Science of the Total Environment 651 (2019) 580–590

Contents lists available at ScienceDirect

Science of the Total Environment

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

The residual tetracycline in pharmaceutical wastewater was effectively

removed by using MnO2/graphene nanocomposite
Zhi Song, Yu-Long Ma ⁎, Cong-Er Li
State Key Laboratory of High-efficiency Coal Utilization and Green Chemical Engineering, College of Chemistry and Chemical Engineering, Ningxia University, Yinchuan 750021, 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

• MnO2/graphene nanocomposite with

highly selective adsorption for TC was
prepared.
• MnO2/graphene nanocomposite has ex-
cellent water solubility.
• The nanocomposite could effectively re-
move 99.4% of TC from wastewater.
• MnO2/graphene can be used for phar-
maceutical wastewater treatment.

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

Article history: The object of this study was to remove the tetracycline (TC) residue in pharmaceutical wastewater after floccu-
Received 27 July 2018 lation treatment. MnO2/graphene nanocomposite was synthesized by an in situ hydrothermal method and its TC
Received in revised form 18 September 2018 removal rate was up to 99.4%. This nanocomposite had excellent water solubility. More importantly, the intro-
Accepted 18 September 2018
duction of MnO2 nanorods allowed the avoidance of excessive stacking of treated graphene sheets during the ad-
Available online 19 September 2018
sorption process, which made the TC molecules to have more opportunities to make contact with the adsorbents.
Editor: Paola Verlicchi In order to eliminate the interference factors, the adsorption isotherm, kinetics, thermodynamics and mechanism
were all studied in TC aqueous solution. The influence of solution pH, contact time, MnO2 loading amount, tem-
Keywords: perature and solution concentration on the adsorption process were also assessed. The main adsorption mecha-
MnO2/graphene nism contributed to the complexation of Mn(IV) and π-π interactions of the benzene ring structure on treated
Adsorption graphene sheets with TC molecules.
Tetracycline © 2018 Elsevier B.V. All rights reserved.
Removal efficiency

1. Introduction
Since the advent of penicillin, great developments in antibiotics have
been achieved, where pharmaceutical antibiotics have been used
worldwide in the treatment of human and animal diseases (Yu et al.,
⁎ Corresponding author at: State Key Laboratory of High-efficiency Utilization of Coal
and Green Chemical Engineering, College of Chemistry and Chemical Engineering,
Ningxia University, Helanshan Rd. 539, Yinchuan 750021, China.
E-mail address: yulongma796@sohu.com (Y.-L. Ma).
2017; Wang et al., 2018; Aghagoli and Shemirani, 2017). Tetracycline
(TC), the second most widely used antibiotics worldwide, has been ex-
tensively used as human medicine, as a veterinary drug, and as a growth
promoter in animal cultivation, with little research carried out over the
past few decades (Huang et al., 2017; Rattanachueskul et al., 2017). TC is
poorly metabolized and absorbed by humans and animals, thus large
fractions are excreted via urine and faeces (He et al., 2017). Therefore,
TC has been frequently detected in soil, surface waters and even drink-
ing water. However, it is very difficult to remove TC from water because
of its variable state, low biodegradability and complex molecular

https://doi.org/10.1016/j.scitotenv.2018.09.240
0048-9697/© 2018 Elsevier B.V. All rights reserved.
 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590 581

structure (Xiong et al., 2018). TC residues in the environment have Table 1

caused a variety of potential adverse effects such as chronic toxicity Water quality parameters of pharmaceutical wastewater and pharmaceutical wastewater
after flocculation treatment.
and dissemination into antibiotic-resistant genes (Zhu et al., 2018).
More importantly, the existence of TC residues can affect drinking Pharmaceutical wastewater Pharmaceutical wastewater after
water quality and create a potential risk for ecosystem and human flocculation treatment

health in the long term (Zhang et al., 2018; Yang et al., 2017; Thi and Parameters Contents Parameters Contents
Lee, 2017). Thus, it is of great importance to develop effective and pH 5 pH 5
low-cost technologies for removal of TC from wastewater (Zhuang COD 2067 mg COD 2003 mg
et al., 2017; Chao et al., 2017; Z. Li et al., 2017; Ahmed et al., 2017). L−1 L−1
At present, there are many methods for the removal of TC, including TOC 922 mg L−1 TOC 894 mg L−1
Protein 0.88 mg L−1 Protein –
adsorption (Zhang et al., 2015; Li et al., 2016), biodegradation (Ahmed Oxalate 4.5 mg Oxalate –
et al., 2015), photodegradation (Liu et al., 2016; Chen and Liu, 2016; mL−1
Norvill et al., 2017; Chen et al., 2017a, 2017b), and oxidative degrada- Amino nitrogen 1.35 mg L−1 Amino nitrogen –
tion (Ziółkowska et al., 2016; Khan et al., 2010). Of these, adsorption is Reducing sugar 1.6 mg L−1 Reducing sugar –
Total sugar 3.2 mg L−1 Total sugar –
a widely used effective technology, due to the advantages of easy oper-
Tetracycline 200 mg L−1 Tetracycline 180 mg L−1
ation, low-cost, high efficiency, and no risk of highly toxic byproduct hydrochloride hydrochloride
(Yu et al., 2017). There are many different adsorbents used to adsorb Pigment Uncertain Pigment Uncertain
and remove TC, including aluminium oxide (Guo et al., 2017; Chen Inorganic ion Uncertain Inorganic ion Uncertain
and Huang, 2010), clays (Akhtar and Amin, 2015), activated carbons Microorganism Uncertain Microorganism Uncertain

