Journal of Colloid and Interface Science 368 (2012) 521527

Contents lists available at SciVerse ScienceDirect

Journal of Colloid and Interface Science

www.elsevier.com/locate/jcis

Preparation of activated carbon derived from cotton linter bers by fused

NaOH activation and its application for oxytetracycline (OTC) adsorption
Yuanyuan Sun a, Qinyan Yue a,, Baoyu Gao b, Qian Li a, Lihui Huang a, Fujiang Yao b, Xing Xu a
a
b

Shandong Provincial Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, Jinan 250100, China
Shandong Haobang Bio-Energy Limited Liability Company, Zouping 256200, China

a r t i c l e

i n f o

Article history:
Received 9 September 2011
Accepted 26 October 2011
Available online 6 November 2011
Keywords:
Activated carbon
Cotton linter bers
Fused NaOH activation
Oxytetracycline

a b s t r a c t
The objective of this research is to produce high surface areaactivated carbon derived from cotton linter
bers by fused NaOH activation and to examine the feasibility of removing oxytetracycline (OTC) from
aqueous solution. The cotton linter bers activated carbon (CLAC) was characterized by N2 adsorption/
desorption isotherms, Fourier transform infrared spectroscopy (FTIR), and scanning electron microscope
(SEM). The results showed that CLAC had a predominantly microporous structure with a large surface
area of 2143 m2/g. The adsorption system followed pseudo-second-order kinetic model, and equilibrium
was achieved within 24 h. The equilibrium data were described well by Langmuir isotherm. Thermodynamic study showed that the adsorption was exothermic reaction at low concentration and became
endothermic nature with the concentration increasing. Competitive adsorption took place in the weakly
acidic to neutral conditions. Under the strong acidity or strong alkaline condition, the adsorption of the
oxytetracycline was hindered by electrostatic repulsion. The adsorption mechanism depended on the pH
of the solutions as well as the pKa of the oxytetracycline.
2011 Elsevier Inc. All rights reserved.

1. Introduction
With the rapid development of Chinese textile industry, more
and more wastes and by-products are generated in the production
process. Cotton linter is the by-product from cotton textile production. As an essentially regenerative natural resource, it has been
applied widely owing to its low cost.
Activated carbon is a promising material for contaminant
adsorption because of its efciency, stability, and potential for reuse. However, the extensive use of activated carbon was restricted
due to the high cost. How to prepare high-quality activated carbon
at low cost became a hot topic. The textual and chemical characteristics of the activated carbon depend on the nature of the precursor
used as well as the methods and conditions of production. Activated carbon can be produced by all kinds of carbonaceous materials as precursors [1]. Production of activated carbons from ber
could be another appropriate solution for cotton linter utilization.
Previous studies showed that NaOH could be successfully used to
develop activated carbons with large surface area and micropore
volumes [2]. The most important advantages over KOH are lower
price and less corrosive behavior. However, none of the researches
have prepared activated carbon by fused NaOH. By this method,

Corresponding author. Fax: +86 531 88364513.

E-mail address: qyyue58@yahoo.com.cn (Q. Yue).
0021-9797/$ - see front matter 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2011.10.067

activated carbon can be obtained at low temperature without N2

protection. Accordingly, the cost can also be reduced, on which
few researches have focused.
Oxytetracycline is one of the most common members of antibiotics. Recently, there has been much concern about the presence
of antibiotics in the aquatic and soil environments because of their
toxicity, disturbance to the ecosystem function, and poor biodegradability. The continuous release of antibiotics into aquatic
environments increases the potential for antibiotic resistance
among microbial populations [3], and the degradation by-products have been proven even more toxic than the parents [4]. Thus,
more researchers try to develop new processes for more effective
treatment for these pollutants. There are many techniques
available for oxytetracycline removal including adsorption [5],
enzymatic degradation [6], oxidation [7], and photochemical
degradation [8]. Among these methods, adsorption technology
provides a practical method for the removal of pollutants from
wastewater in situ [9].
The objective of this study is (1) to prepare and evaluate the
characteristics of CLAC by N2 adsorption/desorption isotherms,
Fourier transform infrared spectroscopy (FTIR), and scanning electron microscope (SEM); (2) to describe the interactions of oxytetracycline with adsorbent as functions of solution pH, ionic strength,
temperature, and desorption study; and (3) to investigate the
adsorption kinetics, isotherms, and thermodynamic properties
involved in oxytetracycline-adsorbent surface reactions.

