Journal of Membrane Science 640 (2021) 119844

Contents lists available at ScienceDirect

Journal of Membrane Science

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

Electrochemical manufacture of graphene oxide/polyaniline conductive

membrane for antibacterial application and electrically enhanced
water permeability
Bojun Li a, b, 1, Wenjing Tang a, c, 1, De Sun a, *, Bingbing Li a, **, Yanxia Ge a, ***, Xin Ye d, Wei Fang d
a
Department of Chemical Engineering, Changchun University of Technology, 2055 Yanan Street, Changchun, 130012, PR China
b
State Key Laboratory of Water Environment Simulation, School of Environment, Beijing Normal University, Beijing, 100875, PR China
c
Beijing Key Laboratory of Farmland Soil Pollution Prevention and Remediation, College of Resources and Environmental Sciences, China Agricultural University, Beijing,
100193, PR China
d
Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun, 130012, PR China

A R T I C L E I N F O A B S T R A C T

Keywords: Electrofiltration, an effective approach for membrane fouling mitigation, is significantly limited by membrane
Graphene oxide properties. A facile electrochemical method was proposed to fabricate the graphene oxide/sulfuric acid-doped
Polyaniline membrane polyaniline (GO/S-PANI) membrane. For which, charging the graphite in H2SO4 (98 wt%) to obtain graphite
Electrochemical preparation
intercalation compound (GIC), then in a mixed electrolyte (H2SO4, CuSO4, (NH4)2SO4), GO was exfoliated and
Water permeability
assembled concurrently on the PANI membrane, which was doped by H2SO4 simultaneously. The introduction of
Antibacterial
Cu2+ made the GO layers on the membrane stable. Also, the GO/S-PANI membrane showed higher conductivity
(55.6 S m− 1) than the PANI membrane (0.019 S m− 1). Moreover, GO/S-PANI membrane possessed a more
applicable pore structure and improved hydrophilicity. As a result, membrane rejection increased, and the
resistance to the negatively charged pollutants was enhanced. For 1 V electrofiltration of yeast suspension, water
permeation was sustainably raised by using GO/S-PANI membrane than PANI membrane. The GO/S-PANI
membrane was more stable with 1 V than without electric fields. The antibacterial rate can reach 92.1% for
the GO/S-PANI membrane against Escherichia coli. Overall, our strategy provides a facile preparation method for
the GO/S-PANI conductive membrane with application potential in electrofiltration and antibacterial fields.

1. Introduction surface and the negatively charged microorganisms when electric fields
are used. Thus, membrane fouling can be mitigated by reducing the
Membrane fouling is a primary problem that limits the wide use of adsorption potential of microorganisms [5,7–9].
membrane separation technology [1]. Typical pollutants in water, such Commonly, conductive membranes are fabricated by conductive
as microbes, can deposit on and adhere to the membrane surface and materials such as carbons (e.g., graphene and carbon nanotube), metals
then grow to form communities known as biofilms, allowing various (e.g., Au, Ag, and stainless steel), and conductive polymers (e.g., Poly­
bacteria to inhabit [2]. As microorganisms with inherent electronega­ aniline and Polypyrrole) [7,10–15]. However, since carbon membranes
tivity can be affected by electrostatic force, recently, inhibiting micro­ or metal membranes are high in cost and complex in preparation, car­
bial contamination of membranes through electrofiltration has become a bon/conductive polymer composites are considered emerging materials
research hotspot [3–6]. The research on electrofiltration proves that for preparing conductive membranes. The membranes can combine the
applying conductive membranes as the electrode with low-intensity advantages of carbon and polymer materials and can be well utilized in
electric fields can markedly enhance water permeation and decrease electrofiltration processes [5,8,16]. It should be noted that, for the water
energy demand. Electrostatic force can be generated between membrane treatment coupled with electric field, conductive membranes must have

* Corresponding author.
** Corresponding author.
*** Corresponding author.
E-mail addresses: [email protected] (D. Sun), [email protected] (B. Li), [email protected] (Y. Ge).
1
Co-first authors: Bojun Li; Wenjing Tang.

https://doi.org/10.1016/j.memsci.2021.119844
Received 17 June 2021; Received in revised form 13 August 2021; Accepted 5 September 2021
Available online 9 September 2021
0376-7388/© 2021 Elsevier B.V. All rights reserved.
B. Li et al. Journal of Membrane Science 640 (2021) 119844