(Martins et al., 2015; Marzbali et al., 2016), carbon nanotubes (Zhang

et al., 2011; Khan et al., 2017), and biochar (Zhou et al., 2017; H. Li
et al., 2017). However, problems such as limited adsorption ability and
high cost are associated with these adsorbents. Hence, development of
an adsorbent with a high adsorption capacity, and low production
cost, that is environmentally friendly, should be conducted and studied
urgently (M. Li et al., 2017).
More recently, nanomaterials show great promise in adsorbing anti-
biotics, due to the large surface area and controllable surface functional-
ities (Yu et al., 2017). Among them, graphene is a two-dimensional
honeycomb lattice structure formed by the hybridization of a single
layer of carbon atoms and possesses many excellent properties such
as a huge surface area, flexibility, and good chemical stability. Based
on the above advantages, graphene has great application potential in
the adsorption field. On the one hand, under certain pH conditions,
the metal oxide can form a strong coordination with the dimethylamino
group in TC to achieve the purpose of adsorbing TC (Lv et al., 2015; Song
et al., 2018). On the other hand, graphene composites combined with
other materials can be applied in the adsorption of various pollutants
from wastewater. In the work of Zhao et al. (2015), a TiO2-graphene
sponge (TiO2-GS) was prepared by the deposition of amorphous TiO2
on graphene oxide (GO) sheets for the adsorption of tetracycline antibi-
otics, where lyophilization was adopted to obtain the porous structure.
Zhang et al. (2017) prepared a Fe3O4@graphene magnetic nanocompos-
ite via an in situ precipitation method for the removal of oxytetracycline
(OTC) and tetracycline (TC) from aqueous solution. The above results
suggest that the prepared composite can be deemed as a promising ma-
terial for the removal of TC antibiotics from aqueous solution. However,
further research is needed in the study of pharmaceutical wastewater
containing TC.
Herein, the MnO2/graphene nanocomposite was synthesized via an
in situ hydrothermal method. Adsorption behaviour and mechanism
of materials on TC were studied in detail. The adsorption kinetics, iso-
therm, thermodynamics, and the impacts, such as, the MnO2 loading
amount, the solution pH, contact time and so on, were also studied.
Based on the above experiments, this material was applied to remove
TC residues in pharmaceutical wastewater after flocculation treatment.
2. Experimental
2.1. Materials
potassium permanganate (AR, 98%), polyaluminium chloride (AR,
98%) and polyferric sulfate (AR, 98%) were all purchased from Damao
chemical reagent factory in Tianjin, China. Sodium hypochlorite (AR,
14%) was purchased from Aladdin Industrial Corporation. Graphene
(CP) was purchased from TuLingJiHuaKeJi Co., LTD in Shenzhen, China.
2.2. Preparation of MnO2/graphene
The graphene (1 g) was placed in a beaker. Sodium hypochlorite
(AR, 14%, graphene: sodium hypochlorite = 1 g: 120 mL) was then
added to the beaker and thoroughly mixed. The mixture was stirred at
50 °C water bath for 24 h. The above product was washed with deion-
ized water until the effluent was colourless and was then dried at 80
°C for 12 h. The pretreated graphene sample (3.5 g) was ultrasonically
dispersed in deionized water (350 mL) for 2 h. Various amounts of
KMnO4 (i.e., 0.88 g, 2.33 g and 4.74 g) were added to the above mixture
and stirred rapidly for 2 h. The mixture was transferred to a Teflon-lined
stainless steel autoclave and reacted at 130 °C for 15 h. It was cooled to
room temperature after the reaction was completed. The product was
rinsed to neutral pH with deionized water and dried at 100 °C for
12 h. Finally, MnO2/graphene powder was obtained (treated graphene
is the abbreviation for sodium hypochlorite treated graphene in this
paper; 40%MnO2/graphene is the abbreviation of 40%MnO2/sodium hy-
pochlorite treated graphene in this paper; 40% refers to the loading
amount of MnO2).
2.3. Flocculation treatment of pharmaceutical wastewater
The flocculation process can effectively remove impurities in phar-
maceutical wastewater and keep TC concentrations relatively stable.
In this way, the adsorption mechanism of TC on MnO2/graphene can
be better studied. The flocculation treatment was carried out according
to the wastewater treatment process provided by the pharmaceutical
factory: The dosage of polyaluminium chloride (PAC) and polymeric
ferric sulfate (PFS) was 0.175 g L−1. The above two drugs were added
to pharmaceutical wastewater and mixed. The rotation speed was
50 r min−1. Both coagulation and settling time were 50 min. Finally,
the pharmaceutical wastewater after flocculation treatment was ob-
tained (pH = 5.0, C0 = 180 mg L−1).

Tetracycline hydrochloride (AR, 98%) was purchased from 2.4. Characterization

QiYuanYaoYe Co., LTD in Ningxia, China. Pharmaceutical wastewater
(pH = 5, C0 = 200 mg L−1) was provided by QiYuanYaoYe production SEM, EDX (Hitachi S-4800) and TEM (FEI Tecnai G20) analyses were
line. We have obtained relevant water quality parameters from phar- used to investigate the particle size, morphology and microstructure of
maceutical companies (Table 1). NaOH (AR, 96%), HCl (AR, 36%), the samples. The adsorption mechanism was studied by FT-IR (Bruker
582 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590

Tensor 27), and the adsorption mechanism was further supplemented
by XPS tests (kratos-ultra DLD). Thermogravimetric (TG) analysis was
carried out on the calcined samples, in air temperatures of up to 1000
°C, using a French SETARAM SETSYS-1750 instrument with a heating
rate of 10 °C min−1. The Raman spectrum (Labram-010, JY, FAR) was
also measured for different samples. Zeta potential measurements
using a zeta analyser (90 Plus Zeta, Brookhaven) were performed with
various pH values. X-ray diffraction (XRD) patterns were collected on
a Rigaku company D/MAX2200 powder diffractometer using a Ni-
filtered Cu Kα radiation source at 40 kV and 30 mA and a Braun position
sensitive detector at a scan rate of 4° min−1 and a step size of 0.02°. N2
adsorption-desorption measurements were performed at 77 K on a
Micromeritics ASAP 2020 instrument. The samples were all out gassed
at 200 °C for 12 h before measurement.
2.5. TC concentration determination
HPLC (CTO-10ASVP) was used to determine the TC concentration.
The detection wavelength of TC was 356 nm. The instrument parame-
ters are as follows: Chromatographic column: HC-C18 (InertSustain,
4.6 × 150 mm, 5 μm). Column temperature: 35 °C. Injection volume:
20 μL. Mobile phase: organic phase (acetonitrile: methanol = 2:1, v:
v), aqueous phase (0.01 mol L−1 oxalic acid). Flow rate: 0.8 mL min−1
(organic phase: 0.28 mL min−1, aqueous phase: 0.56 mL min−1). The