522

Y. Sun et al. / Journal of Colloid and Interface Science 368 (2012) 521527

2. Materials and methods

2.1. Chemicals
All the primary chemicals used in this study were of analytical
grade. Distilled water was utilized throughout the experiments for
solution preparation and glassware cleaning. Oxytetracycline
hydrochloride (USP grade) from BBI Corporation was purchased
from Canada. The main physicochemical property of OTC was
MW: 460.43 g/mol; H-bond acceptors: 11; H-bond donors: 8;
aqueous solubility: very soluble at pH 1, 4, 7, 8, 10; and pKa: 3.3,
7.3, and 9.1 [10].
2.2. Synthesis of activated carbons
Cotton linters (CLs) obtained from Jinan (China) were used as
precursor for the preparation of activated carbon ber. CLs were
washed with distilled water for the removal of impurities and
dried at 100 C for 10 h. NaOH was heated into molten state under
the condition of exposing to air. After being poured into fused
NaOH at 350 C (weight ratio of 1:3 for CLs/NaOH), the sample continued to be heated at a rate of 10 C min1 to 550 C and maintained at this temperature for 1 h. The resultant activated carbon
bers were cooled down to room temperature and were washed
several times with hot water, hydrochloric acid, and distilled water
for the removal of residual NaOH until the pH of ltrate was constant (at about 7.0). The wet mass was dried overnight by a drying
oven at 80 C. Finally, it was stored in desiccator for further
experiments.
2.3. Characterization methods
The morphologies of the samples were characterized by using a
scanning electron microscope (SEM Hitachi S4800, Japan). The BET
surface area and porous property of adsorbent were determined by
nitrogen adsorption/desorption isotherms at 77 K using a surface
area analyzer (Quantachrome Corporation, USA). The surface area
(SBET) and the total pore volume (Vtot) were obtained by the manufacturers software. The specic surface area was derived from the
BrunauerEmmettTeller (BET) theory. The micropore surface area
(Smic), external surface area (Sext) and the micropore volume (Vmic)
were evaluated by the t-plot method. The pore size distribution
was obtained by the Density Functional Theory (DFT) method.
The mean pore diameter (Dp) was obtained from DP = 4Vtot/SBET.
The surface functional groups before and after adsorption were recorded by the KBr technique using a Fourier transform infrared
(FTIR) spectrometer (Fourier-380FT-IR, USA), and the spectra were
recorded in the spectral range 4000400 cm1.

by a UVvis spectrophotometer (UV-754, Shanghai) at the maximum absorption wavelength. To control the experiments, the
OTC solutions without adsorbents were conducted under the same
condition.
In desorption experiments, the spent CLAC in kinetics experiments (OTC initial concentration 530.03 mg/L) was collected. Distilled water, HCl, NaOH, and acetic acid were used as desorption
agent. The percentage of desorption was calculated as follows:

%Desorption

md
 100
ma

where md is the amount of adsorbate desorbed (mg/L) and ma is the

amount of adsorbate adsorbed (mg/L).
3. Results and discussion
3.1. Characteristics of CLAC
3.1.1. SEM
Scanning electron microscopy (SEM) technique was used to
study the surface morphology of the adsorbent. The SEM micrographs of CLAC were presented in Fig. 1. CLAC appeared to have
a brous lament structure with cavities on its surface, indicating
that the wall of cotton linter became open and a wider porosity
was formed though NaOH activation. The irregular cavities created
were related to the dissolved organic matter, and the gas generated
in the activation progress. According to the literature [11,12],
NaOH reacted with ber, leading to the formation of hydrogen
and carbonates 12NaOH + (C6H10O5) + H2O = 6Na2CO3 + 12H2. It
can be obviously seen that big pores on the surface were attributed
to the gas released in the reaction progress, which could burst cotton linter and result in the holes.

