Chemical Engineering Journal 343 (2018) 371–378

Contents lists available at ScienceDirect

Chemical Engineering Journal

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

Adsorption of 17β-estradiol by graphene oxide: Effect of heteroaggregation T

with inorganic nanoparticles
⁎
Luhua Jianga,b, Yunguo Liua,b, , Guangming Zenga,b, Shaobo Liuc, Wei Qued, Jiang Lic,
Meifang Lia,b, Jun Wene
a
College of Environmental Science and Engineering, Hunan University, Changsha 410082, PR China
b
Key Laboratory of Environmental Biology and Pollution Control (Hunan University), Ministry of Education, Changsha 410082, PR China
c
School of Architecture and Art Central South University, Central South University, Changsha 410082, PR China
d
Department of Economics and Trade, Hunan University, Changsha 410082, PR China
e
College of Agriculture, Guangxi University, Nanning 530005, PR China

H I GH L IG H T S

• The presence of INPs significantly inhibited the E2 adsorption on GO.

• E2 adsorption onto GO decreased with increase of INPs/GO ratios.
• The inhibitory effect of Al O onto E2 adsorption by GO was greater than that of SiO .
2 3 2

• Electrostatic attraction between oppositely charged Al O and GO favored their heteroaggregation.

2 3

A R T I C LE I N FO A B S T R A C T

Keywords: Increasingly applications of graphene nanomaterials have resulted in their release into the water environments
Graphene nanomaterial where lot kinds of chemicals and inorganic nanoparticles co-exist. In this work, the influences of two common
Inorganic nanoparticle inorganic nanoparticles (INPs), namely SiO2 and Al2O3, on the adsorption of 17β-estradiol (E2) onto graphene
Adsorption oxide (GO) were examined by using batch adsorption experiments in combination with microscopic, spectro-
17β-estradiol
scopic, and computational methods. Results exhibited that the presence of INPs significantly inhibited the ad-
DLVO theory
sorption and increased the time to reach adsorption equilibrium for the adsorption of E2 onto GO. Besides, the
impact of Al2O3 was greater than that of SiO2 due to the enhanced electrostatic attraction between positively
charged Al2O3 and negatively charged GO. Derjaguin-Landau-Verwey-Overbeek calculation revealed that GO
tended to first homoaggregate and then heteroaggregate with SiO2. However, oppositely charged Al2O3 and GO
were likely to heteroaggregation rather than homoaggregation. This was also demonstrated by the zeta potential
measurement, and scanning electron microscope. These observations highlighted the significant influence of
INPs on the adsorption of contaminants onto GO, and also provided new insights into the fate and transport of
GO and pollutants in natural water environments.

1. Introduction
Graphene, as the first available two-dimensional atomic crystal, is a
single layer of sp2-hybridized conjugated carbon atoms closely packed
into a honeycomb lattice [1,2]. Since discovered in 2004, it has been
receiving much scientific attention because of its unique physico-
chemical properties, e.g., superior thermal conductivity, excellent me-
chanical strength, high electron conductivity, and large specific surface
area [3–5]. As it has been regarded as ideal candidate for a wide range
of commercial applications, industrial scale production of graphene has
been predicted to increase exponentially in the next decade [6,7]. Thus,
graphene will be inevitably released into the environment, which may
lead to various health and environmental risks for plants, animals and
humans [8–10]. Estrogens, as one classes of endocrine disrupting che-
micals, have been identified to possess serious endocrine-disrupting
activity and can lead to negative effects on organisms, such as re-
productive disorders, abnormal development (i.e., malformations,
feminine), and cancers, even at low concentrations [11–13]. Some of
the adverse effects of graphene might be increased due to adsorbing
estrogen contaminants, and the fate and transport of estrogens in the

⁎
Corresponding author at: College of Environmental Science and Engineering, Hunan University, Changsha 410082, PR China.
E-mail address: liuyunguo_hnu@163.com (Y. Liu).