Fig. 1. (a) Raman spectrum of different samples; (b) FTIR spectra of graphene, treated graphene (treated graphene is the abbreviation of sodium hypochlorite treated graphene in this
paper) and 40%MnO2/graphene (40%MnO2/graphene is the abbreviation of 40%MnO2/sodium hypochlorite treated graphene, 40% refers to the loading amount of MnO2); (c) TG
curves of different samples; (d) XRD patterns of different samples; (e) N2 adsorption-desorption isotherms of different samples; (f)–(k) C1s XPS spectra of different samples.
 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590 583

measured concentration was the residual concentration of TC after Langmuir model:

adsorption.
qe ¼ qmax K L C e =ð1 þ K L C e Þ ð2Þ
2.6. Adsorption experiments of MnO2/graphene
Freundlich model:
A designed concentration of TC aqueous solution was added to a
conical bottle with a capacity of 100 mL. The pH value of TC solution qe ¼ k f C e 1=n ð3Þ
was adjusted by sodium hydroxide or hydrochloric acid aqueous solu-
tion. MnO2/graphene (50 mg) was added to the mixture for the static The linear forms of these models are:
adsorption experiment. The conical bottle was stirred (120 rpm) contin-
uously at different temperatures. The TC concentration in the solution 1=qe ¼ 1=ðqmax K L Þ  1=C e þ 1=qmax ð4Þ
was measured by HPLC at fixed time points. The adsorption capacity
lgðqe Þ ¼ lgðk f Þ þ 1=n  lgðC e Þ ð5Þ
was calculated using Eq. (1) as follows:
where qe (mg g−1) is the equilibrium adsorption capacity. qmax
qe ¼ ðC 0 −C e Þ  V=M ð1Þ (mg g−1) is the maximum adsorption capacity. KL is the Langmuir ad-
sorption equilibrium constant and Ce (mg L−1) is the equilibrium solu-
tion concentration. kf is the Freundlich constant. 1/n is the adsorption
where qe (mg g−1) is the equilibrium adsorption capacity of MnO2/ intensity.
graphene, C0 is the initial concentration, Ce (mg L−1) is the equilibrium
concentration; V (L) is the solution volume, and M (g) is the mass of 2.8. Adsorption kinetics of MnO2/graphene
adsorbent.
The pseudo-first-order model is represented by:
2.7. Adsorption isotherms of MnO2/graphene
ln ðqe −qti Þ ¼ ln ðqe Þ−k1 t ð6Þ
At different temperatures, static adsorption tests were carried out
using the static adsorption method described in Section 2.6. To under- The pseudo-second-order model is represented by:
stand the relationship between adsorbent and adsorbate, the adsorp-

tion data was fitted by the Langmuir and Freundlich models: t=qti ¼ 1= k2 qe 2 þ t=qe ð7Þ

Fig. 2. (a) Photographs of water solubility changes for different samples with different time (here, MnO2/sodium hypochlorite treated graphene refers to 40%MnO2/graphene).
584 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590

Fig. 3. (a)–(c) SEM images of MnO2/graphene with different MnO2 loading amount; (a1)–(c1) EDS spectrum of major elements on the MnO2/graphene surface; (d)–(f) EDS mapping
images of major elements on the 40%MnO2/graphene surface.

Fig. 4. TEM images of different samples.

 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590
Fig. 5. (a) Adsorption capacity of treated graphene in TC aqueous solution with different contact time at different pH; (b) adsorption capacity of MnO2/graphene with different MnO2
loading amount in TC aqueous solution (pH = 5); (c) adsorption rate of TC in aqueous solution by 40%MnO2/graphene at different temperatures with different concentrations;
(d) adsorption rate of 40%MnO2/graphene in different solution; (e) TG curve of 40%MnO2/graphene after adsorption of TC; (f) regeneration performance of 40%MnO2/graphene for
pharmaceutical wastewater after flocculation treatment.
where k1 (min−1) is the first-order rate constant; k2 (mg g−1 min−1) is
the second-order rate constant. qe is the equilibrium adsorption capac-
ity. qti (mg g−1) is the amount of TC adsorbed at time t (min).
2.9. Adsorption thermodynamics of MnO2/graphene
The thermodynamic equilibrium constant K for the adsorption was
calculated from the intercept determined by plotting ln(qe/Ce) vs. qe,
using the method of Khan and Singh. The standard free energy change
ΔG was calculated using the equation:
ΔG ¼ −R T ln K
where ΔG is the Gibbs energy change, R is the molar gas constant, and T
is the absolute temperature.
The enthalpy change ΔH and entropy change ΔS were calculated by
according to the follow equation:
ln K ¼ ΔH=ð−RT Þ þ ΔS=R
3. Results and discussion
3.1. Characterization of materials
Raman characterization results are shown in Fig. 1(a); for the MnO2/
graphene samples, the peak at 1355 cm−1 and 1598 cm−1
corresponded to the D band and G band, respectively. The increased in-
tensity ratio of ID/IG from graphene (1.27) to 60%MnO2/graphene (1.52)
indicated that some oxygen-containing functional groups were intro-
duced into the material after the modification process.
Fig. 1(b) shows the FTIR spectra of graphene, treated graphene and
40%MnO2/graphene composite powders. The peaks at 1746 cm−1,
1335 cm−1, and 1026 cm−1 are attributed to stretching of the bonds
of C_O, C\\O\\H, and C\\O, respectively. These peaks increased in
the spectrum after the modification process. This result coincides with
585
the data from Raman characterization. The introduction of a great quan-
tity of hydrophilic oxygen-containing functional groups significantly
improved the water solubility of the samples. As shown in Fig. 2, an
equal amount of the three samples (graphene, treated graphene, and
40%MnO2/graphene) were uniformly dispersed by ultrasound in an
equal volume of deionized water. Compared with graphene, the other
two samples still maintained good water solubility with the extension
of time. This would facilitate the adsorption of TC in solution by
materials.
The TGA profiles of all samples were shown in Fig. 1(c). Detailed re-
sults indicated that the GO were quite stable at temperatures below 320
ð8Þ °C. There was a two-stage process in the temperature-responsive mass
evolution: the adsorbed/hydrated water was lost at temperatures up
to 600 °C for graphene, 315 °C for treated graphene, 320 °C for 20%
MnO2/graphene, 330 °C for 40%MnO2/graphene and 360 °C for 60%
MnO2/graphene; and then the rupture of chains, fragments and mono-
layers of graphene and treated graphene occurred. Using 40%MnO2/
graphene as an example, the TGA profiles exhibited a higher thermal
ð9Þ stability than the treated graphene due to the introduction of MnO2.
This material showed an initial weight loss below 330 °C (13%), which
was attributed to the loss of adsorbed/hydrated water. The second
weight loss (330–530 °C, 55%) showed the collapse of the treated
graphene structure in 40%MnO2/graphene. When the temperature
was between 530 and 1000 °C, there was little change in the TG curve.
In summary, pretreatment with sodium hypochlorite could increase a
certain amount of oxygen-containing functional groups in graphene
and greatly reduce its thermal stability. Furthermore, MnO2 loading on
Table 2
Adsorption kinetics fitted parameters of 40%MnO2/graphene.
C0 (mg L−1) Pseudo-first-order Pseudo-second-order
K1 R2 qe K2 × 10−4 R2
(min−1) (mg g−1) (g mg−1 min−1)
200 0.0025 0.9409 195.31 4.82 0.9998
586 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590