2.4. Adsorption and desorption procedure

3.1.2. Surface area and pore structure

Specic surface area and pore size distribution are signicant
indicators for adsorbent. Fig. 2 depicts the pore size distribution
and N2 adsorption/desorption isotherms of the adsorbent. As
shown in Fig. 2, AC exhibits a narrow pore size distribution and is
essentially microporous (the hysteresis loop of adsorption/desorption isotherms of N2 at 77 K is type H4). The textual parameters
including the BET surface area and pore volume were calculated
and summarized in Table 1. The calculated BET surface area of CLAC
is 2143 m2/g, which is competitive as compared to other adsorbents
such as hemp-derived activated carbon ber (1350 m2/g) [13] and
Kenaf-based activated carbon ber (1031 m2/g) [14]. The pores of
CLAC are primarily distributed in the range of micropores (pore
width <2) with average pore width of 1.77 nm. Of its porosity,
65.7% specic surface area and 92% pore volume derives from the
micropores. CLAC also contains a certain amount of mesopores providing the transport channels for adsorbates adsorption, which is
essential for adsorption process.

The effects of time, solution pH, temperature, and ionic strength

were investigated in batch adsorption experiments. CLAC (0.06 g)
was weighed into 250-mL conical ask and suspended into
100 mL of OTC solution of different initial concentration with background electrolyte (0.01 M NaCl). The initial solution pH (2.012.0)
was adjusted to the required value by adding concentrated HCl or
NaOH solutions, and NaNO3 and CaCl2 were chosen to investigate
the separate effect on OTC adsorption onto CLAC. The asks were
sealed and placed in temperature-controlled water bath oscillator
at an agitation speed of 125 rpm (SHA-B, Shanghai, China) until
equilibrium was reached. Solution sample (OTC solutions 308.2,
363.7, and 530.3 mg/L) was taken off at various time intervals to
determine adsorption kinetics. Finally, the supernatant liquids
were ltered using 0.45 lm millipore membrane lters for analysis

3.1.3. Surface functional groups of adsorbent

Infrared spectra of the adsorbent before and after adsorption
were shown in Fig. 3. Unlike other activated carbon, CLAC did not
have many functional groups such as hydroxyl functional groups
(mOAH) with broad peak at about 3400 cm1 or lactones (mC@O@C)
located at about 1250 cm1. The pristine adsorbent absorbance at
1697.4 and 1559.2 cm1 corresponded to stretching vibrations of
carbonyl double bond (mC@O) and the carbon double bonds (mC@C),
respectively. The two peaks at 1164.2 and 1041.7 cm1 were the
characteristic of carbonoxygen single bond in phenol (mCAO).
The peak occurring at 700 cm1 was ascribed to ACH of aromatic
rings. After adsorbates were adsorbed onto the adsorbents, the
obvious new infrared frequencies detected at 15301560 and
1035.7 cm1 could be assigned to amine groups in OTC formulation

523

Y. Sun et al. / Journal of Colloid and Interface Science 368 (2012) 521527

Fig. 1. SEM of CLAC.

700

Volume adsorbed (cm /g)

1.0

0.4

650
600

%Transmittance

0.6

Dv(d) (cm /nm/g)

0.8

550
500
450
0.0

0.2

0.4

0.6

0.8

1.0

p/p0

0.2

0.0
0

10

12

14

16

18

Pore width (nm)

90
88
86
84
82
80
78
76
74
72
70
68
66
64
62
60
4000

before
after

3000

2000

1000

Wavenumbers (cm-1)
Fig. 2. Pore size distribution curves of the adsorbent and the insert gure is N2
adsorption/desorption isotherms.

Fig. 3. FTIR spectra of CLAC before and after OTC adsorption.

[3] and aliphatic amine (mCAN) (not aromatic amine with peek at
13801250 cm1) due to the attachment of the OTC species and
the adsorbent. This result indicated that OTC was immobilized onto
the AC samples probably by chemical bonding.

solution to the surface of CLAC. The equilibrium was achieved

within 24 h.
Adsorption kinetics is proposed in order to understand the
mechanism involved in the adsorption progress. Numerous theoretical models provide an insight into the mechanism by which
the adsorbate accumulates on the surface of an adsorbent [15]. In
order to appraise the adsorption kinetics of OTC onto CLAC, the
experimental data were simulated by pseudo-rst-order model
(Eq. (2)) [16], pseudo-second-order model (Eq. (3)) [17] and
intra-particle diffusion model (Eq. (4)) [18].