good stability under aqueous and harsh conditions (e.g., high pressure). sulfuric acid simultaneously. Membrane properties were analyzed, and
Thus, only a limited number of conducting polymers (e.g., Polypyrrole, membrane fouling mitigation’s impact on water permeability for the
Polythiophene, and Polyaniline) can be used to fabricate conductive electrofiltration processes was studied. The facile electrochemical
membranes [17]. The cost of pyrrole monomer is high, and the preparation method of the GO/S-PANI membrane has shown multiple
commercially available polythiophenes are few, so their extensive advantages: (1) In contrast to the GO prepared by the Hummers/
application in conductive polymer membranes was limited [16]. Poly­ modified Hummers’ method, there are no explosive reactions, and the
aniline (PANI), a conductive polymer, has many advantages: relatively production of GO is high [36,37]. (2) In the electrochemical intercala­
low cost, product marketization, controllable conductivity, and excel­ tion step, the sulfuric acid electrolyte can be reused until fully
lent environmental stability. It has been extensively studied and fabri­ consumed, as this step generates no contaminants [36]. (3) Contributing
cated into conductive membranes by the nonsolvent induced phase to the forming of the expanded spaces among layers and the large vol­
separation (NIPS) method [12,17–19]. For the preparation of PANI ume of the water channels, the GO layers assembled on the membrane
membranes, sulfur-containing acids can be introduced as acid dopant to will not be compacted as it happens during vacuum filtration [39,40].
enhance membrane conductivity [5]. In terms of this, using poly (4) For GO composite membranes, the fabrication method often requires
(2-acrylamido-2-methyl-1-propanesulfonic acid) as a dopant, Xu et al. two steps (the preparation of GO and then the combination of GO with
[20] have fabricated PANI conductive membranes revealing a conduc­ membranes); for the PANI membranes prepared by the NIPS method, an
tivity of 0.1 S m− 1 for the cross-flow filtration. Dodecylbenzene sulfonic extra step of acid immersion was needed to increase membrane con­
acid was utilized by Wang et al. [12] to manufacture the PANI ductivity [12,31]. By this electrochemical approach, the three processes,
conductive membrane with a conductivity of 2.2 × 10− 2 S m− 1 for including the preparation of GO, the assembly of GO on the PANI
electrofiltration. Apart from these, the sulfur-containing acids can also membrane, and the acid doping of the PANI membrane can be pro­
improve the antifouling behavior and the chlorine resistance of the PANI ceeded concurrently. This study provides an innovative strategy for
membranes and increase membrane hydrophilicity [21,22]. The PANI preparing acid-doped GO/PANI membrane, promisingly used in elec­
membranes, doped by sulfur-containing acids, can also present tunable trofiltration and antibacterial fields.
transport properties for water and solute when additionally doped by
HCl [22]. 2. Experimental methods
Graphene-based materials are among the most widely used carbon
materials to manufacture separation membranes [14,23–26]. Although 2.1. Materials
pristine graphene typically possesses better conductivity than graphene
oxide (GO), due to the existence of functional hydrophilic groups and Commercial PANI polymer (Emeraldine base, Molecular weight:
the physical piercing of edges, GO has exhibited excellent performance 10,000–100000, Purity: ≥ 99.0 wt%, Particle size: less than 30 μm by
for membrane fouling mitigation [7,23]. In antibacterial field, GO also laser particle analyzer; 100 nm by TEM) was purchased from Cool
revealed a better effect than pristine graphene [24]. Additionally, only Chemical Technology (Beijing) Co., Ltd. (China). Ammonium Sulfate,
limited interaction force (e.g., π-π interaction) can be generated when Sulfuric acid, N-methyl-2-pyrrolidone (NMP), 4-methyl piperidine (4-
the pristine graphene was combined with PANI membrane, leading to MP), and Copper (II) sulfate pentahydrate (CuSO4⋅5H2O) were provided
the restricted stability of graphene on the membrane [27]. In contrast, by Sigma-Aldrich (UK). Graphite foil was acquired from Qingdao Herita
GO and PANI can be combined tightly through electrostatic interactions, Graphite Products Co., Ltd. (China). PET non-woven fabric was supplied
hydrogen bonding, and π-π stacking, providing a good foundation for by Haoxin Insulation Materials Co., Ltd. (China). Dry yeast was obtained
preparing composites [28,29]. By way of example, firstly through the from Angel Yeast Co., Ltd. (China). Deionized (DI) water was prepared
Hummers’ method to obtain GO, then the GO/PANI reinforced hollow from the laboratory-made. All chemical reagents were used as received
fiber membranes were prepared for the preconcentration of the Iver­ without any purification.
mectin [30]. Employing a modified Hummers’ method to synthesize
initial GO, then a GO/PANI nanocomposite membrane was developed 2.2. PANI membrane fabrication
for CO2/N2 separation [31]. Through the Hummers’ method combined
with an in-situ polymerization, a PANI/GO nanocomposite was designed PANI membrane was prepared via the NIPS method (Fig. 1a) at room
for room-temperature thermoelectric enhancement [32]. Although temperature [41]. In a nutshell, 2.64 g 4-MP and 15.4 g NMP were
GO/PANI composites have been extensively prepared, currently GO is mixed, then 3.96 g PANI powder was slowly introduced into the
mostly prepared through chemical oxidation, such as the Hummers’ mixture. The mixture was continuously stirred at a high speed of 300
method [33] and the modified Hummers methods [34], which may rpm for 4 h to produce the homogeneous PANI solution. Next, cast the
bring safety risks and cause environmental issues [35]. Most recently, mixed solution on the non-woven fabric, tiled on a smooth glass plate.
through the technique of electrochemical oxidation and exfoliation, GO The casting thickness was 420 μm that the large thickness brought good
production becomes safe in operation and simple in the process [36–38]. mechanical properties. After being exposed to the atmosphere for 30 s,
For instance, after graphite intercalation compounds (GICs) were ob­ the membrane was immersed in the coagulation bath, which used DI
tained by charging the commercial graphite foil in the concentrated water as the nonsolvent and stayed there for at least 24 h. At last, the
H2SO4, Cao et al. [36] produced GO through a short time electro­ generated membrane was rinsed with fresh water and stored in DI water
chemical oxidation and exfoliation of GICs in 0.1 M (NH4)2SO4 solution. for later use.
As an alternative approach, using a similar step to prepare GICs, Pei
et al. selected the diluted H2SO4 (50 wt%) as the electrolyte to oxidize 2.3. Electrochemical preparation of GO/S-PANI membrane
and exfoliate GICs into GO [37]. Nevertheless, very few studies focus on
the fabrication of GO composite membranes directly using graph­ The electrochemical process for preparing graphite intercalation
ite/intercalated graphite as the precursor. In the fields of electro­ compound (GIC) was shown in Fig. 1b. GIC was synthesized by charging
filtration and antibacterial, it is critical to develop a facile a strip of graphite foil (Fig. 1c, 4 × 5 cm2) under 2 V for 15 min. In this
electrochemical method to produce the conductive GO/PANI mem­ process, the graphite foil and platinum sheet were used as the anode and
branes with high performance. the cathode, respectively; concentrated sulfuric acid (98 wt%, 200 ml)
In this study, GO/sulfuric acid-doped PANI (GO/S-PANI) membranes was selected as the electrolyte (Fig. 1d). The intercalation of sulfuric
were developed by a modified electrochemical method. The graphite foil acid made graphite foil expanded and turned into a blue color which
was electrochemically intercalated to prepare GICs, then, GO was pro­ indicated the formation of GIC (Fig. 1e) [36]. The GIC was still robust
duced and assembled on the PANI membrane, which was doped by mechanically allowing it move from one electrode device to another for