Table 3 C_O, and O\\C_O groups species, respectively. Therefore, the 40%
Adsorption isotherms fitted parameters of 40%MnO2/graphene. MnO2/graphene was highly hydrophilic due to a certain amount of
T (K) Langmuir Freundlich oxygen-containing functional groups that were introduced in it, which
qmax (mg g−1) KL (L mg−1) R2 kf n R2
allowed it to be better dispersed in deionized water.
The SEM images of MnO2/graphene with different MnO2 loading
298 1789 0.014 0.9594 25.33 1.21 0.9171
amount are illustrated in Fig. 3(a)–(c). It was apparent that some wrin-
308 1796 0.024 0.9791 59.33 1.13 0.8866
318 1974 0.022 0.9902 116.97 1.62 0.9379 kles appeared on the surface of the MnO2/graphene and formed a scroll
of flexible sheets. To further confirm the differences between the sam-
ples, TEM technology was used to characterize the morphology of the
the treated graphene enhanced its thermal stability, as certified by a rise tested materials. The results are given in Fig. 4(a)–(e). Treatment with
in decomposition temperature (500–530 °C). sodium hypochlorite did not obviously change the microtopography
Comparison of XRD patterns among the graphene, treated graphene, of graphene. When MnO2 was introduced into treated graphene,
and 40%MnO2/graphene indicated that the tested 40%MnO2/graphene many MnO2 nanorods were firmly loaded on both sides of the sheets.
had a good crystallinity, and proved that the composite material had On the one hand, these sheets might hinder the MnO2 nanorods from
been synthesized successfully (Fig. 1(d)). N2 adsorption-desorption aggregation. On the other hand, the MnO2 nanorods could improve
measurements were conducted to examine the porous nature of the the stability of the treated graphene. Moreover, as shown in Fig. 4(a)–
tested adsorbents. Fig. 1(e) shows the N2 adsorption-desorption iso- (e), with the increase of the MnO2 loading amount, the number of
therm of different samples. The isotherms were identified as type-II, MnO2 nanorods on the graphene surface increased.
which corresponds to the reversible adsorption process of multi- The elemental composition of MnO2/graphene was characterized by
molecular layers occurring on the surface of non-porous solids. The in- EDS and EDS mapping (Fig. 3(a1)–(c1) and (d)–(f)). EDS analysis indi-
flection point occurs at P/P0 = 0.02, which is the first steep part of the cated the presence of only C, O and Mn. The mapping results confirmed
isotherm. It not only indicates the saturated adsorption amount of the the presence and uniform distribution of C, O and Mn on the surface of
monolayer, but also indicates the completion of the monolayer adsorp- 40%MnO2/graphene.
tion. The BET specific surface area of the sample was calculated from N2
isotherms at −196.6 °C. The BET specific surface areas of graphene, 3.2. Adsorption performance of MnO2/graphene
treated graphene and 40%MnO2/graphene were 139, 185 and
106 m2 g−1, respectively. The above results indicate that the introduc- 3.2.1. Effects of contact time and pH
tion of MnO2 nanorods leads to a decrease in the specific surface area TC solution (C0 = 200 mg L−1, pH = 4.0), as the research object, was
of the carrier. selected to study the adsorption mechanism of MnO2/graphene. The pH
The XPS spectrum was shown in Fig. 1(f)–(j), during the modifica- value of the TC solution (50 mL) was adjusted to 3.0, 5.0, 7.0, and 9.0, re-
tion process, the number and type of oxygen-containing functional spectively. Then, the adsorption of TC on treated graphene was studied
groups in the sample increased. Using 40%MnO2/graphene as an exam- (Fig. 5(a)). Solution pH is a key factor that could directly affect the TC
ple, the C1s peak included five peaks, at 284.22, 285.07, 286.58, 288.19 adsorption rate by changing the adsorbent surface charge, the specia-
and 291.43 eV, which were attributable to graphene, C\\O, C\\O\\C, tion distribution of the adsorbate and the ionization degree. The TC

