Chemical Engineering Journal 220 (2013) 237–243

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal

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

Electrospun nanofiber membrane of PEO/Chitosan for the adsorption of

nickel, cadmium, lead and copper ions from aqueous solution
Majid Aliabadi a, Mohammad Irani b,⇑, Jabir Ismaeili c, Hossein Piri b, Mohammad Javad Parnian d
a
Department of Chemical Engineering, Islamic Azad University, Birjand Branch, Birjand, Iran
b
Department of Chemical Engineering, Amirkabir University of Technology, Tehran Polytechnic, Hafez Ave., P.O. Box 15875-4413, Tehran, Iran
c
School of Chemical Engineering, Iran University of Science & Technology, Tehran, Iran
d
Department of Chemical Engineering, Faculty of Engineering, University of Tehran, Tehran, Iran

h i g h l i g h t s

" The selectivity adsorption of nanofiber membrane was in order of Pb(II)<Cd(II)<Cu(II)<Ni(II).

" The prepared nanofiber membrane was mesoporous.
" The mechanism of adsorption metal ions by the nanofiber membrane was physically.
" The metals adsorption capacity did not change after five sorption–desorption cycles.

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

Article history: Poly ethylene oxide (PEO)/Chitosan nanofiber membrane was prepared by the electrospinning technique;
Received 16 October 2012 afterward, the potential of the prepared nanofiber membrane for adsorption of nickel (Ni), cadmium (Cd),
Received in revised form 3 January 2013 lead (Pb) and copper (Cu) from aqueous solution was investigated. The applied voltage, tip-collector
Accepted 5 January 2013
distance and solution flow rate of electrospinning process were 20 kV, 13.5 cm and 0.5 mL h1 respec-
Available online 11 January 2013
tively. The prepared nanofibers were characterized by Fourier Transform Infrared (FTIR), Scanning
Electron Microscope (SEM) and Brunauer–Emmert–Teller (BET) analysis. Based on results, the homoge-
Keywords:
neous electrospun nanofibers with the average diameter and surface area of 98 nm and 312.2 m2 g1 were
PEO/Chitosan
Electrospinning
obtained respectively. Then, the adsorption experiments were carried out to investigate the effect of
Nanofiber different adsorption parameters such as contact time, initial concentration and temperature in a batch
Heavy metal system. The sorption selectivity of lead, copper, cadmium and nickel onto the membrane was in order
Adsorption of Pb(II)<Cd(II)<Cu(II)<Ni(II). The kinetic data of metal ions were analyzed by pseudo-first-order and
pseudo-second-order kinetic models. Three isotherm models namely Freundlich, Langmuir and Dubinin–
Radushkevich were applied to describe the equilibrium data of metal ions. The calculated thermodynamic
parameters (DG, DH and DS) showed that the adsorption of metal ions onto the PEO/Chitosan nanofiber
membrane was feasible, spontaneous and endothermic. The reusability of nanofiber membrane for
the removal of different metal ions was also determined after five sorption–desorption cycles.
Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction effective and widely used process due to its simplicity, moderate
operational conditions and economical feasibility [6–8].
Removal of heavy metals from wastewater is a very important The most important property of any adsorbent is the surface
factor with respect to human’s health and environmental consider- area and structure. Furthermore, the chemical nature and polarity
ations [1]. Several methods used for the removal of heavy metal of the adsorbent surface can influence the attractive forces be-
ions from aqueous solutions include chemical precipitation, elec- tween the adsorbent and adsorbate. The nanofiber membranes
trodialysis, membrane process and solvent extraction [2–5]. How- prepared by the electrospinning method have unique properties
ever, these techniques are associated with problems such as such as high specific surface area and high porosity with fine pores.
excessive time requirements, high costs and high energy consump- These properties caused to use electrospun nanofiber membranes
tion. In this way, adsorption method can be considered as an for the removal of heavy metal ions [9–14]. In recent years,
researchers used nanofiber membranes such as polyvinylpyrroli-
⇑ Corresponding author. Tel.: +98 021 88765937. done (PVP)/SiO2 [15], PVA/SiO2 [16] and PAN-oxime [17] for the
E-mail address: irani_mo@ut.ac.ir (M. Irani). removal of heavy metal ions.

