Int. J. Environ. Sci. Technol.

(2013) 10:231–242
DOI 10.1007/s13762-012-0112-0

ORIGINAL PAPER

Adsorption isotherms, kinetics and mechanism for the adsorption

of cationic and anionic dyes onto carbonaceous particles prepared
from Juglans regia shell biomass
S. Nethaji • A. Sivasamy • A. B. Mandal

Received: 4 November 2011 / Revised: 25 May 2012 / Accepted: 2 August 2012 / Published online: 16 November 2012
Ó CEERS, IAU 2012

Abstract In the present study, Juglans regia shells were Keywords Adsorption  Malachite green  Amido black 
used to prepare activated carbon by acid treatment method. Intraparticle diffusion  Boyd plot
J. regia shell-based activated carbon was used for the
adsorption of two synthetic dyes namely, a basic dye mal-
achite green and an acid dye amido black 10B. The prepared Introduction
adsorbent was crushed and sieved to three different mesh
sizes 100, 600 and 1,000 lm. The adsorbent was charac- Underground water is the major source of drinking water.
terized by scanning electron microscopy, surface acidity This underground water is being polluted by the wastewater
and zero-point charge. Batch experiments were carried out containing a large number of contaminants like acids, bases,
by varying the parameters like initial aqueous phase pH, toxic organics, inorganics, heavy metals (Repo et al. 2010),
adsorbent dosage and initial dye concentration. The equi- dissolved solids and colors disposed by leather and textile
librium data were tested with Langmuir, Freundlich, industries. Out of all such contaminants, color is the most
Redlich–Peterson and Sips isotherm at three different visible pollutant and the presence of very minute quantity
temperatures 293, 300 and 313 K and it was found that the makes it undesirable for use. Some of the dyes are toxic,
Freundlich isotherm best fitted the adsorption of both the stable and non-biodegradable (Gupta et al. 2004). Thus, the
dyes. Kinetic data were tested with pseudo first-order model removal of color from dye-bearing effluents becomes a
and pseudo second-order model. The mechanism for the major challenge due to the difficulty in treating such
adsorption of both the dyes onto the adsorbent was studied wastewaters by conventional treatment methods such as
by fitting the kinetic data with intraparticle diffusion model chemical and biological oxidation methods (Vadivelan and
and Boyd plot. External mass transfer was found to be the Vasanth kumar 2005). However, adsorption technique
rate-determining step. Based on the ionic nature of the proved to be an efficient and economical process for the
adsorbates, the extent of film diffusion and intraparticle treatment of these dye-bearing effluents. But the efficiency
diffusion varied; both being system specific. Thermody- of the process lies in choosing the suitable adsorbent. The
namic parameters were also calculated. Finally, the process chosen adsorbent should be easily available, cheap and
parameters of each adsorption system were compared to should have no economic value. For the past two decades, a
develop the understanding of the best suitable system. large variety of waste materials particularly from industrial
and agricultural waste products, whose disposal has been a
problem, have been successfully utilized as adsorbents for
Electronic supplementary material The online version of this treating the industrial effluents (Bhatnagar and Sillanpaa
article (doi:10.1007/s13762-012-0112-0) contains supplementary
material, which is available to authorized users. 2010). Some common waste materials used for this purpose
include rubber seed coat (Rengaraj et al. 2002), pecan shells
S. Nethaji  A. Sivasamy (&)  A. B. Mandal (Shawabke et al. 2002), jute fiber (Senthilkumaar et al.
Chemical Engineering Area, CSIR-Central Leather Research 2005), Indian rosewood sawdust (Garg et al. 2004), olive
Institute (Council of Scientific and Industrial Research),
Adyar, Chennai 600020, India stones (El-Sheikh et al. 2004), pine wood (Tsenga et al.
e-mail: arumugamsivasamy@yahoo.co.in 2003), coir pith (Namasivayam and Kavitha 2002), rice

123
232 Int. J. Environ. Sci. Technol. (2013) 10:231–242

husk (Guo et al. 2005), bamboo (Hameed et al. 2007a), working solutions of the desired concentrations were
rattan sawdust (Hameed et al. 2007a) and oil palm fiber prepared by diluting the stock solutions with distilled
(Tan et al. 2007). water.
The present study is the thriving effort to utilize an
agricultural waste biomass Juglans regia shell as adsor- Analytical measurements
bent. J. regia is mainly used in food industries and its shell
is thrown out as a waste product which is of no economical Unknown concentrations of the dyes were determined by
value. In the present study, this J. regia shells were car- finding out the absorbance at the characteristic wave-
bonized and the activated carbon prepared from J. regia length using a double beam UV–visible spectrophotometer
shells (JSAC) was used for the adsorption of two dyes (Shimadzu UV-2102 PC). Standard calibration charts were
namely malachite green (MG) and amido black 10B (AB). prepared by measuring the absorbance of different dye
MG is a basic dye whereas, AB is an acid dye. The sig- concentrations at (kmax) 619 nm for AB and 655 nm for
nificance of this study is to discuss the variation in mech- MG and unknown concentrations of dyes before and after
anism of the adsorption processes based on the ionic nature adsorption were computed from the calibration charts. The
of the adsorbates. This study was carried out from Jan, pH of the aqueous solutions was measured by Digisun
2011 to May, 2011 at Central Leather Research Institute Electronics System (Digital pH meter model 2001). Sur-
(CSIR-CLRI), Adyar, Chennai, India. face morphology of the activated carbon was studied by
SEM of Hitachi make (model S-3400N).

