You are on page 1of 10

Chemical Engineering Journal 217 (2013) 119–128

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal


journal homepage: www.elsevier.com/locate/cej

Preparation, characterization and adsorption potential of the


NH4Cl-induced activated carbon for the removal of amoxicillin antibiotic
from water
Gholamreza Moussavi a,⇑, Ahamd Alahabadi a, Kamyar Yaghmaeian b, Mahboube Eskandari a
a
Department of Environmental Health Engineering, Tarbiat Modares University, Tehran, Iran
b
Center for Water Quality Research (CWQR), Institute for Environmental Research (IER), Tehran University of Medical Sciences, Tehran, Iran

h i g h l i g h t s

" NH4Cl-induced activated carbon (NAC) was prepared from a waste biomass.
2
" The prepared NAC was a mesoporous material with a BET surface area of 1029 m /g.
" NAC had a high density of hydroxyl, carbonyl, carboxylic, and carboxylate surface functional groups.
" NAC had an amoxicillin adsorption rate of up to 4.3 times greater than the standard AC.
" NAC had an amoxicillin adsorption capacity of 2 times of that of the standard AC.

a r t i c l e i n f o a b s t r a c t

Article history: The preparation, characterization and application of NH4Cl-induced activated carbon (NAC) for amoxicillin
Received 29 October 2012 removal from the contaminated water were studied. The prepared NAC had a specific surface area of
Accepted 14 November 2012 1029 m2/g and a mean pore volume of 2.46 nm. Over 99% of 50 mg/L amoxicillin (AMX) was adsorbed using
Available online 29 November 2012
0.4 g NAC/L at the optimum solution pH of 6; while standard activated carbon (SAC) could only adsorb
around 55% of AMX under similar experimental conditions. Kinetic analysis revealed that adsorption exper-
Keywords: imental data for both NAC and NAC were best fitted by the pseudo-second-order model, with the greater
Waste biomass
rate for NAC than for SAC. Results of equilibrium experiments indicated that adsorption of AMX onto SAC
Activated carbon
Chemical activation
and NAC were better described by the Langmuir model. The maximum adsorption capacity of AMX onto
Amoxicillin SAC and NAC was 262 and 437 mg/g, respectively. AMX adsorption onto SAC increased from 76.8% to
Adsorption 92% with increased temperature from 10 to 35 °C. However, a further increase of temperature to 50 °C
led to declining AMX removal to 78.1%. Overall, these results indicate that developed NAC was an efficient
adsorbent that presents an attractive adsorbent method for application in treating contaminants in water.
Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction compounds and are thus regarded as hazardous chemicals. The re-
lease of pharmaceuticals into the environment thus results in con-
Pharmaceuticals present class of health care products that are tamination of aquatic or terrestrial ecosystems [2]. Effluent
intensively used worldwide to promote human health and well containing antibiotics needs to be treated chemically or physically
being as well applications in animal care and agriculture. Among to prevent the adverse effects from contaminated water. Although
pharmaceuticals, antibiotics are the most widely used drug for chemical processes, especially advanced oxidation processes, can
the prevention or treatment of bacterial infections in humans, ani- often degrade and decompose antibiotics’ molecules into simple
mals and plants [1]. Antibiotics are released into bodies of water compounds and/or mineralize them, these processes are very
mainly through effluent of municipal wastewater treatment plants, expensive and difficult to operate for complete elimination of re-
as well as through effluent from pharmaceutical manufacturing calcitrant compounds including antibiotics. Physical techniques re-
plants. Pharmaceuticals are gaining recognition as being environ- main the most appropriate treatment option and adsorption is
mental contaminants, classified as recalcitrant bio-accumulative among the most efficient of these techniques for removing organic
compounds from industrial effluent [2,3]. Adsorption is efficient,
⇑ Corresponding author. Tel.: +98 21 82883827; fax: +98 21 82883825. simple to design and operate and it is unaffected by toxicity as well
E-mail address: moussavi@modares.ac.ir (G. Moussavi). as being inexpensive [4]. Adsorption therefore presents an efficient

1385-8947/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cej.2012.11.069
120 G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128