https://doi.org/10.1016/j.cej.2018.03.026
Received 28 December 2017; Received in revised form 4 March 2018; Accepted 5 March 2018
Available online 07 March 2018
1385-8947/ © 2018 Elsevier B.V. All rights reserved.
L. Jiang et al.
environment could be altered.
Consequently, it is imperative to investigate the behavior of estro-
gens by graphene nanomaterial under environmentally relevant con-
ditions, which is crucial to assess the environmental risks of this ma-
terial when adsorbed estrogen contaminants. Until now, lots of
researches have demonstrated that graphene oxide (GO) with large
specific surface areas (SSA), abundant functional groups, and strong
π–π interactions possessed high adsorption ability to estrogen con-
taminants from aqueous solution [14–19]. Our previous studies found
that the estrogen adsorption capacity by GO depended on its physico-
chemical properties (e.g., SSA, pore, and surface functional groups) and
solution chemistry (e.g., pH, background electrolyte, ionic strength, and
natural organic matter) in real water environments [20,21]. Besides,
GO showed higher or comparable adsorption abilities to carbon nano-
tubes, biochars, and activated carbons; and the effect of natural organic
matter competition on the uptake of GNs was less severe [22].
Heteroaggregation of GO with inorganic nanoparticles (INPs) is
inevitable once they are released into water environments, which is
important to thoroughly investigate the fate and transport of GO.
Recently, lots of studies have been devoted extensive efforts to de-
termine the heteroaggregation or interactions between INPs and GO.
Results indicated that the aggregation of GO would be affected by INPs
such as metal oxides (Al2O3, ZnO, TiO2, and MgO), clays (kaolinite,
montmorillonite), minerals (SiO2, layered double hydroxides, hematite,
and goethite), and metal–organic frameworks [23–29]. However, to the
best of our knowledge, very limited researches have been conducted to
examine the impacts of INPs on estrogens adsorption onto GO. To date,
only several studies illustrated that the presence of natural minerals
(e.g., goethite, kaolin, and montmorillonite) and sediments in the het-
eroaggregation of GO inhibited the adsorption of estrogen con-
taminants onto GO due to the occupation of the adsorption sites on GO
through electrostatic attraction and hydrogen bonding [18,30]. En-
gineering nanoparticles, e.g., SiO2 and Al2O3, were widely applied in
the fields of engineering materials, machinery, environmental re-
mediation, electronic engineering, and medicine [31,32]. Besides, they
are ubiquitous in almost all natural surface waters [33,34]. Previous
studies indicated that engineering nanoparticles have great impact on
the adsorption of pollutants by carbon nanotubes (CNTs) via altering
the surface chemistry and morphology of CNTs [34–36]. Similarly,
adsorption of estrogens onto GO is also likely affected by engineering
nanoparticles.
Therefore, in this work, the influence of engineering nanoparticles
(SiO2 and Al2O3) on the adsorption of estrogens on GO was in-
vestigated. One commonly estrogens in natural waters, 17β-estradiol
(E2), was selected as target estrogen pollutants. Adsorption kinetics and
isotherms of E2 by GO were conducted at different INPs/GO ratios. The
underlying influencing mechanisms were discussed according to zeta
potentials, scanning electron microscope (SEM), and Derjaguin-Landau-
Verwey-Overbeek (DLVO) calculation.
2. Materials and methods
2.1. Chemicals and materials
E2 (98%) was obtained from Sigma-Aldrich Chemical Co. (St. Louis,
MO, USA). GO was prepared through exfoliation of graphite flakes via
using the modified Hummers’ method [37–39]. Detailed preparation
method for GO is provided in the Supplementary material. SiO2
(purity > 99.9%, particle size 15 nm) was purchased from Aladdin
Bio-Chem Technology Co., LTD (Shanghai, China). Al2O3 (purity >
99%, γ-phase, particle size 20 nm) was provided by Rhawn Reagent
(Shanghai, China). NaOH, HCl, NaCl, and methanol were obtained from
Sinopharm Chemical Reagent Co., Ltd., China and were analytical re-
agent grades. Ultrapure water generated by a Milli-Q water filtration
system (Millipore, Billerica, MA) was used to all of the experiments.
372
Chemical Engineering Journal 343 (2018) 371–378
2.2. Characterization methods
The Brunauer–Emmett–Teller (BET) surface area and the pore size
distribution were examined by using N2 adsorption and desorption
(Micromeritics, Tristar II 3020) at 77 K over a relative pressure ranging
from 0.0955 to 0.993. The C, H, O, N contents of GO were obtained by
using an elemental analyzer (Vario EL III, Elementar, Germany). X-ray
photoelectron spectroscopy (XPS) was determined by using an
ESCALAB 250Xi X-ray photoelectron spectrometer (Thermo Fisher,
USA) with the scanning range of 0–1000 eV. Fourier transform infrared
spectrum (FT-IR) was investigated by a spectrophotometer (Nicolet
5700 Spectrometer) using the KBr pellet technique with the range from
400 to 4000 cm−1. The surface morphologies were examined by a SEM
(Zeiss EVO MA10, Germany) operated at 20 kV. The zeta potentials
were recorded on dynamic light scattering measurements using a
Zetasizer (Nano ZS90, Malvern Instruments Ltd., Malvern, UK).
2.3. Batch adsorption experiments
The E2 stock solutions were prepared through dissolving in me-
thanol. The desired concentrations of E2 work solution were further
obtained by dissolution of stock solutions in ultrapure water and suc-
cessive dilutions. For the adsorption experiments on GO at different
INPs/GO ratios, various amounts of INPs were added into a series of
pre-dispersed suspensions of GO in amber glass bottles to obtain INPs/
GO ratios of 0:1, 10:1, 50:1 and 100:1, respectively. The suspensions
were then placed on a rotary shaker for 24 h at 160 rpm and 25 °C to
achieve sufficient mixing. After that, E2 solutions were added into the
suspensions to obtained initial concentrations of 4 mg/L. All batch ad-
sorption experiments were conducted in pH 6.5 and 0.01 M NaCl pla-
cing in a temperature controlled water bath shaker for 24 h with a
shaking speed of 160 rpm at 25 °C. For adsorption kinetic experiments,
initial E2 solutions (4 mg/L) were taken at specific time intervals from 0
to 48 h. For adsorption isotherm experiments, the initial E2 con-
centrations were in the range of 0.2–8 mg/L. Subsequently, the sus-
pensions were centrifuged at 8000 rpm for 10 min and then the filtered
through 0.45 μm membrane filters. The E2 concentrations were mea-
sured through means of a fluorescence quenching method using a F-
4500 fluorescence spectrophotometer (Hitachi, Japan) as illustrated in
previous studies [19,20]. The adsorbed E2 quantity was calculated from
the difference between the initial and equilibrium concentrations.
2.4. Model of data analysis
The adsorption kinetic data were analyzed by the pseudo-first-order,
pseudo-second-order, and intra-particle diffusion model, which are
commonly presented as follows [40–42]:
Pseudo-first-order model: qt = qe−qe e−k1 t (1)
k2 qe2 t
Pseudo-second-order model qt =
1 + k2 qe t (2)
Intra-particle diffusion model qt = k d t 0.5 + L (3)
where qe (mg/g) and qt (mg/g) are the amount of E2 adsorbed at
equilibrium and different time, respectively; k1 (1/min), k2 (g/mg min),
and kd (mg/g·min0.5) are the rate constants of the pseudo-first-order
model, pseudo-second-order model, and intra-particle diffusion model,
respectively; L (mg/g) is the thickness of boundary layer.
Langmuir and Freundlich model were used to fit the isotherm ex-
perimental data, and their mathematical expressions are given as fol-
lows [43–45]:
KL qm Ce
Langmuir equation qe =
1 + KL Ce (4)
L. Jiang et al. Chemical Engineering Journal 343 (2018) 371–378