1385-8947/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cej.2013.01.021
238 M. Aliabadi et al. / Chemical Engineering Journal 220 (2013) 237–243

In the electrospinning technique, high voltage is applied

between a nozzle and a collector. An electrically charged jet of
polymer or composite solution is made during the process. The
solvent evaporates before reaching the collector and the fibers
with micro or nano-sized diameters collect on a collector. The
parameters which affect electrospinning, include polymer solution
properties, process and ambient conditions [18–20].
Chitosan, due to the presence of hydroxyl and amine groups, is
widely used for adsorption of heavy metal ions [21–23]. However,
the electrospinning of Chitosan due to the low solubility, low sta-
bility and low mechanical properties is extremely difficult. There-
fore, the polymers such as poly(ethylene oxide), poly(vinyl
alcohol), cellulose, zein, poly(lactic acid), poly(caprolactone) and
nylon-6 have been blended to Chitosan to improve the mechanical
properties of Chitosan nanofibers [24].
In the present study, PEO/Chitosan nanofiber membrane was
prepared by the electrospinning method and its application for
the removal of copper, lead, cadmium and nickel ions from aque-
ous solution was investigated. Then, the effects of contact time, ini-
tial metal ion concentration and temperature have been
investigated to obtain the optimum conditions for the maximum
adsorption capacity of metal ions. Furthermore, the nature of the
adsorption process in a batch system with respect to kinetic, equi-
librium and thermodynamic parameters were also evaluated. Fi-
nally, the reusability of PEO/Chitosan nanofiber membrane was
determined after five adsorption–desorption cycles.

Fig. 1. Schematic of the electrospinning process.

2. Experimental

2.1. Materials Spectrum GX FTIR spectrometer. The surface morphology of the

electrospun nanofibers was examined with SEM (SEM, JEOL JSM-
Chitosan (average Mw 200,000) and poly(ethylene oxide) (aver- 6380) after gold coating. The specific surface area and pore volume
age Mw 900,000) were obtained from Sigma–Aldrich. Acetic acid were determined using Brunauer–Emmert–Teller (BET) method
was purchased from Merck. Deionized water was used throughout (BET, ASAP 2020). The average diameter and diameter distribution
this work. of nanofibers were obtained with an image analyzer (Image-
Proplus, Media Cybrernetics).
The concentration of heavy metal ions in the adsorption med-
2.2. Preparation of PEO/Chitosan solution
ium was determined using an inductively coupled plasma atomic
emission spectrophotometer (ICP-AES, Thermo Jarrel Ash, Model
The Chitosan (3.5% w/v) and PEO (3.5% w/v) solutions were ini-
Trace Scan).
tially prepared separately by dissolving Chitosan or PEO in 0.5 M
acetic acid. Then, the PEO/Chitosan solution was obtained by
2.5. Swelling studies of PEO/Chitosan nanofiber membranes
blending of Chitosan and PEO with ratio of 9:1 and stirring them
for 24 h.
For investigation of the stability of PEO/Chitosan in aqueous
solution, the swelling studies were carried out on the pure and
2.3. Electrospinning process GTA cross-linked PEO/Chitosan nanofiber membranes. For this,
dry samples were immersed in 30 °C deionized water for 24 h.
The prepared solution was loaded into a 5 mL plastic syringe The experiments continued until the weight of the sample stay
equipped with a syringe needle. This was placed to a KD program- more or less constant. All experiments were repeated at three
mable syringe pump to control the solution flow rate. Then, a high times. The swelling percentage was calculated by:
voltage was applied between the needle and the collector that PEO/
Chitosan nanofibers were produced on the collector. A voltage of Swelling % ¼ ðW w  W d =W d Þ  100 ð1Þ
20 kV, with a tip-collector distance of 13.5 cm, at a speed of where Ww and Wd was the masses of the swollen and dried nanofi-
0.5 mL h1 was applied to the solution and fibers were collected ber membranes, respectively.
on the cylindrical collector. Then 25% glutaraldehyde (GTA) vapor
was used for cross-linking of PEO/Chitosan nanofiber membrane 2.6. Batch adsorption experiments
for improving of hydrophilic property. The fibers were dried ini-
tially for 12 h at 60 °C in a vacuum chamber. Diagram of the elec- The effects of adsorption experimental parameters such as, ini-
trospinning process is shown in Fig. 1. tial metal ion concentration (50–1000 mg L1), time (0–240 min)
and temperature (25–45 °C) on the removal of lead were studied.
2.4. Measurement and methods Copper, nickel and cadmium ions were studied in a batch mode.
Batch adsorption experiments were carried out by soaking 50 mg
FTIR analysis was used in the range of 400–4000 cm1 to deter- of PEO/Chitosan nanofiber membrane in 100 mL of nitrate solu-
mine the type of functional groups on the prepared nanofibers. tions over 2 h at 25 °C with magnetic stirring. The amount of adsor-
Dried nanofibers were mixed with KBr powder and then pelletized. bate in solid phase (q (mg g1)) is calculated by the following
The IR characterizations were performed using a Perkin–Elmer equation:
 M. Aliabadi et al. / Chemical Engineering Journal 220 (2013) 237–243 239

ðC 0  C eq ÞV segments to fuse to each other and increased water diffusivity

qe ¼ ð2Þ
1000M resistance and decreased the degree of crosslinked nanofibers
Also the removal percentage is calculated from Eq. (3): swelling.

