Chemical Engineering Journal 326 (2017) 141–150

Contents lists available at ScienceDirect

Chemical Engineering Journal

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

Fast and efficient removal of copper using sandwich-like graphene oxide

composite imprinted materials
Delong Kong, Ning Qiao, Huan Liu, Jian Du, Nian Wang, Zhiyong Zhou ⇑, Zhongqi Ren ⇑
College of Chemical Engineering, Beijing University of Chemical Technology, Beijing 100029, PR China

h i g h l i g h t s g r a p h i c a l a b s t r a c t

A sandwich-like graphene oxide

composite copper imprinted polymer
was prepared.

Effect of solvents on polymerization
and adsorption properties was
investigated.

Copper-imprinted polymer has fast
adsorption kinetics and high
adsorption capacity.

Polymers have good specific
identification for Cu(II) in presence of
other ions.

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

Article history: A sandwich-like ion-imprinted polymer (IIP) was prepared using functional graphene oxide (GO) as the
Received 6 April 2017 support, acrylamide as the functional monomer, and ethylene glycol dimethacrylate as the crosslinking
Received in revised form 21 May 2017 agent in the presence of copper ions with various solvents (porogen). The effect of solvent type on poly-
Accepted 22 May 2017
merization process and properties of polymers was investigated. Methanol/acetonitrile was considered as
Available online 24 May 2017
the best porogen, and almost no redundant secondary polymer particles were produced. This can be
explained by the good thermodynamic compatibility and slow phase separation property of monomers
Keywords:
and crosslinking agents in polymerization process. Fourier-transform infrared spectroscopy, Energy dis-
Cu (II)
Ion imprinted
persive X-ray spectroscopy, thermogravimetric analysis, and scanning electron microscopy were used to
Graphene oxide confirm the prepared IIP. The effects of pH, initial concentration of the Cu(II) aqueous solution, and con-
Adsorption tact time on adsorption behavior were investigated. Adsorption by prepared IIP was fast (adsorption
Mechanism equilibrium was reached within 15 min) and followed pseudo-second-order kinetic and Freundlich iso-
therm models. The IIP showed a good imprinted factor and the maximum adsorption capacity was up
to 132.77 mg g1, which is higher than that of IIP prepared by bulk polymerization, owing to the large
surface area of GO and surface imprinted technology. The selectivity factors for Cu/Zn, Cu/Ni, Cu/Co,
and Cu/Cd were 36.33, 8.44, 77.3, and 8.5, respectively. The polymer showed a widely application pro-
spect for recovery of Cu ions from aqueous solutions.
Ó 2017 Published by Elsevier B.V.

1. Introduction

Rapid industrial development in China has led to concern

⇑ Corresponding authors. regarding the environmental impact of the discharge of increasing
E-mail addresses: zhouzy@mail.buct.edu.cn (Z. Zhou), renzq@mail.buct.edu.cn amount of industrial wastewater. In particular, industrial wastew-
(Z. Ren).

