Journal of Colloid and Interface Science 357 (2011) 157162

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

www.elsevier.com/locate/jcis

KOH catalysed preparation of activated carbon aerogels for dye adsorption

Sie King Ling a, H.Y. Tian b, Shaobin Wang a,, Thomas Rufford c, Z.H. Zhu c, C.E. Buckley b
a

Centre for Process Systems Engineering, Department of Chemical Engineering, Curtin University, GPO Box U1987, Perth, WA 6845, Australia
Department of Imaging and Applied Physics, Curtin University, GPO Box U1987, Perth, WA 6845, Australia
c
School of Chemical Engineering, The University of Queensland, St. Lucia 4072, Australia
b

a r t i c l e

i n f o

Article history:
Received 30 August 2010
Accepted 28 January 2011
Available online 3 February 2011
Keywords:
Carbon aerogel
KOH catalysis
Dye adsorption

a b s t r a c t
Organic carbon aerogels (CAs) were prepared by a solgel method from polymerisation of resorcinol,
furfural, and hexamethylenetetramine catalysed by KOH at around pH 9 using ambient pressure drying. The effect of KOH in the solgel on CA synthesis was studied. It was found that addition of KOH
prior to the solgel polymerisation process improved thermal stability of the gel, prevented the crystallinity of the gel to graphite, increased the microporosity of CA and promoted activation of CA. The
CAs prepared using the KOH catalyst exhibited higher porosity than uncatalysed prepared samples.
Activation in CO2 at higher temperature also enhanced the porosity of CAs. Adsorption tests indicated
that the CAs were effective for both basic and acid dye adsorption and the adsorption increased with
increasing surface area and pore volume. The kinetic adsorption of dyes was diffusion control and
could be described by the second-order kinetic model. The equilibrium adsorption of dyes was higher
than activated carbon.
2011 Elsevier Inc. All rights reserved.

1. Introduction
The pollution of water resources due to the indiscriminate
disposal of toxic organic compounds and heavy metals has been
a worldwide issue in the last few decades. Removal of those contaminants from water is a big challenge. The conventional methods for the removal of toxic compounds include adsorption,
precipitation, coagulation/occulation, ion exchange, reverse
osmosis, complexation/sequestration, electrochemical operation
and biological treatment [18]. Adsorption using activated carbon is the most economic and widely used method. Development of efcient carbon adsorbents is the key to adsorptive
treatment of wastewater and gas cleaning. Dye stuffs from various industries such as textile, food processing, and ink production impose a threat to the environment and ecosystem due to
their blockage of sunlight and toxicity. Several different types
of dye are widely used including basic, acid, reactive, and dispersed dyes. Removal of dyes by adsorption using activated carbon and other low cost adsorbents is the simple and effective
method.
Carbon aerogels (CAs) are novel porous carbon materials and
they are of great signicance in terms of scientic and practical
applications because of their unique properties, such as controlla-

Corresponding author.
E-mail address: shaobin.wang@curtin.edu.au (S. Wang).
0021-9797/$ - see front matter 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2011.01.092

ble mass densities, continuous porosities, and high surface areas

[911]. Recently, CA have drawn much attention as adsorbents
[1214], catalyst supports [9,15], and electrodes [16,17].
Synthesis of CAs involves of several processes including sol
gel polymerisation, polycondensation of organic precursors, wet
gel drying and carbonisation. It has been found that addition
of a catalyst in gel solution could promote the colloid formation
and gelation during the synthesis of CAs. In the past, sodium carbonate is the most commonly used alkaline catalyst for the polymerisation reaction [11]. Recently, some investigations have
been reported in using different catalysts, such as acetic acid
[18], oxalic acid [19], HCl [20], ammonium carbonate [21], and
NaOH [22]. However, few investigations were reported in using
KOH as a catalyst for CA synthesis [23].
Previous investigations have also showed that solgel pH is
an important factor in CA synthesis. The typical pH values would
be between 5.4 and 7.6. At pH higher than 7, poorly porous carbons were produced with the surface area diminished completely [11,24]. Wu et al. [22] reported CA synthesis using
NaOH-catalysed polycondensation of resorcinol-furfural and
supercritical drying in ethanol. However, CA could not been synthesised at high pHs and the porous structure could not be
maintained. In this paper, we report an investigation of CA synthesis with high and stable porous structure at much high pH
(pH 9) catalysed by KOH. The effect of KOH on CA textural structure was investigated by comparing the properties of CAs synthesised with KOH and without KOH and their adsorption
behaviour for dye removal from aqueous solution.