C 0  C eq 3.2. Comparison of metal ions adsorption

R% ¼  100 ð3Þ
C0
The potential of PEO/Chitosan nanofiber as an adsorbent for the
where qe is the adsorption capacity in mg g1, C0 and Ceq are the ini-
removal of lead, cadmium, copper and nickel in initial concentra-
tial and equilibrium concentrations of heavy metal ions in the liquid
tion of 100 mg L1, time of 120 min, pH of solutions (lead: 5.5, cad-
phase (mg L1), respectively. V is the volume of the solution (mL)
mium: 5, nickel: 5 and copper: 6) and 25 °C was investigated.
and M is the amount of nanofiber (g). For regeneration of nanofiber
Results showed that the adsorption capacity of different metal ions
membranes, the nanofibers were rapidly washed with 1 mol L1
decreased in order of Pb(II)<Cd(II)<Cu(II)<Ni(II). The ionic radius
HNO3 and then washed several times with deionized water. Suc-
range for these ions are in order of Ni(II)<Cu(II)<Cd(II)<Pb(II). As
ceeding that, the nanofiber membranes were dried in a vacuum
ionic radius increases due to charge density decreases, the
oven at 80 °C.
availability of active sites of nanofiber membranes for adsorption
process decreases; which resulted in decreasing the adsorption
3. Results capacity of different metal ions by increasing the ionic radius.
Therefore, the adsorption percentages of metal ions were obtained
3.1. Characterization of PEO/Chitosan composite nanofiber 89%, 82%, 72% and 68% for the adsorption of nickel, copper,
cadmium and lead, respectively.
The functional groups of PEO, Chitosan and PEO/Chitosan
nanofibers were characterized by Fourier Transform Infrared 3.3. Effect of contact time and kinetic models
(Fig. 2). In PEO/Chitosan nanofiber membrane, a broad bond at
3100–3600 cm1 was assigned to NAH and OAH stretching of The effect of contact time on the adsorption of nickel, copper,
the polysacharide molecules. Two bonds at 1560 and 1650 cm1 cadmium and lead ions onto the PEO/Chitosan nanofiber mem-
were attributed to the amide groups in the structure of PEO/ brane in initial concentration of 100 mg L1, pH of solution and
Chitosan nanofiber. The broad bond appeared at 2900 cm1 was 25 °C is shown in Fig. 5. As shown, the process of different metal
due to the CH2 stretching groups. The CAH group peaks were sorption onto the nanofiber membrane reached the equilibrium
observed at 1340 and 1370 cm1. The absorption bonds around state after approximately 120 min of contact time. More than
1100 cm1 assigned to CAOAC stretching vibration. 80% of total adsorption of metal ions occurred within the first
Based on BJH theory and Brunauer–Emmet–Teller (BET), the ob- 60 min. Therefore, 120 min is selected as equilibrium time for fur-
tained average pore diameter, surface area and pore volume of ther experiments.
nanofibers were 4.12 nm, 312.2 m2 g1 and 0.495 cm3 g1, respec- Kinetic models namely pseudo-first-order and pseudo-second
tively. The value of pore diameter of fibers indicated that the sur- order models were applied to describe the kinetic data of metal
face of PEO/Chitosan nanofiber membrane was mesoporous. ions sorption onto the PEO/Chitosan nanofiber membrane which
The SEM image and distribution diameter of PEO/Chitosan results are shown in Fig. 5.
electrospun nanofiber membrane at applied voltage of 20 kV, The pseudo-first-order kinetic model is given as [25]:
tip-collector distance of 13.5 cm and flow rate of 0.5 mL h1 were
shown in Fig. 3. As shown, the homogeneous fibers with average
qt ¼ qe ð1  expðk1 ÞÞ ð4Þ
diameter of 98 nm were formed on the collector. where qt and qe ðmg g1 Þ are the adsorption capacity at time t and
Fig. 4 shows degree of swelling (%) of PEO/Chitosan (no cross- equilibrium time, respectively. k1 (min1) is the pseudo-first order
linked) and PEO/Chitosan (cross-linked with GTA) in 30 °C deion- model rate constant. The constants of the pseudo-first order kinetic
ized water for 24 h. As can be seen, degrees of membrane swelling model were obtained by nonlinear regression of qt versus t using
of PEO/Chitosan nanofiber is decreased after cross-linking with the MATLAB software (Fig. 5a), and the results are presented in
moist GTA vapor for interval times. The decrease in the degree of Table 1.
swelling of PEO/Chitosan nanofiber membrane upon cross-linking The pseudo second order kinetic model by Ho and McKay [26] is
with GTA could be due to the formation of aldimine linkages given as:
(ACH@NA) between the free amino groups of Chitosan and the
aldehyde groups of GTA. These linkages caused adjacent fiber k2 q2e t
qt ¼ ð5Þ
1 þ k2 qe t