2
B. Li et al. Journal of Membrane Science 640 (2021) 119844

Fig. 1. (a) Schematics of the preparation procedure for PANI membrane. (b) Schematic illustration of the electrochemical intercalation for the production of GIC. (c)
Photographs of graphite (4 × 5 cm2 in the area). (d) Electrode setup for oxidation of GIC. (e) Photographs of graphite foil (4 × 5 cm2 in area and 0.3 mm in thickness)
being charged to 2 V, forming a blue in color GIC. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of
this article.)

further treatment. membrane can also be doped by sulfuric acid [42]. By weighing the mass
Fig. 2a illustrated the electrochemical process for the fabrication of of the original graphite foil and GIC, the calculated sulfuric acid content
the GO/S-PANI membrane. The details are as follows: the newly pre­ in the electrolyte was 1.1 mol/L. During the electrochemical process, the
pared GIC and the PANI membrane (Fig. 2b left, 4 × 5 cm2) were utilized positively charged Cu2+ ions in the electrolyte automatically adsorbed
as the anode and the cathode, respectively. They were placed vertically the negatively charged GO (GO/Cu2+), which made the GO/Cu2+
in 0.1 M (NH4)2SO4 electrolyte with a distance of 1 cm. After that, set the positively charged (Fig. 3a) [43,44]. Under the action of electric fields,
power supply to 5 V and stabilized it for 120 min. The 0.001 M CuSO4 the formed GO/Cu2+ moved onto the cathode and assembled on the
solution was added to the electrolyte every 10 min (Fig. 2d), then the PANI membrane to form the GO/S-PANI membrane. Through electro­
GO/S-PANI membrane (Fig. 2b right) was successfully obtained. static interactions, hydrogen bonding, and π-π stacking, the GO layer
was tightly combined with the PANI membrane (Figs. 2c and 3b) [28,
29]. By this electrochemical strategy, the three processes, including the
2.4. Mechanism of the electrochemical process for GO/S-PANI membrane
preparation of GO, the assembly of GO on the PANI membrane, and the
preparation
acid doping of the PANI membrane can be proceeded concurrently,
improving the experiment safety and simplifying membrane fabrication
When the voltage is applied in the above electrochemical process, the
steps. The copper ions, confirmed as an excellent cross-linking agent for
nucleophilic water molecules could immediately embed into GIC. The
GO [45], could maintain mechanical stability and enhance water
graphene layers within GIC then react with the nucleophilic water
permeability of the GO layer. Meanwhile, the formed GO/Cu2+ on the
molecules and form –OH, C–O-, and C– – O groups. The gaseous oxygen
GO/S-PANI membrane could increase antibacterial activity [46], which
was produced by the oxidation of the intercalated water, leading to the
made contributions to the promotion of membrane mechanical stability
volume expansion of GIC and then the dissociation/exfoliation of GO
and antibacterial ability.
from GIC [36,37]. Additionally, the concentrated sulfuric acid interca­
lated in the GIC can penetrate into the electrolyte. Thereupon, the PANI

3
B. Li et al. Journal of Membrane Science 640 (2021) 119844

Fig. 2. (a) Schematic illustration of the electrochemical oxidation for GO production; assembly process of GO/S-PANI membrane. (b) Photographs of PANI mem­
brane (4 × 5 cm2 in the area) and GO/S-PANI membrane (4 × 5 cm2 in the area). (c) Schematic illustration of the interaction between GO and PANI. (d) Photographs
of the electrochemical preparation process for GO/S-PANI membrane.

Fig. 3. (a) Schematic of GO formed into GO/Cu2+ (b) Binding forces between GO/S-PANI membrane and GO/Cu2+ layer.

2.5. Analytical method of membranes was carried out on a universal testing machine (YHS-
229WG, Shanghai Yihuan Co, China). Pore size and pore size distribu­
Fourier transform infrared spectroscopy (FT-IR, Thermofisher Nico­ tion (PSD) of the membranes were tested by the capillary flow poros­
let IS50FT-IR) was performed with the test wavelength 4000-525 cm− 1. imeter (POROLUX 500, Germany). Dynamic droplet contact angles were
Raman spectroscopy (Horiba Jobin Yvon LabRAM HR Evolution Raman measured using a contact angle meter (HARKE-SPCA Beijing Hake
spectrometer) was recorded from 200 to 4000 cm− 1 using a 532 nm Experimental Instrument Factory).
laser. X-ray Photoelectron Spectroscopy (XPS ESCALAD 250) was tested
by Mg Kα X-ray source (1486 eV). The X-ray diffraction (XRD, Rigaku 2.6. Evaluation of controlling water permeability by fouling mitigation
Smartlab X-ray Diffractometer) was investigated with the radiation of K
beta filter method (45 kV, 200 mA) in a 2θ range of 5–70◦ with a step The change in permeation flux of the yeast suspension was an
size of 0.01 and a sweep rate of 10◦ per minute. The membrane obvious indicator of evaluating the membrane fouling mitigation [8,47].
morphology and the distribution of the elements were recorded by SEM- Fig. S1 showed the electrofiltration setup. The membrane module
EDX (JSM6510 JEOL). Membrane roughness was measured using the (Fig. S2) was made from non-conductive photosensitive resin materials
AFM (Agilent technologies AFM 5500) in tapping mode. The tensile test with an effective permeation area of 3.24 cm2. Fixing the membrane in

