Journal of Colloid and Interface Science 281 (2005) 49–55

www.elsevier.com/locate/jcis

A comparative adsorption study with different industrial wastes

as adsorbents for the removal of cationic dyes from water
Amit Bhatnagar ∗,1 , A.K. Jain
Department of Chemistry, Indian Institute of Technology Roorkee, Roorkee 247667, India
Received 31 May 2004; accepted 10 August 2004
Available online 1 October 2004

Abstract
Four adsorbents have been prepared from industrial wastes obtained from the steel and fertilizer industries and investigated for their utility
to remove cationic dyes. Studies have shown that the adsorbents prepared from blast furnace sludge, dust, and slag have poor porosity and
low surface area, resulting in very low efficiency for the adsorption of dyes. On the other hand, carbonaceous adsorbent prepared from
carbon slurry waste obtained from the fertilizer industry was found to show good porosity and appreciable surface area and consequently
adsorbs dyes to an appreciable extent. The adsorption of two cationic dyes, viz., rhodamine B and Bismark Brown R on carbonaceous
adsorbent conforms to Langmuir equation, is a first-order process and pore diffusion controlled. As the adsorption of dyes investigated was
appreciable on carbonaceous adsorbent, its efficiency was evaluated by comparing the results with those obtained on a standard activated
charcoal sample. It was found that prepared carbonaceous adsorbent exhibits dye removal efficiency that is about 80–90% of that observed
with standard activated charcoal samples. Thus, it can be fruitfully used for the removal of dyes and is a suitable alternative to standard
activated charcoal in view of its cheaper cost.
 2004 Elsevier Inc. All rights reserved.

Keywords: Rhodamine B; Bismark Brown R; Wastewater; Industrial solid wastes; Adsorbents