http://dx.doi.org/10.1016/j.cej.2017.05.140
1385-8947/Ó 2017 Published by Elsevier B.V.
142 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150
ater containing heavy-metal ions, such as copper, zinc, and lead, is
one of the sources of severe pollution. Such wastewater is toxic and
the complex compositions change rapidly. Copper ion is a common
toxic contaminant in wastewater, and excessive amounts of Cu(II)
are harmful to the environment and living organisms [1]. Various
techniques have been developed for removal of copper ions from
wastewater [2,3], e.g., chemical precipitation [4], adsorption [5],
ion exchange [6], membrane filtration [7], and electrochemical
treatment [8]. Among these methods, adsorption is deemed to be
the most widely used and economically feasible alternative due
to high efficiency and low cost. Various adsorbents, such as active
carbon, molecular sieves, resins, and bio-adsorbents, are widely
used for the treatment of wastewater containing heavy metals
[9]. However, most of the adsorption materials are non-selective
and show a low adsorption capacity and slow adsorption kinetic.
Therefore, development of high selective adsorbents for fast
removal of heavy metal ions has become the current research hot.
Ion-imprinting technology (IIT) is attracting increasing atten-
tion because of its structure predictability, recognition specificity,
and universal applicability [10,11]. It is widely used in many fields
including separation, catalysis, and detection. Ion-imprinted poly-
mers (IIPs) are prepared by copolymerization of functional mono-
mers and cross-linkers in presence of a target ion based on
specific coordinative or electrostatic interactions. Removal of the
template ion with an eluent results in the formation of three-
dimensional specific recognition cavities in the polymer matrix
[12,13]. However, some problems such as incomplete removal of
template and slow mass transfer property restrict the use of tradi-
tional IIPs [10,14,15]. The surface imprinting technique, which
makes binding sites locate on the adsorbent surface or in a thin
surface layer, overcomes these problems and gives highly efficient
removal performance of template ions, low mass-transfer
resistance, and good accessibility [16,17]. Li et al. [18] used a
surface-imprinting technique with activated carbon as the support
to prepare a polymer for selective solid-phase extraction of Cu(II);
the maximum static adsorption capacities of the ion-imprinted and
non-imprinted adsorbents were 26.71 and 6.86 mg g1, respec-
tively. Yan [19] synthesized a Pb(II)-IIP using a surface-
imprinting technique; the fast kinetics and the satisfied adsorption
capacity were obtained.
Graphene oxide (GO) with a large surface area and numerous
oxygen atoms is a promising material for heavy-metal removal.
However, the direct use of GO as an adsorbent for heavy-metal ions
is impractical [20]. GO tends to aggregate in a layer-by-layer mode
because of strong interplanar interactions, which is not favorable
for regeneration of adsorbent [21]. In addition, GO is also unstable
at high temperature. For better practical applications, it is there-
fore necessary to modify GO or prepare GO composite materials
to improve the stability of GO [22,23].
Due to the unique surface properties of GO, modified or func-
tionalized GO can be used as a solid support for IIPs. The large
specific surface area and two-dimensional nanosheet structure of
GO are expected to give a high adsorption capacity and fast mass
transfer property. Liu et al. [24] synthesized a novel GO-based
cadmium-imprinted polymer with a high adsorption capacity
and fast adsorption kinetics. Pan et al. [25] prepared a GO-based
molecularly imprinted polymer for fast recognition and capture
of luteolin. Although many GO-based polymers have been pre-
pared, the effect of solvent type on polymerization has not been
studied.
In this study, novel copper-ion-imprinted polymers based on
functionalized GO were synthesized using a surface-imprinting
technique with acrylamide (AM) as the functional monomer. The
combination of IIP and GO using a surface-imprinting technique
provides selective performance with fast mass transfer property
and also can solve some problems associated with practical appli-
cations of GO. For surface-imprinting techniques, the solvent type
affects the inner structure of the IIP and polymer-gel production.
Various polymerization solvents were used to avoid polymer-gel
production. The imprinted polymer was characterized using
Fourier-transfer infrared (FT-IR) spectroscopy, scanning electron
microscopy (SEM), thermogravimetric analysis (TGA), and Energy
Dispersive X-ray spectroscopy (EDX). The effects of initial pH, ini-
tial Cu(II) concentrations, and contact time on the adsorption
behavior of copper were investigated. The thermodynamics and
kinetics of the adsorption process were also studied to explore
the adsorption mechanism. The reusability of the prepared adsor-
bent was also investigated.
2. Experimental
2.1. Materials and analysis
GO was purchased from Nanjing JCNANO Co., Ltd. Ethylene gly-
col dimethacrylate (EGDMA), acrylamide (AM), 3-(Trimethoxysilyl)
propyl methacrylate (MPS) and azodiisobutyronitrile (AIBN) were
obtained from China National Pharmaceutical Group Corp. All
other chemicals were analytical grade and purchased from Beijing
Chemical Factory. Copper (II) aqueous solutions were prepared
using Cu(NO3)23H2O. Deionized water was used to prepare all
solutions. UV–vis spectroscopy (UV-1800, Shimadzu, Japan) was
used to determine the Cu(II) concentrations in aqueous solutions.
A pH-meter (UB-7, Denver, USA) was used to measure the pH of
aqueous solutions.
2.2. Characterization
FT-IR spectra (FT-IR 8000, Shimadzu, Japan) of the IIPs were
recorded before and after elution. GO-IIP was mixed with KBr
and pressed into pellets. The particle morphology was examined
using SEM (S-4700, Hitachi, Japan). TGA was performed on a TG
thermal analyzer (TG209C, Netzsch, Germany) in a dynamic N2
atmosphere from 25 to 800 °C. EDX (S-4700 SEM-EDS, Hitachi,
Japan) was used to detect copper in the IIP before and after elution.
N2 adsorption–desorption measurements were carried out using a
Belsorp-Max surface area and pore size analyzer. Surface areas
were calculated by the Brunauer–Emmett–Teller (BET) method,
and the pore volume and pore size distribution were calculated
using the Barrett–Joyner–Halenda (BJH) model.
2.3. Preparation process of the imprinted polymer
2.3.1. Modification of GO
In a typical procedure, a certain amount of GO was dispersed in
methanol and the mixture was ultrasonicated for 60 min. 3-
(Trimethoxysilyl) propyl methacrylate (4 mL) and NH3H2O
(3 mL) were then added to the above mixture. The reaction was
continued with stirring for 24 h at 60 °C. After that, the modified
GO (GO-MPS) was separated by high-speed centrifugation and
then washed several times with absolute ethanol and dried under
vacuum.
2.3.2. Preparation of Cu(II)-imprinted polymer with GO as support
(GO-IIP)
GO-IIP was prepared by precipitation polymerization using sur-
face IIT. Cu(NO3)23H2O (0.1 mmol) and AM (0.4 mmol) were
added to methanol/acetonitrile. The solution was stirred mechani-
cally for 1 h to enable self-assembly of Cu(II) and AM. GO-MPS was
dispersed in the mixture ultrasonically and then EGDMA (2 mmol)
was added. The above mixture was purged with N2 and then AIBN
was added to the system. The mixture was polymerized in an oil
 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150 143

bath at 65 °C for 8 h. After polymerization, the products were ðC 0  C f ÞV

recovered by centrifugation. The imprinted polymer was washed
several times with methanol/acetonitrile (3:1 v/v) to remove unre-
acted raw materials. The polymer was then treated with
1.0 mol L1 HCl to completely leach coordinated and non-
coordinated Cu(II). The polymer was then washed with distilled
water, and dried at 50 °C under vacuum. A control sample, i.e., a
non-imprinted polymer (GO-NIP), was also prepared using the
same method but without the addition of Cu(NO3)23H2O. To con-
firm the function of GO, the imprinted bulk polymer without GO
(Bulk-IIP) was also prepared by the same method.
The entire process and corresponding mechanism are shown in
Fig. 1.
Q¼ ð1Þ
m
where Q represents the adsorption capacity of polymers, mg/g. C0
and Cf are the initial and final concentrations of metal ion in aque-
ous solution, mg L1, respectively. V is the volume of aqueous solu-
tion, L. m is the weight of the polymers added into the solution, g.
Selectivity experiments were conducted in a batch system. GO-
IIP was added to an aqueous solution containing 20 mg L1 Cu(II)/
Co(II), Cu(II)/Ni(II), Cu(II)/Zn(II), or Cu(II)/Cd(II). The mixture was
stirred magnetically for adsorption at 25 °C. The distribution ratios,
selectivity factors, and relative selectivity factors were calculated
using the following equations,