where k2 (g mg1 min1) is the adsorption rate constant. Fig. 5b

shows the plot of qt versus t to the pseudo-second-order model
a for the adsorption of metal ions onto the PEO/Chitosan nanofiber
Transmittance (%)

membrane. The parameters of pseudo-second-order kinetic model

were also given in Table 1. As shown in Table 1, both equations of
b pseudo-first-order (R2 > 0.99) and pseudo-second-order (R2 > 0.99)
fitted the experimental data well.

c 3.4. Modeling of equilibrium data

In the present study, three well-known adsorption isotherms

namely Langmuir, Freundlich and Dubinin–Radushkevich (D–R)
Wavenumbers (cm -1)
are used to analyze the equilibrium data of nickel, cadmium, cop-
Fig. 2. FTIR spectra of (a) PEO, (b) Chitosan and (c) PEO/Chitosan nanofiber per and lead ions. Langmuir model is based on the physical hypoth-
membranes. esis that the maximum adsorption capacity consists of a monolayer
240 M. Aliabadi et al. / Chemical Engineering Journal 220 (2013) 237–243

Fig. 3. The SEM image and fiber diameter distribution of PEO/Chitosan nanofiber.

100
(a) 180
80 160
Degree of swelling (%)

140
Non-crosslinked PEO/Chitosan nanofibers
60
crosslinked PEO/Chitosan nanofibers qt (mg / g) 120

40 100
Kinetic data-Nickel
80 Kinetic data-Copper
20 Kinetic data-Cadmium
60 Kinetic data-Lead
Pseudo-First-Order-Ni
0 40 Pseudo-First-Order-Cu
0 5 10 15 20 25
Pseudo-First-Order-Cd
Time (hr) Pseudo-First-Order-Pb
20
Fig. 4. Degree of swelling of nanofiber membranes.
0
0 20 40 60 80 100 120
Time (min)
adsorption, that there are no interactions between adsorbed mole-
cules and adsorption energy which is distributed homogenously (b) 180
over the entire coverage surface. The general form of the Langmuir
isotherm is [27]: 160

qmax bC eq 140
qeq ¼ ð6Þ
1 þ bC eq
120
where qeq (mg g1) is the equilibrium metal uptake capacity, Ceq
qt (mg / g)

(mg L1) is the equilibrium metal ion concentration, qmax (mg g1) 100
is the maximum amount of the metal ion per unit weight of adsor-
bent to form a complete monolayer on the surface bond at high Ceq 80
Kinetic data-Nickel
and b (L mg1) is constant which is related to the affinity of the Kinetic data-Copper
60 Kinetic data-Cadmium
binding sites. The non-linear relationship between the amounts of
Kinetic data-Lead
metal ions sorbed per unit mass of nanofiber membrane against 40 Pseudo-Second-Order-Nickel
the concentration of metal ions remaining in solution are shown Pseudo-Second-Order-Copper
in Fig. 6 at different temperatures. The values of qmax, b and corre- 20 Pseudo-Second-Order-Cadmium
Pseudo-Second-Order-Lead
lation coefficients (R2) obtained from non-linear regression of the
Langmuir model onto the equilibrium data are presented in Table 2. 0
0 20 40 60 80 100 120
As shown in Table 2, the high regression correlation coefficients
Time (min)
(>0.96) were found, suggesting that this model is very suitable for
describing equilibrium data of different metal ions by the PEO/ Fig. 5. Adsorption kinetics (a) pseudo-first-order and (b) pseudo-second-order of
Chitosan nanofiber membrane. Furthermore, the maximum metal ions onto the PEO/Chitosan nanofiber membrane.
adsorption capacity (qmax) of metal ions increased with increasing
the temperature. A higher value of b also implied strong bonding of 1
metal ions to nanofiber membrane at studied temperatures. qeq ¼ K f C neq ð7Þ
The Freundlich expression is an empirical equation based on
sorption on a heterogeneous surface and is commonly presented where Kf (mg g1) and n are the Freundlich constants related to the
as follows [28]: adsorption capacity and intensity of the sorbent, respectively.
 M. Aliabadi et al. / Chemical Engineering Journal 220 (2013) 237–243 241

Table 1
Kinetic parameters of metal sorption onto the PEO/Chitosan nanofiber membrane.