Fig. 6. (a) Isotherm graph for TC adsorption; (b)–(d) linear fitted by Langmuir model.
 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590
Table 4
Calculated thermodynamic parameters.
T (K) K ΔG (kJ mol−1) ΔS (J mol−1 K−1) ΔH (kJ mol−1)
298 18.49 −7.23 298.23 81.45
308 67.73 −10.79
318 145.36 −13.16
adsorption capacity increased with the increase of pH value of the solu-
tion (from 3.0 to 5.0). However, when the pH values were higher than
5.0, the adsorption capacity decreased with the increase of pH. At a pH
of 5.0, the adsorption capacity reached its maximum (168 mg g−1).
Therefore, the TC adsorption behaviour of treated graphene showed
clear pH dependence. It was found that the initial pH of the pharmaceu-
tical wastewater after the flocculation treatment was 5.0. Therefore, in
this study, the optimum pH and contact time were 5.0 and 1500 min,
respectively.
3.2.2. Effect of MnO2 loading amount
On this basis, in order to better describe TC adsorption behaviour on
MnO2/graphene, the static adsorption tests were studied using a TC so-
lution with different MnO2 loading amount under optimal pH and con-
tact time. A designed concentration of TC solution (pH = 5.0, 50 mL,
200 mg L−1) was added to the conical flask. Adsorbents (MnO2/
graphene with different MnO2 loading amount and treated graphene,
50 mg) were added to the above solution. As shown in Fig. 5(b), the ad-
sorption rate of TC was more increased over MnO2/grapheme sample
than that of treated graphene. The results showed that the TC adsorp-
tion rate also increased significantly with the increase of MnO2 loading
amount. However, when the MnO2 loading amount was increased from
40% to 60%, the TC adsorption rate showed little change. In summary,
considering the cost and efficiency, the optimal loading amount of
MnO2 in this study was 40%.
3.2.3. Effects of temperature and solution concentration
At different test temperatures, 40%MnO2/graphene (50 mg) was
added to different concentrations of a TC solution (50 mL) for the static
adsorption test. The adsorption rate decreased with an increase of TC so-
lution concentration (Fig. 5(c)). However, with increase of test temper-
ature, the TC adsorption rate showed an upward trend under the same
solution concentration.
Finally, to further test the adsorption performance of the material,
the adsorption capacities of 40%MnO2/graphene to TC in different solu-
tions, such as pharmaceutical wastewater (C0 = 200 mg L−1), pharma-
ceutical wastewater after flocculation treatment (C0 = 180 mg L−1) and
TC solution (C0 = 200 mg L−1), were tested under the same conditions.
As shown in Fig. 5(d), 50 mg of 40%MnO2/graphene was added to the
above solutions with a volume of 50 mL for static adsorption experi-
ments. At a pH of 5.0, the TC adsorption rate by 40%MnO2/graphene in
pharmaceutical wastewater was relatively low (54%). However, the TC
adsorption rate in another two solutions (pharmaceutical wastewater
after flocculation treatment and TC solution) were greatly improved
(99.4% and 99.8%). Therefore, the TC adsorption rate in pharmaceutical
wastewater by 40%MnO2/graphene is lower than that in the other two
solutions. The composition of pharmaceutical wastewater is very com-
plex, and it contains a certain amount of impurities, which will greatly
affect the adsorption process of the adsorbent for the pollutants. There-
fore, in practical applications, the pharmaceutical wastewater is first
Table 5
Adsorption performance parameters of adsorbents.
Samples Water solubility Electrostatic interaction
Graphene − +
Treated graphene ++ + +
40%MnO2/graphene + +
587
flocculated to remove impurities, which can effectively improve the
performance of the adsorbent.
3.2.4. Regeneration performance of 40%MnO2/graphene
As shown in Fig. 5(e), the thermogravimetric characterization of the
40%MnO2/graphene after application was performed. Compared with
the nanocomposites before application, there was a three-stage process
in the temperature-responsive mass evolution: the adsorbed/hydrated
water was lost (below 200 °C); TC was lost (200–350 °C); and then
the rupture of chains, fragments and monolayers of treated graphene
(350–800 °C) occurred. Therefore, in order to remove TC in the adsor-
bent and regenerate the adsorbent, the 40%MnO2/graphene after appli-
cation was calcined at 400 °C for 4 h (heating rate was 2 °C min−1). The
adsorption experiment of regenerated adsorbent was carried out using
the static adsorption method described in Section 2.6. As shown in Fig. 5
(f), the adsorption rate of the adsorbent to TC decreased slightly with
the increase of the number of cycles, indicating that the adsorbent
maintained good stability in the process of recycling.
3.3. Adsorption kinetics
To fully understand the relationship between the adsorption rate
and the adsorption time, the adsorption kinetics of the whole adsorp-
tion process were studied, which was an important parameter for the
adsorption performance of the adsorbent. The time profile for the ad-
sorption of TC by MnO2/graphene in aqueous solution with different
MnO2 loading amount is shown in Fig. 5(b). The adsorption equilibrium
was reached at 1500 min. In addition, the adsorption capacity of 40%
MnO2/graphene increased rapidly in the first hour, and then slowly in-
creased until equilibrium was reached. The kinetic parameters are sum-
marized in Table 2. The results showed that the pseudo-second-order
model was better fitted with the experimental data (R2 = 0.9998). Fur-
thermore, the adsorption capacity calculated qe,cal (195 mg g−1) from
the pseudo-second-order equation was more consistent with the exper-
imental qe,exp (198 mg g−1) value. These results indicate that the ad-
sorption of TC by the material is selective chemisorption. The
adsorption mechanism will be discussed in detail in the subsequent
sections.
3.4. Adsorption isotherm
The study of the adsorption isotherm not only helps to describe how
the adsorbed molecules are distributed on the adsorbent, but also re-
veals the adsorption properties. Isothermal adsorption data can be ex-
plained by Langmuir and Freundlich isotherm models. The adsorption
of TC by 40%MnO2/graphene, in aqueous solution with different concen-
trations, was studied at three different adsorption temperatures (Fig. 5
(c) and Table 3). By comparing the correlation coefficients (R2), a Lang-
muir model is better to describe the TC adsorption isotherms on 40%
MnO2/graphene. Thus, monolayer adsorption dominated the adsorp-
tion process, with contributions from heterogeneous adsorption (Yang
et al., 2017). At different temperatures, the maximum adsorption capac-
ity values determined by Langmuir isotherm were 1789, 1796 and
1974 mg g−1, respectively, which indicated that a higher initial concen-
tration and higher adsorption temperature were conducive to adsorp-
tion (Fig. 6 and Table 3). The KL value from the Langmuir model
increased with the increase of temperature, indicating that the binding
between TC and the active centre of the adsorbent increased with the
H-bonding Coordination effect π-π interactions
− − +
− +
++ + +
588 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590