4
B. Li et al. Journal of Membrane Science 640 (2021) 119844

the membrane module (Fig. S3), titanium plate (3.5 × 3.5 cm2) and represents the effective membrane area (m2), and Δt stands for the
conductive membrane were used as the anode and the cathode. A dis­ permeation time (h).
tance of 1 cm was placed for the two electrodes. A direct-current power The enhanced rate of total permeation flux is calculated based on the
supply was utilized to generate electric fields. Yeasts (Zeta potential: following formula:
− 12.9 mV, size: 2–5 μm, Fig. S4) were dissolved in tap water to prepare
Q2 − Q1
yeast suspension, and 2 L yeast suspension with a concentration of 2 g q= (2)
Q1
L− 1 was used as the feed. With the application of electric fields, the
permeation liquid was collected at 0.04 Mpa, 25 ◦ C, and a feed flow rate In the formula, q is the enhanced rate of total permeation flux (%), Q1
of 1.6 L min− 1. Each set of tests was conducted in three cycles with each is the total permeation flux of the GO/S-PANI membrane without elec­
cycle continuing for 180 min. Each test was repeated three times to tric fields (L), Q2 stands for the total permeation flux of tested mem­
obtain the average value with errors as the final results. After each cycle branes under series of low-intensity electric fields (L).
of testing, using tap water to wash the membrane and the conductivity of Through UV–Vis spectrophotometer (Shanghai YOKE instrument),
the membranes was tested instantly. the yeast concentrations in the solutions are measured at a wavelength
The flux is calculated based on the following equation: of 600 nm, the rejection rate (R) of yeast is calculated based on the
following formula:
Q
J= (1)
A Δt Cf − Cp
R(%) = ( ) × 100% (3)
Cf
− 2 − 1
Where J is flux (L m h ), Q is the volume of collected solution (L), A
In the formula, R represents the rejection rate of membranes, Cp is the

Fig. 4. Spectra of GO, PANI membrane, and GO/S-PANI membrane: (a) FTIR (b) Raman (c) XRD (d) XPS wide scan spectra (e) Cu 2p in XPS (f) C 1s in XPS (g) PANI
converted from emeraldine base into emeraldine salt during the electrochemical process.