1. Introduction streams. The nonbiodegradable, toxic, and inhibitory nature

Dyes are important water pollutants which are generally
present in the effluents of the textile, leather, food process-
ing, dyeing, cosmetics, paper, and dye manufacturing indus-
tries. They are synthetic aromatic compounds which are em-
bodied with various functional groups. The worldwide high
level of production and extensive use of dyes generates col-
ored wastewaters which cause environmental pollution. The
colored dye effluents are generally considered to be highly
toxic to the aquatic biota [1] and affect the symbiotic process
by disturbing the natural equilibrium through reduced pho-
tosynthetic activity due to the coloration of the water in
* Corresponding author. Fax: +41-1-823-5210.
E-mail address: amit.bhatnagar@eawag.ch (A. Bhatnagar).
1 Present address: Water Resources and Drinking Water (W+T), Swiss
Federal Institute for Environmental Science & Technology (EAWAG), Post-
fach 611, Ueberlandstrasse 133, CH-8600, Duebendorf, Switzerland.
0021-9797/$ – see front matter  2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2004.08.076
of spent dye baths has a considerable deleterious effect on
the environmental matrix (water and soil). Some dyes are
reported to cause allergy, dermatitis, skin irritation, cancer,
and mutations in humans [2,3]. Thus, the removal of dyes
from effluents before they are mixed up with unpolluted nat-
ural water bodies is important. Several treatment technolo-
gies exist for dye removal [4], but the process of adsorption
using activated carbon [5–8] has been found to be an effi-
cient technology for decolorization of wastewater. Though
the removal of dyes through adsorption is quite effective,
its use is restricted sometimes due to the higher cost of ac-
tivated carbon and difficulties associated with regeneration.
Attempts have therefore been made to utilize natural as well
as waste materials as alternative adsorbents. But the utiliza-
tion of industrial waste materials is of vital concern over
the past few years because these wastes represent unused
resources and, in many cases, cause serious disposal prob-
lems. The aim behind using waste materials as adsorbents is
50 A. Bhatnagar, A.K. Jain / Journal of Colloid and Interface Science 281 (2005) 49–55
that it will provide a twofold advantage with respect to en-
vironmental pollution. First, the volume of waste materials
could be partly reduced, and second, the low-cost adsorbent,
if developed, can reduce the pollution of wastewaters at a
reasonable cost. In view of the low cost of such adsorbents, it
would not be necessary to regenerate spent materials. Thus,
a number of waste materials from different sources have
been investigated with or without treatment as adsorbents
for the removal of dyes from wastewater. Recently, Garg et
al. [9] have listed materials studied as adsorbents for dye re-
moval from aqueous solutions.
A number of low-cost alternative adsorbents have been
explored but they have not been highly successful. In general
the limitations of such studies have been
(i) the surface area of alternative adsorbents developed has
been low in most cases, resulting in poor adsorptive
power;
(ii) comparative studies on different adsorbents have not
been carried out; and
(iii) the performance of alternative adsorbents has not been
assessed by comparing the results with those obtained
on activated carbon.
It is therefore still important to develop low-cost adsorbents
with good surface area, which may in turn exhibit good ad-
sorption potential for the removal of aqueous pollutants.
Keeping the above points under consideration, a com-
parative adsorption study was carried out by us [10] using
carbon slurry waste obtained from a fertilizer plant and blast
furnace sludge, dust, and slag from steel plant wastes as ad-
sorbents for the removal of two anionic dyes, methyl orange
and brilliant blue G. It was found that carbonaceous adsor-
bent prepared from a fertilizer plant exhibits good potential
for the removal of dyes as compared to the other three ad-
sorbents prepared. In order to further test the versatility of
the carbonaceous adsorbent, investigations with regard to
the adsorption of another class of dyes, cationic dyes, were
also undertaken, and two important and widely used cationic
dyes, rhodamine B and Bismark Brown R, were studied on
all the adsorbents prepared. Further, a comparative study
with standard activated charcoal was also carried out and the
results are incorporated into the present communication.
2. Materials and methods
2.1. Materials
Rhodamine B and Bismark Brown R were procured
from S.D. Fine-Chem. Limited (India). The standard acti-
vated charcoal sample was obtained from E. Merck. Other
reagents used were of AR grade. Double-distilled water was
used in preparing solutions throughout these investigations.
2.2. Preparation of adsorbents
The adsorbents under investigation have been prepared
from fertilizer and steel plant wastes.
2.2.1. Preparation of carbonaceous adsorbent
In India, carbonaceous slurry waste is produced in fer-
tilizer plants using fuel oil/LSHS (low sulfur heavy stock).
This slurry is available at a very cheap rate ($0.007 per kg)
and can act as a good adsorbent in view of its porous na-
ture. The dried carbon slurry was procured from National
Fertilizer Limited (NFL), Panipat (India). It was found to
consist of small, black, greasy granules and treated [10,11]
with H2 O2 to oxidize the adhering organic material. It was
then washed with distilled water and heated at 200 ◦ C till
the evolution of black soot stopped. The activation of this
material was done in the range 300–700 ◦ C in a muffle fur-
nace for 1 h in the presence of air. After the activation, the
ash content was removed by treating the material with 1 M
HCl and washed with distilled water and dried. The surface
area of samples activated at temperatures of 300, 400, 500,
600, and 700 ◦ C was found to be 280, 363, 380, 374, and
367 m2 g−1 , respectively. Thus, the activation at 500 ◦ C im-
parts maximum surface area and it was also seen that the
extent of adsorption was at a maximum for the sample acti-
vated at 500 ◦ C. Therefore, all studies reported in this paper
are on a sample activated at 500 ◦ C. The yield of the finished
product was found to be ∼ 90% and it is now called “car-
bonaceous adsorbent.” Different mesh sizes were obtained
after sieving and kept in a desiccator for further use.
2.2.2. Preparation of blast furnace sludge, dust, and slag
adsorbents
These adsorbents were prepared from wastes obtained
from Malvika Steels, Jagdishpur (India), by the method re-
ported elsewhere [10]. The products were sieved and stored
in a desiccator.
2.3. Instrumentation
The spectrophotometric determination of dyes was done
on a Shimadzu 1601 UV–vis spectrophotometer. The pH of
solutions was measured with an ELICO LI 127 pH meter.
A Vario EL III, C H N S elemental analyzer was used to
determine the carbon content of the adsorbents.
2.4. Adsorption studies
The adsorption of dyes on prepared adsorbents was stud-
ied at room temperature by employing the batch method.
A known volume (10 ml) of dye solutions of varying ini-
tial concentrations, taken in 50-ml stoppered glass tubes, was
shaken with a fixed dose of adsorbent (0.01 g) for a specified
period of contact time in a thermostated shaking assembly.
After equilibrium, the concentration of the adsorbate in the
residual solution was determined spectrophotometrically at
 A. Bhatnagar, A.K. Jain / Journal of Colloid and Interface Science 281 (2005) 49–55 51