2.4. Adsorption experiments C0  Cf

Kd ¼ ðV=mÞ ð2Þ
Cf
The adsorption of Cu(II) from aqueous solution on GO-IIP was
assessed in batch experiments. A GO-IIP sample (10 mg) was dis-
KdðCuðIIÞÞ
persed in Cu(II) aqueous solution (20 mL) at 25 °C using an orbital k¼ ð3Þ
KdðMÞ
shaker. The aqueous solutions with Cu(II) concentration from
20 mg L1 to 200 mg L1 were prepared. The pH was adjusted with
buffer solution, dilute HCl and NaOH aqueous solutions. The con- 0 kimprinted
centration of Cu(II) was determined using UV spectrophotometry k ¼ ð4Þ
knon-imprinted
method.
The adsorption capacity of GO-IIP or GO-NIP was calculated by where Kd is the distribution ratio, M represents other competitive
the following equation, metal ions, k and k0 are the selectivity factor and relative selectivity
factor, respectively.

Fig. 1. The scheme of the whole experimental process and corresponding mechanism.
144 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150
3. Result and Discussion
3.1. FT-IR
FT-IR spectral analysis was employed to confirm that the ion-
imprinted composite materials were prepared successfully, and
the imprinted layer was located on the surface of GO. The FT-IR
spectra of GO, GO-IIP, and GO-NIP are shown in Fig. 2. The GO spec-
trum has strong bands at 3400, 1730 and 1640 cm1, correspond-
ing to AOH, C@O, and C@C groups, respectively. In GO-IIP
spectrum, the peak at 3438 cm1 is the overlapped vibrations peak
of AOH and ANH2. The peak at 1722 cm1 and 1259 cm1 corre-
spond to C@O (ester) stretching vibrations (the C@O group existed
in crosslinking agent EGDMA). There are two absorption bands,
centered at 1635 and 1450 cm1, which correspond to the C@O
and NAH vibration of ACONH2 (the group existed in the monomer
AM), respectively. The peak at 1384 cm1 corresponds to CAN (the
group existed in the monomer AM). The peak at 964 cm1 is
assigned to the quaternary amine group. The FT-IR results showed
that ion imprinted polymer on surface of GO was successfully
synthesized.
3.2. SEM
The morphology of GO-IIP was observed using SEM and the
images are shown in Fig. 3 (the images were showed in different
magnification times). All particles had irregular shapes
(Fig. 3a and b). Fig. 3c shows a structure consisting of many layers,
which confirms that the GO-IIP is a stacked composite material.
Fig. 3d shows that the surface of GO-IIP was very rough after poly-
merization with AM and EGDMA, indicating that many polymers
had been grafted on the functional GO surface. The SEM results also
confirm that IIP was assembled on the GO layers.
3.3. EDX
The EDX analysis was employed to determine the composition
of adsorbent and confirm that Cu (II) was eluted successfully. As
shown in Fig. 4, the presence of C, N, O, and Cu could be confirmed
in the structure of GO-IIP. The percentages detected were not pre-
cise because of the presence of C in the tape and sample holder. Cu
Fig. 2. FTIR spectra of GO, GO-IIP and GO-NIP.
was present in the GO-IIP but not in the GO-NIP, indicating that Cu
(II) had successfully participated in the polymerization of GO-IIP.
The existence of Si came from the group of MPS, which further
indicated that the MPS was grafted to the surface of GO
successfully.
3.4. TGA
TGA curves of GO and GO-IIP are shown in Fig. 5. The weights of
GO and GO-IIP decrease slightly between 0 and 100 °C, mainly
because of dehydration. The weight loss increases sharply from
150 °C to 200 °C, due to the decomposition of the oxygen-
containing groups on the GO surface. The weight loss of GO
increases gradually with increasing the temperature further, which
indicates that the GO materials were oxidized well. GO-IIP was
very stable when the temperature was lower than 300 °C, indicat-
ing that approximately 90% of GO-IIP was polymer. Weight loss of
GO-IIP increases sharply with the increase of temperature from
400 °C to 500 °C, which is mainly attributed to the decomposition
of the imprinted polymer. The thermal stability of GO-IIP is obvi-
ously better than that of GO, which is favorable for practical appli-
cations of GO-IIP. The thermal behavior of GO-IIP is probably
associated to the high crosslink density of imprinted layer on the
GO surface.
3.5. Preparation and batch adsorption experiments
3.5.1. The porous properties and adsorption capacities of different
polymers
Solvent significantly affects the performance of the prepared IIP,
since it acts as a porogen in preparation process. In surface-
imprinting technique, factors such as solvent polarity and solubil-
ity of the product, significantly affect the bonding strength, IIP
inner structure, and polymer-gel production. The surface-
imprinted layer should be located on the surface and polymer-
gel should be avoided. Therefore, the effect of solvent type, i.e.,
methanol/H2O and methanol/acetonitrile, on the polymerization
reaction was investigated. Methanol and H2O were selected as sol-
vents because they were used in GO modification process (as
described in section Modification of GO) and gave a good dispersity
of GO. Acetonitrile was used because it had high imprinting effi-
ciency due to its low polarity.
The surface structure was determined by the Brunauer, Emmett
and Teller (BET) method. The porosity of the prepared materials
was revealed by the N2 adsorption isotherms. The surface area,
pore volume and average pore diameter of polymer prepared with
methanol/H2O (GO-IIP-1) are 73.588 m2 g1, 0.303 cm3 g1 and
16.44 nm, respectively. The surface area, pore volume and average
pore diameter of polymer prepared with methanol/acetonitrile
(GO-IIP-2) are 19.829 m2 g1, 0.150 cm3 g1 and 12.00 nm, respec-
tively. Although GO-IIP-1 has larger surface areas and pore vol-
umes than that of GO-IIP-2, the adsorption capacity of GO-IIP-2
(22.07 mg g1) is even a little higher than that of GO-IIP-1
(21.3 mg g1) in an aqueous solution with 200 mg L1 initial Cu
(II) concentration. This is ascribed to that more imprinted sites
may be obtained in the imprinted layer of GO-IIP-2.
The patterns of polymer prepared with different solvents are
shown in Fig. S1 (see Supporting Information). The colour of poly-
mer prepared with methanol/acetonitrile is black brown, which is
similar with the modified GO. However, the colour of polymer pre-
pared with methanol/H2O is hoar, due to the formation of polymer-
gel. This phenomenon can be explained by that monomer,
crosslinking agent and double bond in the surface of modified
GO have good thermodynamic compatibility and slow phase sepa-
ration rate in the methanol/acetonitrile system. Therefore, metha-
 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150 145

