxic Substances and
Disease Registry (http://www.atsdr.cdc.gov), U.S. Department of Health
and Human Services, Atlanta, GA, 1999.
[5] ATSDR, Draft Toxicological profile for nickel, Agency for Toxic Sub-
stances and Disease Registry (http://www.atsdr.cdc.gov), U.S. Department
of Health and Human Services, Atlanta, GA, 2003.
[6] H.H. Tran, F.A. Roddick, J.A. O’Donnell, Water Res. 33 (1999) 2992.
[7] V.K. Gupta, M. Gupta, S. Sharma, Water Res. 35 (2001) 1125.
[8] R.L. Ramos, J.L.A. Bernal, B.J. Mendoza, R.L. Fuentes, C.R.M. Guerrero,
J. Hazard. Mater. B 90 (2002) 27.
[9] Y.S. Ho, J.F. Porter, G. McKay, Water Air Soil Pollut. 141 (2002) 1.
[10] S.-C. Pan, C.-C. Lin, D.-H. Tseng, Resour. Conserv. Recycl. 39 (2003) 79.
[11] L.J. Yu, S.S. Shukla, K.L. Dorris, A. Shukla, J.L. Margrave, J. Hazard.
Mater. B 100 (2003) 53.
[12] N. Chiron, R. Guilet, E. Deydier, Water Res. 37 (2003) 3079.
423
[13] D.C.K. Ko, C.W. Cheung, K.K.H. Choy, J.F. Porter, G. McKay, Chemo-
sphere 54 (2004) 273.
[14] C.R.T. Tarley, M.A.Z. Arruda, Chemosphere 54 (2004) 987.
[15] V.K. Gupta, I. Ali, J. Colloid Interface Sci. 271 (2004) 321.
[16] V.K. Gupta, P. Singh, N. Rahman, J. Colloid Interface Sci. 275 (2004) 398.
[17] S. Maity, S. Chakravarty, S. Bhattacharjee, B.C. Roy, Water Res. 39
(2005) 2579.
[18] U. Wingenfelder, B. Nowack, G. Furrer, R. Schulin, Water Res. 39 (2005)
3287.
[19] V.K. Gupta, A. Rastogi, V.K. Saini, N. Jain, J. Colloid Interface Sci. 296
(2006) 59.
[20] K. Tanabe, in: J.R. Anderson, M. Boudart (Eds.), Catalysis—Science and
Technology, Springer-Verlag, New York, 1981.
[21] R. Naseem, S.S. Tahir, Water Res. 35 (2001) 3982.
[22] A. Mellah, S. Chegrouche, Water Res. 31 (1997) 621.
[23] S.A. Khan, R. Rehman, M.A. Khan, Waste Manage. 15 (1995) 271.
[24] A.G. Sanchez, E.A. Ayuso, O.J. De Blas, Clay Miner. 34 (1999) 469.
[25] O. Yuvaz, Y. Altunkayank, F. Guzel, Water Res. 37 (2003) 948.
[26] E. Alvarez-Ayuso, A. Garcia-Sanchez, Clays Clay Miner. 51 (2003) 475.
[27] D.G. Strawn, N.E. Palmer, L.J. Furnare, C. Goodell, J.E. Amonette, R.K.
Kukkadapu, Clays Clay Miner. 52 (2004) 321.
[28] S.-H. Lin, R.-S. Juang, J. Hazard. Mater. B 92 (2002) 315.
[29] K.G. Bhattacharyya, S. Sen Gupta, Ind. Eng. Chem. Res. 45 (2006) 7232.
[30] A.G. Espantaleon, J.A. Nieto, M. Fernandez, A. Marsal, Appl. Clay Sci.
24 (2003) 105.
[31] G.W. Sears, Anal. Chem. 28 (1956) 1981.
[32] F. Bergaya, M. Vayer, Appl. Clay Sci. 12 (1997) 275.
[33] H.M.F. Freundlich, J. Phys. Chem. 57 (1906) 385.
[34] I. Langmuir, J. Am. Chem. Soc. 40 (1918) 1361.
[35] S. Lagergren, Kungl. Svenska Vetensk. Handl. 24 (1898) 1.
[36] Y.S. Ho, Scientometrics 59 (2004) 171.
[37] Y.-S. Ho, C.-C. Chiang, Y.-C. Hsu, Sep. Sci. Technol. 36 (2001) 2473.
[38] W.J. Weber, J.C. Morris, J. Sanit. Engg. Div. Am. Soc. Civ. Eng. 89 (1963)
31.
[39] G.E. Boyd, A.W. Adamson, L.S. Myers Jr., J. Am. Chem. Soc. 69 (1947)
2836.
[40] W.J. Thomas, B. Crittenden, in: Adsorption Technology and Design,
Butterworths–Heinemann, Oxford, 1998, chap. 1.