λmax of 554.0 nm for rhodamine B and 460.0 nm for Bis- Table 1

mark Brown R. The experiments were repeated a number Characteristics of adsorbents used
of times and average values are reported. Standard devia- Standard Carbonaceous BF sludge BF dust BF slag
tions were found to be within ±3.0%. Furthermore, error activated adsorbent
bars for the figures were smaller than the symbols used to charcoal
plot the graphs and hence are not shown. The pH of the so- Surface area 710 380 28 13 4
(m2 g−1 )
lutions was adjusted using dilute sulfuric acid and sodium
Methylene 198 90 6 3 2
hydroxide solutions. The pH of the solutions was measured blue number
before and after the equilibration and a slight change in Iodine number 635 330 24 11 3
pH was observed. However, after adjustment of the pH of
the adsorbent–adsorbate system no further change was ob-
served. The pH of solutions in contact with adsorbents was paratively higher surface area. Further, the methylene blue
found to be in the range from 5.5 to 6.5 for both dyes. The number and iodine number, important characteristics of an
adsorption was studied as a function of contact time, con- adsorbent, also confirm that carbonaceous adsorbent would
centration, particle size, and temperature. Kinetic studies of be a better adsorbent for pollutants (dyes) removal than the
adsorption were also carried out at two concentrations of the other three adsorbents.
adsorbates where the extent of adsorption was investigated
as a function of time. 3.2. Effect of contact time and concentration