and economical technique widely used to accumulate organic first debarked, then crushed into small parts and ground. The wood
compounds from contaminated streams onto solid material [2]. A particles were carbonized at 700 °C for 1 h in an oven under a N2 gas
saturated adsorbent is easily separated from a treated stream for stream. The carbonized granules were then soaked in the NH4Cl
regeneration or disposal in an environmentally acceptable way. solution with different concentrations (NH4Cl to char ratio was
The performance of an adsorption process is affected by an adsor- 0–10 w%); each solution was shaken for 24 h at room temperature.
bent’s characteristics as well as by the composition of a waste The mixture was then sieved to separate the char, and the sepa-
stream (mainly by the structural properties of the contaminant(s) rated char was then oven-dried at 105 °C for 24 h. Pre-treated char
and the composition of the waste stream) [5]. was finally activated by oven-heating at various temperatures
Several materials including activated carbon, waste material ranging from 500 to 900 °C and activation time between 0.5 and
and minerals either plain or modified have been examined so far 2.5 h under N2 atmosphere. Accordingly, the prepared NH4Cl-in-
as adsorbents for organic contaminants. Due to its porosity and duced activated carbon (NAC) was ground to particles according
very high specific surface area and capacity, activated carbon is to a mesh size of 20/16 for use in the adsorption experiments.
the most widely used adsorbent to remove various classes of com- The SAC and NAC were characterized for surface morphology,
pounds from contaminated streams [6]. However, the main disad- specific surface area, pore volume and size, pH of point of zero
vantages of activated carbon are high production and regeneration charge (pHpzc) and surface functional groups. The surface mor-
costs [5]. Interestingly, it has been found that adsorption of antibi- phology of the adsorbent was observed using the scanning electron
otics onto activated carbon is significantly influenced by character- microscopy (SEM) technique (Philips-XL30 Electron Microscope).
istics of the activated carbon [7]. Therefore, the method of carbon Evaluations for BET (Brunauer–Emmett–Teller) specific surface
activation aimed at improving its adsorption capability and thus area and pore volume and size were determined by a nitrogen
reducing the rate of activated carbon consumption to reduce the gas adsorption analyzer (BELSORP-miniII instrument). The pHpzc
cost of activated carbon adsorption making more it cost-effective. of adsorbents was determined according to the pH drift procedure.
Several chemicals have been used in the process of microwave Surface functional groups of the selected adsorbents were deter-
assisted activation of carbon, and the best activation has been mined using Fourier Transform Infrared (FTIR) spectroscopy with
obtained with the presence of alkaline hydroxide of KOH [8]. a Nicolet spectrometer.
However, the main challenge with this technique of microwave-
assisted modification of activated carbon using metal salts is the
metal compound formed during the activation process [8]. These 2.3. Adsorption experiments
metal compounds are deposited in the internal structure and pores
of the activated carbon, thereby reducing its pore volume and 2.3.1. Parametric and kinetic experiments
specific surface area. Accordingly, non-metal hydroxides might Parametric experiments were conducted in the stirred
be used as an alternative chemical to activate carbon. (100 rpm) glass reactor in a temperature-controlled incubator. In
This study was set up to examine the NH4Cl-induced activation of each test, 50 mL of contaminated water containing given concen-
carbon prepared from pomegranate wood, an agricultural waste, to trations of AMX was transferred into the reactor, its pH was regu-
compare the capability of NH4Cl-induced activated carbon (NAC) lated (with 0.1 N HCl or NaOH) at the desired level, the
with that of standard activated carbon (SAC) in adsorption of amox- predetermined amounts of activated carbon (NAC or SAC) were
icillin (AMX) as a model of an antibiotic, from contaminated water. added to the reactor and the suspension was mixed at 100 rpm.
AMX, a b-lactam antibiotic, is the drug of choice prescribed for the When the predetermined mixing time was up, the suspension
treatment of a number of human bacterial infections, as a growth was filtered using a cellulose acetate filter with 0.2 lm pore size
promoter for animals and in agriculture to prevent infections [9]. and the filtrate was analyzed for residual AMX. The concentration
Influences of the following basic variables were evaluated in the of AMX in a solution was measured using a Knauer HPLC (C18 ODS
tests on adsorption of AMX; pH of the solution, activated carbon con- column; 250  4.6  5) with a UV detector 2006 at a wavelength of
centration, AMX concentration, reaction time and solution temper- 190 nm. The mobile phase was a mixture of buffer phosphate with
ature. The equilibrium, kinetic, mechanism and thermodynamics of pH = 4.8 and acetonitrile with a volumetric ratio of 60/40 with an
AMX adsorption onto SAC and NAC were also evaluated. injection flow rate of 1 mL/min. The pH level of samples was mea-
sured using a pH meter (Sense Ion 378, Hack). The temperature of
2. Materials and methods solutions was measured using a mercury thermometer. Parameters
at this stage were as follows; water pH (2–9), initial AMX concen-
2.1. Material tration (10–100 mg/L), solution temperature (10–50 °C), and
reaction time (0.3–100 min) in kinetic experiment and 6 h in
Analytical grade AMX (chemical formula: C16H19N3O5S, molar equilibrium tests). Experimental runs and conditions are given in
mass: 365.4 g/mol) was purchased from Sigma Aldrich Co., and used Table 1. All tests were performed in duplicate to ensure reproduc-
as received. As stated above, two types of activated carbon were used ibility of the results; the mean of these two measurements was ta-
in this study, NAC and SAC. The NAC was prepared from pomegranate ken to represent each evaluation. Calculations of amounts of
wood waste according to the procedure explained in the following sec- adsorption of AMX onto prepared NAC and SAC were based on re-
tion. Activated carbon obtained from Merk Co. was used as the SAC to moval percentages (Eq. (1)) and adsorption capacity (Eq. (2)).
make a comparison with and to highlight the adsorption potential of
the NAC prepared in this study. The contaminated water used in the ðC 0  C t Þ
experiments was prepared from mixing aliquots of AMX stock solu- Amoxicillin removal ð%Þ ¼  100 ð1Þ
C0
tions (1000 mg/L) with distilled water. A stock solution of AMX was
made by dissolving 1 g of analytical grade AMX in distilled water.
VðC 0  C t Þ
Adsorption capacity ðmg=gÞ ¼ ð2Þ
2.2. Activated carbon preparation and characterization m

Dried pomegranate wood was used as the base material for the where C0 and Ct represent AMX concentration at initial and time t of
preparation of activated carbon. The pomegranate wood was ob- the contact time, respectively. V is the volume of the AMX solution
tained from a local farm as discarded agricultural waste and was and is the mass of adsorbent added.
G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128 121

Table 1
The experimental runs and conditions.

Run Experiment AMX concentration (mg/L) pH Adsorbent Contact time (min) Temperature (°C)
concentration
(g/L)
SACa NACb
1 Adsorption potential of NAB activated under different conditions 50 Natural (6.8) – 0.4 10 25
2 Effect of solution pH 50 2–9 0.8 0.4 10 25
3 Effect of AMX concentration and contact time 10–100 6 0.8 0.8 0.3–100 25
4 Equilibrium adsorption 50–500 6 0.8 0.8 360 25
5 Effect of temperature 50 6 0.8 0.4 10 10–50
a
Standard activated carbon.
b
NH4Cl-induced activated carbon.