Fig. 3. SEM images of GO-IIP.

nol/acetonitrile was used as the solvent in the both preparation GO-IIP prepared in this work is faster than that of nearly all other
processes of GO-IIP and GO-NIP. adsorbents listed in Table 1.
This fast adsorption process is attributed to the formation of
3.5.2. Effect of pH surface ion imprinting binding sites on a thin surface film of the
The pH of aqueous solution affects the ion form and the number GO-IIP and good complexation performance of the Cu(II) with
of protonated nitrogen in the bonding sites of GO-IIPs. The adsorp- functional monomer. The GO-IIP prepared in our study is therefore
tion property of GO-IIPs is dependent on the interaction between a good adsorbent for application in the analysis and detection of Cu
the template ion and functional group. Thus, the effect of pH on (II).
adsorption performances was investigated. Pseudo-first-order and pseudo-second-order models were used
Since the pKa of acrylamide is 4.25, the pH value which is higher to evaluate the adsorption kinetics.
than 4.25 can realize the deprotonation of monomer, enhancing The above two models are expressed as the following equations
the attraction for Cu(II). In addition, the precipitation formed with [26,27],
Cu(II) and OH- will appear in the solution when the pH value is The pseudo-first-order kinetic model:
higher than 7. Therefore, effect of pH ranging from 4.0 to 7.0 on  
k1
adsorption capacity of GO-IIP was investigated. As shown in logðqe1  qt1 Þ ¼ log qe1  t ð5Þ
2:303
Fig. 6, the adsorption capacity increases with increasing pH from
4 to 7. The protonation level of amino groups weakens with The pseudo-second-order kinetic model:
increasing pH, resulting in the increase of adsorption capacity.
t 1 t
¼ þ ð6Þ
qt2 k2 q2e2 qe2
3.5.3. Effect of contact time
The study of adsorption kinetics was conducted at 25 °C and the where qe1 and qe2 are the experimental amount of Cu(II) adsorbed
results are shown in Fig. 7. The adsorption rate was fast and at equilibrium state, mg g1; qt1 and qt2 are the amount of Cu(II)
adsorption equilibrium was reached within 15 min. The compar- adsorbed at time t, mg g1; k1 and k2 are the adsorption rate con-
ison of the performance of different copper(II)-imprinted polymers stant of the pseudo-first-order and pseudo-second-order adsorp-
is listed in Table 1. It can be seen that the adsorption process by tion, min1 and g mg1 min1.
146 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150

Fig. 6. Effect of pH on adsorption capacity of GO-IIP (10 mg adsorbent, 20 mL

20 mg L1 copper solution).

Fig. 7. Effect of contact time on adsorption capacity of GO-IIP. (10 mg adsorbent,

20 mL 50 mg L1 copper solution, pH 6.0).
Fig. 4. (a) EDX spectra of GO-IIP (b) EDX spectra of GO-NIP.

Table 1
Comparison of the performance of different copper(II)-imprinted polymers.

Adsorbent Equilibrium Q/mgg1 Reference

time
Poly(EGDMA-MAH/CuSO4) 60 min 48 [28]
CuSO4-IIP(MAA/4-VP) 60 min 15 [29]
P(TSC-Cu(NO3)2) 20 min 38.8 [30]
Poly(EGDMA-AQ/Cu(NO3)2) 30 min 4.7 [31]
Cu-IIP(Cu(II)-PAR/salicylic 60 min 0.3 [32]
acid/formaldehyde)
Cu-IIP (TEPA) – 33.33 [33]
CuSO4-IIP(MAA/diethylenetriamine) – 76.5 [34]
Cu((DBDA15C4)/2-vinylpyridine)-IIP 45 min 4.8 [35]
Carbon-Cu-IIP 10 min 26.71 [36]
Magnetic-Cu-IIP 60 min 42.2 [37]
(Cu-PEI)-imprinted hydrogel 36 h 40.00 [38]
GO-IIP 15 min 132.77 This work

The calculated kinetic parameters are listed in Table 2. Fig. S2

shows the fitting plots of log (qe  qt) (pseudo-first order equation)
or t/q (pseudo-second order equation) on the y-axis versus t on the
Fig. 5. TGA curves of GO and GO-IIP. x-axis. The correlation coefficient (R2) of the pseudo-first-order
 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150 147

Table 2
Comparison of pseudo-first-order and pseudo-second-order adsorption kinetic rate constants.