The effect of contact time on the adsorption of Bismark

3. Results and discussion Brown R and rhodamine B is shown in Fig. 1. It is seen that
nearly 25 min are required for the equilibrium adsorption
3.1. Characterization of the prepared adsorbents to be attained for both the dyes. As such for all adsorp-
tion studies, the equilibration time was kept 3 h. It is fur-
The BF sludge, dust, and slag were chemically analyzed ther seen from Fig. 1 that adsorption is very fast initially,
and it was found that BF slag is basically inorganic in nature; showing that 50% adsorption in the case of both the dyes
i.e., it contains mainly inorganic constituents such as silica is completed in 6 min. After that it slowly attains equilib-
(32.7%), calcium oxide (31.7%), magnesium oxide (6.8%), rium adsorption. The effect of concentration on equilibrium
and R2 O3 ; R = Fe, Al (22.8%). On the other hand, silica, adsorption time was also investigated as a function of ini-
R2 O3 (mainly iron oxide), and carbon (due to the presence of tial dye concentration and the results are shown in Fig. 1 (in
coke particles) are the prominent components of BF dust and the case of rhodamine B). Similar plots were also obtained
sludge. Although the inorganic constituents such as silica for other dyes. These plots show that the time of equilibrium
(15.8% and 12.7%), R2 O3 (44.9% and 35.4%), MgO (4.2%
and 3.0%), and CaO (4.7% and 3.5%) in BF dust and sludge,
respectively, are present in major amounts, but the presence
of appreciable carbon content impart a partial organic char-
acter to BF sludge and dust. Of course, the partial organic
character of BF sludge (carbon content = 35.0%) is more
than that of BF dust (carbon content = 21.7%). We may
therefore say that BF sludge and dust is of mixed character,
with inorganic nature predominating over organic. However,
as compared to slag, dust, and sludge, the analysis of carbon
slurry indicates 0.9% ash and 89.8% carbon. Thus, due to
the presence of high carbon content, the carbon slurry may
be treated as organic in nature.
In order to assess the adsorption characteristics of these
four adsorbents, their surface area was determined by N2 gas
adsorption and given in Table 1 with other important adsorp-
tion characteristics of the adsorbents used. It is seen from Ta-
ble 1 that surface area increases as carbon content increases
from BF slag to carbonaceous adsorbent, indicating that the
porosity of carbon is responsible for higher surface area. To
confirm it, SEM photographs of the four adsorbents were
taken which clearly showed that BF sludge, dust, and slag Fig. 1. Effect of contact time on uptake of cationic dyes on carbonaceous ad-
possess very poor porosity whereas carbonaceous adsorbent sorbent at different initial concentrations (temperature 25 ◦ C; particle size
exhibits a significantly porous structure, giving rise to a com- 200–250 mesh).
52 A. Bhatnagar, A.K. Jain / Journal of Colloid and Interface Science 281 (2005) 49–55
Fig. 2. Effect of particle size on the uptake of rhodamine B on carbonaceous
adsorbent (temperature 25 ◦ C).
adsorption as well as the time required to achieve a definite
fraction of equilibrium adsorption is independent of the ini-
tial dye concentration. This type of behavior indicates that
the adsorption process is first-order, which is confirmed by
Lagergren’s plots discussed later under dynamic modeling.
These results are similar to our previous studies with anionic
dyes.
3.3. Effect of particle size
The adsorption of both the dyes was also investigated at
three particle sizes, 100–150, 150–200, and 200–250 British
Standard Sieve (BSS) mesh, and the results are shown in
Fig. 2 (in the case of rhodamine B). Similar plots were also
obtained for other dyes. It is seen that with decreased particle
size, the adsorption increases to some extent. This could not
be due to any substantial increase in surface area [12]. It
appears that large molecules of dyes are not able to penetrate
into some of the interior pores of the particles when their
size is large. Apparently access to all pores is facilitated as
particle size becomes smaller, resulting in higher adsorption.
Similar observations were also made by McKay et al. [13].
As particles of size 200–250 BSS mesh exhibit maximum
adsorption capacity, all further detailed studies were done
with this fraction only.
3.4. Adsorption isotherms
To assess the adsorption potential of the four prepared ad-
sorbents for cationic dyes, their adsorption isotherms were
drawn (Fig. 3 for rhodamine B). Similar type of plots were
also obtained for another dye. It is seen from these plots
that the extent of adsorption is in the order carbonaceous
adsorbent > BF sludge > BF dust > BF slag for both the
cationic dyes, which is parallel to the surface area of the ad-
sorbents. Further, the maximum adsorption on carbonaceous
adsorbent, BF sludge, dust, and slag is found to be 71.5,
8.2, 4.6, and 2.7 mg g−1 for Bismark Brown R and 82.8,
Fig. 3. Adsorption isotherms of rhodamine B on different adsorbents at
25 ◦ C (×, standard activated charcoal; 5, carbonaceous adsorbent; *, BF
sludge; P, BF dust; ', BF slag).
10.0, 7.1, and 3.4 mg g−1 for rhodamine B on carbonaceous
adsorbent, BF sludge, BF dust, and BF slag, respectively.
Broadly, this type of adsorption behavior is similar to that of
anionic dyes. However, the adsorption of cationic dyes onto
carbonaceous adsorbent and sludge is smaller than the ad-
sorption of anionic dyes reported earlier. On the other hand,
the adsorption of these dyes onto BF dust and slag, though
very small, is found to be slightly higher than that observed
for anionic dyes. The slightly higher adsorption on these two
adsorbents could be due to some partially negative charge on
metal oxide constituents present in BF dust and slag. Thus,
these results again show that carbonaceous adsorbent, which
is of an organic nature, having higher porosity and sur-
face area, is a suitable adsorbent for the removal of another
class of dyes, i.e., cationic dyes, whereas the other three ad-
sorbents, which are mainly of an inorganic nature, having
poor porosity and surface area, are poor adsorbents. Out of
the four adsorbents prepared, only carbonaceous adsorbent
shows substantial potential for the removal of cationic dyes
(organics); its performance was evaluated by comparing the
results with those obtained on standard activated charcoal.
The adsorption isotherms of both the cationic dyes on stan-
dard activated charcoal were also determined and shown in
Fig. 3. A comparison of maximum adsorption capacity (as
obtained from adsorption isotherms) of the dyes on carbona-
ceous adsorbent and standard activated charcoal shows that
carbonaceous adsorbent is about 80–90% as efficient as stan-
dard activated charcoal. Thus, it is reasonable to conclude
that prepared carbonaceous adsorbent is appreciably effi-
cient in removing dyes from aqueous solutions and, there-
fore, all further studies were carried out on this adsorbent
only.
 A. Bhatnagar, A.K. Jain / Journal of Colloid and Interface Science 281 (2005) 49–55 53

Fig. 5. Langmuir adsorption isotherms of rhodamine B on carbonaceous

adsorbent at different temperatures.