5
B. Li et al. Journal of Membrane Science 640 (2021) 119844

concentrations of the yeast in the permeation, and Cf is the concentra­ perfect formation of the hybrid system [32]. The peak appeared at 2θ =
tions of the yeast in the feed. As schemed in Fig. S5, the standard curve (a 25.2◦ in PANI was shifted to 2θ = 25.8◦ in GO/S-PANI, corresponding to
wavelength of 600 nm) was plotted for 2 g L− 1 yeast suspension. the π-π stacking of the benzene ring and GO [32,62].
To analyze the elemental composition of GO/S-PANI membrane and
3. Results and discussion the combination state of carbon and copper, further XPS spectra
(Fig. 4d) were tested. Obvious peaks representing the elements C, O, N,
3.1. Membrane characterization S, and Cu were shown in GO/S-PANI, in which the existed S (168.0 eV)
reflected the acid doping of PANI in GO/S-PANI. Besides, the XPS of C 1s
3.1.1. Membrane chemical analysis (284.6 eV), N 1s (399.9 eV), and O 1s (531.8 eV) has appeared in GO/S-
The FTIR spectra were represented in Fig. 4a. For GO samples, the PANI. As shown in Fig. 4e, two signal peaks of Cu (933.1) and Cu2+
absorption peak centered at 3446 cm− 1 was attributed to the O–H (935.1) were observed, reflecting the combination of Cu2+ and GO [63].
stretching, and the other four peaks at 1629 cm− 1, 1384 cm− 1, 1250 Meanwhile, the C 1s contained five signal peaks in Fig. 4f: C–
– C (284.2),
cm− 1, and 1050 cm− 1 belonged to the C– – C, C–OH, C–O–C, and C–O C–C (284.6), C–O (284.9), C–N (286.2eV), and O–C– – O (288.5) [64],
stretching vibrations, respectively. Besides, the weak peak near 1730 demonstrating the existence of the abundant oxygen-containing func­
cm− 1 was related to the C– – O in the carbonyl and carboxyl groups. The tional groups on GO. To avoid the agglomeration caused by the coor­
above peaks were consistent with the classic map, indicating the suc­ dination between the COO− group on GO sheets and the positively
cessful synthesis of GO [48,49]. According to PANI spectra, the peaks at charged Cu2+, the added amount of copper ions must be sufficiently low.
1595 cm− 1 and 1495 cm− 1 originated from the C– – C and C– –N The Cu:C ratio was 0.072. The C:O ratio of the XPS has been widely used
stretching vibrations of the quinoid ring, respectively, revealing the to evaluate the oxidation degree of GO, and the obtained ratio was 3:1,
PANI chains in emeraldine base form. The peak at 1302 cm− 1 was which confirmed that the GO with a small size and high oxidation degree
related to the C–N stretching mode of the benzene unit, and 1166 cm− 1 was successfully prepared [65].
was attributed to the in-plane bending vibration of the benzene ring. In
addition, the peak at 953 cm− 1 was caused by the out-of-plane C–H 3.1.2. Membrane morphology and elemental analysis
deformation vibration of 1, 2, 4-disubstituted benzene [50,51]. By The SEM images in Fig. 5 presented the surface morphology and the
comparison, for the GO/S-PANI spectra, a broad peak was found at 3221 cross-section of the PANI and GO/S-PANI membranes. Both membranes
cm− 1 attributing to the -NH2+ stretching, which demonstrated the pro­ shared similar cross-section structures, containing a dense surface layer
tonation of PANI. The weak but identifiable peak at 1054 cm− 1 repre­ and a sublayer with finger-like pores. The top surface morphology of the
sented the S– – O stretching and the 708 cm− 1 was related to the S–O PANI membrane (PANI-a) revealed big crack-like pores (red circles) on
stretching, confirming the existence of the sulfuric acid doping in the the membrane surface. It can be observed clearly from PANI-b (magni­
PANI of GO/S-PANI [52]. Meanwhile, the peaks at 1578, 1489, and fication 2000 × ) and PANI-c (magnification 5000 × ). After electro­
1294 cm− 1 were redshifted by 17, 6, and 8 cm− 1, respectively, and the chemically prepared and assembled on the PANI membrane, the GO
peak at 1166 cm− 1 disappeared compared with PANI, further reflecting layer was manifested on the PANI membrane surface (GO/S-PANI-a).
GO/S-PANI membrane was in the emeraldine salt form [53,54]. Addi­ The 2000 × magnified image (GO/S-PANI-b) showed that the GO in
tionally, the peak of the GO/S-PANI membrane at 953 cm− 1 dis­ flake form completely covered the membrane surface. The oblique
appeared, implying a bonding interaction between the nitrogen atom of stacking of GO sheets can be observed clearly from the 5000 × magni­
PANI and the C– – O group on GO [55]. A redshift from 1720 cm− 1 in fied image (GO/S-PANI-c), providing a favorable spatial structure layer
PANI to 1715 cm− 1 in GO/S-PANI also confirmed that the carboxyl with a thickness of 182 nm (Fig. S6) for water penetration and rejection
group of GO was connected to the PANI through the nitrogen atom of the enhancement of contaminant.
PANI main chain [56]. The EDX analysis of GO/S-PANI membrane (GO/S-PANI-d) helps to
As illustrated in Fig. 4b, for Raman spectra, the centers of the two track the distribution of GO. The mass percentages (wt%) of C and N
characteristic D and G bands associated with the defect density in GO elements obtained at line data 1, line data 2, and line data 3 (GO/S-
and the sp2 bonding carbon bonds of the graphite sheet were at 1348 and PANI-d) were summarized in Table 1. The acquired data for the mem­
1580 cm− 1, respectively. The ID/IG ratio of Raman spectra was widely brane surface (Line data 1) is of the highest carbon-nitrogen (C/N) ratio
used to evaluate the quality of carbon materials. The ratio of GO was of 8.3, achieving a higher value than the pure PANI of 5.0, further
0.81, indicating the existence of a few defects in the GO [57]. It can be reflecting the uniform coverage of the GO layer on the membrane sur­
observed that the D band at 1343 cm− 1 for the PANI spectra was moved face. The C/N ratios measured by line data 2 and line data 3 were 5.6
to a higher wavenumber of 1350 cm− 1 for the GO/S-PANI spectra. It and 5.2, respectively, also confirmed that GO got partially into the
resulted from the interaction between the cationic radical species finger-like pore layer of the PANI membrane, which benefits the tight
(C–N+) of PANI and the anionic radical species (COO-) of GO [58]. binding of the GO layer with the PANI membrane.
Furthermore, compared with PANI, the G band of the GO/S-PANI was Surface roughness is crucial in controlling membrane fouling miti­
sharper and moved from 1585 cm− 1 to 1573 cm− 1 caused by the reso­ gation, and it can be provided by AFM characterization. As illustrated in
nance of the paired electrons N atoms in PANI with the adjacent benzene Fig. 5, significant differences existed in the 3D AFM images of the PANI
structures in GO [59]. Based on the above discussions, it can be and GO/S-PANI membranes. For PANI membrane, a uniform ridge-and-
confirmed that GO was successfully fabricated and assembled on the valley morphology could be seen (Fig. 5a), and it was the typical pattern
PANI membrane tightly. Besides, PANI was also converted from a of PANI membrane fabricated by the NIPS method [18]; its
semiconductor (emeraldine base) into a conductor (emeraldine salt) root-mean-square roughness (rms) and the average roughness (Ra) were
during the electrochemical process (Fig. 4g). ~ 31.3 nm and ~ 27.8 nm, respectively. When GO was combined with
Phase structures of GO, PANI, and GO/S-PANI were investigated by the PANI membrane, a rugged ridge-and-valley morphology for
XRD, as shown in Fig. 4c. The sharp peak for GO at 2θ = 11◦ corre­ GO/S-PANI membrane (Fig. 5b) appeared, and the rms roughness and
sponded to the GO (001) crystal plane, indicating the presence of GO the Ra roughness were increased to ~ 47.5 nm and ~ 40.2 nm,
[48]. Furthermore, the peak observed at 2θ = 26.5◦ (002) suggested the respectively. In the cross-flow mode, protein pollution tends to deposit
unoxidized state of graphene [60]. The XRD structure of the PANI on hydrophobic surfaces and results in reduced membrane reversibility,
sample was similar to that prepared by Sanches et al. [61]. For thus great hydrophilicity can help to alleviate membrane fouling [66].
GO/S-PANI, the diffraction peaks at 2θ = 18.0◦ , 2θ = 22.8◦ , and 2θ = As confirmed by the Wenzel relation [67], roughness can enhance
25.8◦ were exhibited, corresponding to the (011), (020), and (200) wettability that a hydrophilic solid (θ < 90◦ ) turns into more hydrophilic
crystal planes of the emeraldine salt form of PANI, confirming the when rough. Therefore, the hydrophilicity can be improved after the