Fig. 7. XPS patterns of (a)–(b) 40%MnO2/graphene and (c)–(d) TC-40%MnO2/graphene.

increase in temperature, which also indicated that the adsorption pro-
cess was endothermic (Yang et al., 2017).
3.5. Adsorption thermodynamics
The negative ΔG value indicated that the adsorption of 40%MnO2/
graphene to TC was a spontaneous adsorption process. The positive
ΔH value indicated that the adsorption reaction had typical endother-
mic properties. The results showed that the higher temperature pro-
moted the adsorption of TC by 40%MnO2/graphene (Figs. 5(c) and 6,
Table 4). The positive ΔS value indicated an increase in the degree of
randomness at the adsorbate-adsorbent interface. In solution, TC was
surrounded by a tightly bound hydration layer, where water molecules
were more highly ordered than in the bulk water (M. Li et al., 2017).
When the TC molecules came into close interaction with the hydration
surface of 40%MnO2/graphene, the ordered water molecules in these
two hydration layers were compelled and disturbed, which would con-
tribute to the increase of entropy (M. Li et al., 2017). Although the ad-
sorption of TC molecules on the 40%MnO2/graphene surface caused
the degree of freedom of TC molecules to decrease, it seemed likely
that positive entropy related to the adsorption of TC on 40%MnO2/
graphene was due to the entropy increase of water molecules,
outweighing the entropy decrease of TC molecules (M. Li et al., 2017).
3.6. Adsorption mechanism
In summary, there are four major interactions between TC and 40%
MnO2/graphene, namely electrostatic interaction, H-bonding, coordina-
tion effect and π-π interactions (Table 5).

Fig. 8. Zeta potential of adsorbents at different pH values.