Metal qexp (mg/g) Pseudo-first-order model Pseudo-second-order model

2
qeq (mg/g) k1 (min1
) R qeq (mg/g) k2 (g/mg min) R2
Nickel 175.1 183.4 0.02638 0.998 242.9 0.000097 0.996
Copper 163.7 172.3 0.02593 0.998 229.2 0.000100 0.996
Cadmium 143.8 150.3 0.02783 0.997 198.6 0.000125 0.992
Lead 135.4 143.4 0.02403 0.998 195.1 0.000104 0.994

(a) 350 (c) 250

300 200

250

qe (mg / g)
qe (mg/g)

150 Exprimental data-Cadmium-T=25 C

Exprimental data-Nickel-T=25 C
200 Exprimental data-Nickel-T=35 C Exprimental data-Cadmium-T=35 C
Exprimental data-Nickel-T=45 C Exprimental data-Cadmium-T=45 C
100
150 Langmuir isotherm-T=25 C
Langmuir isotherm-T=25 C
Langmuir isotherm-T=35 C Langmuir isotherm-T=35 C
100 50 Langmuir isotherm-T=45 C
Langmuir isotherm-T=45 C

50
0
0 200 400 600 800 0 200 400 600 800
Equilibrium concentration (mg/L) Equilibrium concentration (mg / L)

(b) 300 (d)

200
250

Exprimental data-Lead-T=25 C
qe (mg / g)
qe (mg / g)

200 150
Exprimental data-Copper-T=25 C
Exprimental data-Lead-T=35 C
Exprimental data-Copper-T=35 C
150 Exprimental data-Lead-T=45 C
Exprimental data-Copper-T=45 C 100 Langmuir isotherm-T=25 C
Langmuir isotherm-T=25 C
100 Langmuir isotherm-T=35 C
Langmuir isotherm-T=35 C
50 Langmuir isotherm-T=45 C
50 Langmuir isotherm-T=45 C

0 0
0 200 400 600 800 0 200 400 600 800
Equilibrium concentration (mg / L) Equilibrium concentration (mg / L)

Fig. 6. Langmuir isotherm plots for (a) nickel (b) copper (c) cadmium and (d) lead sorption onto the nanofiber membrane.

The values of n, Kf and R2 obtained from nonlinear regression qeq ¼ qDR expðbDR e2 Þ ð8Þ
by MATLAB software are given in Table 2. As can be seen in Ta-
ble 2, the n1 values at all studied temperatures were between where qeq is the amount of metal ions adsorbed on per unit weight of
0 and 1 indicating that the adsorption of different metal ions adsorbent (mmol g1), qmax is the maximum adsorption capacity
using PEO/Chitosan nanofiber membrane was favorable at studied (mmol g1), b is the activity coefficient related to adsorption free
conditions. However, compared to the R2 values of the Freundlich energy (mol2 J2) and e is the polanyi potential (e = RT ln(1 + 1/Ceq)),
(R2 > 0.82) and Langmuir (R2 > 0.96) models it can be concluded R is the gas constant (8.314 J mol1 K1) and T is the absolute
that the Langmuir isotherm model fitted better with the equilib- temperature (K). The Polanyi sorption theory assumes a fixed
rium data of metal ions. The applicability of the Langmuir iso- volume of sorption space close to the sorbent surface and the
therm model for metal ions-Chitosan adsorption system implies existence of a sorption potential over these spaces. The mean free
that monolayer adsorption conditions exit under the experimen- energy of adsorption (E, J mol1) can be calculated from the follow-
tal conditions. ing equation:
The D–R isotherm model was also performed to the equilibrium 1
data in order to determine the nature of adsorption processes as E ¼ pffiffiffiffiffiffi ð9Þ
2b
physical or chemical. The D–R sorption isotherm is more general
than the Langmuir isotherm, as its derivation is not based on ideal The mean free energy of adsorption gives information about
assumptions such as equipotent of sorption sites, absence of stoic adsorption mechanism. If it lies between 8 and 16 kJ mol1, the
hindrance between sorbed and incoming particles and surface adsorption process takes place chemically and while E < 8 kJ mol1,
homogeneity on microscopic level. The general form of the D–R the adsorption process proceeds physically [30]. The values of qmax
isotherm equation is expressed as follows [29]: and b obtained from nonlinear regression of equilibrium data by
242 M. Aliabadi et al. / Chemical Engineering Journal 220 (2013) 237–243

Table 2
Isotherm parameters for metal adsorption onto the PEO/Chitosan nanofiber membrane.