6
B. Li et al. Journal of Membrane Science 640 (2021) 119844

Fig. 5. SEM images of the PANI and GO/S-PANI membranes: top surface magnification 500 × (a), 2000 × (b), 5000 × (c), and cross-section magnification 250 × (d).
EDX spectrum analysis: surface layer (Line data 1), finger-like pore up-layer (Line data 2), finger-like pore sub-layer (Line data 3). 3D AFM images of the membranes:
PANI membrane (a), GO/S-PANI membrane (b).

layers can enhance the overall membrane mechanical properties.

Table 1
Pore size and pore size distribution (PSD) of the PANI and GO/S-
Membrane cross-sectional EDX spectrum line scanning data.
PANI membranes were given in Table 2 and Fig. 6b, respectively. As
Mass percentage Pure PANI GO/S-PANI the crack-like pores existed on the PANI membrane surface, a wide pore
Position Sample Line data 1 Line data 2 Line data 3 size distribution ranged from 0.27 μm to 6.11 μm were detected.
C (wt%) 83.2 89.2 84.9 83.9 Nonetheless, the average pore diameter for the PANI membrane was
N (wt%) 16.8 10.8 15.1 16.1 only 1.18 μm. On the other hand, impacted by the combined GO, a
C/N ratio 5.0 8.3 5.6 5.2 constrictive pore size distribution from 0.16 μm to 2.31 μm with an
average pore diameter of 0.84 μm was acquired by the GO/S-PANI
membrane. Once the GO was assembled on the PANI membrane and
PANI membrane (θ = 75.9◦ ) was utilized to prepare the GO/S-PANI
formed a layer, the surface crack-like pores were significantly reduced,
membrane, as the rugged ridge-and-valley morphology on GO/S-PANI
which improves the membrane rejection of yeast pollutants and brings
can enhance the membrane surface roughness. Moreover, the rugged
about a more stable pore structure. Besides, the introduction of GO did
ridge-and-valley morphology can generate local agitation during
not affect porosity, and it only reduced from 82.94% for the PANI
cross-flow filtration, further limiting blockage and increasing the flux by
membrane to 78.24% for the GO/S-PANI membrane.
creating instability [68].
Membrane hydrophilicity can be obtained by the dynamic contact
angle of water on the membrane. The difference in change of water
3.1.3. Membrane properties
contact angles with time gives a quantitative indication of the mem­
The tensile strength and elongation at break of the PANI and GO/S-
branes. Based on Fig. 6c, the water contact angle changed from 81.3◦ to
PANI membranes were shown in Fig. 6a. Compared to the PANI mem­
75.9◦ for PANI membrane and from 40.2◦ to 37.1◦ for GO/S-PANI
brane, a slight decrease in mechanical strength for the GO/S-PANI
membrane. The S-PANI membrane was also utilized to study the effect
membrane was caused by the acid doping in the electrochemical pro­
of acid doping on hydrophilicity, and the obtained result was 65.4◦ at
cess, which was confirmed by Xu et al. [69] that acid doping could affect
the original time and 60.7◦ in the end (Fig. S8a). By comparing the final
the intermolecular reconfiguration of the PANI backbone, causing the
hydrophilicity of the S-PANI membrane (60.7◦ ) and GO/S-PANI mem­
reduction in membrane strength. The PANI membrane, which was only
brane (37.1◦ ), a conclusion can be drawn that the GO layer was highly
immersed in sulfuric acid (1.1 mol/L) for 2 h, was named the S-PANI
hydrophilic and had a strong impact on the hydrophilicity of the GO/S-
membrane and was also tested as a comparison (Fig. S7). The obtained
PANI membrane. According to AFM, GO layers made the GO/S-PANI
results for the tensile strength and elongation at break were lower than
membrane surface rougher than that of the PANI membrane, which
that of the GO/S-PANI membrane, indicating that the composite GO

7
B. Li et al. Journal of Membrane Science 640 (2021) 119844

Fig. 6. Properties of PANI and GO/S-PANI membranes: (a) Mechanical properties: Tensile strength and elongation at break (%). (b) Pore size distribution. (c)
Dynamic water contact angles. (d) Membrane conductivity.