Materials and methods Equilibrium experiments

Preparation of JSAC Adsorption equilibrium experiments were carried out by

varying the initial concentrations of dyes as 1, 10, 15, 25,
J. regia shells used in the present study were procured from 50, 100, 250, 500, 700 and 1,000 mg/L. The required
small scale food industries in Chennai, India and were amount of the adsorbent was added and the solutions were
crushed into small pieces. The crushed shells were washed agitated for 24 h at 100 rpm at 300 K. The same experi-
with deionized water until all the leachable impurities were mental procedures were repeated at 293 and 313 K.
removed. Then the materials were dried in sunlight for 48 h The amount of adsorbed dyes at equilibrium, qe (mg/g)
to remove moisture. The dried shells were carbonized by was calculated by (Repo et al. 2011a)
treating with concentrated H2SO4 (1:1 ratio) for 48 h. The ðC0  Ce Þv
carbonized J. regia shells were then washed with water qe ¼ ð1Þ
w
until it became neutralized. Then the neutralized activated
carbon was dried in a hot air oven at 140 °C for 48 h. The where Co and Ce (mg/L) are the liquid phase initial and
dried J. regia shell-based activated carbon (JSAC) was equilibrium concentrations of the dyes, respectively. V is
then crushed and sieved to three mesh sizes (100, 600 and the volume of the solution (L), and W is the mass of dry
1,000 lm). adsorbent used (g).
Langmuir and Freundlich isotherm parameters were
Characterization of activated carbon determined by fitting the data with linear regression anal-
ysis using Origin 7.0 software whereas non-linear regres-
The zero-point charge of the prepared activated carbon was sion analysis by Microsoft Office Excel 2007 solver was
determined by solid addition technique (Balistrieri and applied to estimate Sips and Redlich–Peterson isotherm
Murray 1981). The surface functionality of JSAC was parameters.
evaluated by Boehm titration method (Boehm 1966) and
the surface morphology of JSAC was studied by scanning Kinetic experiments
electron microscope.
The kinetics of adsorption of dyes onto JSAC was carried
Preparation of adsorbate solutions out by withdrawing and analyzing the samples at the time
interval of every 5 min for the first 30 min and later at
MG (C52H54N4O12) and AB 10B (C4OH28N7NaO13S4) every 10 min until the consecutive residual dye concen-
were purchased from M/s. s.d. fine chemicals and used trations became closer. The kinetic experiments were car-
without further purification. The stock solution (1,000 mg/L) ried out separately for four different initial dye
of the dyes was prepared by dissolving accurately concentrations such as 25, 50, 75 and 100 mg/L for all the
weighed amount of the dyes in distilled water. All three particle sizes at 300 K.

123
Int. J. Environ. Sci. Technol. (2013) 10:231–242 233

The amount of adsorption at time t was calculated by Influence of initial pH

(Bulut and Ozacar (2008)
The adsorption of MG and AB onto JSAC were studied at
ðC0  Ct Þv
qt ¼ ð2Þ different pH varying from 2.3 to 10.5 for all three particle
w sizes (100, 600 and 1,000 lm). It was found that the
where C0 and Ct (mg/L) are the liquid phase concentrations maximum adsorption occurred at pH 6–7 for MG and at pH
of dyes at initial time and time t, respectively. V is the 2.3 for AB dye. The surface of the adsorbent will be
volume of the solution (L), and W is the mass of dry negatively charged above pHzpc and positively charged
adsorbent used (g). below pHzpc (2.5) (Preethi and Sivasamy 2006). At low pH,
MG will get protonated as –N?(CH3)2, and H? ions from
the acidic solution compete with the positively charged dye
Results and discussions molecules thus inhibiting the adsorption process. But, at
pH range between 6 and 7, the H? ions from the aqueous
Characterization of the adsorbent solution decreased and hence favouring the adsorption of
protonated MG dyes onto the negatively charged adsorbent
The surface acidity of JSAC was determined by Boehm surface. Similar result was reported elsewhere by Hameed
titration method and the total surface acidity was calculated and El-Khaiary (2008). In case of azo dye AB, if the
as 0.3503 meq g-1 with the maximum composition of electrostatic interaction was the only step involved in the
phenolic group (0.270871 meq g-1), with traces of lactonic adsorption process, then, the % adsorption would have
(0.0321 meq g-1) and carboxylic (0.0473 meq g-1) been maximum at pH ranging from 2.5 to 5.5 where the
groups. The zero-point charge (pHzpc) was found to occur adsorbent surface will be negatively charged and the dye
at pH 2.5. The surface morphology of the JSAC was will be protonated as –SO3H. But in the observed process,
studied by SEM and the images were displayed in Fig. 1. JSAC might have interacted with the dye molecules
From these micrographs, pores and scale like structures are through hydrogen bonding and hydrophobic–hydrophobic
evident on the surface of the prepared activated carbon. mechanisms as reported elsewhere by Al-Degs et al. (2008).