Freundlich equation qe = KF CeN (5)

where qm (mg/g) is the maximum adsorption amount at saturation

state; Ce (mg/L) is the equilibrium solution phase concentration; KL (L/
g) and KF [(mg/g)/(mg/L)N] are the Langmuir constant and the
Freundlich affinity coefficient, respectively; N is the exponential coef-
ficient.

2.5. DLVO calculations

In this study, DLVO theory was used to determine the interaction

energy between GO and INPs in aqueous solution. The DLVO total in-
teraction energy (Φtot) is considered as the summation of van der Waals
attraction (ΦvdW) and electrical double layer repulsion (Φedl) [46].
Details of the DLVO calculations are provided in Supplementary ma-
terial.

3. Results and discussion

3.1. Characterization of GO

BET analysis showed that the BET surface areas (SABET) and mi-
croporosity (Vmicro) of GO were 132 m2/g and 0.023 cm3/g, respec-
tively. The SABET of GO was much lower than the theoretical maximum
value of thin graphene (2630 m2/g) [47], which could be ascribed to
the incomplete exfoliation and the aggregation of graphene layers
during the preparation process. The elemental composition of GO was C
84.57%, H 0.86%, O 14.32%, and N 0.25%. FT-IR spectroscopy was
further used to analyze the functional groups on the GO surface. As
illustrated in Fig. 1a, the band around 3430 cm−1 could be ascribed to
the skeletal vibration of OeH groups [48]. The band around 1571 cm−1

Fig. 2. Adsorption kinetics of E2 onto GO in the presence of INPs.

corresponded to C]O bonds in carboxylic acid and carbonyl moieties

[49]. The band around 1024 cm−1 could be ascribed to the appearance
of alkoxy (CeO) bonds [50]. The band around 1404 cm−1 could be
attributed to the appearance of carboxyl (OeC]O) bonds [51]. Be-
sides, the surface functional groups were also investigated by XPS
(Fig. 1b). The deconvolution of the C 1s peak of GO exhibited four peaks
at 284.4, 285.3, 286.5, and 288.4 eV. This could correlate to the carbon
in C]C, CeO, C]O, and OeC]O, respectively [52–54].

3.2. Adsorption kinetics

As illustrated in Fig. 2, the adsorbed amount of E2 increased with

increase of contact time, and the adsorption equilibrium was obtained
in less than 480 min. Comparison to homogeneous system of GO, the
presence of INPs led to increasing adsorption equilibrium time. Besides,
the equilibrium time of E2 increased from 120 min to 330 min with
increase of INPs/GO from 0:1 to 100:1. This indicated that the presence
of INPs decelerated the adsorption of E2 onto GO. Furthermore, the
equilibrium time under SiO2 preloading was lower than that under
Al2O3 preloading, which suggested that the decelerated effect is higher
for Al2O3 than SiO2.
Two kinetic models, e.g., pseudo-first-order and pseudo-second-
order model (Eqs. (1) and (2)), were used to describe the experimental
data. As reflected in Table 1, the value of determination coefficients
Fig. 1. FT-IR spectra (a) and C 1s XPS spectra (b) of GO.
(R2) in pseudo-second-order model was much higher than that in the

373
L. Jiang et al. Chemical Engineering Journal 343 (2018) 371–378

Table 1
Kinetic parameters of E2 adsorption onto GO in the presence of INPs.

Adsorbent Pseudo-first-order Pseudo-second-order

qe (mg/g) k1 (1/min) R2 qe (mg/g) k2 (g/mg min) R2

SiO2-GO 0:1 47.33 0.0042 0.9801 53.25 1.01 × 10−4 0.9818

10:1 42.45 0.0033 0.9631 48.03 0.88 × 10−4 0.9662
50:1 37.73 0.0025 0.9656 44.07 0.64 × 10−4 0.9780
100:1 30.09 0.0022 0.9624 34.88 0.43 × 10−4 0.9764

Al2O3-GO 0:1 47.33 0.0042 0.9801 53.25 1.01 × 10−4 0.9818

10:1 40.39 0.0029 0.9765 46.45 0.77 × 10−4 0.9822
50:1 33.03 0.0023 0.9449 38.95 0.57 × 10−4 0.9602
100:1 25.92 0.0021 0.9677 30.50 0.33 × 10−4 0.9730