3.2. Adsorption kinetics

Fig. 4 shows the effect of initial OTC concentration, Co, on the
kinetics of adsorption of the OTC at natural pH (3.103.36), CLAC
dose 1.0 g/L, and 25 C. It can be observed that the adsorption
was found to be extremely rapid at the initial stage and slowed
down as the adsorption proceeded. In the rst 30 min, the percentages of total amounts adsorbed are about 83%, 72%, and 53% for initial OTC concentrations 308, 364, and 530 mg/L, respectively. This
was attributed to the high concentration gradient in the beginning,
which exhibited a high driving force for the migration of OTC from

lnqe  qt ln qe  k1 t
t
1
1

t
qt k2 q2e qe

2
3

qt kpi t 1=2 C i

Table 1
Porous structure parameters of CLAC.
SBET

Smic

Vtot

Vmic

(m2/g)

(m2/g)

(%)

(m2/g)

Sext
(%)

(cm3/g)

(cm3/g)

(%)

Vext
(cm3/g)

(%)

Dp
(nm)

2143

1408.1

65.7

734.9

34.3

0.948

0.875

92

0.073

1.77

SBET: BET surface area, Smic: micropore surface area, Sext: external surface area, Vt: total pore volume, Vmic: micropore volume, Vmec: external volume, Dp: mean pore diameter
(mode).

524

Y. Sun et al. / Journal of Colloid and Interface Science 368 (2012) 521527

800
750

Adsorption capacity (mg/g)

700
650
600
550
500
450
400

308 mg/L
364 mg/L
530 mg/L

350
300
250

procedure [20]. It also can be seen that as the initial concentration

of OTC solution changed from 308.2 to 530.3 mg/L, the amount adsorbed increased from 512.6 to 738.5 mg/g while the adsorption
reaction rate decreased from 0.440 to 0.044 L/min by comparing
of the values of the second-order rate constants (k2). Two dominant factors were suggested that: (i) the increase in the initial
OTC concentration could increase the driving force of the concentration gradient between the adsorbate in solution and adsorbate
in adsorbent and result in the higher OTC uptake, and (ii) the competition between adsorbate molecules for active sites at higher
concentration was larger, and long time was required to reach
equilibrium.

200
150
100
-200

3.3. Adsorption isotherms

0

200 400 600 800 1000 1200 1400 1600 1800 2000

time (min)
Fig. 4. Adsorption of OTC on CLAC as a function of contact time at different
concentrations. T = 25 C, natural pH (3.103.36) and CLAC dose 1 g/L.

where qe and qt (mg/g) are the amounts of OTC adsorbed at equilibrium and at time t (min); k1 (g/(mg min)), k2 (g/(mg min)), and kpi
are the pseudo-rst-order rate constants, pseudo-second-order rate
constants, and intra-particle diffusion rate constant of stage i,
respectively; Ci is the intercept at stage i. The plots of log (qeqt) versus t, t/qt versus t and qt versus t1/2 give the values of k1 and qe, k2
and qe, and kpi and Ci.
The correlated parameters at three concentrations were summarized in Table 2. It was observed that OTC adsorption onto CLAC
followed the pseudo-second-order reaction with high correlation
coefcients (R2 > 0.999), and the adsorption capacity (qe,cal) from
the pseudo-second-order model was much closer to the experimental data (qe,exp), which suggested that the process controlling
the rate may be a chemical sorption [19]. These observations were
consistent with sorption mechanisms, which proposed latter that
strong electrostatic interaction or ion exchange occured on the
adsorbent surface. The plots tted by intra-particle diffusion model
presented multi-linearity, which revealed that three steps occurred
in the adsorption process: (i) fast initial adsorption of the adsorbate from the bulk to outer surfaces of the adsorbent, (ii) penetration by slow diffusion into interlayer of the adsorbent, and (iii)
interactions with the surface atoms of the solid leading to chemisorption (strong adsorbateadsorbent interactions equivalent to
covalent bond formation) or weak adsorption (weak adsorbate
adsorbent interactions, very similar to van der Waals forces); in
the latter case, desorption may be the ultimate result [15]. This
result was in keeping with the universally recognized adsorption