3.2. Electrofiltration experiment

Table 2
Pore sizes and Porosity of the PANI and GO/S-PANI membranes.
3.2.1. Effect of different low voltages on water permeability and rejection
Pore size Maximum (μm) Mean (μm) Minimum (μm) Porosity (%) rate
Membrane
Contrast experiments for membrane permeability have been
PANI 6.11 ± 0.21 1.18 ± 0.09 0.27 ± 0.04 82.94 ± 1.73 explored at low voltages of 1 V, 2 V, and 3 V. Experiments were carried
GO/S-PANI 2.31 ± 0.14 0.84 ± 0.11 0.16 ± 0.07 78.24 ± 2.18 out using 2 g L− 1 yeast suspension as the feed. From Fig. 7a, for GO/S-
PANI membrane under 1 V, 2 V, and 3 V, the average permeation
might be another reason for the increase in membrane hydrophilicity. fluxes were 190.16 L m− 2 h− 1, 194.78 L m− 2 h− 1, and 203.06 L m− 2 h− 1,
The conductivities of the membranes wetted with deionized water which were higher than obtained without voltages applied (68.37 L m− 2
have been shown in Fig. 6d. The membrane reaching high conductivity h− 1). The rejection rate is also an important index to the membrane
was mainly provided by acid doping, and this could be confirmed from performance and is summarized in Fig. 7b. Due to the deformation under
the comparison of conductivities: the conductivity was enhanced from pressure filtration, yeast pollutants can pass through the membrane
0.019 S m− 1 for the PANI membrane to 58.1 S m− 1 for the S-PANI pores which were smaller than themselves. For the PANI membrane,
membrane (Fig. S8b), after GO was assembled on the membrane, the resulting from the existence of crack-like pores on the membrane sur­
conductivity was slightly decreased to 55.6 S m− 1. face, the rejection rates of yeast particles were only 70.4% under 0 V and
Cyclic voltammetry (CV) was carried out to determine whether redox 75.2% under 1 V. For GO/S-PANI membrane, the rejection rates had
reactions occurred on the membrane surface. The CV scan for the GO/S- been significantly improved to 87.5% when no voltage applied, and
PANI membrane in Fig. S9a revealed that no redox peak, proving that slightly increased to 90.2%, 92.0%, and 93.1% under 1 V, 2 V, and 3 V
the electrooxidation did not happen on the membrane surface during respectively, indicating that the electrochemically designed GO layer
electrofiltration. As schemed in Fig. S9b, the obtained surface zeta po­ can enhance membrane rejection for yeasts.
tential was − 35.4 ± 1.4 mV for the PANI membrane and − 52.9 ± 5.1
mV for the GO/S-PANI membrane indicating that the GO/S-PANI 3.2.2. Effect of applied and unapplied low voltage on water permeability
membrane owned favorable electronegativity which could generate Considering the requirements of rejection and energy-efficient,
electrostatic repulsion to yeast pollutants (Zeta potential: − 12.9 mV). water permeability under 1 V applied or 0 V was depicted in Figs. 7c-
f. As observed in Fig. 7c, semi-conducting PANI membrane limited its
use in electrofiltration, resulting in insignificant changes. The total
permeation flux for 1 V applied was only 1.10 times of those without
voltage. Besides, for the tests without and with 1 V applied, the lowest
fluxes in the third cycle were 7.81 L m− 2 h− 1 and 12.33 L m− 2 h− 1,

8
B. Li et al. Journal of Membrane Science 640 (2021) 119844

Fig. 7. (a) Variation of permeation flux of GO/S-PANI membrane in three cycles under applied 1 V, 2 V, and 3 V, respectively. (b) The rejection rate of membranes
under applied different electric fields. (c–f) Variation of permeation flux of PANI and GO/S-PANI membranes, with 0 V or 1 V applied in three cycles. (g) Schematic
illustration of electrical repulsion and water permeability enhancement.

respectively. Membrane permeability decreased with the clogging of the 177.01%. As observed through the q value of each cycle in Table 3, the
yeasts to the membranes over time. Consequently, poor membrane permeation stability for the GO/S-PANI membrane demonstrated better
conductivity led to the unpractical efficacy of the electrofiltration. While results.
the GO/S-PANI membrane showed better results as presented in Fig. 7d.
The average flux increased greatly from 68.37 L m− 2 h− 1 without elec­ 3.2.3. Effect of GO layer on GO/S-PANI membrane
tric field to 190.16 L m− 2 h− 1 with 1 V. It could still reach 115.41 L m− 2 The S-PANI membrane was also tested to filter 2 g L− 1 yeast sus­
h− 1 at the end of the third cycle, indicating the enhancement of mem­ pension as a comparison. For the S-PANI membrane and GO/S-PANI
brane permeability during the electrofiltration process. membrane, an inconspicuous difference was displayed between their
As shown in Fig. 7e, the downtrend was analogous without electric permeation fluxes (Fig. 8a). However, the rejection rates were signifi­
fields for both membranes. Although the crack-like pores and the rela­ cantly different (Fig. 8c). The rejection rate for the GO/S-PANI mem­
tively higher porosity could be helpful to gain a greater flux for the PANI brane was 87.5%, and it was 72.9% for the S-PANI membrane, proving
membrane, the rejection rate obtained was only 70.4%. Better hydro­ the improvement in the rejection rate by the formed GO layer on the
philicity and larger electronegativity for the GO/S-PANI membrane
could enhance its resistance to yeast pollutants, impelling the perme­
Table 3
ation flux of the GO/S-PANI membrane to maintain relative stability.
The increased rate of total permeation flux (q) for membranes.
Also, benefit from GO layers, the rejection rate could reach 87.5%.
Fig. 7f contrasted the flux curves under 1 V for the PANI membrane Membrane 1st cycle 2nd cycle 3rd cycle Average
sample
and GO/S-PANI membrane. The average permeation flux of the GO/S-
PANI membrane was 1.97 times that of the PANI membrane. GO/S-PANI 191.07% ± 176.66% ± 163.30% ± 177.01% ±
membrane 14.12% 18.24% 12.27% 14.87%
Compared to the PANI membrane with a q value of 38.85%, the
PANI 69.38% ± 38.11% ± 9.05% ± 38.85% ±
permeability for the GO/S-PANI membrane gained a q value of membrane 11.65% 12.72% 7.94% 10.77%

9
B. Li et al. Journal of Membrane Science 640 (2021) 119844

Fig. 8. (a) Variation of permeation flux of S-PANI and GO/S-PANI membranes under 0 V. (b) Variation of permeation flux of S-PANI and GO/S-PANI membranes
under 1 V. (c) The rejection rate of membranes under 0 V. (d) The rejection rate of membranes under 1 V.