158

S.K. Ling et al. / Journal of Colloid and Interface Science 357 (2011) 157162

2. Experimentals

3. Results and discussion

2.1. Carbon aerogel synthesis and activation

3.1. Characterisation of carbon aerogels

CAs were synthesised from resorcinol (R) (99.0%), furfural (F)

(99.0%), and hexamethylenetetramine (HMTA) (99.0%), which were
all obtained from SigmaAldrich, Australia. During the synthesis,
two different CAs were prepared with or without KOH addition
in gel formation. For KOH catalysed CA, the KOH was used to adjust
the pH value to 8.9 from 6.5 of the precursor solution and the molar ratio of resorcinol to KOH was about 20. The resultant solutions
were put in an oven at 7580 C for 37 days after being rst stabilised for at least 24 h at room temperature. The organic wet gels
were then transferred into a chamber and were dried for 10 h. The
aerogel samples were carbonised at 900 C at a heating rate of 3 C/
min for 3 h under nitrogen atmosphere (N2 ow rate = 1.5 L/min)
and then activated at different temperatures for 1 h under CO2
atmosphere. The samples will be referred as CA-K and CA for
KOH catalysed and uncatalysed samples, respectively, followed
by AT, representing activating temperature, T in C, e.g., CA-KA1000. For a comparison, one inactivated sample with KOH (CAK-900) was also prepared without CO2 activation.

Fig. 1 shows FT-IR spectra of dry gel of CAs synthesised with and
without KOH. The peak positions were almost the same for the two
samples due to the same composition and the addition of KOH
would not result in a signicant change in chemical structure of
the gels. However, there was an increase in the transmittance (%)
for KOH-catalysed sample, indicating the formation of more porous
structure in the CA catalysed by KOH. The FT-IR spectra showed a
broad peak at 3450 cm1, characteristic of OAH stretching vibrations, ascribed to adsorbed water. The other peaks between 1400
and 1800 cm1 at 1664, 1606, 1503, and 1469 cm1 were associated with aromatic and furfural ring stretching vibrations of
O@C, aromatic ring, OACH2, and CACH2AC, respectively. Meanwhile, the bands at 1012 and 960 cm1 and the bands at
737 cm1 and its accompanying peaks at 803 and 839 cm1 were
assigned to aromatic and furfural CH out-of plane bending vibrations. These IR peaks suggested that the aromatic ring structures
would give rise to the skeleton formation of CAs.
Fig. 2 shows TGA/DTA proles of as-prepared CAs without KOH
and KOH-catalysed. For the two samples, the TG proles were similar showing a three-stage variation. The rst stage of weight loss
was at 25161.2 C, attributing to the evaporation of surface adsorbed water and residual organic reagents. In this stage, KOH-catalysed sample presented less weight loss. At the second stage of
weight loss between 161.2 and 796.6 C, KOH-catalysed sample
exhibited higher weight loss. This stage was the major process of
carbonisation of wet gels. In the last stage, much less weight loss
was observed for both samples compared with the other two processes, which was caused by the further pyrolysis of carbon at high
temperatures. The overall weight loss from the two samples suggested that the gel with KOH would be more stable during the carbonisation process and more organic network would be
transformed to CA. The DTA proles of the two samples also
showed the instability of the uncatalysed gel. A differentiated
curve presented at the top line of Fig. 2 showed the exothermic
and endothermic peaks during the course of carbonisations. In
the second stage of carbonisation, the uncatalysed gel presented
several endothermic and exothermic peaks between 220 and
420 C and also another endothermic peak occurred at 916.6 C
in the last stage, suggesting more violent chemical reactions occurring during the carbonisation, resulting in more loss of carbon.
However, no signicant reaction occurred on KOH-catalysed gel
in the same stages.
Fig. 3 shows the XRD proles of CAs synthesised with or without KOH. Two CAs presented amorphous phase with a broad peak
between 40 and 47. The CA without KOH catalyst also showed
two sharp peaks of graphite (i.e. 21.5 and 23.9). This suggested
that without KOH, amorphous carbon networks would be crystallised to form graphite and the porous structure could be partially
lost. With KOH, no signicant crystallisation to graphite occurred
during carbonisation.
Fig. 4 displays N2 adsorption isotherms of inactivated CA and
activated CAs at varying temperatures prepared by uncatalysed
and KOH-catalysed methods. The surface areas, pore volume and
average pore radius of those samples are presented in Table 1.
The inactivated CA (CA-K-900) had the lowest N2 adsorption. Under the same activation conditions, the uncatalysed CAs also had
lower N2 adsorption than KOH-catalysed CAs. The surface area
and pore volume of the inactivated CA (CA-K-900) were 680 m2/g
and 0.40 cm3/g, respectively. After activation at 900 and 1000 C,
the surface area and pore volume were signicantly increased.
Meanwhile, the activated CAs without KOH also showed higher