2.3.2. Adsorption equilibrium experiments and isotherm modeling activation temperature, and activation time was evaluated on the
Test to evaluate an adsorption equilibrium onto prepared NAC activation process. The activation performance index for each acti-
and SAC were done by adding a given mass of 0.04 g SAC or NAC vation condition was the capability of the prepared activated carbon
into a series of Erlenmeyer flasks, containing 50 mL of various con- in adsorption of AMX tested under experimental conditions given in
centrations, in the range of 50–500 mg/L, with a pH of 6 (optimum Table 1 (run 1). Fig. 1 shows the effect of NH4Cl to char ratio ranging
pH). Flasks were then stirred for 6 h at a constant temperature from 0 to 10 wt% on the adsorption potential of the produced carbon
(25 °C) to attain equilibrium, after which the suspension was ana- activated at 800 °C for 2 h. As seen in Fig. 1, carbon activated without
lyzed as described in Section 2.3.1. The equilibrium adsorption of NH4Cl induction removed only 56% of the AMX under the selected
AMX onto prepared SAC and NAC was modeled using isotherm conditions. The percentage of AMX removal increased when the
models Langmuir, Freundlich, and Dubinin-Radushkevich by fitting amount of NH4Cl was increased to 2% at which around 87% removal
results from the experiment on to the models. of AMX was observed. However, a further increase of NH4Cl to
10 wt% resulted in decreasing the amount of AMX that was removed.
2.3.3. Adsorption thermodynamic and modeling This can be attributed to destruction of the carbon texture likely due
The thermodynamics of AMX adsorption onto the prepared NAC to generation of the excessive gas from the NH4Cl combustion. Over-
and SAC was evaluated using adsorption experiments performed by all, it is seen that the when the activation of char was induced with
an Erlenmeyer flask, containing 50 mL of 50 mg/L solution with the NH4Cl at optimum ratio of 2 wt%, the prepared activated carbon had
optimum pH of 6 into which amounts of 0.04 g SAC or NAC were an AMX adsorption potential considerably (31%) greater than the
added and shaken in a temperature-controlled shaker incubator. carbon activated without NH4Cl induction. This finding confirms
This experiment was conducted at various temperatures between the efficacy of the selected method of carbon activation.
10 and 50 °C. At the end of each test the suspension was analyzed The effect of activation temperature ranging from 500 to 900 °C
as described in Section 2.3.1. The thermodynamics of AMX adsorp- on the quality of the prepared carbon is indicated in Fig. 2. Accord-
tion onto NAC and SAC was analyzed using an estimated change in ing to the results given in Fig. 2, the carbon activated at 800 °C at-
adsorption free energy (DG), adsorption enthalpy (DH), and tained the highest AMX adsorption. Fig. 3 depicts the results of the
adsorption entropy (DS) as shown in the following equations [10]. effect of activation time between 0.5 and 2.5 h on the quality of the
prepared activated carbon. It is observed in Fig. 5 that the carbon
3. Results and discussion activated for 2 h at 800 °C achieved the highest AMX removal of
86.6%. Accordingly, the optimum conditions for the activation of
3.1. The effect of activation conditions on the adsorption potential of prepared carbon in these tests were found to be a NH4Cl to carbon
NAC ratio of 2 wt%, temperature of 800 °C, and activation time of 2 h
was selected as the optimum activation temperature for the rest
At the beginning of the study, in order to obtain the optimum of experiments.
activation conditions, the effect of NH4Cl to carbon ratios,

Fig. 1. The effect of NH4Cl concentration in activation process on the adsorption Fig. 2. The effect of activation temperature (500–900 °C) on the adsorption
potential of prepared carbon (activation temperature and time was 800 °C and 2 h, potential of prepared carbon (NH4Cl concentration and activation time was 2%
respectively). and 2 h, respectively).
122 G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128

Textural characteristics of SAC and NAC are given in Table 2. As


seen in Table 2, the activated carbon prepared in this work (NAC)
had a high BET specific surface area of 1029 m2/g, which is close
to that of SAC supplied by Merck Co. (Table 2). Both SAC and
NAC are mesoporous materials. However, the total volume and size
of pores in NAC were around 10% greater than those in SAC. The
larger pore volume and size of NAC compared with SAC can be re-
lated to the generation and release of gases due to the evaporation
and volatilization of ammonium impregnated onto carbon at a
high activation temperature (800 °C). Overall, it was demonstrated
that NAC prepared from pomegranate wood chips had comparable
textural characteristics with the high-grade standard carbon. The
pHpzc of SAC and NAC were 7.4 and 6.6, respectively; the surface
of carbons would be positively charged at pH below pHpzc and
negatively charged at solution pH greater pHpzc.

Fig. 3. The effect of activation time (0.5–2.5 h) on the adsorption potential of


prepared carbon (NH4Cl concentration and activation temperature was 2% and
800 °C, respectively).

3.2. Activated carbon characterization

The FTIR spectra of NAC and SAC are shown in Fig. 4, indicating
several intense peaks revealing the presence of a number of impor-
tant functional groups on the surface of the carbon. An absorption
band is observed at wave numbers between 3200 and 3600 cm1
for both SAC and NAC (with maximums of 3425 and 3443 cm1
for SAC and NAC, respectively), which is representative of hydroxyl
functional groups on the surface [11–13] of the selected activated
carbons. Absorption at wave numbers of 2929 cm1 for SAC and
2922 cm1 for NAC are attributed to aliphatic CAH interaction with
the surface of the adsorbent [13,14]. Absorption at wave number of
2350 cm1 is attributed to C@O asymmetric stretching vibration of
CO2 [15]. Absorption at wave number of 1604 cm1 observed at
FTIR spectra of SAC is attributed to C@O stretching vibration of car-
bonyl groups [16]. NAC had no absorption at 1594 cm1. The
strong absorption bands at wave numbers of from 1420 to
1090 cm1 is also observed in the FTIR spectra of both SAC and
NAC, demonstrating that oxygen was functioning to stretch car-
boxylic groups and that carboxylate groups were present on the
surface of the activated carbon [17,18]. According to Fig. 4, the FTIR
spectrum of NAC shows lower transmission (thus higher absorp-
tion) percentages at all observed bands compared to SAC, implying
greater density of functional groups on the surface of NAC than on
SAC.

Fig. 4. The FTIR spectra of NAC and SAC. Fig. 5. The SEM images of (a) NAC and (b) SAC.
G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128 123