Fig. 1 SEM micrograph of

JSAC

123
234 Int. J. Environ. Sci. Technol. (2013) 10:231–242

Hence the adsorption of AB onto JSAC was maximum at Langmuir isotherm

pH 2.3. But the maximum % adsorption was observed to be
higher for MG than AB. The Langmuir equation is applicable to homogeneous
adsorption where the adsorption of each adsorbate mole-
Influence of adsorbent dosage cule on to the surface has equal sorption activation energy.
The linear form of this isotherm is represented by the
Effect of variation of mass of JSAC was studied by expression (Langmuir 1916):
increasing the adsorbent dosage from 0.001/10 to 1 1 1
0.5 g/10 mL for all the three particle sizes. The results ¼ þ ð3Þ
qe qm K L C e qm
showed that % adsorption of both dyes, either basic or acid,
increased with the increase in the adsorbent dosage. The where qe (mg/g) and Ce (mg/L) are the amount of adsorbed
increase in adsorbent dosage appeared to promote more adsorbate per unit weight of adsorbent and unadsorbed
active sites, which facilitates the adsorption of dyes. adsorbate concentration in solution at equilibrium,
However, for the adsorption of MG, the % adsorption respectively. The constant KL (L/g) is the Langmuir equi-
decreased after certain point. This may be due to the librium constant and qm is the theoretical monolayer
decrease in total adsorption surface area available to MG saturation capacity.
resulting from overlapping or aggregation of adsorption The essential feature of the Langmuir isotherm can be
sites (Zhang et al. 2008). It was found that 3 g/L of JSAC expressed in terms of a dimensionless constant called
adsorbed 100 % of 25 mg/L of MG whereas the separation factor (RL, also called equilibrium parameter)
same amount of JSAC could adsorb only around 90 % of which is defined by the following equation:
25 mg/L of AB. Since JSAC was acid treated, it contained 1
RL ¼ ð4Þ
more acidic functional groups on its surface. The basic dye 1 þ KL C0
MG had a good affinity toward the acidic adsorbent surface
where C0 (mg/L) is the initial adsorbate concentration.
when compared to the acid dye AB. Hence the % adsorp-
The value of RL indicates the shape of the isotherms to be
tion of MG was higher than AB.
either unfavorable (RL [ 1), linear (RL = 1), favorable
(0 \ RL \ 1) or irreversible (RL = 0).
Effect of initial dye concentration at different The equilibrium data for the adsorption of both the dyes
temperatures at all three temperatures for all the three particle sizes (100,
600 and 1,000 lm) were fitted with the Langmuir isotherm.
The effect of initial dye concentrations was studied for both It can be inferred from Table 1 that the monolayer satu-
the dyes by varying the dye concentrations as 1, 10, 15, 25, ration capacity (qm) and the Langmuir constant (KL)
50, 100, 250, 500, 700 and 1,000 mg/L at different tem- increased with the increase in temperature (Mittal 2006)
peratures 293, 300 and 313 K and the result suggested that and decreased with the increase in the particle sizes of the
the % adsorption decreased with the increase in initial dye adsorbent (Patil et al. 2011). It can also be found that the
concentrations for both the dyes. Furthermore, adsorption values of qm and KL were higher for the adsorption of MG
increased with an increase in temperature for both the dyes, than AB. This was because; values of qm and KL are
indicating that the processes were endothermic. directly proportional to % adsorption. Since decrease in
the particle size of the adsorbent and the increase in tem-
perature favors % adsorption, the values of qm and KL
Adsorption equilibrium
increased. It was also found that the separation factor (RL)
was between 0 and 1 for both the dyes for all the three
To optimize the design of the adsorption processes, it is
particle sizes indicating that the adsorption processes were
important to establish the most appropriate correlation
favorable.
for the equilibrium curves. Various isotherms such as
Langmuir, Freundlich, Sips and Redlich–Peterson isotherms
Freundlich isotherm
had been used to evaluate the equilibrium characteristics
of the adsorption processes. The parameters of the iso-
The most important multisite adsorption isotherm for het-
therm equations were calculated by linear regression
erogeneous surfaces is the Freundlich adsorption isotherm
analysis. The calculated adsorption parameters and the
and the linear form of this isotherm is expressed as
correlation coefficient (r2) for all the isotherms at 293, 300
(Freundlich 1906):
and 313 K for all the three particle sizes of JSAC for the
adsorption of MG and AB onto JSAC are summarized in 1
log qe ¼ log KF þ log Ce ð5Þ
Tables 1, 2. n

123
 Table 1 Analysis of Langmuir and Freundlich adsorption isotherm parameters for three different particle sizes by linear regression method
Isotherm parameters 100 lm 600 lm 1,000 lm
293 K 300 K 313 K 293 K 300 K 313 K 293 K 300 K 313 K

Adsorption of mg onto JSAC

Langmuir isotherm
Int. J. Environ. Sci. Technol. (2013) 10:231–242

qm (mg/g) 8.947 25.31 29.74 6.45 7.57 7.64 4.53 5.59 7.026
KL 1.09 2.98 7.43 1.04 2.16 6.75 0.08 0.339 1.21
RL 0.477–0.0009 0.504–0.001 0.225–0.0002 0.919–0.011 0.316–0.0004 0.128–0.0001 0.923–0.011 0.746–0.002 0.45–0.0008
2
r 0.9477 0.899 0.9391 0.9191 0.9756 0.9786 0.8982 0.94841 0.9153
Freundlich isotherm
KF (L/g) 2.8942 6.3277 8.8214 0.4566 3.2388 5.1007 1 0.9265 1.367
n 1.8457 2.1699 3.1261 1.4087 2.0539 2.1986 1 1.6581 1.6952
r2 0.9711 0.9589 0.9772 0.9937 0.9956 0.9765 1 0.9913 0.9906
Adsorption of AB10B onto JSAC
Langmuir isotherm
qm (mg/g) 3.20 4.13 4.37 3.04 3.83 4.25 2.64 2.82 3.06
KL 0.3281 1.0141 1.4791 0.2374 0.5903 1.0082 0.1291 0.3887 0.907
RL 0.8387–0.0051 0.4417–0.0007 0.4589–0.0008 0.5755–0.0013 0.5056–0.0010 0.4979–0.0009 0.55464–0.0012 0.3116–0.0004 0.3562–0.0005
r2 0.9435 0.8951 0.9082 0.994 0.9989 0.9942 0.9911 0.9847 0.9884
Freundlich isotherm
KF (L/g) 0.8841 1.2029 1.2502 0.7659 1.0018 1.1256 0.7184 0.8438 0.9214
n 1.7403 1.8929 1.8436 1.7384 1.822 1.8251 1.7134 1.7254 1.7524
r2 0.9985 0.994 0.995 0.9971 0.9899 0.9947 0.998 0.9988 0.999
235