Temperature is one of the most signicant factors that inuence

adsorption of organic compounds from aqueous solutions onto
activated carbon. Adsorption isotherms were studied at three different temperatures. The mathematical correlation of adsorption
isotherms is usually depicted by graphically expression of the solid
phase against the equilibrium concentration. Langmuir, Freundlich,
and Temkin model is widely used for adsorption isotherms.
Langmuir isotherm can be applied to homogeneous adsorption
(the adsorbed layer is one molecule in thickness). This empirical
model assumes that monolayer adsorption occurs at a nite number of denite localized sites, and there is no interaction between
the adsorbed molecules, even on adjacent sites [21]. Freundlich
formulation [22] was developed in 1906 to present the adsorption
properties of animal charcoal. This model refers to nonideal sorption on heterogeneous surfaces as well as multi-layer sorption,
with non-uniform distribution of adsorption heat and afnities
over the heterogeneous [23]. The Temkin isotherm [24] assumes
that the heat of adsorption (function of temperature) of all molecules in the layer would decrease linearly rather than logarithmically with coverage due to adsorbentadsorbate interactions, and
that the adsorption is characterized by a uniform distribution of
the binding energies, up to some maximum binding energy.
The Langmuir (Eq. (5)), Freundlich (Eq. (6)), and Temkin (Eq. (7))
adsorption isotherm have the following linear equations:

Ce
1
1

Ce
qe qm K l qm
1
ln qe ln kF ln C e
n
qe A ln K T A ln C e

5
6
7

where Ce (mg/L) is the equilibrium concentration of OTC, qe (mg/g)

the amount of OTC adsorbed under equilibrium, qm (mg/g) the

Table 2
Estimated kinetic model constants for OTC adsorption.
Kinetic models

Parameters

C0 (mg/L)
308.2

363.7

530.3

Pseudo-rst-order parameters

qe,exp (mg/g)
qe,cal (mg/g)
K1 (g/(mg min))
R2

512.6
144.6
0.00807
0.972

603.1
203.0
0.00503
0.974

738.5
389.30
0.00388
0.9778

Pseudo-second-order parameters

qe,exp (mg/g)
qe,cal (mg/g)
K2 (g/(mg min))
R2

512.6
515.5
0.00044
1

603.1
606.1
0.00016
0.99996

738.5
746.3
0.000044
0.9998

Intra-particle diffusion parameters

Kp1 (mg/g min0.5)

C1
(R1)2
Kp2 (mg/g min0.5)
C2
(R2)2

59.59
120.35
0.983
8.48
401.65
0.965

56.31
142.78
0.986
13.92
380.70
0.999

34.78
176.76
0.9855
5.748
557.37
0.980

525

Y. Sun et al. / Journal of Colloid and Interface Science 368 (2012) 521527

Table 3
Isotherm parameters for the removal of OTC by CLAC at different temperatures.

1.0
OTC

OTC

OTC

Parameters

Temperature (C)
19

35

50

Langmuir

qm (mg/g)
Kl (L/mg)
R2

869.57
0.1171
0.9993

1025.18
0.0791
0.9986

1340.82
0.0572
0.9989

Freundich

KF (mg/g (L/mg)1/n)
n
R2

433.31
8.791
0.9065

391.24
6.401
0.9136

293.92
3.820
0.938

Temkin

KT (L/mg)
B
R2

830.58
66.74
0.9395

36.30
102.62
0.9571

1.801
200.83
0.9798

Mass fraction OTC

0.8

Isotherms

0.6

0.4
OTC

-2

0.2

0.0
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0

pH

DH (KJ/mol)

DS (J/mol K)