pseudo-first-order model, which indicated that the pseudo-second order
model fitted the adsorption process slight better than pseudo-first-order
model. Thus, the chemisorption might be the rate-limiting step of E2
adsorption and the reaction rate was proportional to the number of
unoccupied adsorption sites on the GO surface. In addition, the kinetic
rate constant of the pseudo-second-order model, k2, decreased with
increase of INPs/GO ratios. The low value of k2 revealed that more
adsorption sites of GO surface might be occupied or/and blocked by the
INPs with the increase of INPs/GO ratios. In other previous literatures,
Yang et al. showed that k2 for the Cu(II) adsorption onto the hetero-
geneous system of SiO2-GO also decreased with increasing SiO2-GO
ratios [55].
The intra-particle diffusion model (Eq. (3)) was used to examine the
diffusion mechanism involved in the adsorption process. As exhibited in
Fig. 3, the plots of qt versus t0.5 was found to be non-linear, suggesting
multiple processes taking place over the whole adsorption process. The
stage I (rapid adsorption) was a film diffusion attributed to the external
mass transport of E2 from the bulk solution to the adsorbent surface;
the stage II showed a gradual adsorption stage corresponding to the
intra-particle diffusion of E2 molecules from the adsorbent surface into
the adsorbent pores [56,57]. Therefore, both the film diffusion and
intra-particle diffusion were simultaneously involved in the diffusion
process. Table 2 provides the parameters of intra-particle diffusion
model for E2 adsorption onto the heterogeneous system of INPs/GO. As
known, the values of intercept L give information regarding the
boundary layer effect [58]. As reflected by Table 2, L values in Stage I
decreased with increasing INPs/GO ratios, which revealed that the
presence of INPs contributed a negative effect on the initial film dif-
fusion of E2 onto GO as a result of inhibiting E2 adsorption.

3.3. Adsorption isotherms

The adsorption isotherms of E2 onto GO in the presence of INPs

exhibited that the adsorption isotherms of E2 were well described by
both Langmuir and Freundlich models with R2 > 0.96 (Table 3). Be-
sides, the presence of INPs inhibited the E2 uptake by GO and the in-
hibitory effect was concentration-dependent as reflected by Fig. 4.
Clearly, the KF values decreased from 32.51 to 26.15 (mg/g)/(mg/L)N
with increase of Al2O3/GO ratio from 0:1 to 100:1; meanwhile, the KF
values decreased from 32.51 to 18.82 (mg/g)/(mg/L)N with increase of
SiO2/GO ratio from 0:1 to 100:1. That is to say, the inhibitory effect of
SiO2 is lower than that of Al2O3. The same phenomenon was reflected
by values of qm as showed in Table 3. Similarly, Sun et al. also observed
that the presence of INPs would lead to a decrease in adsorption ca-
pacities of estrogen contaminants on CNTs; and the inhibitory effects of
INPs on the adsorption of contaminants were exhibited to be relied on
their surface properties [36].
The inhibitory effects of INPs on E2 adsorption by GO could be
attributed to following two possibilities: (i) direct competition between
INPs with E2 for the available uptake sites on GO, and (ii) increased GO
aggregation due to INPs preloading, resulting in decreasing in the
available adsorption sites [24,25]. It is worth noting that E2 adsorption
onto GO was mainly via hydrophobic interactions; whereas the het-
eroaggregation between GO and INPs occur mainly via van der Waals
attractions and electrostatic interactions [59,60]. Thus, hetero-
aggregation might be the main factor causing the inhibitory effects of
INPs on E2 adsorption by GO. Furthermore, based on the above results
(Figs. 2 and 4), the inhibitory effect relied not only on the type of INPs,
but also the preloading amount of the INPs. Thus, the key point to
explain the mechanisms of the inhibitory effect was to determine how
the INPs impacts their interactions with GO.

Fig. 3. Intraparticle diffusion model for the E2 adsorption onto GO in the presence of
INPs.

374
L. Jiang et al. Chemical Engineering Journal 343 (2018) 371–378

Table 2
Parameters calculated from intra-particle diffusion model for E2 adsorption onto GO in the presence of INPs.

Adsorbent Stage I Stage II

kd1 (mg/g min0.5) L1 (mg/g) R2 kd2 [mg/g min0.5] L2 (mg/g) R2

SiO2-GO 0:1 1.84 −0.97 0.9951 0.17 39.64 0.7317

10:1 1.49 −0.55 0.9927 0.07 38.73 0.3436
50:1 1.05 0.24 0.9699 0.05 33.89 0.8785
100:1 0.79 0.66 0.9556 0.04 26.64 0.5687

Al2O3-GO 0:1 1.84 −1.97 0.9951 0.17 39.64 0.7317

10:1 1.41 −1.76 0.9934 0.11 34.65 0.4908
50:1 1.03 −1.21 0.9324 0.05 29.54 0.5747
100:1 0.69 −0.07 0.9713 0.02 24.37 0.5878

Table 3
Isotherm parameters of E2 adsorption onto GO in the presence of INPs.

Adsorbent Langmuir Freundlich

KL (L/g) qm (mg/g) R2 KF [(mg/ N R2

g)/(mg/
L)N]

SiO2-GO 0:1 0.309 141.89 0.9852 32.51 0.669 0.9856

10:1 0.304 131.36 0.9739 30.01 0.657 0.9751
50:1 0.342 114.45 0.9896 28.39 0.628 0.9845
100:1 0.360 101.48 0.9754 26.15 0.607 0.9610

Al2O3-GO 0:1 0.309 141.89 0.9852 32.51 0.669 0.9856

10:1 0.331 118.13 0.9803 28.74 0.636 0.9831
50:1 0.308 101.92 0.9871 23.66 0.629 0.9786
100:1 0.267 89.51 0.9812 18.82 0.603 0.9898