Table 2
Langmuir constants and separation factor for the adsorption of cationic dyes
on carbonaceous adsorbent at different temperatures
Fig. 4. Adsorption isotherms of dyes on carbonaceous adsorbent at 45 ◦ C.
Dyes Temperature qm b RL
(◦ C) (mg g−1 ) (l mol−1 )
3.5. Effect of temperature
Bismark Brown R 25 85.0 4.4 × 104 7.0 × 10−1
45 73.2 4.3 × 104 7.1 × 10−1
To determine the effect of temperature on the adsorp-
Rhodamine B 25 91.1 7.9 × 104 5.7 × 10−1
tion of dyes, experiments were also conducted at 45 ◦ C and 45 83.3 7.6 × 104 5.8 × 10−1
results are shown in Fig. 4. A comparison of adsorption
isotherms at 25 and 45 ◦ C shows that adsorption decreases
with increased temperature, indicating that the adsorption is Table 3
exothermic. The adsorption data were further analyzed and Thermodynamic parameters for the adsorption of cationic dyes on carbona-
found to conform best to the Langmuir equation ceous adsorbent at different temperatures
Dyes Temperature −G0 S 0 −H 0
1 1 1
= + , (1) (◦ C) (kJ mol−1 ) (J mol−1 K−1 ) (kJ mol−1 )
qe qm qm bCe
Bismark Brown R 25 26.5 86.2
0.81
where qe is the amount adsorbed at equilibrium concentra- 45 28.2 86.2
tion Ce , qm the Langmuir constant representing maximum Rhodamine B 25 27.9 89.9
1.12
45 29.7 89.9
monolayer capacity, and b the Langmuir constant related to
energy of adsorption. The plots between 1/qe and 1/Ce for
the adsorption of rhodamine B are drawn in Fig. 5. Simi-
The values of RL calculated per the above equation are incor-
lar plots were also obtained for other dye. The values of
porated into Table 2. As the RL values lie between 0 and 1,
monolayer capacity (qm ) and Langmuir constant (b) have
the adsorption isotherm is favorable [14].
been evaluated from the intercept and slope of these plots
The values of free energy change (G0 ), enthalpy
and given in Table 2. A perusal of Table 2 shows that mono-
change (H 0 ), and entropy change (S 0 ) for the adsorp-
layer capacity (qm ) of the adsorbent for the dyes is compa-
tion process were calculated using the equations
rable to the maximum adsorption obtained from adsorption
isotherms (Fig. 3). As b values reflect equilibrium constant G0 = −RT ln(b), (3)
for the adsorption process, it shows the affinity of the adsor-

H0 1 1
bent for dye. ln(b2 /b1 ) = − − , (4)
The effect of isotherm shape has been discussed [14] with R T2 T1
a view to predicting whether an adsorption system is favor- G0 = H 0 − T S 0 (5)
able or unfavorable. The essential feature of the Langmuir
isotherm can be expressed in terms of RL , a dimensionless and are summarized in Table 3. The small negative values
constant referred to as separation factor or equilibrium para- of H 0 indicate that the adsorption is physical in nature,
meter. RL is calculated using the following equation: involving weak forces of attraction. Further, negative G0
values indicate the spontaneous nature of the adsorption
1 process and positive S 0 values indicate the affinity of the
RL = . (2)
1 + bC0 adsorbent for dyes.
54 A. Bhatnagar, A.K. Jain / Journal of Colloid and Interface Science 281 (2005) 49–55
Fig. 6. Lagergren plot for dyes on carbonaceous adsorbent.
3.6. Dynamic modeling
Kinetics of sorption is one of the important characteristics
in defining the efficiency of sorption. Various kinetic models
have been proposed by different workers where the adsorp-
tion has been treated as first-order [15,16], pseudo-first-order
[17,18], and pseudo-second-order processes [19]. Different
systems conform to different models. The Lagergren rate
equation [20] is the one most widely used [15,16,21] for the
sorption of a solute from a liquid solution. Thus this first-
order equation,
kads
log(qe − q) = log qe − t, (6)
2.303
where qe and q are the amounts of the dye adsorbed at equi-
librium and at time t, in mg g−1 , and k ads the first-order rate
constant, was applied to the present studies of dye adsorp-
tion. As such, the values of log(qe − q) were calculated from
the kinetic data of Fig. 1 and plotted against time in Fig. 6.
The plots are found to be linear, indicating that Lagergren’s
equation is applicable to the dye adsorption on carbonaceous
adsorbent and the adsorption is a first-order process. The
first-order rate constants calculated from the slope of the
plots (Fig. 6) are found to be 1.24 × 10−1 min−1 for rho-
damine B and 1.54 × 10−1 min−1 for Bismark Brown R.
The kinetic data were further used to learn about the slow
step occurring in the present adsorption system. The ap-
plicability of Bangham’s equation [22] to the present dye
Fig. 7. Bangham plot for dyes on carbonaceous adsorbent.
adsorption studies was tested,