123
236 Int. J. Environ. Sci. Technol. (2013) 10:231–242

Table 2 Analysis of Sips and Redlich–Peterson adsorption isotherm parameters for three different particle sizes by non-linear regression method
Isotherm parameters 100 lm 600 lm 1,000 lm
293 K 300 K 313 K 293 K 300 K 313 K 293 K 300 K 313 K

Adsorption of MG onto JSAC

Redlich–Peterson isotherm
qm (mg/g) 8.11 24.84 25.82 6.07 7.38 7.71 4.29 5.93 6.99
KKR 1.25 3.02 8.29 2.78 3.82 5.72 3.62 4.62 1.19
NKR 0.18 0.39 0.16 0.42 0.29 0.39 0.41 0.53 0.41
v2 0.232 0.422 1.453 2.634 12.432 0.344 1.323 5.322 5.432
Sips isotherm
qm (mg/g) 12.54 30.18 35.72 11.94 14.82 16.82 10.82 12.75 19.73
KS 3.92 4.02 4.92 1.15 3.28 4.97 1.06 1.26 1.47
nS 0.38 0.48 0.51 0.16 0.29 0.42 0.38 0.27 0.11
v2 2.233 1.343 12.453 23.42 12.42 0.432 2.435 1.343 0.435
Adsorption of AB10B onto JSAC
Redlich–Peterson isotherm
qm (mg/g) 3.01 3.92 4.29 2.98 3.63 4.11 2.58 2.61 2.99
KKR 0.86 1.02 1.26 0.62 0.68 1.01 0.16 0.29 0.94
NKR 0.28 0.36 0.18 0.51 0.29 0.39 0.32 0.11 0.42
v2 1.343 14.432 0.422 32.422 0.432 3.12 23.334 0.12 2.41
Sips isotherm
qm (mg/g) 7.92 8.03 10.98 4.82 9.28 12.85 5.02 6.72 7.08
KS 0.78 0.94 1.05 0.62 0.92 1.07 0.69 0.74 0.85
nS 0.41 0.29 0.41 0.49 0.27 0.32 0.29 0.52 0.39
v2 10.83 25.92 11.73 14.63 25.98 11.73 17.38 1.43 0.51

where KF (L/g) is the Freundlich constant and n (g/L) is the isotherm returns to the Langmuir isotherm and predicts
Freundlich exponent. Therefore, a plot of log qe versus log homogeneous adsorption. On the other hand, deviation of
Ce enables the constant and exponent n to be determined. nS value from the unity indicates heterogeneous surface.
Freundlich isotherm was tested for both the dyes at the Adsorption data of both the dyes onto JSAC were tested
three temperatures for all the three adsorbent particle sizes. with Sips isotherm and the parameters obtained are given
From Table 1, it was found that the Freundlich exponent in Table 2. The Sips isotherm constants increased with the
(n) and Freundlich constant (KF) increased with the increase increase in temperature and decrease in particle size. Also
in temperature and decrease in the adsorbent particle sizes. nS deviated widely from unity which again denotes the
It was also inferred that the values of n and KF were higher heterogeneity of the surface of the adsorbent.
for the adsorption of MG than AB. Here the value of n was
between 1 and 10 which again proved that the processes are Redlich–Peterson isotherm
favorable at all the temperature for all three particle sizes.
The Redlich–Peterson isotherm combines the features of
Sips isotherm Langmuir and Freundlich isotherms. It can be represented
as (Repo et al. 2011a):
The Sips isotherm is a combination of the Langmuir and qm ðKRP Ce Þ
Freundlich isotherms and can be derived using either equi- qe ¼ ð7Þ
1 þ ðKRP Ce ÞnRP
librium or thermodynamic approach (Repo et al. 2011a).
where KRP (L/mg) and nRP are Redlich–Peterson constants.
qm ðKRP Ce ÞnS
qe ¼ ð6Þ Redlich–Peterson isotherm parameters are similar to that of
1 þ ðKRP Ce ÞnRP
Sips isotherm parameters.
where KS (L/mg) is the affinity constant and nS describes Redlich–Peterson isotherm was also used to fit the
the surface heterogeneity. When nS equals unity, the Sips equilibrium data of the adsorption of MG and AB onto

123
Int. J. Environ. Sci. Technol. (2013) 10:231–242 237

Table 3 Adsorption kinetic parameters

Kinetic model 100 lm 600 lm 1,000 lm
parameters
25 50 75 100 25 50 75 100 25 50 75 100
mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L

Adsorption of MG onto JSAC

Pseudo first-order model
k1 (1/min) 0.5051 0.2334 0.2065 0.239 0.193 0.0954 0.0822 0.0619 0.0678 0.0348 0.0327 0.0289
qe (mg/g) 0.6342 0.8184 2.1679 8.9182 7.6668 5.7017 17.8749 19.2574 3.7013 5.8784 10.3119 13.6549
r2 1 0.9612 0.9569 0.9726 0.9644 0.8421 0.973 0.9703 0.8768 0.9859 0.9366 0.9184
Pseudo second-order model
k2 (g/mg min) 2.75 0.5246 0.17773 0.0617 0.0365 0.0254 0.0058 0.0022 0.0252 0.0071 0.0022 0.0014
qe (mg/g) 2.5282 5.0684 7.6511 10.3626 3.0703 5.5096 12.4069 13.2978 2.6041 5.2438 8.9928 11.8063
r2 0.9999 0.9999 0.9998 0.9992 0.9955 0.9982 0.9943 0.98 0.9914 0.9939 0.9747 0.985
Intraparticle diffusion model
ki (mg/g min1/2) 0.0319 0.1023 0.2292 0.4772 0.3246 0.3927 0.8719 0.9576 0.1853 0.3023 0.5081 0.6396
C 0.077 0.103 0.274 0.572 0.551 0.743 0.942 1.183 0.584 0.921 1.034 1.324
Adsorption of AB10B onto JSAC
Pseudo first-order model
k1 (1/min) 0.2661 0.1464 0.1025 0.0577 0.061 0.0665 0.0368 0.026 0.0735 0.0383 0.026 0.0238
qe (mg/g) 8.6175 10.6642 13.8101 9.4111 1.8266 4.7077 4.694 5.1008 2.2875 3.1096 5.699 5.6916
2
r 0.99628 0.9459 0.9686 0.987 0.9968 0.9981 0.9924 0.9792 0.9913 0.9964 0.9675 0.9913
Pseudo second-order model
k2 (g/mg min) 0.0579 0.0136 0.0093 0.0066 0.0457 0.0168 0.0112 0.0075 0.0495 0.0179 0.006 0.0063
qe (mg/g) 2.5549 5.1679 6.9252 8.6132 2.386 4.4642 6.1538 7.8492 2.2011 4.1684 6.0679 7.55585
r2 0.996 0.9922 0.9958 0.99965 0.9993 0.9977 0.9996 0.9991 0.999 0.9998 0.9998 0.9995
Intraparticle diffusion model
ki (mg/g min1/2) 0.2817 0.5179 0.5582 0.5852 0.1628 0.3134 0.3101 0.3705 0.1236 0.2093 0.3147 0.3388
C 0.568 0.5207 1.1326 1.7113 0.7842 1.0779 2.0317 2.4952 0.8391 1.193 1.374 2.2129