Pseudo-first-order kinetic models Pseudo-second-order kinetic models

1 1
qexp (mg g ) qe1 (mg g ) k1 R 2
qexp (mg g1) qe2 (mg g1) k2 R2
22.43 6.33 0.144 0.9129 22.43 23.36 0.036 0.9992

tion isotherms at different concentrations were recorded. The ini-

tial Cu(II) concentration of the aqueous solution affects mass
transfer property during adsorption. The adsorption capacities of
the GO-IIP and GO-NIP both increase with increasing initial con-
centration of Cu(II); the maximum adsorption capacity of GO-IIP
(38.45 mg g1) was achieved with an initial Cu(II) concentration
of 200.0 mg L1, as shown in Fig. 8(a). The adsorption capacity of
GO-IIP is higher than that of GO-NIP at the same concentration.
In addition, the adsorption capacity of GO-NIP does not change
when the Cu(II) concentration is higher than 100 mg L1. This
results proved that the imprinted binding sites enhanced Cu(II)
adsorption by the IIP. This is mainly because imprinting provides
specific binding cavities in the polymer and enables strong chela-
tion function between the polymer surface groups and Cu(II).
In order to know the function of GO, the adsorption perfor-
mances of polymer prepared with GO and without GO were com-
pared. The ion imprinted polymer without GO was defined as the
conventional polymer prepared by bulk polymerization. The
imprinted bulk polymer without GO (Bulk-IIP) was prepared by
the same condition as the GO-IIP. The comparison results are
showed in Fig. 8(a). The results show that the maximum adsorp-
tion capacity of GO-IIP (38.45 mg g1) is much higher than that
of polymer prepared without GO (11.53 mg g1). Though the poly-
mer without GO was not prepared by the same polymerization
conditions reported in the previous study, the bulk polymer was
reported previously [39] and could be used to further confirm
the function of GO in our study. The copper imprinted cross-
linked poly(AM-co-EGDMA) was synthesized by bulk polymeriza-
tion, and the maximum adsorption capacity of IIP was just
1 mg g1 [39]. In addition, the surface area of GO-IIP is higher than
that of the bulk IIP [39] because of the existence of GO. The adsorp-
tion capacity of GO-IIP could be significantly improved by the com-
bination of GO and imprinted technology. Both the large surface of
GO and the numerous surface binding sites of imprinted layer con-
tribute to the high adsorption capacity.
Langmuir and Freundlich adsorption isotherm models were
used to study the adsorption thermodynamic properties. The
expression of the above two models are as following,
Langmuir adsorption isotherm equation:

Fig. 8. (a) Effect of initial Cu(II) concentration on the adsorption capacity. (10 mg
C e =Q e ¼ C e =Q max þ 1=ðQ max  bÞ ð7Þ
adsorbent, 20 mL copper solution, pH 6.0) (b) Effect of initial Cu(II) concentration on
Freundlich adsorption isotherm equation:
the adsorption capacity of GO-IIP. (10 mg adsorbent, 20 mL copper solution, pH
6.5).
 
1
InQ e ¼ InK F þ InC e ð8Þ
n

equation is 0.9129 and the corresponding calculated value of qe1 is
significantly lower than the experimental value. The experimental
data fits the pseudo-second-order kinetic equation well with a cor-
relation coefficient R2 of 0.9992 and the calculated value of qe2 is
very close to the experimental result. The results show that
adsorption on the GO-IIP can be considered as a pseudo-second-
order kinetic adsorption process.
3.5.4. Effect of initial Cu(II) concentration
To evaluate the adsorption performance of GO-IIP and the
imprinting efficiency, the effect of the initial Cu(II) concentration
on adsorption were investigated at pH 6.0 and 25 °C. The adsorp-
where Qe and Ce are the experimental amount of Cu(II) adsorbed by
the GO-IIP and Cu(II) concentration at equilibrium state, mg g1 and
mg L1; Qmax is the maximum amount of adsorption, mg g1; b rep-
resents the constant in Langmuir adsorption isotherm model,
L mg1; KF and n are both constants in Freundlich adsorption iso-
therm model.
Langmuir and Freundlich adsorption isotherm models were
used to investigate the adsorption thermodynamics. The curves
obtained from plots of Ce/Qe or ln Qe on the y-axis versus Ce or ln
Ce on the x-axis are shown in Fig. S3. The calculated parameters
of two models are listed in Table 3. Both the adsorption capacities
of GO-IIP and GO-NIP calculated by the Langmuir model don’t
agree with the experimental data. Therefore, the results of lang-
148 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150

Table 3a
Comparison of Langmuir and Freundlich isotherm constants-GO-IIP.

Langmuir Freundlich
2
Qexp (mg/g) Qmax (mg/g) b R KF n R2
38.45 13.32 0.206 0.9157 2.01 1.69 0.9821

Table 3b
Comparison of Langmuir and Freundlich isotherm constants-GO-NIP.

Langmuir Freundlich
Qexp (mg/g) Qmax (mg/g) b R2 KF n R2
16.15 49.26 0.0175 0.9897 2.00 1.81 0.995

Table 4
Selectivity parameters of GO-IIP and GO-NIP for Cu(II) in the presence of competitive metal ions.