 
C0  k 0 m
log log = log + α log t, (7)
C0 − q  m 2.303V
where C0 is the initial concentration of the adsorbate in
solution (mmol l−1 ), V the volume of the solution (ml),
m the weight of adsorbent used per liter of solution (g l−1 ),
q  (mmol g−1 ) the amount of adsorbate retained at time t,
and α(< 1) and k0 are constants. As such log log[C0 /(C0 −
q  m )] was plotted against log t in Fig. 7. The linearity of
these plots confirms the applicability of Bangham’s equation
and indicates that the adsorption is pore-diffusion-controlled
[17,22].
4. Conclusions
It is reasonable to conclude from the present investiga-
tions on the adsorption of cationic dyes on some adsorbents
prepared from industrial wastes that (i) carbonaceous adsor-
bent possessing an organic nature and consequently having
higher surface area and porosity is efficient for the removal
of dyes, whereas the other three adsorbents, BF sludge, dust,
and slag, possessing inorganic nature resulting in lower sur-
face area and porosity, are poor materials for this purpose;
(ii) the adsorption of dyes on carbonaceous adsorbent is first-
order and pore-diffusion-controlled; and (iii) carbonaceous
adsorbent is about 80–90% as efficient as standard activated
charcoal and thus can be used in its place for the removal of
dyes from effluents in view of its cheaper cost ($0.10 per kg).
References
[1] G.E. Walsh, L.H. Bahner, Environ. Pollut. Ser. A 21 (1980) 169.
[2] P.K. Ray, J. Sci. Ind. Res. India 45 (1986) 370.
[3] K.G. Bhattacharyya, A. Sharma, J. Environ. Manag. 71 (2004) 217.
 A. Bhatnagar, A.K. Jain / Journal of Colloid and Interface Science 281 (2005) 49–55 55

[4] E. Forgacs, T. Csehàti, G. Oros, Environ. Int. (2004), in press. [13] G. McKay, M.S. Otterburn, A.G. Sweeny, Water Res. 14 (1980) 21.
[5] G.M. Walker, L.R. Weatherley, Water Res. 31 (1997) 2093. [14] T.W. Weber, R.K. Chakravorti, J. Am. Inst. Chem. Eng. 20 (1974) 228.
[6] G.M. Walker, L.R. Weatherley, Environ. Pollut. 99 (1998) 133. [15] E. Haribabu, Y.D. Upadhya, S.N. Upadhyay, Int. J. Environ. Stud. 43
[7] J.-M. Chern, C.-Y. Wu, Water Res. 35 (2001) 4159. (1993) 169.
[8] S.J. Allen, in: G. McKay (Ed.), Use of Adsorbents for the Removal of [16] K.K. Panday, G. Prasad, V.N. Singh, Water Res. 19 (1985) 869.
Pollutants from Wastewaters, CRC Press, Boca Raton, FL, 1996. [17] E. Tutem, R. Apak, C.F. Unal, Water Res. 32 (1998) 2315.
[9] V.K. Garg, M. Amita, R. Kumar, R. Gupta, Dyes Pigments 63 (2004)
[18] Y.S. Ho, G. McKay, Water Res. 33 (1999) 578.
243.
[19] Y.S. Ho, J.C.Y. Ng, G. McKay, Sep. Sci. Technol. 36 (2001) 241.
[10] A.K. Jain, Suhas, S. Jain, A. Bhatnagar, Toxicol. Environ. Chem. 84
(2001) 41. [20] S. Lagergren, Vetensk. Handl. 24 (1898) 1.
[11] S.K. Srivastava, N. Pant, N. Pal, Water Res. 21 (1987) 1389. [21] K. Periasamy, C. Namasivayam, Ind. Eng. Chem. Res. 33 (1994) 317.
[12] F.W. Pontius, Water Quality and Treatment, fourth ed., McGraw–Hill, [22] C. Aharoni, M. Ungarish, J. Chem. Soc., Faraday Trans. I 73 (1977)
New York, 1990. 456.