Table 2 used. Based on data plotted in Fig. 6, the adsorption of AMX onto
The main properties of the NAC and SAC. NAC improved from 76.8% to over 91% when the solution pH was
Parameter Unit Value increased from 2 to 6. A further increase of pH to 9 resulted in a
SACa NACb reduction of AMX to 75.4%.
2 2
The relationship between effects of the selected adsorbents in
BET (m /g) m /g 1024 1029
Total pore volume (P/P0 = 0.990) cm3/g 0.572 0.633
AMX adsorption and solution pH can be explained by considering
Vm (monolayer volume) cm3/g 235.3 236.4 both the surface charge of the adsorbent and the dissociation con-
BET constant C – 691.8 1088.7 stant (pKa) of AMX. Since the pHpzc of SAC and NAC were 7.4 and
Mean pore diameter nm 2.23 2.46 6.6, respectively, a positive charge developed on their surfaces at
Pores structure – Mesopore Mesopore
pH below pHpzc. AMX has pKa of 2.4 (carboxyl), 7.4 (amine), and
pHpzc – 7.4 6.6
9.6 (phenol) [19]. Therefore, in an acidic solution (pHs 2–6), the
a
Standard activated carbon. carboxyl group (ACOOH) present in the AMX molecule is dissoci-
b
NH4Cl-induced activated carbon.
ated into carboxylate (ACOO) [20]. Accordingly, electrostatic
attraction between the AMX (anionic) molecules (through COO
functional group) and the surface of the activated carbons (posi-
The surface morphology of NAC was compared to that of SAC in
tively charged at pHs below 7.4 for SAC and 6.6 for NAC) was most
terms of SEM images as shown in Fig. 5a. As can be seen in Fig. 5a,
likely the predominant adsorption mechanism [21]. However, the
NAC has a fibrous structure composed of a series of long parallel
reduced AMX removal with an increased pH beyond 6 can be jus-
similar shaped channels with a smooth surface. Ioannidou et al.
tified by an increase of OH formation in the solution and subse-
[14] also reported similar morphology for activated carbon pre-
quent competition with the COO anions on AMX molecules for
pared from rapeseed stalks using steam activation. Fig. 5b shows
active adsorption sites on the surface of the activated carbon.
the SEM image of SAC, indicating a smooth surface with scattered
The reduction of AMX adsorption at alkaline solution pH
cavities of various sizes on the surface of the carbon. The difference
(pH > pHpzc) can be related to a negative charge on the surface
in surface morphology of NAC and SAC can be attributed to specific
of the carbon and thereby activating electrostatic repulsion of the
base materials as well as to the activation procedure. Accordingly,
anions to adsorb on the adsorption sites, onto the activated carbon
preparation of the mesoporous activated carbon with high pore
[22]. Putra et al. [23] reported the maximum adsorption of AMX
volume and specific surface area from low/no cost agricultural bio-
onto a commercial activated carbon at adsorbent pHpzc (around
mass presents an economically viable application to control water
4.9), although they tested the limits of levels of solution pH (only
pollution with appropriate benefits to the environment.
2, 5, and 7). Therefore, the pH of 6 was selected as the optimum va-
lue of pH in the next experiments.
3.3. Parametric experiments As stated above, although NAC concentration used in this exper-
iment was half that of SAC, the percentages of AMX adsorption
3.3.1. Effect of solution pH and adsorption mechanism onto NAC was considerably greater than that onto SAC. For in-
To find the optimum pH for maximum adsorption of AMX onto stance, MAX removal by NAC was 18.1% greater than that by SAC
SAC and NAC, the effect of AMX solution pH level ranging from 2 to in optimum pH (6) and similar contact time. This indicates the
9 was investigated under the conditions given in Table 1 (run 2). greater adsorption potential of prepared NAC compared with
The adsorption experiments for each adsorbent at each pH were SAC. The greater performance of NAC than that of SAC relates to
carried out in duplicate and the average results are presented in the surface characteristics of adsorbents, which is discussed later
Fig. 6. As seen in Fig. 6, the maximum AMX adsorption was ob- in this text.
tained at solution pH of around 6 for both SAC and NAC adsorbents.
According to Fig. 6, the percentage of AMX adsorption onto SAC in- 3.3.2. Dynamic of AMX adsorption and its kinetic modeling
creased from 53.9% to a maximum of 73% when the pH was raised The effect of various concentrations of AMX (10, 50, and
from 2 to 6. However, AMX removal percentage decreased consid- 100 mg/L) on its adsorption onto SAC and NAC was investigated
erably with a further increase of pH to 9. The same trend was ob- under conditions given in Table 1 (run 3). Data obtained from the
served for adsorption of AMX onto NAC, although with greater experiment were evaluated based on both removal efficiency and
percentages, while lower NAC concentration (half of SAC) was adsorption kinetics. Fig. 7 shows removal efficiency at various con-
centrations of AMX using adsorption onto SAC and NAC as a func-
tion of contact time; significantly greater adsorption percentages
and rates were recorded for NAC than for SAC for all tested concen-
trations. As can be seen in Fig. 7, the complete adsorption of 10 mg/
L AMX onto NAC was obtained at a contact time of 2 min. More-
over, the complete removal of medium (50 mg/L) and high
(100 mg/L) concentrations of AMX were obtained by adsorption
onto a small dosage of NAC (0.8 g/L) at contact times of 20 and
40 min, respectively. In comparison, SAC could only completely re-
move 10 mg/L AMX in 30 min contact time. Extending the contact
time for adsorption of medium (50 mg/L) and high (100 mg/L) con-
centrations of AMX onto SAC up to 100 min achieved 95.1% and
98.1% removal percentages, respectively. These findings reveal
the high affinity of prepared NAC to adsorb AMX molecules indi-
cating that the adsorption of AMX onto NAC is a rapid process
occurring within a wide rage in a reasonable contact time. The ra-
pid AMX adsorption onto NAC can be attributed to surface mass
transfer that will be addressed below.
Fig. 6. The effect of solution pH (2–9) on the adsorption of amoxicillin onto NAC Kinetics of adsorption of various concentrations (10, 50 and
and SAC (see Table 1 for the experimental conditions). 100 mg/L) AMX onto NAC and SAC was analyzed by fitting data
124 G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P
(a) ½ðqexp  qmodel Þ=qexp 2
Dq ¼ ð5Þ
n1

where qexp and qmodel are the experimental and model estimated
adsorption capacities, respectively, and n represents the number
of the tests.
Information on the kinetics of AMX adsorption onto NAC and
SAC is given in Table 3. The higher values of R2 and lower values
of Dq for PSO model than for PFO one for adsorption of all the
tested concentrations of AMX onto both NAC and SAC indicates
that results were best described with the PSO kinetic model. Avail-
able literature on adsorption of AMX is limited. Putra et al. [23]
also reported the better fitness of adsorption of AMX onto both
commercial activated carbon and bentonite to the PSO model.
The better fitness of data from the experiment to the PSO model
implies that the rate of AMX adsorption onto SAC/NAC was affected
mainly by the availability of active adsorption sites on the adsor-
(b) bent rather than the concentration of the AMX in the solution
[25,26]. It is also reconfirmed that adsorption of AMX onto both
SAC and NAC is attained through the electrostatic attraction be-
tween carboxyl anions in the structure of antibiotic molecules
and positively charged functional groups (adsorption sites) on
the cationic surfactant of the adsorbent. A similar conclusion was
made by Shaarani and Hameed [27] for adsorption of 2,4-dichloro-
pheno onto ammonia-modified activated carbon.
Two important points can be derived from the information on
kinetics given in Table 3. Firstly, values of k2 for adsorption of all
tested concentrations of AMX onto NAC are greater (between 2.7
and 4.3 times) than those onto SAC, verifying the higher affinity
of NAC than SAC for interaction with AMX that resulted in a higher
adsorption potential, which means higher removal percentages
were obtained using NAC in a shorter contact time (Fig. 7). The
higher affinity of AMX toward NAC than toward SAC (at similar
experimental conditions) was related to the surface characteristics
Fig. 7. The adsorption of various concentrations of amoxicillin (10–100 mg/L) onto of carbons to the greater number of functional groups on the sur-
(a) NAC and (b) SAC as a function of contact time (see Table 1 for the experimental face of NAC than SAC (see Fig. 4 and Section 3.2). The greater num-
conditions). ber of surface functional groups means a greater number of
adsorption active sites and a higher interaction of AMX molecules
from the tests of this step with pseudo first order (PFO) and pseudo
with the surface of the adsorbent surface thus a higher adsorption
second order (PSO) models.
rate. Secondly, values k2 for AMX adsorption onto NAC decreased
PFO equation : lnðqe  qt Þ ¼ ln qe  k1 t ð3Þ with an increased concentration of AMX. Decreasing k2 values with
an increase of AMX concentration in the adsorption onto NAC de-
t 1 t picts that there has not been any limitation to adsorption sites
PSO equation : ¼ þ ð4Þ
qt k2 q2e qe and that adsorption is limited by the concentration of AMX [8]
and thus by the mass transfer rate. Indeed, the increase of initial
where k1 and k2 are constants of adsorption rate, qt is adsorption
AMX concentration resulted in increase of driving force [28], there-
capacity at time t, qe is adsorption capacity at equilibrium condition.
by improving the mass transfer rate.
The determination coefficient (R2) and the standard deviation
The two main steps involved in mass transfer in the adsorption
relative error (Dq) were used as criteria of goodness of fitness.
of organic molecules onto adsorbents are the liquid phase mass
The Dq was calculated according to the Eq. (6) as reported by Basha
transport (film diffusion) and the intra particle mass transport
et al. [24].
(pore diffusion or intraparticle diffusion) phase [29]. To deduce
whether adsorption of AMX onto NAC has been limited by film