GO-IIP GO-NIP
Kd(Cu(II)) Kd(ion(II)) k Kd(Cu(II)) Kd(ion(II)) k k’
Cu/Zn 1.333 0.0367 36.33 0.247 0.030 8.28 4.39
Cu/Co 0.839 0.0108 77.3 0.440 0.146 4.37 17.69
Cu/Cd 0.690 0.080 8.5 0.276 0.0483 5.7 1.50
Cu/Ni 0.427 0.050 8.44 0.242 0.04171 5.82 1.45

muir (Qmax = 13.32 mg g1) is not consistent with the experimental which may affect the adsorption and diffusion behaviors of Cu(II)
results (38.45 mg g1). Meanwhile, the results of R2 show that the in Cu(II) imprinted cavities. Based on the Hard and Soft Acids
Freundlich adsorption isotherm model is more appropriate for the and Bases theory, Co(II) is a kind of borderline acid metal ion.
description of adsorption process by GO-IIP than the Langmuir The acidity of Co(II) is the strongest among Cu(II), Zn(II), Co(II)
adsorption isotherm model, which indicates that the adsorption and Ni(II). AM is a kind of weak-base monomers and the complex
process of Cu(II) by the prepared GO-IIP is a multiple-layer adsorp- formed by AM and Co(II) is not very stable in aqueous solution.
tion process. Based on our previous studies on Ni-IIP and Cu-IIP, AM has a better
The effect of the initial Cu(II) concentration on the adsorption compatible chelate ability of Cu(II) and Ni(II) than that of Co(II).
capacity at pH 6.5 was investigated too. Fig. 8(b) shows that the Meanwhile, due to the imprinting effect, Cu(II) could preferentially
maximum adsorption capacity was 132.77 mg g1. The comparison occupy the imprinted cavity and the mass transfer property of Cu
of the adsorption capacity of different copper(II)-imprinted poly- (II) is not obviously affected by the adsorption of Co(II). Therefore,
mers are previously listed in Table 1. It can be seen that the GO- the GO-IIP showed a good selectivity for Cu/Co. Cd(II) is a typical
IIP prepared in this study has the highest adsorption capacity. soft acid ion and has a stronger hydrolysis ability than other metal
ions. In the same adsorption conditions, much hydrated Cd ions
3.6. Adsorption selectivity adsorbed on GO-IIP occupy the adsorption site and affect the mass
transfer and diffusion process of Cu(II), due to the steric effect
The adsorption selectivity of Cu(II)-IIP was investigated using of hydrated Cd ions, leading to the low adsorption capacity of
binary mixtures. Simulated wastewaters containing two metal Cu(II), resulting in low selectivity for Cu/Cd.
ions, i.e., Cu(II)/Zn(II), Cu(II)/Ni(II), Cu(II)/Co(II), or Cu(II)/Cd(II)
were prepared. The concentration of both metal ions were
20.0 mg L1. These ions were selected as competitive ions because
their ionic radius and chemical behaviors are similar to those of Cu
(II) or they are likely to coexist in real wastewaters. The selectivity
coefficient, selectivity factor and relative selectivity factors in each
binary system were calculated using Eqs. (2)-(4) and the results are
listed in Table 4. The selectivity factors of Cu(II)/Zn(II), Cu(II)/Ni(II),
Cu(II)/Co(II) and Cu(II)/Cd(II) are 36.33, 8.44, 77.3, and 8.5, respec-
tively. The selectivity factors of GO-IIP prepared in this work are
higher than those of IIP-prepared by bulk polymerization (selectiv-
ity factors of Cu(II)/Cd(II), Cu(II)/Zn(II) and Cu(II)/Co(II) were 3.60,
6.05 and 5.05, respectively) [39,40]. The relative selectivity factors
are much higher than one because of the specific recognition abil-
ity for Cu(II) during imprinting process.
In our study, GO-IIP has a better adsorption selective for Cu/Zn
and Cu/Co. Zn(II), Ni(II) and Cu(II) have similar chemical property.
The ion radius of Zn(II) is larger than that of Cu(II). Hence, Zn(II)
doesn’t match the imprinted cavities after the elution of Cu(II). In
this case, the adsorption of Cu(II) is not affected by the existence
of Zn(II) and GO-IIP show high adsorption selectivity for Cu/Zn. Fig. 9. Regeneration of GO-IIP (10 mg adsorbent, 20 mL 50 mg L1 copper solution,
However, the ion radius of Ni(II) is smaller than that of Cu(II), pH 6.0).
 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150
3.7. Reusability and stability
Regeneration is an important index in evaluating adsorbents for
practical applications. GO-IIP was eluted with 1.0 M HCl solution
until Cu(II) could not be detected. An aqueous solution with the
same Cu(II) concentration as that originally used was then added.
The process was repeated five times and each adsorption capacity
was recorded. The results are shown in Fig. 9. It can be seen that
just a little decrease of the adsorption capacity of GO-IIP could
be observed after 5 regeneration cycles, indicating that elution pro-
cess did not affect the polymer cavity structure and chemical prop-
erties. GO-IIP has good stability and regeneration properties.
4. Conclusion
A novel IIP was prepared by surface IIT. Functional GO was used
as the imprinted polymer support because of its large surface area.
AM, EGDMA and AIBN were used as the functional monomer,
cross-linker and initiator, respectively. Methanol/acetonitrile was
used as the polymerization solvent to reduce the production of sec-
ondary polymer particles. The prepared IIP was characterized using
FT-IR, TGA, and SEM. Adsorption process was fast (adsorption equi-
librium was reached within 15 min) and followed a pseudo-
second-order kinetic equation and Freundlich isotherm adsorption
model. The adsorption capacity of the GO-IIP (38.45 mg g1) was
higher than that of the GO-NIP (16.15 mg g1) at pH 6.0. The max-
imum adsorption capacity of the GO-IIP (132.77 mg g1) was
obtained with an initial Cu(II) concentration of 200 mg L1 at pH
6.5. The selectivity factors of Cu(II)/Zn(II), Cu(II)/Ni(II), Cu(II)/Co
(II), and Cu(II)/Cd(II) pairs were 36.33, 8.44, 77.3, and 8.5, respec-
tively. The good selectivity performances are attributed to the cre-
ation of specific recognition sites for Cu(II) during imprinting
process. The polymer shows a widely application prospect for
recovery of Cu from aqueous solutions.
Acknowledgements
This work was supported by the National Natural Science Foun-
dation of China (21576010 and U1607107), Supported by Beijing
Natural Science Foundation (2172043), Fundamental Research
Funds for the Central Universities (BUCTRC-201515) and Higher
Education and High-quality and World-class Universities
(PY201607). The authors gratefully acknowledge these grants.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.cej.2017.05.140.
References
[1] D. Kolodyńka, R. Wne˛trzak, J.J. Leahy, M.H.B. Hayes, W. Kwapiński, Z. Hubicki,
Kinetic and adsorptive characterization of biochar in metal ions removal,
Chem. Eng. J. 197 (2012) 295–305.
[2] N.B. Chang, C. Houmann, K.S. Lin, M. Wanielist, Fate and transport with
material response characterization of green sorption media for copper removal
via adsorption process, Chemosphere 144 (2016) 1280–1289.