JSAC. Redlich–Peterson isotherm constants also varied in The kinetic data for the adsorption of MG and AB onto
the same fashion as that of the Sips isotherm constants. The JSAC of three different particle sizes with the four initial
parameters for Redlich–Peterson isotherm for all the three dye concentrations (25, 50, 75 and 100 mg/L) were tested
particle sizes at three temperatures are given in Table 2. with the well-known kinetic models namely pseudo first-
By testing with both the isotherm models, it can be order model and pseudo second-order model. The mecha-
inferred that the value of separation factor (RL) fell nism of the adsorption processes were studied by plotting
between 0 and 1 and the value of the Freundlich exponent the data with intraparticle diffusion model and Boyd
(n) from the Freundlich isotherm was between 1 and 10 plot. The parameters of these kinetic models are given in
which proved that the conducted adsorption processes were Table 3.
favorable. Both these adsorption processes followed the
Freundlich isotherm since the r2 values were closer. Fur- Pseudo first-order equation
ther Redlich–Peterson and Sips isotherm constants (nRP
and nS) also confirmed that the surface of the prepared Lagergren proposed a method for adsorption analysis
adsorbent was heterogeneous in nature. namely pseudo first-order kinetic equation. The linear form
of this equation is (Santhy and Selvapathy 2006)
Adsorption kinetics lnðqe  qt Þ ¼ ln qe  k1 t ð8Þ

The kinetic studies were carried out by withdrawing and where qe (mg/g) and qt (mg/g) are the amounts of adsorbed
analyzing the samples at the time interval of every 5 min adsorbate at equilibrium and at time t, respectively, and
for the first 30 min and later at every 10 min until the k1 (min-1) is the rate constant of pseudo first-order
consecutive residual dye concentrations became closer. adsorption.

123
238 Int. J. Environ. Sci. Technol. (2013) 10:231–242

As seen from Table 3, the pseudo first-order rate con-

stant k1 increased with the decrease in the dye concentra-
tions (Abechi et al. 2011) and decreased with the increase
in particle size of the adsorbent (Ho and McKay 1988) for
both the studied adsorption processes. It was also evident
that there was a considerable difference between calculated
qe values and the experimental qe. The values of k1 were
found to be slightly higher for the adsorption of MG
than AB.

Pseudo second-order equation

The sorption kinetics can also be described by pseudo

second-order model. The linear form of pseudo second-
order equation is expressed as (Bulut and Ozacar 2008).
Fig. 2 Intraparticle diffusion model for the adsorption of MG onto
t 1 1 JSAC for 100 lm (ads. dosage = 1.5 g/150 ml; pH 6.87; T = 300 K;
¼ þ t ð9Þ
qt k2 q2e qe agitation = 150 rpm)

where k2 (g/mg min) is the equilibrium rate constant of

pseudo second-order adsorption. Equation (9) does not
adsorption processes were not only by intraparticle diffu-
have the problem of assigning an effective qe. If the pseudo
sion but the film diffusion also played a role in both the
second-order kinetic equation is applicable, the plot of t/qt
observed processes (Vasanth kumar et al. 2005).
against t should give a linear relationship, from which qe
and k2 can be determined from the slope and intercept of
Boyd plot
the plot.
From Table 3, it was clear that the pseudo second-order
The Boyd plot predicts the actual slow step involved in the
constant k2 decreased with the increase in the dye con-
adsorption process. The Boyd kinetic expression is given
centration and also increased with the decrease in particle
by (Vadivelan and Vasanth kumar 2005)
size of the adsorbent for both the adsorption processes.
From this table it can also be found that the calculated qe F ¼ 1  ð6=pð22Þ2Þ expðBt Þ ð11Þ
values were on par with the experimental qe values for both and
the observed processes unlike the pseudo first-order equa-
tion. But the value of k2 was found to vary in the same F ¼ qt =q0 ð12Þ
fashion as the value of k1 of pseudo first-order model. where q0 is the amount of adsorbates adsorbed at infinite
time (mg/g) and qt represents the amount of dyes adsorbed
Intraparticle diffusion model at any time t (min), F represents the fraction of solute
adsorbed at any time t, and Bt is a mathematical function
The intraparticle diffusion equation is given as (Cheung of F.
et al. 2007): Substituting Eq. 10 in Eq. 9,
1
qt ¼ k i t 2 þ c ð10Þ 1  F ¼ ð6=pð24Þ2Þ expðBt Þ ð13Þ
where qt is the amount of solute on the surface of the or
sorbent at time t (mg/g) and ki is the intraparticle diffusion
Bt ¼ 0:4977  lnð1  FÞ ð14Þ
rate constant (mg/g min1/2). When intraparticle diffusion
alone is the rate limiting step, then the plot of qt versus t1/2 The Bt values at different contact times can be
passes through the origin. When film diffusion is also calculated using Eq. (14) for various time intervals. The
taking place then the intercept is C, which gives the idea on calculated Bt values were plotted against time t.
the thickness of the boundary layer. The Boyd plots are shown in Figs. 4 and 5 for the
From the intraparticle diffusion plots shown in Figs. 2 adsorption of MG and AB onto JSAC, respectively. From
and 3, it was evident that the adsorption processes followed these figures, it was clear that the plots were linear for both
two steps. The first linear portion followed the boundary the adsorbates at all the studied adsorbate concentrations.
layer diffusion followed by another linear portion which Though r2 values of both pseudo first-order equation and
represents the intra particle diffusion. This shows that the pseudo second-order equation lie closer to unity, the