Table 3
The kinetic information of amoxicillin adsorption onto NAC and SAC.

Adsorbent Amoxicillin concentration (mg/L) qexp (mg/g) Pseudo first order kinetic Pseudo second order kinetic
R2 qe (mg/g) k1 (min1) Dq R2 qe (mg/g) k2 (g/mg min) Dq
SACa 10 12.3 0.931 3.6 0.185 0.70 0.999 12.7 0.122 0.370
50 54.9 0.871 30 0.083 0.39 0.999 58.8 0.008 0.650
100 110.6 0.983 83.3 0.044 0.59 0.998 125 0.001 1.170
NACb 10 12.5 0.958 1.55 1.064 1.005 0.999 12.8 2.035 0.092
50 62.5 0.912 12.85 0.311 0.827 0.999 63.3 0.078 0.212
100 126 0.961 40.29 0.099 0.626 0.999 126.6 0.013 0.472
a
Standard activated carbon.
b
NH4Cl-induced activated carbon.
G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128 125

Table 4
The film diffusion and intraparticle diffusion information of amoxicillin adsorption onto NAB and SAC.

Type of diffusion SACa, amoxicillin concentration (mg/L) NACb, amoxicillin concentration (mg/L)
10 50 100 10 50 100
Film diffusion
R2 0.896 0.891 0.938 0.914 0.988 0.965
kfd (min1) 0.11 0.032 0.018 0.608 0.264 0.095
Intraparticle diffusion
R2 0.64 0.813 0.912 0.87 0.854 0.818
kid (g/mg h) 1.26 4.01 8.97 1.227 3.939 9.377
C 7.54 27.6 31.89 10.86 45.56 69.46
a
Standard activated carbon.
b
NH4Cl-induced activated carbon.

diffusion or by intra particle diffusion, adsorption data from the


experiment was fitted with the intra particle diffusion model
(a)
[30] and the liquid film diffusion model [31], given in Eqs. (6)
and (7), respectively.

lnð1  qt =qe Þ ¼ K fd t ð6Þ

qt ¼ K id t0:5 þ C ð7Þ

where Kfd and Kid represent constants of liquid film diffusion and in-
tra particle diffusion, respectively.
Constants of the fitted models are given in Table 4. As seen in
Table 4, R2 values of the film diffusion model for adsorption of
AMX particularly onto NAC at all of the tested concentrations
was higher than those of the intra particle diffusion model. This
finding suggests that diffusion through the boundary liquid layer (b)
surrounding the NAC particles was the main step controlling the
rate of AMX adsorption [29]. A greater initial AMX concentration
causes a greater concentration gradient thus a higher rate of
AMX molecules diffuse through the boundary layer around the
adsorbent, which in turn stimulates a higher rate of adsorption.

3.3.3. Equilibrium of AMX adsorption and its isotherm modeling


The equilibrium of adsorption of various concentrations of AMX
onto SAC and NAC was investigated under experimental conditions
given in Table 1 (run 4). Results are shown in Fig. 8. According to
plots shown in Fig. 8, the adsorption efficiency and capacity of
AMX was considerably greater for adsorption onto NAC that onto
SAC particularly at higher AMX concentrations, reconfirming the
better performance of NAC to adsorb AMX compared with SAC. Re-
moval of AMX at the initial concentration of 50 mg/L by SAC and
NAC was 100% and 99.3%; removal efficiencies decreased to
41.7% (by SAC) and 68.3% (by NAC) when the initial AMX concen- Fig. 8. The adsorption percentage and capacity of various concentrations of
tration was increased step wise to 500 mg/L. In contrast, adsorp- amoxicillin (50–500 mg/L) onto (a) SAC and (b) NAC under equilibrium condition
(see Table 1 for the experimental conditions).
tion capacity in the tests increased with increased concentration
of AMX and reached 260.4 and 426.9 mg/g for SAC and NAC,
greater removal rate and capacity than SAC for removal of AMX
respectively, under the similar conditions. Considering that the
from aqueous solutions, confirming that the activation method
concentration of adsorbents (SAC or NAC) was constant during
used was an effective and promising method in modification of
equilibrium adsorption tests (0.8 g/L), the decrease of removal effi-
activated carbon and thus improvement of the favorability the
ciency with increased AMX concentration was related to limited
resultant adsorbent for the removal of antibiotics from contami-
availability of adsorption sites for increased amounts of AMX mol-
nated streams. The equilibrium adsorption results were further
ecules [32]. The increased adsorption capacity as a function of AMX
analyzed with isotherm models of Langmuir, Freundlich, and Dub-
concentration under the conditions of these tests can is also dem-
inin-Radushkevich as given in following equations [34]:
onstrated by the constant state of the concentration of the adsor-
bent; therefore, increasing the concentration of AMX led to an
 Langmuir:
increased mass transfer of AMX molecules to a given mass of
adsorbent, thus increasing the adsorption capacity. These findings Ce 1 Ce
¼ þ ð8Þ
are in accordance with those of other research. For instance, Wang qe bqmax qmax
et al. [33] reported a similar trend of changing adsorption rate and
where the qmax (mg/g) is the maximum adsorption capacity. The b
capacity as a function of initial adsorbate concentration, although
parameter is the Langmuir constant (L/mg). Values of qmax and b
tests were done with a different adsorbent (walnut hull) and ads-
are determined from the linear regression plot of Cqee versus Ce.
robate (Cr(VI)). Overall, it has been demonstrated that NAC had a
126 G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128