3.4. Influencing mechanisms of INPs

3.4.1. Zeta potential measurement

Fig. 5 illustrates graphically the measured zeta potential values of
GO and INPs at various pH values. As exhibited, both GO and SiO2 were
negatively charged in the examined pH range of 2–11. However, Al2O3
was positively charged at lower pH and negatively charged at higher
pH. Al2O3 had a point of zero charge (pHPZC) value of ∼8.8, while SiO2
had a pHPZC value of ∼2. Similar findings have been reported in the
previous literatures by several other researchers [24,26]. At the pH of
6.5 conducted in the batch adsorption experiments, the zeta potentials
of SiO2 and GO were −34.1 mV and −30.3 mV, respectively, whereas
the zeta potential of Al2O3 was 22.1 mV.
GO possesses an open-layered structure and are easily dispersed in
aqueous media due to the abundant oxygen-containing functional
groups on surface [61]. Thus, the heteroaggregation between GO and
INPs could occur due to the folding of GO’s sheets via van der Waals
force. However, when the GO nanosheets were exposed to Al2O3, the
heteroaggregation of GO could be promoted through the ligand com-
plexation and electrostatic attraction between GO and Al2O3. The li-
gand complexation occurred via an exchange between the oxygen-
containing groups of GO and the surface functional groups of Al2O3, Fig. 4. Adsorption isotherm of E2 onto GO in the presence of INPs.
which has been demonstrated by other researchers [24]. The electro-
static attraction could be attributed to positively charged Al2O3 and
the surface. In the presence of INPs (Fig. 6b and c), it could observed
negatively charged GO as reflected by their zeta potentials (Fig. 6).
that the stonelike or flowerlike INPs attached onto the surface of GO.
Thence, the heteroaggregation between GO and Al2O3 were greater
Compared with SiO2/GO heteroaggregates (Fig. 6b), more wrinkles and
than SiO2. And then, the available adsorption sites of GO then de-
larger particles of Al2O3/GO heteroaggregates (Fig. 6c and d) were
creased due to direct competition for the adsorption sites and the de-
found. As well known, the hydroxyl and epoxide groups are on the basal
crease of surface areas. This led to the stronger inhibitory effect of
plane of GO nanosheets, whereas a number of carboxylic groups locate
Al2O3 on E2 adsorption onto GO as illustrated in Fig. 2.
on its edges [62]. Carboxylic groups with pKa of 4.3 are much earlier to
be ionized than hydroxyl groups with pKa of 9.8 [63]. At the experi-
3.4.2. SEM analysis mental pH of 6.5, the Al2O3 particles would primarily associate with the
The morphologies of the INPs/GO heteroaggregates were char- edges of GO nanosheets and then the excess negatively charged GO
acterized by using SEM (Fig. 6). The SEM image of GO (Fig. 6a) ex- edges interacted with another side of Al2O3 or other Al2O3 particles in
hibited that GO sheets were not rigid with some ripples presenting on