[41] M. Radojevc, V. Jovic, D. Karaulic, D. Vitorovic, J. Serb. Chem. Soc. 67
(2002) 499.
[42] G. Jozefaciuk, G. Bowanko, Clays Clay Miner. 50 (2002) 771.
[43] F.R.V. Diaz, P.D.S. Santos, Quim. Nova 24 (2001) 345.
[44] D.S. Fanning, V.Z. Keramidas, M.A. El-Desosky, in: J.B. Dixon, S.B.
Weed (Eds.), Minerals in Soil Environments, in: SSSA Book Series, vol. 1,
Soil Science Society of America, Medison, WI, 1989.
[45] V.C. Farmer, in: H. van Olphen, J.J. Fripiat (Eds.), Data Handbook for
Clay Materials and Other Non-Metallic Minerals, Pergamon Press, Ox-
ford, 1979.
[46] M.L. Hair, Infrared Spectroscopy in Surface Chemistry, Marcel Dekker,
New York, 1967.
[47] H. Suquet, S. Chevalier, C. Marcilly, D. Barthomeuf, Clay Miner. 26
(1991) 49.
[48] J. Ravichandran, B. Sivasankar, Clays Clay Miner. 45 (1997) 854.
[49] M. Kara, H. Yuzer, E. Sabah, M.S. Celik, Water Res. 37 (2003) 224.
[50] R.E. Grim, Clay Mineralogy, McGraw Hill, New York, 1968.
[51] M.G.F. Rodrigues, Ceramica 49 (2003) 146.
[52] C.-H. Lai, C.-Y. Chen, B.-L. Wei, S.-H. Yeh, Water Res. 36 (2002) 4943.
[53] Y.-H. Wang, S.-H. Lin, R.-S. Juang, J. Hazard. Mater. B 102 (2003) 291.
[54] S.T. Bosso, J. Enzweiler, Water Res. 36 (2002) 4795.
[55] V.C. Taty-Costodes, H. Fauduet, C. Porte, A. Delacroix, J. Hazard. Mater.
B 105 (2003) 121.
[56] B. Bayat, J. Hazard. Mater. B 95 (2002) 251.
[57] Y. Nuhoglu, E. Oguz, Process Biochem. 38 (2003) 1627.
[58] C.K. Jain, D. Ram, Water Res. 31 (1997) 154.
[59] T. Mathialagan, T. Viraraghavan, J. Hazard. Mater. 94 (2002) 291.
[60] V. Padmavathy, P. Vasudevan, S.C. Dhingra, Process Biochem. 38 (2003)
1389.
424 K.G. Bhattacharyya, S.S. Gupta / Journal of Colloid and Interface Science 310 (2007) 411–424
[61] B. Yu, Y. Zhang, A. Shukla, S.S. Shukla, K.L. Dorris, J. Hazard. Mater.
B 80 (2000) 33.
[62] Y.S. Ho, G. McKay, Adsorp. Sci. Technol. 20 (2002) 797.
[63] S. Rengaraj, K.-H. Yeon, S.-Y. Kang, J.-U. Lee, K.-W. Kim, S.-H. Moon,
J. Hazard. Mater. B 92 (2002) 185.
[64] B. Yu, Y. Zhang, A. Shukla, S.S. Shukla, K.L. Dorris, J. Hazard. Mater.
B 84 (2001) 83.
[65] C.K. Jain, M.K. Sharma, Water Air Soil Pollut. 137 (2002) 1.
[66] H. Ucun, Y.K. Bayhan, Y. Kaya, A. Cakici, O.F. Algurb, Desalination 154
(2003) 233.
[67] K. Kadirvalu, C. Namasivayam, Adv. Environ. Res. 7 (2003) 471.
[68] V.K. Gupta, C.K. Jain, I. Ali, M. Sharma, V.K. Saini, Water Res. 37 (2003)
4038.
[69] E. Alvarez-Ayuso, A. Garcia-Sanchez, X. Querol, Water Res. 37 (2003)
4855.
[70] L. Curkovic, S. Cerjan-Stefanovic, A. Rastovean-Mioe, Water Res. 35
(2001) 3436.
[71] E. Lopez, B. Soto, M. Arias, A. Nunez, D. Rubinos, M.T. Barral, Water
Res. 32 (1998) 1314.
[72] A. Shukla, Y.-H. Zhang, P. Dubey, J.L. Margrave, S.S. Shukla, J. Hazard.
Mater. 95 (2002) 137.
[73] Y.-S. Ho, Water Res. 37 (2003) 2323.
[74] M.M. Abou-Mesalam, Colloids Surf. A Physicochem. Eng. Aspects 225
(2003) 85.
[75] J. Echeverria, J. Indurain, E. Churio, J. Garrido, Colloids Surf. A Physic-
ochem. Eng. Aspects 218 (2003) 175.