123
Int. J. Environ. Sci. Technol. (2013) 10:231–242 239

Fig. 5 Boyd plot for the adsorption of AB10B onto JSAC for 100 lm
(ads. dosage = 1.5 g/150 ml; pH 2.3; T = 300 K; agitation =
Fig. 3 Intraparticle diffusion model for the adsorption of AB10B
150 rpm)
onto JSAC for 100 lm (ads. dosage = 1.5 g/150 ml; pH 2.3;
T = 300 K; agitation = 150 rpm)
activated, there may be the presence of some irremovable
acid groups on the surface which might have been involved
in bonding with the basic dye (MG) molecules which may
not be that significant in case of acid dye (AB) molecules.
The plots of the intraparticle diffusion did not pass
through the origin indicating that the adsorption processes
not only followed the intraparticle diffusion but the film
diffusion also played an important role in both the studied
adsorption processes. It was also in coincidence with the fact
that the processes followed the pseudo second-order model.
The fact that film diffusion also played a major role in the
studied adsorption processes suggested that the adsorption
was mainly by covalent bonding by the surface acidic
functional groups. Also, the Boyd plots suggested that the
rate-determining step is the external mass transfer since the
plots were linear and does not pass through the origin.

Comparison of mechanisms of adsorption of acid dye

Fig. 4 Boyd plot for the adsorption of MG onto JSAC for 100 lm and basic dye onto JSAC
(ads. dosage = 1.5 g/150 ml; pH 6.87; T = 300 K; agitation =
150 rpm)
In general, the mechanism for dye removal by adsorption
on a sorbent material may be assumed to involve the fol-
calculated qe values of the pseudo first-order equation were lowing four steps (Cheung et al. 2007): (1) migration of
not in par with the experimental qe values. But, in case of dye from bulk of the solution to the surface of the adsor-
pseudo second-order equation, the calculated qe values are bent (bulk diffusion); (2) diffusion of dye through the
in par with the experimental qe values. So, it is clear from boundary layer to the surface of the adsorbent (film dif-
the accuracy of the model that the adsorption of both the fusion); (3) transport of the dye from the surface to the
dyes followed pseudo second-order chemical reaction. It interior pores of the particle (intraparticle diffusion or pore
was already demonstrated that the pseudo second-order diffusion); (4) adsorption of dye at an active site on the
kinetic equation for adsorption is much similar to the surface of material (chemical reaction via ion-exchange,
universal rate law for a chemical reaction (Liu 2008). Since complexation and/or chelation).
the processes followed the pseudo second-order equation, it The dye sorption is governed usually by either the liquid
literally suggests that the adsorption is mainly by simple phase mass transport rate or the intraparticle mass transport
chemical reaction between the dye molecules and the rate. Hence diffusive mass transfer is incorporated into the
surface functional groups on JSAC. Since JSAC was acid adsorption process. In diffusion studies, the rate can be

123
240 Int. J. Environ. Sci. Technol. (2013) 10:231–242

Table 4 Thermodynamic parameters

Temperature (K) 100 lm 600 lm 1,000 lm
DG DH DS DG DH DS DG DH DS
(kJ/mol) (kJ/mol) (J/mol K) (kJ/mol) (kJ/mol) (J/mol K) (kJ/mol) (kJ/mol) (J/mol K)

Adsorption of MG onto JSAC

293 -0.223 -0.118 6.074
300 -2.724 70.322 241.61 -1.921 153.53 508.85 2.697 98.746 317.448
313 -5.221 -4.972 -0.51
Adsorption of AB10B onto JSAC
293 2.761 3.502 4.986
300 -0.035 53.66 175.59 1.314 52.067 166.911 0.859 71.035 226.805
313 -1.018 -0.021 0.253