 Freundlich: tion rate and capacity. As above indicated, the adsorption of AMX
molecules onto SAC and NAC adsorbs mainly as a monolayer on
1
ln qe ¼ ln K F þ ln C e ð9Þ homogenously distributed adsorption sites on the surfaces of
n
selected adsorbents. The values of C constant for SAC and NAC
where KF and n are constants from the Freundlich equation. The are 692 and 1089, respectively, demonstrating that AMX had a
constant KF represents the capacity of the adsorbent for the adsor- significantly greater interaction with the surface of NAC that that
bate, and 1/n is the reciprocal of reaction order. A linear regression of SAC resulting in adsorption of a greater number of AMX
plot of ln qe versus ln Ce gives the KF and n values. molecules onto NAC than onto SAC. The greater interaction
between AMX and the NAC than the SAC can be related to the
 Dubinin-Radushkevich (D-R): higher density of charge and surface functional groups in NAC
compared to that in SAC (Fig. 4). It means the greater available
ln qe ¼ ln qm  K DR e2 ð10Þ sites on the surface of NAC than on the SAC for interaction with
2 2 the AMX molecules.
where KDR is the D-R constant (mol /kJ ) and e represents the Pola-
This finding indicates that the activated carbon prepared in this
nyi potential (J/mol), which is determined by:
study had a maximum adsorption capacity considerably greater
(67%) than SAC, presenting NAC as a promising candidate for appli-
e ¼ RT ln½1 þ 1=C e  ð11Þ cation as an adsorbent to control environmental pollution. There is
limited scientific research on adsorption of AMX from water. The
KDR is related to the mean of adsorption free energy E (kJ/mol)
efficacy of a few adsorbents have been investigated, examples as
by the following equation:
follows; chitosan beads [35]; bentonite and activated carbon
[23]; and almond shell ash [20]. These studies investigated adsorp-
1
E ¼ pffiffiffiffiffiffiffiffiffiffiffi ð12Þ tion of AMX antibiotic from contaminated water; the maximum
2K DR
AMX adsorption capacity reported for a commercial activated car-
Information on the isotherm analysis is given in Table 5. Based bon was with the value of around 222 mg/g [23]. Putra et al. [23]
on data given in Table 5, the R2 values of the Langmuir isotherm reported that both Langmuir and Freundlich models could equally
were greater than those of the other isotherms. This means that well-fit the results of tests of adsorption of AMX onto a commercial
adsorption of AMX molecules onto SAC and NAC particles is better activated carbon with the maximum adsorption capacity of around
described by the Langmuir model, which implies that AMX molec- 222 mg/g at an optimum acidic pH of 4.98. Accordingly, the devel-
ular adsorption occurs as a monolayer adsorb on homogenously oped adsorbent in the present work achieved an adsorption capac-
distributed adsorption sites on the surfaces of the selected adsor- ity significantly greater than adsorbents previously tested, and
bents. As seen in Table 5, the value of RL (a dimensionless param- thus presents an effective adsorbent for elimination of antibiotics
eter derived from the Langmuir equation) falls between 0 and 1, from contaminated streams. Accordingly, the high adsorption
indicating that AMX adsorption onto the prepared adsorbent was capacity, profound availability and ease of accessibility of the base
favorable. material in nature, as well as simplicity of preparation, presents
The maximum monolayer adsorption capacity of AMX onto SAC NAC as a promising, efficient and economically viable adsorbent
and NAC based on Langmuir model was around 262 and 437 mg/g, for eliminating antibiotics from contaminated streams.
respectively. The greater AMX adsorption capacity of NAC than SAC
(at similar experimental conditions) can further be explained by 3.3.4. Thermodynamic of adsorption and its modeling
considering the BET constant, term C, summarized in Table 2 as To investigate the thermodynamics of AMX adsorption onto
well as the best fitted isotherm (Langmuir isotherm) to data of SAC and NAC, the adsorption of AMX was investigated at various
the experiment. The BET constant, term C, is attributed to the solution temperatures between 10 and 50 °C under the conditions
adsorption energy in the first adsorbed layer. Accordingly, the va- specified in Table 1 (run 5). The results in terms of AMX
lue of C indicates the magnitude of interactions between adsorbate removal percentage as a function of solution temperature are
and adsorbent; the greater the C value the greater the interactions shown in Fig. 9. As seen in Fig. 9, the AMX adsorption percentages
between adsorbate and adsorbent and thus the higher the adsorp- onto SAC increased from 67.4% to 81.5% with the increase of

Table 5
The isotherm information of amoxicillin adsorption onto NAC and SAC.

Isotherm Adsorbent
NACa SACb
Langmuir
R2 0.994 0.998
qmax 438.6 261.8
b 0.128 0.166
1
RL ¼ ð1þbC 0.02 0.01

Freundlich
R2 0.959 0.96
KF 76.25 89.48
n 2.7 4.95
Dubinin-Radushkevich
R2 0.831 0.819
KDR 0.0016 0.0001
E 17.7 70.7
a
NH4Cl-induced activated carbon. Fig. 9. The effect of solution temperature (10–50 °C) on adsorption of amoxicillin
b
Standard activated carbon. onto PAC and NAC (see Table 1 for the experimental conditions).
G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128 127