[3] N.K. Srivastava, C.B. Majumder, Novel biofiltration methods for the treatment
of heavy metals from industrial wastewater. Removal of heavy metal ions from
wastewaters: a review, J. Environ. Manage. 92 (2011) 407–418.
[4] F. Akbal, S. Camci, Copper, chromium and nickel removal from metal plating
wastewater by electrocoagulation, Desalination 269 (2011) 214–222.
[5] M. Bilal, J.A. Shah, T. Ashfaq, Waste biomass adsorbents for copper removal
from industrial wastewater - a review, J Hazard. Mater. 263 (2013) 322–333.
[6] Q. Liu, Q.Z. Liu, Z.N. Ruan, X.Q. Chang, J.S. Yao, Removal of Cu(II) from aqueous
solution using synthetic poly (catechol-diethylenetriamine–p
phenylenediamine) particles, Ecotoxicol. Environ. Saf. 129 (2016) 273–281.
[7] M. Mullett, R. Fornarelli, D. Raplh, Nanofiltration of mine water: impact of feed
pH and membrane charge on resource recovery and water discharge,
Membranes 4 (2014) 163–180.
149
[8] M.A. Gonzalez, R. Trocoli, I. Pavlovic, C. Barriga, F.L. Mantia, Capturing Cd(II)
and Pb(II) from contaminated water sources by electro-deposition on
hydrotalcite-like compounds, Phys. Chem. Chem. Phys. 18 (2016) 1838–1845.
[9] D. Arán, J. Antelo, S. Fiol, F. Macías, Influence of feedstock on the copper
removal capacity of waste-derived biochars, Biol. Technol. 212 (2016) 199–
206.
[10] T.P. Rao, R. Kala, S. Daniel, Metal ion-imprinted polymers-novel materials for
selective recognition of inorganics, Anal. Chim. Acta 578 (2006) 105–116.
[11] J.Q. Fu, L.X. Chen, J.H. Li, Z. Zhang, Current status and challenges of ion
imprinting, J. Mater. Chem. A 3 (2015) 13598–13627.
[12] M.J. Kuras, E. Wie˛ckowska, Synthesis and characterization of a new copper(II)
ion-imprinted polymer, Polym. Bull. 72 (2015) 3227–3240.
[13] M.R. Younis, S.Z. Bajwa, P.A. Lieberzeit, W.S. Khan, A. Mujahid, A. Ihsan, A.
Rehman, Molecularly imprinted porous beads for the selective removal of
copper ions, J. Sep. Sci. 39 (2016) 793–798.
[14] D.K. Singh, S. Mishra, Synthesis, characterization and analytical applications of
Ni(II)-ion imprinted polymer, Appl. Surf. Sci. 256 (2010) 7632–7637.
[15] D.L. Kong, F. Zhang, K.Y. Wang, Z.Q. Ren, W.D. Zhang, Fast removal of Cr(VI)
from aqueous solution using Cr(VI)-imprinted polymer particles, Ind. Eng.
Chem. Res. 53 (2014) 4434–4441.
[16] M. Monier, A.A. Ibrahim, M.M. Metwally, D.S. Badawy, Surface ion-imprinted
amino-functionalized cellulosic cotton fibers for selective extraction of Cu(II)
ions, Int. J. Biol. Macromol. 81 (2015) 736–746.
[17] Y. Liu, X.G. Meng, M. Luo, M.J. Meng, L. Ni, J. Qiu, Z.Y. Hu, F.F. Liu, G.X. Zhong, Z.
C. Liu, Y.S. Yan, Synthesis of hydrophilic surface ion-imprinted polymer based
on graphene oxide for removal of strontium from aqueous solution, J. Mater.
Chem. A 3 (2014) 1287–1297.
[18] Z.H. Li, J.W. Li, Y.B.Y.J. Wei Wang, Synthesis and application of surface-
imprinted activated carbon sorbent for solid-phase extraction and
determination of copper (II), Spectrochim. Acta A 117 (2014) 422–427.
[19] Y. Liu, Z.C. Liu, J. Gao, J.D. Dai, J. Han, Y. Wang, J.M. Xie, Y.S. Yan, Selective
adsorption behavior of Pb(II) by mesoporous silica SBA-15-supported Pb(II)-
imprinted polymer based on surface molecularly imprinting technique, J.
Hazard. Mater. 1 (2011) 197–205.
[20] G.X. Zhao, T. Wen, C.L. Chen, X.K. Wang, Synthesis of graphene-based
nanomaterials and their application in energy-related and environmental
related areas, RSC Adv. 2 (2012) 9286–9303.
[21] S. Luo, X. Xu, G. Zhou, C. Liu, Y. Tang, Amino siloxaneoligomer-linked
grapheme oxide as an efficient adsorbent for removal of Pb(II) from
wastewater, J. Hazard. Mater. 274 (2014) 145–155.
[22] H.T. Xing, J.H. Chen, X. Sun, Y.H. Huang, Z.B. Su, S.R. Hu, W. Weng, S.X. Li, H.X.
Guo, W.B. Wu, Y.S. He, F.M. Li, Y. Huang, NH2-rich polymer/graphene oxide use
as a novel adsorbent for removal of Cu(II) from aqueous solution, Chem. Eng. J.
263 (2015) 280–289.
[23] D. Vilela, J. Parmar, Y.F. Zeng, Y.L. Zhao, S. Sánchez, Graphene-based microbots
for toxic heavy metal removal and recovery from water, Nano Lett. 16 (2016)
2860–2866.
[24] Y. Liu, X. Hu, M.J. Meng, Z.C. Liu, L. Ni, X.G. Meng, J. Qiu, RAFT-mediated
microemulsion polymerization to synthesize a novel high-performance
graphene oxide-based cadmium imprinted polymer, Chem. Eng. J. 302
(2016) 609–618.
[25] S.C. Liu, J.M. Pan, H.J. Zhu, G.Q. Pan, F.X. Qiu, M.J. Meng, J.T. Yao, D. Yu,
Graphene oxide based molecularly imprinted polymers with double
recognition abilities: the combination of covalent boronic acid and
traditional non-covalent monomers, Chem. Eng. J. 290 (2016) 220–231.
[26] Y.S. Ho, G. McKay, Kinetic models for the sorption of dye from aqueous
solution by wood, Process Saf. Environ. Prot. 76 (1998) 332–340.
[27] Y.S. Ho, G. McKay, The kinetics of sorption of divalent metal ions onto
sphagnum moss peat, Water Res. 34 (2000) 735–742.
[28] R. Say, E. Birlik, A. ErsӦz, F. Yilmaz, T. Gedikbey, A. Denizli, Preconcentration of
copper on ion-selective imprinted polymer microbeads, Anal. Chim. Acta 480
(2003) 251–258.
[29] N.T. Hoai, D.K. Yoo, D. Kim, Batch and column separation characteristics of
copper-imprinted porous polymer micro-beads synthesized by a direct
imprinting method, J. Hazard. Mater. 173 (2010) 462–467.
[30] M. Roushani, S. Abbasi, H. Khani, Synthesis and application of ion-imprinted
polymer nanoparticles for the extraction and preconcentration of copper ions
in environmental water samples, Environ. Monit. Assess. 187 (2015) 219.
[31] M. Shamsipur, A. Besharati-Seidani, J. Fasihi, H. Sharghi, Synthesis and
characterization of novel ion-imprinted polymeric nanoparticles for very fast
and highly selective recognition of copper(II) ions, Talanta 83 (2010) 674–681.
[32] D.K. Singh, S. Mishra, Synthesis of a new Cu(II)-ion imprinted polymer for solid
phase extraction and preconcentration of Cu(II), Chromatographia 70 (2009)
1539–1545.
[33] W. Peng, Z.Z. Xie, G. Cheng, L. Shi, Y.B. Zhang, Amino-functionalized adsorbent
prepared by means of Cu(II) imprinted method and its selective removal of
copper from aqueous solutions, J. Hazard. Mater. 294 (2015) 9–16.
[34] S. Wang, R.F. Zhang, Selective solid-phase extraction of trace copper ions in
aqueous solution with a Cu(II)-imprinted interpenetrating polymer network
gel prepared by ionic imprinted polymer (IIP) technique, Microchim. Acta 154
(2006) 73–80.
[35] M. Shamsipur, J. Fasihi, A. Khanchi, R. Hassani, K. Alizadeh, H. Shamsipur, A
stoichiometric imprinted chelating resin for selective recognition of copper(II)
ions in aqueous media, Anal. Chim. Acta 599 (2007) 294–301.
[36] Z.H. Li, J.W. Li, Y.B. Wang, Y.J. Wei, Synthesis and application of surface-
imprinted activated carbon sorbent for solid-phase extraction and
150 D. Kong et al. / Chemical Engineering Journal 326 (2017) 141–150