expressed in terms of the square root time. The mathe- Thermodynamic parameters obtained are shown in
matical dependence of qt versus t1/2 is obtained if the Table 4. From this table, it was evident that the adsorptions
process is considered to be influenced by diffusion in the of both the dyes were endothermic since the value of DH is
particles and convective diffusion in the solution. positive. Irrespective of the nature of the dyes, the studied
But from Figs. 2 and 3, it was evident that the plots did adsorption processes were endothermic. Similar results
not pass through the origin, this is indicative of some for endothermic adsorption were also observed on
degree of boundary layer control and these further shows adsorption on bentonite (Mall et al. 2005), activated carbon
that the intraparticle diffusion is not the sole rate control- prepared from deoiled soya (Onal et al. 2006), activated
ling step, but other processes may also control the rate of carbon prepared from Tuncbilek lignite (Mittal 2006) and
adsorption. From Fig. 2, for the adsorption of basic dye, hen feathers (Mall et al. 2005). The positive value of
MG onto JSAC, one can clearly witness that diffusive mass DS showed the increased randomness of the adsorbate
transfer (first linear portion of the curve) occurred up to the molecules on the solid surfaces than in the solution for both
t1/2 of 3.8 min which is followed by intraparticle diffusion acid and basic dyes. The adsorption of MG was feasible
(second linear portion of the curve). But in case of Fig. 3, and spontaneous at all the three temperatures for the
for the adsorption of acid dye, AB onto JSAC, the diffusive adsorbent particle sizes of 100 and 600 lm and only at
mass transfer occurred up to the t1/2 of around 6 min. This 313 K for 1,000 lm adsorbent particles. However, in the
suggests that since MG is a basic dye, a sort of chemical case of AB, it was feasible and spontaneous only for
reaction occurred between the cationic dye molecules and 100 lm particle size of the adsorbent at 300 and 313 K.
negatively charged adsorbent surfaces with the aid of the The adsorption of AB was not feasible at lower tempera-
irremovable acidic functional groups in the observed pH. ture and larger adsorbent particles. Hence the thermody-
Whereas, since AB is an anionic dye, there is not much namic studies too suggest that the adsorption of the basic
intraparticle diffusion due to the presence of some acidic dye, MG onto the acid-activated JSAC was well feasible
functional groups on the surface of JSAC and hence the % compared to the acid dye, AB.
adsorption also decreased considerably for the adsorption
of acid dye onto acid-activated carbon when compared to
the basic dye. Conclusion
It is important to find the slowest step which is the rate-
determining step. It was proved by the Boyd plot that Adsorption studies had been carried out for the removal of
external mass transfer is the rate-determining step for both Malachite Green (MG) dye and Amido Black 10B (AB)
the adsorbates. But the extent of film diffusion and dye by acid-activated carbons derived from J. regia shells
intraparticle diffusion varied based on the ionic nature of (JSAC). The % adsorption of MG dye was maximum at the
the adsorbates. pH ranging from 6 to 8 whereas for AB dye, it was
maximum at pH 2.3. Equilibrium data were tested with
Thermodynamic properties Langmuir and Freundlich isotherm model equations.
Adsorption of both the dyes followed Freundlich isotherm
Thermodynamic parameters which include Gibbs free and hence the prepared activated carbons had heteroge-
energy change (DG), enthalpy change (DH) and entropy neous surfaces. This heterogeneity was also confirmed
change (DS) were calculated using Langmuir constant (KL). by the constants of Redlich–Peterson and Sips isotherms.

123
Int. J. Environ. Sci. Technol. (2013) 10:231–242 241

The favorability of the studied adsorption processes were Bulut E, Ozacar M (2008) Adsorption of malachite green onto
proved by the separation factor (RL) of the Langmuir iso- bentonite: equilibrium and kinetic study and process design.
Micropor Mesopor Mat 115:234–246
therm and the Freundlich exponent (n) of the Freundlich Cheung WH, Szeto YS, McKay G (2007) Intraparticle diffusion
isotherm. The kinetic data were tested with pseudo first- processes during acid dye adsorption onto chitosan. Bioresour
order and pseudo second-order models. It was concluded Technol 98:2897–2904
that the adsorption of both the dyes followed the pseudo El-Sheikh AH, Newman AP, Al-Daffaee HK, Phull S, Cresswell N
(2004) Characterization of activated carbon prepared from a
second-order kinetics. The nature and the mechanism of the single cultivar of Jordanian Olive stones by chemical and
adsorption processes were studied by employing intrapar- physicochemical techniques. J Anal Appl Pyrolysis 71:151–164
ticle diffusion model and Boyd plot. External mass transfer Freundlich HMF (1906) Over the adsorption in solution. J Phys Chem
was the rate-determining step for the adsorption of both the 57:385–470
Garg KV, Amita M, Kumar R, Gupta R (2004) Basic dye (methylene
adsorbates. The adsorption is not only by intraparticle dif- blue) removal from simulated wastewater by adsorption using
fusion but the film diffusion also played an important role Indian Rosewood sawdust: a timber industry waste. Dyes
for all three particle sizes with varying initial dye concen- Pigments 63:243–250
trations. The extent of film diffusion and intraparticle dif- Guo Y, Zhao J, Zhang H, Yang S, Qi J, Wang Z, Xu H (2005) Use of
rice husk-based porous carbon for adsorption of Rhodamine B
fusion depended on the ionic state of the adsorbates. from aqueous solutions. Dyes Pigments 66:123–128
Thermodynamic analyses showed that both the processes Gupta VK, Mittal A, Krishnan L, Gajbe V (2004) Adsorption kinetics
were endothermic in nature and there were increased ran- and column operations for the removal and recovery of
domness of the adsorbate molecules on the solid surfaces malachite green from wastewater using bottom ash. Sep Purif
Technol 40:87–96
than in the solution. It was also found that the adsorption of Hameed BH, El-Khaiary MI (2008) Malachite green adsorption by
MG dye was much more an efficient process than the rattan sawdust: isotherm, kinetic and mechanism modeling.
adsorption of AB dye onto JSAC. This may be due to the J Hazard Mater 159:574–579
reason that since JSAC is acid activated; there may be Hameed BH, Ahmad AL, Latiff KNA (2007a) Adsorption of basic
dye (methylene blue) onto activated carbon prepared from rattan
the presence of some irremovable acid groups on the sur- sawdust. Dyes Pigments 75:143–149
face of JSAC which would have been involved in formation Hameed BH, Din ATM, Ahmad AL (2007b) Adsorption of methylene
of covalent bond with the basic dye molecules effectively blue onto bamboo-based activated carbon: kinetics and equilib-
when compared to the acid dye molecules. The fact that the rium studies. J Hazard Mater 141:819–825
Ho S, McKay G (1988) Kinetic models for the sorption of dye from
studied adsorption processes were not only by intraparticle the aqueous solution by wood. Trans IChemE 76:183–191
diffusion but the film diffusion also played an important Langmuir I (1916) The adsorption of gases on plane surface of glass,
role also suggests the same. From this study, it can be mica and platinum. J Am Chem Soc 40:1361–1403
concluded that the acid-activated carbon can be effectively Liu Y (2008) New insight to pseudo second-order kinetic equation for
adsorption. Colloid Surface A 320:275–278
used for the adsorption of basic dyes when compared to the Mall ID, Srivastava VC, Agarwal NK, Mishra IM (2005) Adsorptive
acid dyes. These J. regia shells literally of no economical removal of malachite green dye from aqueous solution by
value, proved to have a great potential in environmental bagasse fly ash and activated carbon-kinetic study and equilib-
remediation processes. rium isotherm analyses. Colloids Surf A: Physicochem Eng
Aspects 264:17–28
Mittal A (2006) Adsorption kinetics of removal of a toxic dye,
Acknowledgments The authors are thankful to Council for Scien- malachite green from waste water using hen feathers. J Hazard
tific and Industrial Research (CSIR), New Delhi, India for providing Mater B 133:196–202
the necessary funding and facilities to carry out this work. Namasivayam C, Kavitha D (2002) Removal of Congo red from
water by adsorption onto activated carbon prepared from coir
pith, an agricultural solid waste. Dyes Pigments 54:47–58
Onal Y, Akmil-Basar C, Eren D, Depci T (2006) Removal of
References malachite green using carbon-based adsorbents. J Hazard Mater
B 128:150–157
Abechi ES, Gimba CE, Uzairu A, Kagbu JA (2011) Kinetics of Patil S, Renukdas S, Patel N (2011) Removal of methylene blue,
adsorption of methylene blue onto activated carbon prepared a basic dye from aqueous solutions by adsorption using teak
from palm kernel shell. Arch Appl Sci Res 3:154–164 tree (Tectona grandis) bark powder. Int J Environ Sci 1:711–
Al-Degs YS, El-Barghouthi MI, El-Sheikh AH, Walker GM (2008) 726
Effect of solution pH, ionic strength, and temperature on Preethi S, Sivasamy A (2006) Removal of safranin basic dye from
adsorption behavior of reactive dyes on activated carbon. Dyes aqueous solutions by adsorption onto corncob activated carbon.
Pigments 77:16–23 Ind Eng Chem Res 45:7627–7632
Balistrieri LS, Murray JW (1981) The surface chemistry of goethite Rengaraj S, Moona SH, Sivabalan R, Arabindoo B, Murugesan V
(-FeOOH) in major ion sea water. Am J Sci 281:788–806 (2002) Removal of phenol from aqueous solution and resin
Bhatnagar A, Sillanpaa M (2010) Utilization of agro-industrial and manufacturing industry wastewater using an agricultural waste:
municipal waste materials as potential adsorbents for water rubber seed coat. J Hazard Mater B 89:185–196
treatment—a review. Chem Eng J 157:277–296 Repo E, Kurniawan TA, Warchol JK, Sillanpaa MET (2010) Adsorption
Boehm HP (1966) Chemical identification of surface groups. Adv of Co(II) and Ni(II) on EDTA- and/or DTPA-modified chitosan:
Catal 16:179–274 kinetic and equilibrium modeling. Chem Eng J 161:73–82