2.2. Characterisation of CAs

The pyrolysis of as-prepared CAs was investigated by TGA and
DTA using a SDT V3.0F (2960) instrument under nitrogen ow.
The crystalline structure of CAs was measured with X-ray diffraction (XRD) on a Bruker D8 instrument scanning at 2h = 2060.
FT-IR spectra of carbon gel were also collected on a FT-IR spectrometer (Bruker IFS 66). Each of the IR spectra was the average of
32 scans at a speed of 2 s per scan in the range of 4004000 cm1.
The resolution of the spectrometer was set to 4 cm1. TEM image
was performed with a JOEL-2000FX microscope operating at
200 kV. Some powders were scraped from the bulk CA samples
and dispersed with ethanol onto a copper grid for microscopic
examination.
The textural structure of CAs was determined by nitrogen
adsorption at liquid nitrogen temperature. The nitrogen adsorption
isotherm was measured on an Autosorb-1 (Quantachrome, USA) at
196 C. The samples were evacuated at 200 C for 24 h prior to
the adsorption measurements. The specic surface area obtained
via N2 adsorption was determined by the standard Brunauer, Emmett and Teller (BET) analysis and the density function theory
(DFT) method was used to determine pore size distribution. The
micropore volume was calculated by the DFT and DubininRadushkevich (DR) methods. The average micropore radius (RGmic)
was also obtained by the DR model [10].

2.3. Adsorption tests

Adsorption tests of CAs were carried out for dye removal from
aqueous solution. In the tests, two different types of dyes were employed, one was basic blue 9 (BB9, CI 52015, C16H18N3ClS) and the
other was acid red 183 (AR183, CI 18800, C16H11ClN4Na2O8S2), supplied by Aldrich. In each run, 0.01 g of CA was mixed with 500 mL
dye solution in an Erlenmeyer ask and stirred at 700 rpm and
26 C. The initial concentration of BB9 was 10 mg/L and that of
AR 183 was 150 mg/L. At certain time interval, 5 mL solution was
withdrawn and separated using a centrifuge. The clear solution
was analysed for dye concentration on a spectrophotometer at
the wavelength of 663 and 494 nm for BB9 and AR183,
respectively.

159

S.K. Ling et al. / Journal of Colloid and Interface Science 357 (2011) 157162

Fig. 1. FT-IR of as-prepared CA samples: (a) noncatalysed sample and (b) KOH-catalysed sample. The full spectra between 400 and 4000 cm1 were shown in (A) and the
spectra between 400 and 2000 cm1 were shown in (B).

Fig. 2. TGA/DTA proles of CAs prepared with KOH and without KOH.

N2 adsorption (STD, cm /g)

3000
2500
2000

CA-K-900
CA-K-A900
CA-K-A1000
CA-A900
CA-A1000

1500
1000
500
0
0.0

0.2

0.4

0.6

0.8

1.0

p/p0
Fig. 4. N2 adsorption isotherms of various CAs.
Fig. 3. XRD proles of CAs prepared with KOH and without KOH.

surface area and pore volume than the inactivated CAs but the surface area and pore volume were lower than the corresponding
KOH-catalysed samples. The order of surface area and pore volume
of various CAs would be CA-K-A1000 > CA-A1000 > CA-KA900 > CA-A900 > CA-K-900.
The pore size distribution proles of various CAs are illustrated
in Fig. 5. The inactivated CA (CA-K-900) presented microporous
structure with pore size centred at 6 and 10 . CO2 activation could
produce larger pores. The sample CA-A900 exhibited a similar pro-

le to CA-K-900 while CA-K-A900 showed more pores at 11.8 . At

higher temperature activation, the porous structure was changed
signicantly by shifting micropores to mesopores. For CA-A1000
and CA-K-A1000, micropores at 6.3 were reduced and a major
pore size centred at 11.5 was appearing. Moreover, larger pores
at 25 were occurred. Based on the above pore size, it was deduced that KOH would cause the generation of larger pores and
highly developed porous structure.
In this investigation, it has been shown that CA could be
prepared at high solgel pH at round 9. CA-K-900 possesses a

160

S.K. Ling et al. / Journal of Colloid and Interface Science 357 (2011) 157162

Table 1
Textural properties of various CAs prepared.