212.43
419.10
621.07
827.74
1040.67
1247.34

44.54
20.63
34.28
20.40
14.27
11.17

106.75
34.59
155.53
103.21
80.72
69.09

DG (KJ/mol)
292 K

308 K

323 K

13.44
10.20
11.38
9.92
9.40
9.09

11.50
10.75
13.03
10.95
10.35
9.91

10.15
9.02
16.28
13.18
11.94
11.26

maximum adsorption amount, Kl (L/mg), KF (mg/g (L/mg)1/n), KT

(L/mg) the Langmuir, Freundlich and Temkin isotherm constant,
respectively. A = RT/b, R (8.314 J/(mol K)) the gas constant, T the
absolute temperature (K).
The calculated parameters of the models and the correlation
coefcients (R2) were given in Table 3. The results showed that
the afnity of OTC onto CLAC was dependent on the temperature
of the isotherm. The uptake of OTC was higher at higher temperatures. The increase in temperature seemed to provide the potential
activation energy required to change the reactants into the activated complex, thus enhancing the adsorption capacity by the activated adsorption. In addition, higher temperature may have a
swelling effect within the internal structure of the CLAC enabling
large OTC molecules to penetrate further [25]. Bouberka et al.
[26] reported that adsorption capacity of yellow 4GL (acid dye)
onto modied clays increased with the temperature increasing
from 30 to 50 C. Hameed et al. [25] also reported that the amount
of acid blue 62 dye absorbed on sepiolite at 50 C was higher than
that at 30 C. The maximum adsorption capacity at 292, 308, and
323 K calculated by Langmuir model was 869.57, 1025.18, and
1340 mg/g, respectively. CLAC had a large micropore area (87.5%)
and small average pore diameter (1.77 nm). Therefore, micropore
lling could play a dominant role in OTC adsorption onto CLAC. It
was also observed that Langmuir model gave the best description
among the three models with high R2 values (>0.99), indicating
the homogeneity of the surface active sites on CLAC and the monolayer adsorption of OTC. The values of n give an indication of the
favorability of the adsorbentadsorbate system.
3.4. Thermodynamics
For understanding the thermodynamic performances of OTC
adsorption onto CLAC, thermodynamic considerations were evaluated. The enthalpy change (DH), Gibbs free energy change (DG),
and entropy change (DS) can be calculated from the following
Gibbs free energy equations [27,28]:

Adsorption capacity (mg/g)

C0 (mg/L)

750

700

650

600

550

500

450
2

10

12

Initial pH
Fig. 5. OTC chemistry and solution speciation (a), and pH-sorption prole of OTC
onto CLAC (b).

800

Adsorption capacity (mg/g)

Table 4
Thermodynamic parameters for the uptake of OTC on CLAC calculated under standard
conditions.

700

600

NaNO3

500

CaCl2

400
0.00

0.02

0.04

0.06

0.08

0.10

C (mol/L)
Fig. 6. Effect of ion strength for the adsorption of OTC onto CLAC (C0 = 575 mg/L,
CLAC dosage = 0.6 g/L, pH = 3.03, temperature = 293 K, t = 24 h).

C Ad
Ce
DG RT ln K c
DS DH

ln K c
R
RT
Kc

8
9
10

where Kc is the equilibrium constant, CAd is the concentration of OTC

adsorbed on solid at equilibrium (mg/L), Ce is the equilibrium

526

Y. Sun et al. / Journal of Colloid and Interface Science 368 (2012) 521527

a
H+

OTC +00

OTC +0

H+

OTC +0

OTC +00

H+

OTC +00

OTC +0

H+

Adsorption

OTC +0

H+
H+

H+

H+

OTC +00

H+

H+

Adsorbent surface

Adsorbent surface

b
OTC +00

H+

OTC +0

OTC +

H+

OTC

OTC +0

OTC +0

H+

OHAdsorption

OH_

OTC +00
_

Adsorbent surface

OTC +0
_

Adsorbent surface

Fig. 7. Schematic diagrams for ion exchange and electrostatic attraction interactions between OTC and CLAC surface at pH below 4.8 (a) and at pH above 4.8 (b).

concentration of OTC in the solution (mg/L), R is the gas constant

(8.314 J/(mol K)), and T is the temperature in Kelvin. The values of
DH and DS can be obtained from the slope and intercept of Vant
Hoff plot of ln Kc versus 1/T. Table 4 showed the calculated data. The
thermodynamic parameters showed that the adsorption progress at
low concentration was different from that at high concentration.
Although the all negative values of DG veried that the adsorption
processes of OTC onto CLAC were spontaneous and feasible. At low
concentration, the increased DG values suggested that adsorption
was less spontaneous at higher temperature; at high concentration,
the highly negative values of DG at higher temperature indicated
that adsorption was more spontaneous. They have the opposite
trend. The value of DH was negative at low concentration and
positive at high concentration, which indicated that the interaction
between adsorbent and OTC was an exothermic reaction at low concentration and became endothermic with the concentration
increasing. The negative values of DS at low concentration indicated the decreasing randomness at the solidsolution interface
during the adsorption of OTC on CLAC. The positive values of DS
at high concentration indicated the increasing randomness and
the good binding afnity of OTC toward the adsorbent.

3.5. Effect of initial solution pH and ionic strength on adsorption of

oxytetracycline in CLAC
Both the adsorption capacity and adsorption mechanism were
strongly inuenced by the pH of the solutions. pH might change
the surface charge properties of OTC and CLAC, resulting in substantial differences in adsorption capacity. Fig. 5b displayed the
effects of the initial solution pH. As can be seen, the OTC uptake increased with increasing pH when pH was below 3, and the removal
capacity of OTC was not markedly affected by the increase in pH
from 3 to 5.8. However, the adsorption capacity decreased sharply
from 717.6 to 528.4 mg/g with the increase in pH from 5.8 to 10.6.