Metal T (°C) Freundlich isotherm Langmuir isotherm D–R isotherm

2 2
KF (mg/g) n R qmax (mg/g) KL (L/mg) R qDR (mmol/g) BDR (mol2/J2) E (kJ/mol) R2
8
Nickel 25 93.88 4.965 0.942 338.6 0.06542 0.971 5.494 3.697  10 3.6776 0.937
35 109.8 5.451 0.917 350.2 0.08876 0.982 5.755 2.796  108 4.2288 0.959
45 122.9 5.879 0.902 357.1 0.12040 0.983 5.928 2.099  108 4.8807 0.969
Copper 25 88.27 5.496 0.902 285.1 0.06443 0.991 4.298 3.755  108 3.6491 0.971
35 94.95 5.525 0.904 302.4 0.07348 0.987 4.577 3.098  108 4.0174 0.969
45 101.3 5.675 0.904 310.2 0.08533 0.986 4.726 2.559  108 4.4203 0.969
Cadmium 25 69.19 5.371 0.929 232.3 0.05040 0.979 1.981 2.847  108 4.1907 0.955
35 84.09 6.078 0.889 242.5 0.07322 0.980 2.101 2.056  108 4.9309 0.971
45 101.2 7.117 0.821 248.1 0.12070 0.963 2.177 1.369  108 6.0434 0.936
Lead 25 64.23 5.434 0.909 214.8 0.04519 0.983 1.007 1.981  108 5.0231 0.975
35 71.39 5.639 0.912 226.2 0.05413 0.987 1.066 1.636  108 5.5283 0.983
45 81.52 6.023 0.885 237.2 0.07084 0.991 1.128 1.258  108 6.3044 0.989

Table 3
Thermodynamic parameters for metal adsorption onto the PEO/Chitosan nanofiber membrane.

Metal KC DH (kJ/mol) DS (kJ/mol K) DG (kJ/mol)

25 °C 35 °C 45 °C 25 °C 35 °C 45°
Nickel 8.965 15.69 21.87 35.210 0.13668 5.434 7.050 8.156
Copper 7.215 8.356 9.952 12.654 0.05884 4.896 5.436 6.075
Cadmium 4.851 6.088 7.602 17.692 0.07249 3.912 4.625 5.363
Lead 3.346 4.812 7.412 31.294 0.11490 2.993 4.023 5.296

MATLAB software are given in Table 2. The relatively high R2 values Thermodynamic parameters for the adsorption process of nickel,
for the D–R model (R2 > 0.93) showed that the model was able to cadmium, copper and lead ions by the nanofiber membrane are
adequately describe the relationship between the amounts of presented in Table 3.
sorbed metal ions and their equilibrium concentration in the solu- As shown in Table 3, the DG value was negative for different
tion. Also, the free adsorption energy for adsorption of different metals at studied conditions. This indicates the feasibility and
metal ions onto the nanofiber membrane was found to be lower spontaneity of different metal sorption by the PEO/Chitosan nano-
than 8 kJ mol1. These results denote that the adsorption mecha- fiber membrane. Also, the amount of DG for different metal sorp-
nism of nickel, cadmium, copper and lead ions by the nanofiber tion onto the membrane inclined to be more negative with
membrane was physical. By comparing the correlation coefficients, increasing temperature which showed that the higher temperature
it was found that the Langmuir isotherm model fitted the equilib- was more feasible for the adsorption of metal ions onto the nano-
rium data of different metal adsorption onto the electrospun com- fiber membrane.
posite nanofiber membrane better than Freundlich and D–R The positive value of DH proved the endothermic nature of the
isotherm models. different metal sorption onto the nanofiber membrane. The posi-
tive value of DS confirmed the increased randomness of solid-
3.5. Thermodynamic parameters solution interface during the adsorption process.

Thermodynamic parameters play an important role in adsorp- 3.6. Regeneration of PEO/Chitosan nanofiber membrane
tion process which determines the degrees of spontaneity and fea-
sibility of process by calculating values of Gibbs free energy and The regeneration of adsorbent is a very important factor in the
entropy change. adsorption process. Because the reuse of the adsorbent is a key fac-
The Gibbs free energy change of adsorption process is calcu- tor in improving process economy. Furthermore, adsorption/
lated by following equation:
200
DG ¼ RT ln kc ð10Þ Nickel
Copper
C es Cadmium
kC ¼ lim ð11Þ 180
C el !0 C el Lead
qe (mg / g)

where R is the gas constant (8.314 J mol1 K1), T is an absolute

temperature (K) and kC is the adsorption equilibrium constant. Ces 160
and Cel are the values of solid and liquid phase concentration in
equilibrium (mg L1) respectively. The relationship between the
equilibrium constant (kc) and temperature is given by the Van’t Hoff 140
equation (Eq. (12)):
DS DH
ln kC ¼  ð12Þ 120
R RT
1 2 3 4 5
The entropy change of adsorption (DS) and the enthalpy Number of cycles
change of adsorption (DH) are obtained from the slope and inter-
cept of a Van’t Hoff plot of ln kC versus 1/T (figure is not presented). Fig. 7. Five cycles of metal ions adsorption–desorption with 1 M HNO3.
 M. Aliabadi et al. / Chemical Engineering Journal 220 (2013) 237–243 243