Table 2
Kinetic parameters of dye adsorption on various CAs.

Sample

SBET
(m2/g)

Vtotal
(cm3/g)

RGmic
(nm)

Vmica
(cm3/g)

Vmicb
(cm3/g)

CA-K-900
CA-K-A900
CA-K-A1000
CA- A900
CA- A1000

680
1403
2796
1280
2139

0.40
0.64
1.89
0.53
1.00

0.62
0.67
0.94
0.61
0.78

0.39
0.58
1.33
0.50
0.81

0.34
0.52
1.06
0.45
0.75

Sample

CA-K-900
CA-K-A900
CA-K-A1000
CA-A900
CA-A1000
CA-K-900
CA-K-A900
CA-K-A1000
CA-A900
CA-A1000

CA-K-900
CA-A900
CA-K-A900
CA-A1000
CA-K-A1000

dV/dR (cc/A)

0.20

6
10 12

0.00
10

20

30

40

Pore size (A)

Fig. 5. The pore size distribution proles of various CAs.

well-developed porous structure. This suggested that CA precursors, F and R, could react with KOH as a catalyst to form a crosslinking network at high pH and the pores was dominated by
micropores. The characterisation by TGA/DTA showed that addition of KOH in the precursors would reduce the weight loss and
maintain stable carbonisation. XRD showed that less graphite
would be formed and CA would be more inclined to amorphous
phase. N2 adsorption also indicated that CAs synthesised with
addition of KOH catalyst would have higher surface area and larger
pore volume. These results suggested that the addition of KOH was
important in synthesis of porous CAs. During the synthesis of gels
and carbonisation, KOH would promote the formation of stable
networks and prevent the graphitisation. In addition, it is well
know that KOH is a commonly used reagent for carbon activation
in steam or CO2. Thus, it was also deduced that during the activation of CAs, the presence of KOH in carbon gels would also favour
the development of porous structure, resulting in higher surface
area and larger pore volume.

AR183

8.90  10
4.04  105
2.85  105
7.96  105
5.87  105
1.69  103
1.49  104
4.60  105
2.97  105
2.29  105

q2e k2 t
1 qe k2 t

3000

200
CA-K-900
CA-A900
CA-K-A900
CA-A1000
CA-K-A1000
Second-order kinetics

CA-K-900
CA-A900
CA-K-A900
CA-A1000
CA-K-A1000
Second-order kinetics

2000
1500
1000
500
0

50

100

150

200

250

0.988
0.986
0.998
0.980
0.995
0.923
0.969
0.970
0.960
0.986

(B)

2500

100

232.6
301.6
399.2
360.8
405.6
352.9
490.9
890.1
1262.9
2686.9

R2

where qe is the adsorption equilibrium capacity (mg/g), qt is the

amount of dye adsorbed (mg/g) at various time t, and k2 is the rate
constant of the pseudo-second order equation (g/mg min).

(A)

300

0
0

qt

Adsorption (mg/g)

Adsorption (mg/g)

400

5

Fig. 6 shows the adsorption of two dyes on various CAs at varying time. For both dyes, CAs showed a fast increase in dye adsorption within 50 min and the adsorption approached to the
equilibrium at around 300 min. In general, activated CAs exhibited
higher adsorption and KOH-catalysed CAs presented higher
adsorption than uncatalysed samples. The order of adsorption followed
CA-K-A1000 > CA-A1000 > CA-K-A900 > CA-A900 > CA-K900, which was similar to the order of surface area and pore volume. This suggested that dye adsorption was strongly dependent
on the porous structure of CAs. At equilibrium, CA-K-900 would
have BB9 and AR183 adsorption at 200 and 280 mg/g, respectively.
CA-K-A1000 could have adsorption of BB9 and AR183 at 360 and
2650 mg/g, respectively.
In general, adsorption is dependent on porous structure and
surface functional properties. As discussed above, the presence of
KOH in CA gels promoted the formation of stable porosity and
development of larger pores during CO2 activation, which were
evidenced in Table 1. The higher surface area and pore volume of
KOH-catalysed CAs will provide more active sites for adsorption
of dye molecules. Also the presence of KOH could helps to produce
and modify the surface functional groups. These functional groups
with either positive charge or negative charge will favour the
adsorption of charged dye molecules.
The dynamic adsorption of both dyes was also investigated by
pseudo-second-order kinetics, which is presented in the following
equation.