375
L. Jiang et al.
Fig. 5. Zeta potentials of GO and INPs.
the Al2O3 aggregates. This might contribute to the larger particles of
Al2O3/GO heteroaggregates.
3.4.3. DLVO calculation
The parameters used for DLVO theory analysis are exhibited in
Table S1. The Φtot between GO/GO, INPs/INPs and INPs/GO particles
are showed in Fig. 7. Clearly, in the homogeneous system, the Φtot
between GO/GO was lower than those of between SiO2/SiO2 and
Al2O3/Al2O3. This indicated that GO should be more prone to aggregate
than SiO2 and Al2O3 due to hydrophobic forces. In heterogeneous sys-
tems of GO/INPs, the Φtot between identical particles of SiO2 and Al2O3
are much greater than those between pairs of SiO2/GO and Al2O3/GO,
respectively. This suggested that the SiO2 and Al2O3 were prone to
heteroaggregate with GO rather than to homoaggregate. Besides, pairs
of GO particles had a lower Φtot than pairs of SiO2/GO particles, which
Fig. 6. INPs/GO heteroaggregation in the binary system. (a) SEM image of GO; (b) SEM image of SiO2/GO heteroaggregate; (c) and (d) SEM image of Al2O3/GO heteroaggregate.
376
Chemical Engineering Journal 343 (2018) 371–378
Fig. 7. Interaction energies between identical (a) and different (b) GO and INPs calcu-
lated by DLVO theory (GO/INPs = 1:10, T = 25 °C, I = 0.01 M NaCl, pH = 6.5).
might indicate that the GO primarily homoaggregated and then het-
eroaggregated with SiO2 in its binary systems because of the electro-
static repulsion between same charged GO and SiO2. However, in the
heterogeneous systems of Al2O3/GO, the electrostatic attraction pro-
moted the heteroaggregation between oppositely charged GO and
Al2O3 resulting the GO and Al2O3 were more prone to heteroaggrega-
tion rather than homoaggregation.
4. Conclusions
In this work, the adsorption of E2 by GO in the presence of INPs was
examined. Results showed that the adsorption equilibrium time of E2
by GO was increased in the presence of INPs, and the E2 adsorption
decreased with increase of INPs/GO ratios because of INPs’ competition
L. Jiang et al.
to the active sites on GO surface. Besides, the inhibitory effect of Al2O3
was greater than that of SiO2. The heteroaggregation between GO and
SiO2 could occur due to the van der Waals force. However, the het-
eroaggregation between oppositely charged Al2O3 and GO could be
promoted by ligand complexation and electrostatic attraction besides
van der Waals force. SEM images revealed that heteroaggregation of
INPs and GO resulted in the much larger particle sizes for the Al2O3/GO
aggregation than SiO2/GO. DLVO calculation exhibited that GO tended
to first homoaggregate and then heteroaggregate with SiO2.
Nevertheless, the electrostatic attraction between oppositely charged
Al2O3 and GO promoted them to heteroaggregation rather than
homoaggregation. Thus, more adsorptions sites were occupied by the
Al2O3 than SiO2, resulting in the inhibitory effect of Al2O3 was higher.
This investigation revealed that INPs in the natural water environments
could interact with GO and alter the adsorption behavior of con-
taminations by GO.
Acknowledgments
This work was supported by the National Natural Science
Foundation of China (Grants 51521006 and 51609268), the Key Project
of Technological Innovation in the Field of Social Development of
Hunan Province, China (Grants 2016SK2010 and 2016SK2001), and the
Hunan Provincial Innovation Foundation for Postgraduate (Grant
CX2016B135).
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.2018.03.026.
References
[1] M.J. Allen, V.C. Tung, R.B. Kaner, Honeycomb carbon: a review of graphene, Chem.
Rev. 110 (2010) 132–145.
[2] X. Hu, H. Wang, Y. Liu, Statistical analysis of main and interaction effects on Cu (II)
and Cr (VI) decontamination by nitrogen-doped magnetic graphene oxide, Sci. Rep.
6 (2016) 34378.
[3] K.S. Novoselov, V.I. Fal, L. Colombo, P.R. Gellert, M.G. Schwab, K. Kim, A roadmap
for graphene, Nature 490 (2012) 192.
[4] T. Ramanathan, A.A. Abdala, S. Stankovich, D.A. Dikin, M. Herrera-Alonso,
R.D. Piner, et al., Functionalized graphene sheets for polymer nanocomposites, Nat.
Nanotechnol. 3 (2008) 327–331.
[5] X. Li, Y. Zhu, W. Cai, M. Borysiak, B. Han, D. Chen, et al., Transfer of large-area
graphene films for high-performance transparent conductive electrodes, Nano Lett.
9 (2009) 4359–4363.
[6] A.K. Geim, K.S. Novoselov, The rise of graphene, Nat. Mater. 6 (2007) 183–191.
[7] Y. Zhou, O.G. Apul, T. Karanfil, Adsorption of halogenated aliphatic contaminants
by graphene nanomaterials, Water Res. 79 (2015) 57–67.
[8] R. Arvidsson, S. Molander, B.A. Sandén, Review of potential environmental and
health risks of the nanomaterial graphene, Hum. Ecol. Risk Assess. 19 (2013)
873–887.
[9] X. Hu, Q. Zhou, Health and ecosystem risks of graphene, Chem. Rev. 113 (2013)
3815–3835.
[10] K. He, G. Chen, G. Zeng, M. Peng, Z. Huang, J. Shi, et al., Stability, transport and
ecosystem effects of graphene in water and soil environments, Nanoscale 9 (2017)
5370–5388.
[11] A. Tabata, S. Kashiwada, Y. Ohnishi, H. Ishikawa, N. Miyamoto, M. Itoh, et al.,
Estrogenic influences of estradiol-17β, p-nonylphenol and bis-phenol-A on Japanese
medaka (Oryzias latipes) at detected environmental concentrations, Water Sci.
Technol. 43 (2001) 109–116.
[12] E. Diamantikandarakis, J.P. Bourguignon, L.C. Giudice, R. Hauser, G.S. Prins,