determination of copper (II), Spectrochim. Acta Part A Mol. Biomol. Spectrosc.
117 (2014) 422–427.
[37] H. He, D.L. Xiao, J. He, H. Li, H. He, H. Dai, J. Peng, Preparation of a core-shell
magnetic ion-imprinted polymer via a sol-gel process for selective extraction
of Cu(II) from herbal medicines, Analyst 139 (2014) 2459–2466.
[38] J.J. Wang, Z.K. Li, Enhanced selective removal of Cu(II) from aqueous solution
by novel polyethylenimine-functionalized ion imprinted hydrogel: behaviors
and mechanisms, J Hazard. Mater. 300 (2015) 18–28.
[39] T.O. Germiniano, M.Z. Corazza, M.G. Segatelli, E.S. Ribeiro, M.J.S. Yabe, E.
Galunina, C.R.T. Tarleya, Synthesis of novel copper ion-selective material based
on hierarchically imprinted cross-linked poly (acrylamide-co-ethylene glycol
dimethacrylate), React. Funct. Polym. 82 (2014) 72–80.
[40] A. Baghel, M. Boopathi, B. Singh, P. Pandey, T.H. Mahato, P.K. Gutch, K. Sekhar,
Synthesis and characterization of metal ion imprinted nano-porous polymer
for the selective recognition of copper, Biosen. Bioelectron. 22 (2007) 3326–
3334.