0.10

BB9

K2 (g/mg min)

3.2. Adsorption of dyes

0.15

0.05

Second-order kinetics
qe (mg/g)

RGmic: average micropore radius.

a
Vmic: the total micropore volume as calculated by the D-R model.
b
Vmic: the total micropore volume as calculated by the DFT method.

0.25

Dye

300

350

50

100

Time (min)
Fig. 6. Dye adsorption on various CAs. (A) BB9, (B) AR183.

150

200

Time (min)

250

300

350

161

S.K. Ling et al. / Journal of Colloid and Interface Science 357 (2011) 157162

300

3000

(A)

CA-K-900
CA-A900
CA-K-A900
CA-A1000
CA-K-A1000

200

100

(B)

CA-K-900
CA-A900
CA-KA900
CA-A1000
CA-K-A1000

2500

Adsorption (mg/g)

Adsorption (mg/g)

400

2000
1500
1000
500

10

12

14

16

18

20

t1/2 (min1/2)

1/2

10

12

14

16

18

20

1/2

(min )

Fig. 7. Intraparticle diffusion plots of dye adsorption on various CAs. (A) BB9, (B) AR183.

From the curve tting (Fig. 6) and the kinetic parameters (Table
2) obtained from Eq. (1), the dynamic adsorption of two dyes on
various CAs was tted well to the second-order kinetics based on
equilibrium adsorption and regression coefcients. It is known that
the pseudo-second order equation assumes a chemical adsorption
mechanism. For CAs, several functional groups are usually presented such as OH, COOH and they can act as binding sites for
dye adsorption. For BB9 and AR183, the mechanism may be proposed as follows.

COO NCS BB9 ! COOSCN

COH CSO AR183 ! COSC OH

From a mechanistic viewpoint to interpret the experimental

data, it is necessary to identify the steps involved during adsorption, described by external mass transfer (boundary layer diffusion) and intraparticle diffusion. An intraparticle diffusion model
is dened as follows:

qt ki t 1=2

where the parameter, ki (mg/g min1/2), is the diffusion coefcient. In

theory the plot between qt and t1/2 is given by multiple regions representing the external mass transfer followed by intraparticle diffusion in macro, meso, and micropore. Fig. 7 shows that there were
two linear portions for the two dye adsorption, suggesting the dye
adsorption on CAs is diffusion control and proceeds in two steps.
Previously, different activated carbon samples have been tested
for BB9 adsorption. Wang et al. [25] reported BB9 adsorption on
commercial activated carbons with capacity of 200300 mg/g at
10 ppm BB9 in solution. El Qada et al. [26] investigated BB9
adsorption onto activated carbon produced from steam activated
bituminous coal and reported the adsorption capacity of 230
340 mg/g. Hameed et al. [27] also reported several activated carbons from different sources and reported the adsorption capacity
of 140470 mg/g at BB9 concentration of 100500 mg/l. For
AR183, few investigations have been reported. Aydn and Yavuz
[28] studied the adsorption of AR183 onto activated carbon, raw
kaolinite and montmorillonite. The adsorption capacity was
1495, 111, 29 and 19 mg/g for the commercial activated carbon,
activated carbon obtained from hazelnut, raw kaolinite and montmorillonite at 25 C, respectively. Thus it is seen that the activated
CAs exhibited much higher adsorption capacity of both acid and
basic dyes than other adsorbents reported previously and they
could be used as effective adsorbents for dye removal in water
treatment.