A.M. Soto, et al., Endocrine-disrupting chemicals: an endocrine society scientific
statement, Endocr. Rev. 30 (2009) 293–342.
[13] B. Han, M. Zhang, D. Zhao, Y. Feng, Degradation of aqueous and soil-sorbed es-
tradiol using a new class of stabilized manganese oxide nanoparticles, Water Res. 70
(2015) 288–299.
[14] X. Bai, R. Feng, Z. Hua, L. Zhou, H. Shi, Adsorption of 17β-Estradiol (E2) and Pb(II)
on Fe3O4/graphene oxide (Fe3O4/GO) nanocomposites, Environ. Eng. Sci. 32
(2015) 370–378.
[15] X. Bai, C. Qin, R. Feng, Binary adsorption of 17β-estradiol and bisphenol A on su-
perparamagnetic amino-functionalized graphene oxide nanocomposites, Mater.
Chem. Phys. 189 (2017) 96–104.
[16] L.K. Boateng, J. Heo, J.R.V. Flora, Y.-G. Park, Y. Yoon, Molecular level simulation of
the adsorption of bisphenol A and 17α-ethinyl estradiol onto carbon nanomaterials,
377
Chemical Engineering Journal 343 (2018) 371–378
Sep. Purif. Technol. 116 (2013) 471–478.
[17] Y. Wen, Z. Niu, Y. Ma, J. Ma, L. Chen, Graphene oxide-based microspheres for the
dispersive solid-phase extraction of non-steroidal estrogens from water samples, J.
Chromatogr. A. 1368 (2014) 18–25.
[18] W. Sun, M. Li, W. Zhang, J. Wei, B. Chen, C. Wang, Sediments inhibit adsorption of
17β-estradiol and 17α-ethinylestradiol to carbon nanotubes and graphene oxide,
Environ. Sci. Nano 4 (2017) 1900–1910.
[19] L. Jiang, Y. Liu, S. Liu, X. Hu, G. Zeng, X. Hu, et al., Fabrication of β-cyclodextrin/
poly (L-glutamic acid) supported magnetic graphene oxide and its adsorption be-
havior for 17β-estradiol, Chem. Eng. J. 308 (2017) 597–605.
[20] L. Jiang, Y. Liu, G. Zeng, F. Xiao, X. Hu, X. Hu, et al., Removal of 17β-estradiol by
few-layered graphene oxide nanosheets from aqueous solutions: external influence
and adsorption mechanism, Chem. Eng. J. 284 (2016) 93–102.
[21] L. Jiang, Y. Liu, G. Zeng, S. Liu, X. Hu, L. Zhou, et al., Adsorption of estrogen
contaminants (17β-estradiol and 17α-ethynylestradiol) by graphene nanosheets
from water: effects of graphene characteristics and solution chemistry, Chem. Eng.
J. (2017), http://dx.doi.org/10.1016/j.cej.2017.12.034.
[22] L. Jiang, Y. Liu, S. Liu, G. Zeng, X. Hu, X. Hu, et al., Adsorption of estrogen con-
taminants by graphene nanomaterials under natural organic matter preloading:
comparison to carbon nanotube, biochar, and activated carbon, Environ. Sci.
Technol. 51 (2017) 6352–6359.
[23] S. Yu, X. Wang, R. Zhang, T. Yang, Y. Ai, T. Wen, et al., Complex roles of solution
chemistry on graphene oxide coagulation onto titanium dioxide: batch experiments,
spectroscopy analysis and theoretical calculation, Sci. Rep. 7 (2017) 39625.
[24] X. Ren, J. Li, X. Tan, W. Shi, C. Chen, D. Shao, et al., Impact of Al2O3 on the
aggregation and deposition of graphene oxide, Environ. Sci. Technol. 48 (2014)
5493–5500.
[25] N.P. Sotirelis, C.V. Chrysikopoulos, Interaction between graphene oxide nano-
particles and quartz sand, Environ. Sci. Technol. 49 (2015) 13413–13421.
[26] J. Zhao, F. Liu, Z. Wang, X. Cao, B. Xing, Heteroaggregation of graphene oxide with
minerals in aqueous phase, Environ. Sci. Technol. 49 (2015) 2849–2857.
[27] Y. Feng, X. Liu, K.A. Huynh, J.M. Mccaffery, M. Liang, S. Gao, et al.,
Heteroaggregation of graphene oxide with nanometer- and micrometer-sized he-
matite colloids: influence on nanohybrid aggregation and microparticle sedi-
mentation, Environ. Sci. Technol. 51 (2017) 6821–6828.
[28] J. Li, Q. Wu, X. Wang, Z. Chai, W. Shi, J. Hou, et al., Heteroaggregation behavior of
graphene oxide on Zr-based metal–organic frameworks in aqueous solutions: a
combined experimental and theoretical study, J. Mater. Chem. A. 5 (2017)
20398–20406.
[29] J. Wang, S. Yu, Y. Zhao, X. Wang, T. Wen, T. Yang, et al., Experimental and the-
oretical studies of ZnO and MgO for the rapid coagulation of graphene oxide from
aqueous solutions, Sep. Purif. Technol. 184 (2017) 88–96.
[30] W. Sun, C. Wang, W. Pan, S. Li, B. Chen, Effects of natural minerals on the ad-
sorption of 17β-estradiol and bisphenol A on graphene oxide and reduced graphene
oxide, Environ. Sci. Nano 4 (2017) 1377–1388.
[31] H.F. Krug, P. Klug, Impact of nanotechnological developments on the environment,
Nanotechnology (2008).
[32] H.-E. Schaefer, Nanoscience: The Science of the Small in Physics, Engineering,
Chemistry, Biology and Medicine, Springer Science & Business Media, 2010.
[33] M. Gan, J. Li, S. Sun, Y. Cao, Z. Zheng, J. Zhu, et al., The enhanced effect of
Acidithiobacillus ferrooxidans on pyrite based Cr (VI) reduction, Chem. Eng. J. 341
(2018) 27–36.
[34] B. Chen, W. Sun, C. Wang, X. Guo, Size-dependent impact of inorganic nano-
particles on sulfamethoxazole adsorption by carbon nanotubes, Chem. Eng. J. 316
(2017) 160–170.
[35] W. Shi, S. Li, B. Chen, C. Wang, W. Sun, Effects of Fe2O3 and ZnO nanoparticles on
17β-estradiol adsorption to carbon nanotubes, Chem. Eng. J. 326 (2017)
1134–1144.
[36] W. Sun, C. Zhang, N. Xu, J. Ni, Effect of inorganic nanoparticles on 17β-estradiol
and 17α-ethynylestradiol adsorption by multi-walled carbon nanotubes, Environ.
Pollut. 205 (2015) 111–120.
[37] X. Hu, Y. Liu, H. Wang, G. Zeng, X. Hu, Y. Guo, et al., Adsorption of copper by
magnetic graphene oxide-supported β-cyclodextrin: effects of pH, ionic strength,
background electrolytes, and citric acid, Chem. Eng. Res. Des. 93 (2015) 675–683.
[38] M. Li, Y. Liu, S. Liu, D. Shu, G. Zeng, X. Hu, et al., Cu (II)-influenced adsorption of