temperature from 10 to 25 °C (equilibrium temperature). Further DG and the positive value of DH reveal that adsorption of AMX
increase of temperature to 50 °C resulted in decrease of AMX onto SAC (at temperatures up to 25 °C) and NAC (at temperatures
removal by SAC to 61.5%. For AMX adsorption onto NAC, AMX re- up to 35 °C) are a spontaneous and endothermic favorable process
moval improved from 76.8% to 92% when the temperature of the [29,39]. In addition, the greater positive value of DS for NAC than
solution increased from 10 to 35 °C (equilibrium temperature). SAC indicates the higher affinity of NAC toward AMX molecules
However, a further increase of temperature to 50 °C led to a declin- and increasing randomness at the interface of NAC–AMX solution
ing AMX removal percentage to 78.1%. The improvement of AMX during the interaction of AMX molecules with the active adsorp-
adsorption with increased temperature is attributed to enhance- tion sites on NAC [18] compared with SAC. Moreover, the values
ment of the interaction rate between AMX molecules and the of distribution coefficient (Kd = qt/Ct; mL/g) at different tested tem-
adsorption sites with increased temperature [36]. Moreover, the peratures for AMX adsorption onto SAC/NAC are given in Table 6.
increase of AMX adsorption percentages as a function of tempera- As seen in Table 6, values of Kd for adsorption onto both SAC and
ture reveals that AMX adsorption onto SAC and NAC is an endo- NAC increased with increasing temperature, confirming that AMX
thermic process under the selected conditions of this experiment adsorption onto both SAC and NAC is an endothermic process.
[37]. Hence, the increase of temperature up to the optimum level
accelerated AMX mass transfer through the boundary liquid layer 4. Conclusion
thereby a higher amount of AMX molecules were transferred onto
the surface of SAC/NAC particles and thus greater removal percent- In this study, an activated carbon was prepared from agricul-
ages were achieved. Meanwhile, the decrease of AMX with in- tural waste using innovated NH4Cl-induced activation method
creased temperature related to desorption of AMX adsorbed and (NAC) and its adsorption potential was compared with standard
its release into the solution [38], suggesting that MAX adsorption activated carbon (SAC) supplied by Merck Co. in adsorption of
is an exothermic process at an elevated temperature. Another AMX. NAC was a mesoporous particle with a higher volume of
important point is made in Fig. 9; the equilibrium temperature pores, greater surface area of functional groups, and greater inter-
for AMX adsorption onto NAC (35 °C) is significantly greater than action with AMX compared with SAC. These features resulted in a
that of SAC (25 °C). This is an important feature for NAC from the much greater AMX adsorption rate and capacity than were
point of view of application; NAC can be efficiently applied for a achieved by SAC under similar conditions. Accordingly, NAC is
wider range of water temperatures compared to SAC. hereby presented as a mesoporous adsorbent with high pore vol-
The thermodynamics of AMX adsorption on the prepared adsor- ume and specific surface area from a low/no cost agricultural bio-
bents were analyzed using the following equations [10]: mass that is promising for application in the control of water
DG ¼ RT ln K ð13Þ pollution and presents a very valuable economic with benefits to
the environment.
DH  DS
ln K ¼  þ ð14Þ Acknowledgement
RT R

DG ¼ DH  T DS ð15Þ The authors are grateful to the Tarbiat Modares University for
providing financial and technical support.
The obtained thermodynamics information is given in Table 6.
Based on the information presented in Table 6, values of DG for
References
the temperatures tested for AMX adsorption were calculated to
be between 2.25 and 4.08 kJ/mol for SAC and between 4.97 [1] J.L. Martinez, Environmental pollution by antibiotics and by antibiotic
and 8.60 kJ/mol for NAC. Also, the DH and DS for AMX adsorp- resistance determinants, Environ. Pollut. 157 (2009) 2893–2902.
tion onto SAC and NAC adsorption 31.5 kJ/mol and 119.3 J/k mol, [2] V. Homem, L. Santos, Degradation and removal methods of antibiotics from
aqueous matrices: a review, J. Environ. Manage. 92 (2011) 2304–2347.
and 37.1 kJ/mol and 148.5 J/k mol. Overall, the negative values of [3] S.-H. Lin, R.-S. Juang, Adsorption of phenol and its derivatives from water using
synthetic resins and low-cost natural adsorbents: a review, J. Environ. Manage.
90 (2009) 1336–1349.
[4] R. Han, D. Ding, Y. Xu, W. Zou, Y. Wang, Y. Li, L. Zou, Use of rice husk for the
Table 6 adsorption of congo red from aqueous solution in column mode, Bioresource
The thermodynamic information of amoxicillin adsorption onto SAC and NAC. Technol. 99 (2008) 2938–2946.
[5] Z. Aksu, Ö. Tunç, Application of biosorption for penicillin G removal:
Adsorbent Temperature Kd DG  R2 D H DS comparison with activated carbon, Process Biochem. 40 (2005) 831–847.
(K) (mL/g) (kJ/mol) (kJ/mol) (J/K mol) [6] A. Demirbas, Agricultural based activated carbons for the removal of dyes from
SACa 283 2.61 2.25 0.994 31.5 119.3 aqueous solutions: a review, J. Hazard. Mater. 167 (2009) 1–9.
288 3.45 2.97 [7] K.Y. Foo, B.H. Hameed, Coconut husk derived activated carbon via microwave
induced activation: effects of activation agents, preparation parameters and
293 4.13 3.45
adsorption performance, Chem. Eng. J. 184 (2011) 57–65.
298 5.19 4.08
[8] J. Rivera-Utrilla, G. Prados-Joya, M. Sánchez-Polo, M.A. Ferro-García, I. Bautista-
298 5.19 4.08 0.987 32 93.3
Toledo, Removal of nitroimidazole antibiotics from aqueous solution by
303 4.70 3.90 adsorption/bioadsorption on activated carbon, J. Hazard. Mater. 170 (2009)
308 3.59 3.27 298–305.
313 2.86 2.73 [9] D. Dimitrakopoulou, I. Rethemiotaki, Z. Frontistis, N.P. Xekoukoulotakis, D.
323 2.00 1.86 Venieri, D. Mantzavinos, Degradation, mineralization and antibiotic
inactivation of amoxicillin by UV-A/TiO2 photocatalysis, J. Environ. Manage.
NACb 283 8.28 4.97 0.992 37.1 148.5
98 (2012) 168–174.
288 11.01 5.74
[10] Y. Zhao, B. Zhang, X. Zhang, J. Wang, J. Liu, R. Chen, Preparation of highly
293 14.17 6.46
ordered cubic NaA zeolite from halloysite mineral for adsorption of
298 17.66 7.11 ammonium ions, J. Hazard. Mater. 178 (2010) 658–664.
303 25.28 8.14 [11] I. Puziy, O. Poddubnaya, A. Martínez-Alonso, F. Suárez- García, J. Tascón,
308 28.75 8.60 Synthetic carbons activated with phosphoric acid III. Carbons prepared in air,
308 28.75 8.60 0.991 63.52 178.58 Carbon 41 (2003) 1181–1191.
313 17.50 7.45 [12] T. Yang, A. Lua, Characteristics of activated carbons prepared from pistachio-
323 8.92 5.88 nut shells by physical activation, J. Colloid Interface Sci. 267 (2003) 408–417.
[13] D. Cuhadaroglu, O.A. Uygun, Production and characterization of activated
a
Standard activated carbon. carbon from a bituminous coal by chemical activation, Afr. J. Biotechnol. 7
b
NH4Cl-induced activated carbon. (2008) 3703–3710.
128 G. Moussavi et al. / Chemical Engineering Journal 217 (2013) 119–128