4. Conclusion
Activated CAs were synthesised with addition of KOH catalyst
in a solgel solution and under CO2 activation. The structure and
adsorption properties were compared with CAs without KOH
catalysis. The KOH catalyst was found to play an important role
in improving the gel stability during carbonisation process. The
addition of KOH would also improve the development of porous
structure of CAs by improving larger pores. Activation with CO2
at a high temperature also resulted in higher surface area and
larger pore volume. In adsorption of dyes in aqueous solution,
CAs were much effective for both basic and acid dye adsorption
and the porous structure was the major factor inuencing the
adsorption. The dynamic adsorption of dyes on CAs could be
described by the second-order kinetic model and the equilibrium
adsorption was higher compared with some activated carbons.
Acknowledgement
Partial support from ARC LIEF Grant LE0775551 (XRD) is
acknowledged.
References
[1] A. Dabrowski, P. Podkoscielny, Z. Hubicki, M. Barczak, Chemosphere 58 (2005)
10491070.
[2] Y. Oren, Desalination 228 (2008) 1029.
[3] H. Guo, Y. Wyart, J. Perot, F. Nauleau, P. Moulin, Water Res. 44 (2010) 4157.
[4] S. Wang, Y. Peng, Chem. Eng. J. 156 (2010) 1124.
[5] J.J. Heijnen, A. Mulder, W. Enger, F. Hoeks, Chem. Eng. J. 41 (1989) B37B50.
[6] O. Lefebvre, R. Moletta, Water Res. 40 (2006) 36713682.
[7] S. Wang, H.M. Ang, M.O. Tad, Chemosphere 72 (2008) 16211635.
[8] S. Wang, H. Wu, J. Hazard. Mater. 136 (2006) 482501.
[9] C. Moreno-Castilla, F.J. Maldonado-Hodar, Carbon 43 (2005) 455465.
[10] H.Y. Tian, C.E. Buckley, S. Mule, M. Paskevicius, B.B. Dhal, Nanotechnology 19
(2008) 475605.
[11] S.A. Al-Muhtaseb, J.A. Ritter, Adv. Mater. 15 (2003) 101114.
[12] J. Goel, K. Kadirvelu, C. Rajagopal, V.K. Garg, J. Chem. Technol. Biotechnol. 80
(2005) 469476.
[13] M. Sanchez-Polo, J. Rivera-Utrilla, J. Mendez-Diaz, J. Lopez-Penalver, Ind. Eng.
Chem. Res. 47 (2008) 60016005.
[14] K. Kadirvelu, J. Goel, C. Rajagopal, J. Hazard. Mater. 153 (2008) 502507.
[15] J.H. Ramirez, F.J. Maldonado-Hodar, A.F. Perez-Cadenas, C. Moreno-Castilla,
C.A. Costa, L.M. Madeira, Appl. Catal. B 75 (2007) 312323.
[16] W.C. Li, G. Reichenauer, J. Fricke, Carbon 40 (2002) 29552959.
[17] H.J. Kim, W.I. Kim, T.J. Park, H.S. Park, D.J. Suh, Carbon 46 (2008) 1393
1400.
[18] R. Brandt, R. Petricevic, H. Probstle, J. Fricke, J. Porous. Mater. 10 (2003) 171
178.
[19] D. Fairen-Jimenez, F. Carrasco-Marin, C. Moreno-Castilla, Carbon 44 (2006)
23012307.
[20] M. Reuss, L. Ratke, J. SolGel Sci. Technol. 47 (2008) 7480.
[21] M. Reuss, L. Ratke, J. SolGel Sci. Technol. 53 (2010) 8592.
[22] D.C. Wu, R.W. Fu, Z.Q. Yu, J. Appl. Polym. Sci. 96 (2005) 14291435.
[23] H.Y. Tian, C.E. Buckley, S.B. Wang, M.F. Zhou, Carbon 47 (2009) 21282130.

162

S.K. Ling et al. / Journal of Colloid and Interface Science 357 (2011) 157162

[24] L. Zubizarreta, A. Arenillas, A. Dominguez, J.A. Menendez, J.J. Pis, J. Non-Cryst.

Solids 354 (2008) 817825.
[25] S.B. Wang, Z.H. Zhu, A. Coomes, F. Haghseresht, G.Q. Lu, J. Colloid Interface Sci.
284 (2005) 440446.

[26] E.N. El Qada, S.J. Allen, G.M. Walker, Chem. Eng. J. 124 (2006) 103110.
[27] B.H. Hameed, A.T.M. Din, A.L. Ahmad, J. Hazard. Mater. 141 (2007) 819825.
[28] A.H. Aydn, . Yavuz, Indian J. Chem. Technol. 11 (2004) 8994.