The cationic form of OTC+00 was primarily present at pH below

3.3. The zwitterion (+0) was the major form at pH between 3.3
and 7.3 and transformed into deprotonation species (+) with
pH increasing (Fig. 5a). Zeta potential showed that the surface
charge on CLAC was zero at pH about 4.8. The surface charge
was positive at pH < pHzpc and negative at pH > pHzpc. When the
solution pH was below 5.5, the dominant forms of OTC were
OTC+00 and OTC+0. The increasing solution pH increased the negative charge density of CLAC, thus enhancing the electrostatic
attraction between OTC+00 or OTC+0 and the positively or weakly
negatively charged CLAC surface, which would increase the
adsorption of OTC onto CLAC. However, the dominant species of
OTC were OTC+ and OTC0 when pH was above 5.5. The
increasing solution pH resulted in more negative charges on CLAC,
enhancing the electrostatic repulsion between OTC+ or OTC0
and CLAC surface charges. Therefore, it would decrease the adsorption of OTC onto CLAC. These ndings indicated that the oxytetracycline cation and zwitterion species exhibited a stronger
interaction with the CLAC than the oxytetracycline anionic species.
On the other hand, the changes in pH of the solution before and
after adsorption were due to the following reasons: (i) the OTC
concentration became lower and (ii) ion exchange might be one
of the mechanisms. The typical environmental pH of antibiotic
water environment (weakly acidic) was favorable for OTC
adsorption.
Effects of ionic competition on OTC adsorption were investigated
at pH 3.03 (natural pH without being adjusted) by conducting
adsorption experiments at two different salt concentrations (NaNO3
and CaCl2). CLAC adsorption capacity exhibited similar decreasing
trends at all ionic competitions (see Fig. 6). Ca2+ and Na+ ions might
compete with OTC+00 or OTC+0 for binding sites on CLAC surface,
which would negatively affect the interactions between OTC and
CLAC. The decrease in adsorption, to some extent, has been
interpreted to indicate the formation of outer sphere complexes in
the adsorption process. OTCadsorbent interactions such as

Y. Sun et al. / Journal of Colloid and Interface Science 368 (2012) 521527

electrostatic attraction, hydrogen bonding, or hydrophobic attraction were responsible for the adsorption. It can also be seen that
the ionic strength from 0 to 0.01 mol/L led to a sharp decrease in
the OTC uptake and that the increase in ionic strength from 0.01 to
0.1 mol/L did not result in a signicant decrease in OTC adsorption.
This phenomenon can be explained by the fact that a certain amount
of surface interactions were in the form of outer sphere complexes,
and there was little change to the inner-sphere surface complexes
that also existed in the surface interactions when the ionic strength
increased. In addition, the effect of Ca2+ was more apparent than that
of Na+. This was because Ca2+ had more positive charge, thus exhibiting a higher inhibition of OTC adsorption than Na+.

527

of OTC onto CLAC can be viewed as a spontaneous and preferential

process. OTC can be absorbed by CLAC through ion exchange, electrostatic attraction, and van der Waals forces. The mechanism for
OTC sorption on CLAC was illustrated in Fig. 7.
Acknowledgments
This research was supported by the National Natural Science
Foundation of China (50878121 and 21007034), Natural Science
Foundation of Shandong Province (ZR2010EQ031), Foundation for
Young Excellent Scientists of Shandong Province (BS2009NY005).
References

3.6. Desorption study

Desorption studies facilitate the further study of the adsorption
mechanism of OTC onto CLAC. Namasivayam and Yamuna [29]
have suggested that whether the adsorbate on the solid surface
could be desorbed by water, physisorption such as van der Waals
forces could occur, to some degree, between adsorbate and adsorbent surface. It was found that the desorption percentage of water
was 8.2%. Additionally, when HCl (0.1 M) was used for desorption,
very little (2.5%) was desorbed, which indicated the possibility of
ion-exchange adsorption [29]. As expected, in 0.1 M NaOH solution, 72% OTC was desorbed. At higher pH, the adsorbent surface
was negatively charged, which could lead to desorption of negatively charged OTC. Thus, NaOH could be chosen as an ideal chemical for OTC regeneration of on CLAC.
As mentioned above, OTC showed high afnity toward CLAC,
which probably accounted for the electrostatic attraction, ion
exchange, and van der Waals forces between the adsorbent
surface and the target adsorbates in solution. It was due to the
existence of the above-mentioned three interactions.
4. Conclusions
The prepared activated carbons bers by fused NaOH activation
had a predominantly microporous structure and a large surface
area. Pseudo-second-order kinetic model was found to provide better correlation of the adsorption kinetics data than rst-order kinetic model. The Langmuir model tted test data best. The OTC
uptake was found to be signicantly inuenced by the pH of the
solutions and weakly dependent on ionic strength. The adsorption