123
242 Int. J. Environ. Sci. Technol. (2013) 10:231–242

Repo E, Malinen L, Koivula R, Harjula R, Sillanpaa MET (2011a) Tan IAW, Hameed BH, Ahmad AL (2007) Equilibrium and kinetic
Capture of Co(II) from its aqueous EDTA chelate by DTPA- studies on basic dye adsorption by oil palm fibre activated
modified silica gel and chitosan. J Hazard Mater 187:122–132 carbon. Chem Eng J 127:111–119
Repo E, Warchol JK, Bhatnagar A, Sillanpaa MET (2011b) Heavy Tsenga RL, Wub FC, Juang RS (2003) Liquid-phase adsorption of
Metal Adsorption by Novel EDTA-Modified Chitosan-Silica dyes and phenols using pinewood-based activated carbons.
Hybrid Materials. J Colloid Interface Sci 358:261–267 Carbon 41:487–495
Repo E, Petrus R, Sillanpaa MET, Warchol JK (2011c) Equilibrium Vadivelan V, Vasanth Kumar K (2005) Equilibrium, kinetics,
studies on the adsorption of Co(II) and Ni(II) by modified silica gels: mechanism, and process design for the sorption of methylene
one-component and binary systems. Chem Eng J 172:376–385 blue onto rice husk. J Colloid Interface Sci 286:90–100
Santhy K, Selvapathy P (2006) Removal of reactive dyes from Vasanth kumar K, Ramamurthi V, Sivanesan S (2005) Modeling the
wastewater by adsorption on coir pith activated carbon. Biore- mechanism involved during the sorption of methylene blue onto
sour Technol 97:1329–1336 fly ash. J Colloid Interface Sci 284:14–21
Senthilkumaar S, Varadarajan PR, Porkodi PK, Subbhuraam CV Zhang J, Li Y, Zhang C, Jing Y (2008) Adsorption of malachite green
(2005) Adsorption of methylene blue onto jute fiber carbon: from aqueous solution onto carbon prepared from Arundo donax
kinetics and equilibrium studies. J Colloid Interface Sci root. J Hazard Mater 150:774–782
284:78–82 Zhang J, Fu H, Lu X, Tang J, Xu X (2011) Removal of Cu (II) from
Shawabke RA, Rockstraw DA, Bhada RK (2002) Copper and aqueous solution using the rice husk carbons prepared by the
strontium adsorption by a novel carbon material manufactured physical activation process. Biomass Bioenerg 35:464–472
from pecan shells. Carbon 40:781–786

123