[14] O.A. Ioannidou, A.A. Zabaniotou, G.G. Stavropoulos, Md.A. Islam, T.A. Albanis, [28] D. Rameshraja, V.C. Srivastava, J.P. Kushwaha, I.D. Mall, Quinoline adsorption
Preparation of activated carbons from agricultural residues for pesticide onto granular activated carbon and bagasse fly ash, Chem. Eng. J. 181–182
adsorption, Chemosphere 80 (2010) 1328–1336. (2012) 343–351.
[15] V. Gómez-Serrano, M. Acedo-Ramos, C. Valenzuela-Calahorro, A.J. López- [29] S. Nethaji, A. Sivasamy, Adsorptive removal of an acid dye by lignocellulosic
Peinado, Regeneration of activated carbon after contact with sulfuric acid waste biomass activated carbon: equilibrium and kinetic studies,
solution, J. Chem. Technol. Biotechnol. 75 (2000) 835–839. Chemosphere 82 (2011) 1367–1372.
[16] W. Tongpoothorn, M. Sriuttha, P. Homchan, S. Chanthai, C. Ruangviriyachai, [30] W.J. Weber Jr., J.C. Morris, Kinetics of adsorption on carbon from solution, J.
Preparation of activated carbon derived from Jatropha curcas fruit shell by Sanit. Eng. Div. 89 (1963) 31–60.
simple thermo-chemical activation and characterization of their physico- [31] G.E. Boyd, A.W. Adamson, L.S. Myers, The exchange adsorption of ions from
chemical properties, Chem. Eng. Res. Des. 89 (2011) 335–340. aqueous solutions by organic zeolites. II. Kinetics, J. Am. Chem. Soc. 69 (1947)
[17] R. Fu, L. Liu, W. Huang, P. Sun, Studies on the structure of activated carbon 2836–2848.
fibers activated by phosphoric acid, J. Appl. Polym. Sci. 87 (2003) 2253–2261. [32] G. Moussavi, R. Khosravi, Removal of cyanide from wastewater by adsorption
[18] K.Y. Foo, B.H. Hameed, Adsorption characteristics of industrial solid waste onto pistachio hull wastes: parametric experiments, kinetics and equilibrium
derived activated carbon prepared by microwave heating for methylene blue, analysis, J. Hazard. Mater. 183 (2010) 724–730.
Fuel Process. Technol. 99 (2012) 103–109. [33] X.S. Wang, Z.Z. Li, S.R. Tao, Removal of chromium (VI) from aqueous solution
[19] G.N. Rolinson, Laboratory evaluation of AMX, J. Infect. Dis. 129 (1974) 143. using walnut hull, J. Environ. Manage. 90 (2009) 721–729.
[20] V. Homem, A. Alves, L. Santos, Amoxicillin removal from aqueous matrices by [34] B.H. Hameed, Evaluation of papaya seeds as a novel non-conventional low-cost
sorption with almond shell ashes, Int. J. Environ. Anal. Chem. 90 (2010) 1063–1084. adsorbent for removal of methylene blue, J. Hazard. Mater. 162 (2009) 939–
[21] G. Moussavi, M. Mahmoudi, Removal of azo and anthraquinone reactive dyes 944.
from industrial wastewaters using MgO nanoparticles, J. Hazard. Mater. 168 [35] W.S. Adriano, V. Veredas, C.C. Santana, L.R.B. Gonçalves, Adsorption of
(2009) 806–812. amoxicillin on chitosan beads: kinetics, equilibrium and validation of finite
[22] E. Çalısßkan, S. Göktür, Adsorption characteristics of sulfamethoxazole and bath models, Biochem. Eng. J. 27 (2005) (2005) 132–137.
metronidazole on activated carbon, Sep. Sci. Technol. 45 (2010) 244–255. [36] G. Moussavi, S. Talebi, M. Farrokhi, R. Mojtabaee Sabouti, The investigation of
[23] E.K. Putra, R. Pranowo, J. Sunarso, N. Indraswati, S. Ismadji, Performance of mechanism, kinetic and isotherm of ammonia and humic acid co-adsorption
activated carbon and bentonite for adsorption of amoxicillin from wastewater: onto natural zeolite, Chem. Eng. J. 171 (2011) 1159–1169.
mechanisms, isotherms and kinetics, Water Res. 43 (2009) 2419–2430. [37] H. Zheng, L. Han, H. Ma, Y. Zheng, H. Zhang, D. Liu, S. Liang, Adsorption
[24] S. Basha, D. Keane, A. Morrissey, K. Nolan, M. Oelgemöller, J. Tobin, Studies on characteristics of ammonium ion by zeolite 13X, J. Hazard. Mater. 158 (2008)
the adsorption and kinetics of photodegradation of pharmaceutical compound, 577–584.
indomethacin using novel photocatalytic adsorbents (IPCAs), Ind. Eng. Chem. [38] D. Karadag, Y. Koc, M. Turan, B. Armagan, Removal of ammonium ion from
Res. 49 (2010) 11302–11309. aqueous solution using natural Turkish clinoptilolite, J. Hazard. Mater. B136
[25] Y. Liu, New insights into pseudo-second-order kinetic equation for adsorption, (2006) 604–609.
Colloid Surf. A. 320 (2008) 275–278. [39] F.M. Pachado, C.P. Bergmann, T.H.M. Fernandes, E.C. Lima, B. Royer, T. Calvete,
[26] J.M. Salman, V.O. Njoku, B.H. Hameed, Bentazon and carbofuran adsorption S.B. Fagan, Adsorption of reactive red M-2BE dye from water solutions by
onto date seed activated carbon: kinetics and equilibrium, Chem. Eng. J. 173 multi-walled carbon nanotubes and activated carbon, J. Hazard. Mater. 192
(2011) 361–368. (2011) 1122–1131.
[27] F.W. Shaarani, B.H. Hameed, Ammonia-modified activated carbon for the
adsorption of 2,4-dichlorophenol, Chem. Eng. J. 169 (2011) 180–185.

You might also like