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]

L. Ren, J. Zhang, Y. Li, C. Zhang, Chem. Eng. J. 168 (2011) 53.

Y. Guo, J. Qi, S. Yang, K. Yu, Z. Wang, H. Xu, Mater. Chem. Phys. 78 (2003) 132.
T. Paul, P.L. Miller, T.J. Strathmann, Environ. Sci. Technol. 41 (2007) 4720.
C. Zhao, H. Deng, Y. Li, Z. Liu, J. Hazard. Mater. 176 (2010) 884.
J.P. Chen, L. Yang, Langmuir 22 (2006) 8906.
X. Wen, Y. Jia, J. Li, J. Hazard. Mater. 177 (2010) 924.
L. Hu, A.M. Stemig, K.H. Wammer, T.J. Strathmann, Environ. Sci. Technol. 45
(2011) 3635.
M. Sturini, A. Speltini, F. Maraschi, A. Profumo, L. Pretali, E. Fasani, A. Albini,
Environ. Sci. Technol. 44 (2010) 4564.
L. Wang, J. Zhang, R. Zhao, Y. Li, C. Li, C. Zhang, Bioresour. Technol. 101 (2010)
5808.
P. Kulshrestha, R.F. Giese, D.S. Aga, Environ. Sci. Technol. 38 (2004) 4097.
M. Ishida, K. Otsuka, S. Takenaka, I. Yamanaka, J. Chem. Technol. Biotechnol. 80
(2005) 281.
M. Ishida, S. Takenaka, I. Yamanaka, K. Otsuka, Energy fuels 20 (2006) 748.
J.M. Rosas, J. Bedia, J. Rodrguez-Mirasol, T. Cordero, Fuel 88 (2009) 19.
E.M. Cuerda-Correa, A. Macas-Garca, M.A.D. Dez, A.L. Ortiz, Micropor.
Mesopor. Mater. 111 (2008) 523.
S. Sen Gupta, K.G. Bhattacharyya, Adv. Colloid Interface Sci. 162 (2011) 39.
Y. Yao, F. Xu, M. Chen, Z. Xu, Z. Zhu, Bioresour. Technol. 101 (2009) 3040.
J. Zhang, Q. Shi, C. Zhang, J. Xu, B. Zhai, B. Zhang, Bioresour. Technol. 99 (2008)
8974.
D. Kavitha, C. Namasivayam, Bioresour. Technol. 98 (2007) 14.
L. Wang, L. Yang, Y. Li, Y. Zhang, X. Ma, Z. Ye, Chem. Eng. J. 163 (2010) 364.
M. Dogan, Y. zdemir, M. Alkan, Dyes Pigm. 75 (2007) 701713.
I. Langmuir, J. Am. Chem. Soc. 40 (1918) 1361.
H.M.F. Freundlich, J. Chem. Phys. 57 (1906) 385.
K. Vijayaraghavan, T.V.N. Padmesh, K. Palanivelu, M. Velan, J. Hazard. Mater.
133 (2006) 304.
M.I. Tempkin, V. Pyzhev, Acta Phys. Chim. USSR 12 (1940) 327.
B.H. Hameed, A.A. Ahmad, N. Aziz, Chem. Eng. J. 133 (2007) 195.
Z. Bouberka, S. Kacha, M. Kameche, S. Elmaleh, Z. Derriche, J. Hazard. Mater.
119 (2005) 117.
G. McKay, M.J. Bino, A.R. Altamemi, Water Res. 19 (1985) 491.
D. Mohan, V.K. Gupta, S.K. Srivastava, S. Chander, Colloids Surf. A. 177 (2000)
169.
C. Namasivayam, R.T. Yamuna, Water Air Soil Pollut. 65 (1992) 101.