desorption cycles are repeated to examine the reusability and me- [6] M. Irani, M. Amjadi, M.A. Mousavian, Comparative study of lead sorption
onto natural perlite, dolomite and diatomite, Chem. Eng. J. 178 (2011)
tal recovery efficiency of the adsorbent. The results of five cycles of
317–323.
adsorption/desorption of different metal ions including lead, cop- [7] A. Sari, M. Tuzen, M. Soylak, Adsorption of Pb(II) and Cr(III) from aqueous
per, cadmium and nickel onto the PEO/Chitosan nanofiber mem- solution on Celtek clay, J. Hazard. Mater. 144 (2007) 41–46.
brane are shown in Fig. 7. As can be seen, adsorption capacity of [8] Md. Salim, Y. Munekage, Lead removal from aqueous solution using silica
ceramic: adsorption kinetics and equilibrium studies, Int. J. Chem. 1 (2009)
the membrane for metal ions was gradually reduced after five 23–30.
adsorption–desorption cycles. It could be due to the loosing of [9] J. Deng, X. Kang, L. Chen, Y. Wang, Z. Gu, Z. Lu, A nanofiber functionalized with
some functional groups of membrane by the acid cleavage. Results dithizone by co-electrospinning for lead (II) adsorption from aqueous media, J.
Hazard. Mater. 196 (2011) 187–193.
showed that the PEO/Chitosan nanofiber membrane can be reused [10] C.S. Ki, E.H. Gang, I.C. Um, Y.H. Park, Nanofibrous membrane of wool keratose/
frequently without almost significant loss in adsorption perfor- silk fibroin blend for heavy metal ion adsorption, J. Memb. Sci. 302 (2007) 20–
mance and can be utilized extensively in industrial activities. 26.
[11] S. Xiao, H. Ma, M. Shen, S. Wang, Q. Huang, X. Shi, Excellent copper(II) removal
using zero-valent iron nanoparticle-immobilized hybrid electrospun polymer
4. Conclusion nanofibrous mats, Colloid. Surf. A: Physicochem. Eng. Asp. 381 (2011)
48–54.
[12] M. Teng, H. Wang, F. Li, B. Zhang, Thioether-functionalized mesoporous fiber
In the present study, PEO/Chitosan nanofiber membrane was membranes: sol–gel combined electrospun fabrication and their applications
prepared by electrospinning technique and the potential of PEO/ for Hg2+ removal, J. Colloid. Interf. Sci. 355 (2011) 23–28.
Chitosan nanofibers as an adsorbent for the removal of nickel, cad- [13] S. Haider, S.-Y. Park, Preparation of the electrospun chitosan nanofibers and
their applications to the adsorption of Cu(II) and Pb(II) ions from an aqueous
mium, copper and lead ions from aqueous solution was investi- solution, J. Memb. Sci. 328 (2009) 90–96.
gated. Based on results, the maximum adsorption capacity of [14] M. Irani, A.R. Keshtkar, M.A. Moosavian, Removal of cadmium from aqueous
nickel, copper, cadmium and lead ions by the PEO/Chitosan nano- solution using mesoporous PVA/TEOS/APTES composite nanofiber prepared by
sol–gel/electrospinning, Chem. Eng. J. 200–202 (2012) 192–201.
fiber membrane followed the descending order: nickel(II)>cop- [15] M. Teng, H. Wang, F. Li, B. Zhang, Thioether-functionalized mesoporous fiber
per(II)>cadmium(II)>lead(II). The pore volume calculated by BJH membranes: sol–gel combined electrospun fabrication and their applications
model and its results showed that the structure of PEO/Chitosan for Hg2+ removal, J. Colloid. Interface Sci. 355 (2011) 23–28.
[16] S. Wu, F. Li, H. Wang, L. Fu, B. Zhang, G. Li, Effects of poly(vinyl alcohol) (PVA)
nanofiber membrane was mesoporous. The kinetic data of metal content on preparation of novel thiol-functionalized mesoporous PVA/SiO2
sorption onto the nanofiber membrane were fitted well by both composite nanofiber membranes and their application for adsorption of heavy
of the pseudo-first-order and pseudo-second-order kinetic models. metal ions from aqueous solution, Polymer 51 (2010) 6203–6211.
[17] K. Saeed, S. Haider, T.J. Oh, S.Y. Park, Preparation of amidoxime-modified
The comparison of isotherm models showed that the Langmuir
polyacrylonitrile (PAN-oxime) nanofibers and their applications to metal ions