ciprofloxacin from aqueous solutions by magnetic graphene oxide/nitrilotriacetic
acid nanocomposite: competition and enhancement mechanisms, Chem. Eng. J. 319
(2017) 219–228.
[39] M. Li, Y. Liu, G. Zeng, S. Liu, X. Hu, D. Shu, et al., Tetracycline absorbed onto
nitrilotriacetic acid-functionalized magnetic graphene oxide: influencing factors
and uptake mechanism, J. Colloid Interface Sci. 485 (2017) 269–279.
[40] Z. Yan, Y. Liu, X. Tan, S. Liu, G. Zeng, L. Jiang, et al., Immobilization of aqueous and
sediment-sorbed ciprofloxacin by stabilized Fe-Mn binary oxide nanoparticles: in-
fluencing factors and reaction mechanisms, Chem. Eng. J. 314 (2017) 612–621.
[41] Y.Y. Zhou, X.C. Liu, Y.J. Xiang, P. Wang, J.C. Zhang, F.F. Zhang, et al., Modification
of biochar derived from sawdust and its application in removal of tetracycline and
copper from aqueous solution: adsorption mechanism and modelling, Bioresour.
Technol. 245 (2017) 266–273.
[42] W. Zeng, Y. Liu, X. Hu, S. Liu, G. Zeng, B. Zheng, et al., Decontamination of me-
thylene blue from aqueous solution by magnetic chitosan lignosulfonate grafted
with graphene oxide: effects of environmental conditions and surfactant, RSC Adv.
6 (2016) 19298–19307.
[43] B. Huang, Y. Liu, B. Li, G. Zeng, X. Hu, B. Zheng, et al., Synthesis of graphene oxide
decorated with core@double-shell nanoparticles and application for Cr (VI) re-
moval, RSC Adv. 5 (2015) 106339–106349.
[44] S. Zhu, Y. Liu, S. Liu, G. Zeng, L. Jiang, X. Tan, et al., Adsorption of emerging
L. Jiang et al.
contaminant metformin using graphene oxide, Chemosphere. 179 (2017) 20–28.
[45] L. Jiang, S. Liu, Y. Liu, G. Zeng, Y. Guo, Y. Yin, et al., Enhanced adsorption of
hexavalent chromium by a biochar derived from ramie biomass (Boehmeria nivea
(L.) Gaud.) modified with β-cyclodextrin/poly(L-glutamic acid), Environ. Sci.
Pollut. Res. (2017) 1–10.
[46] M. Elimelech, J. Gregory, X. Jia, Particle Deposition and Aggregation:
Measurement, Modelling and Simulation, Butterworth-Heinemann, 2013.
[47] T. Kim, G. Jung, S. Yoo, K.S. Suh, R.S. Ruoff, Activated graphene-based carbons as
supercapacitor electrodes with macro-and mesopores, ACS Nano 7 (2013)
6899–6905.
[48] W. Sajomsang, S. Tantayanon, V. Tangpasuthadol, W.H. Daly, Quaternization of N-
aryl chitosan derivatives: synthesis, characterization, and antibacterial activity,
Carbohydr. Res. 344 (2009) 2502–2511.
[49] M.A. Bharudin, S. Zakaria, C.H. Chia, Condensed tannins from Acacia mangium
bark: characterization by spot tests and FTIR, AIP Conf. Proc. 1571 (2013) 153–157.
[50] Z. Movasaghi, S. Rehman, D.I. ur Rehman, Fourier transform infrared (FTIR)
spectroscopy of biological tissues, Appl. Spectrosc. Rev. 43 (2008) 134–179.
[51] B. Grabowska, Examining the thermal degradation of polymer binders using FTIR
spectroscopy, Arch. Foundry Eng. 10 (2010) 43–46.
[52] M. Li, J.P. Cheng, F. Liu, X.B. Zhang, 3D-architectured nickel–cobalt–manganese
layered double hydroxide/reduced graphene oxide composite for high-performance
supercapacitor, Chem. Phys. Lett. 640 (2015) 5–10.
[53] M. Wang, C. Wang, Y. Song, H. Xie, Y. Huang, One-pot in situ polymerization of
graphene oxide nanosheets and poly (p-phenylenebenzobisoxazole) with enhanced
mechanical and thermal properties, Compos. Sci. Technol. 141 (2017) 16–23.
[54] Z. Luo, S. Lim, Z. Tian, J. Shang, L. Lai, B. MacDonald, et al., Pyridinic N doped
graphene: synthesis, electronic structure, and electrocatalytic property, J. Mater.
Chem. 21 (2011) 8038–8044.
378
Chemical Engineering Journal 343 (2018) 371–378
[55] L. Yang, J. Hu, W. Wu, J. Tang, K. Ding, J. Li, In situ NH2-functionalized graphene
oxide/SiO2 composites to improve Cu(II) removal from ammoniacal solutions,
Chem. Eng. J. 306 (2016) 77–85.
[56] S. Dawood, T.K. Sen, Removal of anionic dye Congo red from aqueous solution by
raw pine and acid-treated pine cone powder as adsorbent: equilibrium, thermo-
dynamic, kinetics, mechanism and process design, Water Res. 46 (2012)
4316–4317.
[57] Z. Wu, H. Zhong, X. Yuan, H. Wang, L. Wang, X. Chen, et al., Adsorptive removal of
methylene blue by rhamnolipid-functionalized graphene oxide from wastewater,
Water Res. 67 (2014) 330–344.
[58] V. Vimonses, S. Lei, J. Bo, C.W.K. Chow, C. Saint, Kinetic study and equilibrium
isotherm analysis of Congo Red adsorption by clay materials, Chem. Eng. J. 148
(2009) 354–364.
[59] L. Wu, L. Liu, B. Gao, R. Muñozcarpena, M. Zhang, H. Chen, et al., Aggregation
kinetics of graphene oxides in aqueous solutions: experiments, mechanisms, and
modeling, Langmuir ACS J. Surf. Colloids 29 (2013) 15174–15181.
[60] Y. Zou, X. Wang, Y. Ai, Y. Liu, J. Li, Y. Ji, et al., Coagulation behavior of graphene
oxide on nanocrystallined Mg/Al layered double hydroxides: batch experimental
and theoretical calculation study, Environ. Sci. Technol. 50 (2016) 3658.
[61] J. Wang, Z. Chen, B. Chen, Adsorption of polycyclic aromatic hydrocarbons by
graphene and graphene oxide nanosheets, Environ. Sci. Technol. 48 (2014)
4817–4825.
[62] O.C. Compton, S.B.T. Nguyen, Graphene oxide, highly reduced graphene oxide, and
graphene: versatile building blocks for carbon-based materials, Small 6 (2010)
711–723.
[63] B. Konkena, S. Vasudevan, Understanding aqueous dispersibility of graphene oxide
and reduced graphene oxide through pKa measurements, J. Phys. Chem. Lett. 3
(2012) 867.