 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590
In Section 3.2.1 (Fig. 5(a)), under different acid dissociation con-
stants (pKa = 3.3, 7.7, and 9.7), TC exists as a cationic (TCH3+), zwitter-
ionic (TCH20), and anionic (TCH− or TC2−) species, respectively. When
the pH of the solution was 3.0, TC molecules existed in the form of cat-
ions. The Zeta potential characterization showed that the treated
graphene was negative under the same pH value (Fig. 8). Therefore,
there was a strong electrostatic attraction between the treated
graphene and TC. When solution pH was 5.0, TC molecules existed as
zwitterionic species. At this time, the treated graphene was still nega-
tive. There was a certain electrostatic repulsion between TC and treated
graphene, which made the TC adsorption rate decrease. When the pH
value of solution was higher than 5.0, TC molecules existed in the
form of anions. The treated graphene was still negative at this time.
The electrostatic repulsion between TC and treated graphene was fur-
ther enhanced, which caused the adsorption rate of TC to reduce further.
Therefore, as a carrier of the MnO2 nanorods, treated graphene could
generate an electrostatic interaction with the TC molecule, such that
the TC molecule was adsorbed by the adsorbent.
In Section 3.1 (Fig. 1), a series of characterizations (Raman spectrum,
FTIR spectra and C1s XPS spectra) of the adsorbents were sufficient to
demonstrate that a large number of oxygen-containing functional
groups were introduced into the treated graphene during the modifica-
tion process. The C-OR groups on the surface of treated graphene sheets
can form H-bonding with –OH groups on the TC molecule.
We used XPS characterization to analyse the adsorbents before and
after application to better explain the adsorption mechanism of TC on
40%MnO2/graphene. The XPS characterization results showed that the
characteristic peak of the C-OR group appeared in the C1s spectrum of
the adsorbents after application, compared with the adsorbents before
application. Moreover, the C-OR group belongs to the TC molecule.
This indicated that TC was successfully adsorbed on the surface of the
adsorbent (Fig. 7(a) and (c)). As shown in Fig. 7(b) and (d), XPS charac-
terization results showed that the binding energy of Mn did not change
before or after adsorption. However, compared with the Mn species be-
fore application, the peak area of the Mn species decreased, indicating
that TC was adsorbed onto the MnO2. In other words, a large number
of TC molecules were adsorbed by MnO2 nanorods, loaded on both
sides of graphene nanosheets and covering their surface, which made
the characteristic peak area of the Mn species in the XPS spectra weaker.
As mentioned in Section 3.3, the study of adsorption kinetics showed
that the pseudo-second-order kinetic model was able to better describe
the adsorption process and proved that the adsorption process was se-
lective chemisorption.
As shown in Fig. 5(c), the adsorption capacity of 40%MnO2/graphene
for TC was 198 mg g−1, even higher than that of free treated graphene
sheets (170 mg g−1). The increase of adsorption capacity is likely due
to the MnO2 nanorods, where nanorods avoided the over stacking of
treated graphene sheets during adsorption. Therefore, TC molecules
have more opportunities to interact with benzene ring structures on
treated graphene sheets to form π-π interactions, which is advanta-
geous to adsorption.
4. Conclusion
In this study, a MnO2/graphene nanocomposite was prepared via an
in situ hydrothermal method. To eliminate interference, pharmaceutical
wastewater, as a research object, was flocculated before the adsorption
test. The results showed that the TC removal rate in pharmaceutical
wastewater after flocculation was 99.4%. The pseudo-second-order ki-
netic and Langmuir models were fitted well in the adsorption process.
For TC residues, the different adsorbents followed the trend: MnO2/
graphene ˃ treated graphene ˃ graphene. In summary, loading MnO2
on the surface of treated graphene nanosheets to form MnO2/graphene
nanocomposites will provide new ideas for the efficient removal of re-
sidual TC in pharmaceutical wastewater.
589
Acknowledgements
The authors would like to thank the National Natural Science Foun-
dation of China (No. 21467023), the National First-rate Discipline Con-
struction Project of Ningxia (NXYLXK2017A04), and the Science and
Technology Support Program of Ningxia for financial support.
References
Aghagoli, M.J., Shemirani, F., 2017. Application of molybdenum disulfide nanosheets for
adsorption of tetracycline in water samples. Desalin. Water Treat. 70, 250–260.
Ahmed, M.B., Zhou, J.L., Ngo, H.H., Guo, W., 2015. Adsorptive removal of antibiotics from
water and wastewater: Progress and challenges. Sci. Total Environ. 532, 112–126.
Ahmed, M.J., Islam, M.A., Asif, M., Hameed, B.H., 2017. Human hair-derived high surface
area porous carbon material for the adsorption isotherm and kinetics of tetracycline
antibiotics. Bioresour. Technol. 243, 778–784.
Akhtar, J., Amin, N.A.S., 2015. A review on removal of pharmaceuticals from water by ad-
sorption. Desalin. Water Treat. 42, 1–19.
Chao, Y., Yang, L., Ji, H., Zhu, W., Pang, J., Han, C., Li, H., 2017. Graphene-analogue molyb-
denum disulfide for adsorptive removal of tetracycline from aqueous solution equi-
librium, kinetic, and thermodynamic studies. Environ. Prog. Sustain. 36, 815–821.
Chen, W., Huang, C., 2010. Adsorption and transformation of tetracycline antibiotics with
aluminum oxide. Chemosphere 79, 779–785.
Chen, Y., Liu, K., 2016. Preparation and characterization of nitrogen-doped TiO2/diatomite
integrated photocatalytic pellet for the adsorption-degradation of tetracycline hydro-
chloride using visible light. Chem. Eng. J. 302, 682–696.
Chen, F., Yang, Q., Wang, Y., Zhao, J., Wang, D., Li, X., Guo, Z., Wang, H., Deng, Y., Niu, C.,
Zeng, G., 2017a. Novel ternary heterojunction photcocatalyst of Ag nanoparticles
and g-C3N4 nanosheets co-modified BiVO4 for wider spectrum visible-light photo-
catalytic degradation of refractory pollutant. Appl. Catal. B Environ. 205, 133–147.
Chen, F., Yang, Q., Li, X., Zeng, G., Wang, D., Niu, C., Zhao, J., An, H., Xie, T., Deng, Y., 2017b.
Hierarchical assembly of graphene-bridged Ag3PO4/Ag/BiVO4 (040) Z-scheme
photocatalyst: an efficient, sustainable and heterogeneous catalyst with enhanced
visible-light photoactivity towards tetracycline degradation under visible light irradi-
ation. Appl. Catal. B Environ. 200, 330–342.
Guo, Y., Huang, W., Chen, B., Zhao, Y., Liu, D., Sun, Y., Gong, B., 2017. Removal of tetracy-
cline from aqueous solution by MCM-41-zeolite A loaded nano zero valent iron: syn-