model described the equilibrium data of nickel, cadmium, copper adsorption, J. Memb. Sci. 322 (2008) 400–405.
and lead sorption better than the Freundlich and the Dubinin– [18] P. Heikkila, A. Harlin, Parameter study of electrospinning of polyamide-6, Eur.
Polym. J. 44 (2008) 3067–3079.
Radushkevich (D–R) isotherm models. The mean free energy values
[19] U.S. Sajeev, K.A. Anand, D. Menon, S. Nasir, Control of nanostructures in PVA,
calculated from the D–R model indicated that the adsorption PVA/chitosan blends and PCL through electrospinning, Bull. Mater. Sci. 31
mechanism of different metal ions using PEO/Chitosan nanofiber (2008) 343–351.
membrane was physical adsorption. The calculated thermody- [20] S.A. Tehron, E. Zussman, A.L. Yarin, Experimental investigation of the
governing parameters in the electrospinning of polymer solutions, Polymer 5
namic parameters using the equilibrium data showed the feasibil- (2004) 2017–2030.
ity, endothermic and spontaneous nature of the metal ions [21] N. Gupta, A.K. Kushwah, M.C. Chattopadhyay, Adsorptive removal of Pb2+, Co2+
adsorption at studied conditions. Decreasing of DG values with and Ni2+ by hydroxyapatite/chitosan composite from aqueous solution, J.
Taiwan Inst. Chem. Eng. 43 (2012) 125–131.
increasing temperature showed an increase in feasibility of adsorp- [22] W.S. WanNgah, C.S. Endud, R. Mayanar, Removal of copper(II) ions frOITI
tion process at higher temperatures. The reusability of nanofiber aqueous solution onto chitosan and cross-linked chitosan beads, React. Funct.
membrane was determined after five adsorption–desorption cycles Polym. 50 (2002) 181–190.
[23] A.H. Chen, S.C. Liu, C.Y. Chen, Comparative adsorption of Cu(II), Zn(II), and
and the results showed that PEO/Chitosan nanofiber membrane Pb(II) ions in aqueous solution on the crosslinked chitosan with
could be utilized extensively in industrial activities. epichlorohydrin, J. Hazard. Mater. 154 (2008) 184–191.
[24] K.T. Shalumon, K.H. Anulekha, C.M. Girish, R. Prasanth, R.J. Nair, Single step
electrospinning of chitosan/poly(caprolactone) nanofibers using formic acid/
References
acetone solvent mixture, Carbohydr. Polym. 80 (2010) 413–419.
[25] S. Lagergren, Zur theorie der sogenannten adsorption geloster stoffe, Kungliga
[1] M. Nourbakhsh, S. Illhan, H. Ozdag, Biosorption of Cr6+, Pb2+ and Cu2+ ions in Sevenska Vetenskapsakademiens, Handlingar 24 (1898) 1–39.
industrial waste water on Bacillus sp, Chem. Eng. J. 85 (2002) 351–355. [26] Y.S. Ho, G. McKay, Pseudo-second-order model for sorption processes, Process
[2] M.M. Matlock, B.S. Howerton, J.D. Robertson, D.A. Atwood, Gold ore column Biochem. 34 (1999) 451–465.
studies with a new mercury precipitant, Ind. Eng. Chem. Res. 41 (2002) 5278– [27] I. Langmuir, The constitution and fundamental properties of solids and liquids,
5282. J. Am. Chem. Soc. 38 (1916) 2221.
[3] A. Smara, R. Delimi, E. Chainet, J. Sandeaut, Removal of heavy metals from [28] H.M.F. Freundlich, Uber die adsorption in losungen, J. Am. Chem. Soc. 57
diluted mixtures by a hybrid ion-exchange/electrodialysis process, Sep. Purif. (1906) 385–471.
Technol. 57 (2007) 103–110. [29] M.M. Dubinin, E.D. Zaverina, L.V. Radushkevich, Sorption and structure of
[4] C. Böchler, J. Dora, V. Mavrov, H. Chmiel, N.K. Lavaridis, K.A. Matis, Hybrid active carbons. I. Adsorption of organic vapors, Zhurnal Fizicheskoi Khimii 21
flotation—membrane filtration process for the removal of heavy metal ions (1947) 1351–1362.
from wastewater, Water. Res. 37 (2003) 4018–4026. [30] M. Irani, A.R. Keshtkar, M.A. Mousavian, Removal of Cd(II) and Ni(II) from
[5] B.R. Reddy, D.N. Priya, Chloride leaching and solvent extraction of cadmium, aqueous solution by PVA/TEOS/TMPTMS hybrid membrane, Chem. Eng. J. 175
cobalt and nickel from spent nickel–cadmium, batteries using Cyanex 923 and (2011) 251–259.
272, J. Power Sources 161 (2006) 1428–1434.