thesis, characteristic, adsorption performance and mechanism. J. Hazard. Mater. 339,
22–32.
He, L., Liu, F., Zhao, M., Qi, Z., Sun, X., Afzal, M.Z., Sun, X., Li, Y., Hao, J., Wang, S., 2017.
Electronic-property dependent interactions between tetracycline and graphene
nanomaterials in aqueous solution. J. Environ. Sci. (China) 66, 286–294.
Huang, B., Liu, Y., Li, B., Liu, S., Zeng, G., Zeng, Z., Wang, X., Ning, Q., Zheng, B., Yang, C.,
2017. Effect of Cu(II) ions on the enhancement of tetracycline adsorption by
Fe3O4@SiO2-Chitosan/graphene oxide nanocomposite. Carbohydr. Polym. 157,
576–585.
Khan, M.H., Bae, H., Jung, J., 2010. Tetracycline degradation by ozonation in the aqueous
phase: proposed degradation intermediates and pathway. J. Hazard. Mater. 181,
659–665.
Khan, A., Wang, J., Li, J., Wang, X., Chen, Z., Alsaedi, A., Hayat, T., Chen, Y., Wang, X., 2017.
The role of graphene oxide and graphene oxide-based nanomaterials in the removal
of pharmaceuticals from aqueous media: a review. Environ. Sci. Pollut. Res. 24,
7938–7958.
Li, N., Yang, S., Chen, J., Gao, J., He, H., Sun, C., 2016. Electro-adsorption of tetracycline from
aqueous solution by carbonized pomelo peel and composite with aniline. Appl. Surf.
Sci. 386, 460–466.
Li, H., Hu, J., Meng, Y., Su, J., Wang, X., 2017a. An investigation into the rapid removal of
tetracycline using multilayered graphene-phase biochar derived from waste chicken
feather. Sci. Total Environ. 603–604, 39–48.
Li, M., Liu, Y., Zeng, G., Liu, S., Hu, X., Shu, D., Jiang, L., Tan, X., Cai, X., Yan, Z., 2017b. Tetra-
cycline absorbed onto nitrilotriacetic acid-functionalized magnetic graphene oxide:
influencing factors and uptake mechanism. J. Colloid Interface Sci. 485, 269–279.
Li, Z., Qi, M., Tu, C., Wang, W., Chen, J., Wang, A., 2017c. Highly efficient removal of
chlorotetracycline from aqueous solution using graphene oxide TiO2 composite prop-
erties and mechanism. Appl. Surf. Sci. 425, 765–775.
Liu, M., Hou, L., Xi, B., Li, Q., Hu, X., Yu, S., 2016. Magnetically separable Ag/AgCl-zero
valent iron particles modified zeolite X heterogeneous photocatalysts for tetracycline
degradation under visible light. Chem. Eng. J. 302, 475–484.
Lv, J.M., Ma, Y.L., Chang, X., Fan, S.B., 2015. Removal and removing mechanism of tetracy-
cline residue from aqueous solution by using Cu-13X. Chem. Eng. J. 273, 247–253.
Martins, A.C., Pezoti, O., Cazetta, A.L., Bedin, K.C., Yamazaki, D.A.S., Bandoch, G.F.G., Asefa,
T., Visentainer, J.V., Almeida, V.C., 2015. Removal of tetracycline by NaOH-activated
carbon produced from macadamia nut shells: kinetic and equilibrium studies.
Chem. Eng. J. 260, 291–299.
Marzbali, M.H., Esmaieli, M., Abolghasemi, H., Marzbali, M.H., 2016. Tetracycline adsorp-
tion by H3PO4-activated carbon produced from apricot nut shells: a batch study. Pro-
cess. Saf. Environ. 102, 700–709.
Norvill, Z.N., Toledo-Cervantes, A., Blanco, S., Shilton, A., Guieysse, B., Munoz, R., 2017.
Photodegradation and sorption govern tetracycline removal during wastewater
treatment in algal ponds. Bioresour. Technol. 232, 35–43.
Rattanachueskul, N., Saning, A., Kaowphong, S., Chumha, N., Chuenchom, L., 2017. Mag-
netic carbon composites with a hierarchical structure for adsorption of tetracycline,
prepared from sugarcane bagasse via hydrothermal carbonization coupled with sim-
ple heat treatment process. Bioresour. Technol. 226, 164–172.
590 Z. Song et al. / Science of the Total Environment 651 (2019) 580–590
Song, Z., Ma, Y.L., Li, C.E., Xu, M., 2018. Removal of tetracycline residue from pharmaceu-
tical wastewater by using 3D composite film. Chem. Eng. J. 348, 898–907.
Thi, V.H.T., Lee, B., 2017. Great improvement on tetracycline removal using ZnO rod-
activated carbon fiber composite prepared with a facile microwave method.
J. Hazard. Mater. 324, 329–339.
Wang, D., Jia, F., Wang, H., Chen, F., Fang, Y., Dong, W., Zeng, G., Li, X., Yang, Q., Yuan, X.,
2018. Simultaneously efficient adsorption and photocatalytic degradation of tetracy-
cline by Fe-based MOFs. J. Colloid Interface Sci. 519, 273–284.
Xiong, W., Zeng, G., Yang, Z., Zhou, Y., Zhang, C., Cheng, M., Liu, Y., Hu, L., Wan, J., Zhou, C.,
Xu, R., Li, X., 2018. Adsorption of tetracycline antibiotics from aqueous solutions on
nanocomposite multi-walled carbon nanotube functionalized MIL-53(Fe) as new ad-
sorbent. Sci. Total Environ. 627, 235–244.
Yang, Y., Hu, X., Zhao, Y., Cui, L., Huang, Z., Long, J., Xu, J., Deng, J., Wu, C., Liao, W., 2017.
Decontamination of tetracycline by thiourea-dioxide–reduced magnetic graphene
oxide: effects of pH, ionic strength, and humic acid concentration. J. Colloid Interface
Sci. 495, 68–77.
Yu, B., Bai, Y., Ming, Z., Yang, H., Chen, L., Hu, X., Feng, S., Yang, S., 2017. Adsorption behav-
iors of tetracycline on magnetic graphene oxide sponge. Mater. Chem. Phys. 198,
283–290.
Zhang, L., Song, X., Liu, X., Yang, L., Pan, F., Lv, J., 2011. Studies on the removal of tetracy-
cline by multi-walled carbon nanotubes. Chem. Eng. J. 178, 26–33.
Zhang, Z., Liu, H., Wu, L., Lan, H., Qu, J., 2015. Preparation of amino-Fe(III) functionalized
mesoporous silica for synergistic adsorption of tetracycline and copper. Chemosphere
138, 625–632.
Zhang, Y., Jiao, Z., Hu, Y., Lv, S., Fan, H., Zeng, Y., Hu, J., Wang, M., 2017. Removal of tetra-
cycline and oxytetracycline from water by magnetic Fe3O4@graphene. Environ. Sci.
Pollut. Res. 24, 2987–2995.
Zhang, Y., Zuo, S., Zhou, M., Liang, L., Ren, G., 2018. Removal of tetracycline by coupling of
flow-through electro-Fenton and in-situ regenerative active carbon felt adsorption.
Chem. Eng. J. 335, 685–692.
Zhao, L., Xue, F., Yu, B., Xie, J., Zhang, X., Wu, R., Wang, R., Hu, Z., Yang, S., Luo, J., 2015.
TiO2-graphene sponge for the removal of tetracycline. J. Nanopart. Res. 17, 1–9.
Zhou, Y., Liu, X., Xiang, Y., Wang, P., Zhang, J., Zhang, F., Wei, J., Luo, L., Lei, M., Tang, L.,
2017. Modification of biochar derived from sawdust and its application in removal
of tetracycline and copper from aqueous solution: adsorption mechanism and model-
ing. Bioresour. Technol. 245, 266–273.
Zhu, H., Chen, T., Liu, J., Li, D., 2018. Adsorption of tetracycline antibiotics from an aqueous
solution onto graphene oxide/calcium alginate composite fibers. RSC Adv. 8,
2616–2621.
Zhuang, Y., Yu, F., Ma, J., Chen, J., 2017. Enhanced adsorption removal of antibiotics from
aqueous solutions by modified alginate graphene double network porous hydrogel.
J. Colloid Interface Sci. 507, 250–259.
Ziółkowska, A., Margas, M., Grajek, H., Wasilewski, J., Adomas, B., Michalczyk, D.,
Piotrowicz-Cieślak, A.I., 2016. Comparison of UV-C irradiation, ozonation, and iron
chelates treatments for degradation of tetracycline in water. Int. J. Environ. Sci.
Technol. 13, 1335–1346.