GO/S-PANI membrane. generated; the wider the channel is, the larger the volume for the
According to Fig. 8b, by adjusting the source to 1 V, the permeation containment of water molecules is. The great volume can weaken the
flux of the GO/S-PANI membrane was 1.30 times that of the S-PANI viscous interfacial resistance to water transport. In our electrochemical
membrane. The results can be attributed to two aspects: (1) the hydro­ strategy, the GO sheets in membrane fabrication were self-assembled,
philic groups on GO allowed water molecules to embed the GO inter­ making GO sheets stacked in the loosest manner of the three ways (i.
layer structure quickly [70]; (2) the generated electrostatic repulsion e., oxygen-containing groups face to face, pristine graphene facing each
assisted in alleviating pollution [5]. With 1 V applied (Fig. 8d), the other, and oxygen-containing groups face to pristine graphene [40]).
rejection rate for the S-PANI membrane was only 76.1% and could still The interlayer structure of GO layers was mainly determined by the
be around 90.2% for the GO/S-PANI membrane, indicating the stable existing oxygen-containing groups, and this stacking way could form
rejection rate for the GO/S-PANI membrane. Compared to the S-PANI expanded layer spacing and more significant volume nanochannels.
membrane, the GO/S-PANI membrane displayed greater appositeness Also, these hydrophilic groups on GO allowed water molecules to embed
towards electrofiltration and showed improved performance. the GO interlayer structure quickly, and water molecules could go
through GO by taking a flexuous pathway inside the vacant space of
3.2.4. Description of factors influencing and controlling water permeability oxidized regions within GO sheets [70,71]. Besides, the GO layer
Since the CV scan showed no redox peaks, it suggested that the induced an increase in the electronegativity of the membrane surface,
electrooxidation did not occurred while the electrical repulsion played a thereby enhancing the electrostatic repulsion to yeasts. Based on the
decisive role in the electrofiltration process. During this process, the above conditions, the as-prepared GO/S-PANI membrane presented a
membrane surface exhibited negatively charged when the cathode good water permeability and contaminant rejection.
conductor was powered up. Thus, the electrostatic repulsive force was
generated between membrane surface and yeast pollutants, resulting in 3.2.5. Stability of GO/S-PANI membrane
less readily adsorbed/remained yeasts on the membrane surface, the The GO/S-PANI membranes were achieved after electrofiltration
permeation flux was enhanced (Fig. 7g). Moreover, the GO layer was full tests under 0 V and 1 V, and membrane stability was investigated from
of hydrophilic hydroxyl and carboxyl groups, forming nanochannels four aspects: SEM image, water contact angle, pore size, and conduc­
between the GO sheets and allowing water molecules to pass through tivity. Compared with at 1 V (Fig. 9b), the GO/S-PANI membrane at 0 V
them quickly [14,23]. According to the literature [40], the stacking (Fig. 9a) showed more distinct yeast aggregates (green circles). Besides,
method of adjacent GO sheets affects the interlayer nanostructure, GO sheets were partially lost at 0 V, leading to the appearance of crack-
which can control water transport among neighboring GO layers. The like holes (red circles). The above conclusions were also confirmed by
larger the interlayer spacing is, the more water permeation channels are membrane hydrophilicity (Fig. 9c) and membrane pore sizes (Fig. 9d);

10
B. Li et al. Journal of Membrane Science 640 (2021) 119844

Fig. 9. (a) SEM image of the GO/S-PANI membrane magnification 500 × after tests under 0 V. (b) SEM image of the GO/S-PANI membrane magnification 500 ×
after tests under 1 V. (c) Water contact angles of the GO/S-PANI membrane after tests under 0 V and 1 V. (d) Pore sizes of the GO/S-PANI membrane after tests under
0 V and 1 V.

both values increased compared with the original membrane. The water adsorb the GO sheets on the membrane. Acted by the tangential force of
contact angle for 1 V was 41.2◦ and was 50.4◦ for 0 V; meanwhile, the water flow, the yeast-adhered GO sheets were partly lost from the
pore size (Maximum) showed 2.84 μm and 3.25 μm respectively, sug­ membrane surface. In the absence of electric fields, membrane hydro­
gesting the membrane was relatively stable at 1 V. To verify the possible philicity and electrostatic repulsion can partially mitigate yeast
acid leaching from the membrane, the conductivities of the GO/S-PANI adsorption. While as the gradual increase in the loss of GO sheets, above
membrane were tested after each cycle (Fig. S10). Reduced conductiv­ membrane properties weakened, resulting in membrane fouling and
ities were observed at both 0 V and 1 V with the increase of cycle stability decreasing. Under the electric field, the continuous electric
numbers, indicating acid leaching occurred during the electrofiltration repulsion greatly reduced yeast adsorption. The added copper ions also
processes. Membrane conductivities decreased more when at 0 V than at stabilized the GO layers [45], contributing to GO layers being more
1 V, reflecting that the conductive stability for the GO/S-PANI mem­ stable.
brane was better with the utilization of voltage.
The GO/S-PANI membrane, with rugged ridge-and-valley 3.3. Antibacterial activity of membrane
morphology (Fig. 5), was conducive to the formation of local agita­
tion, which could reduce the adsorption potential of yeasts [68]. How­ The antibacterial ability of membranes was tested using Escherichia
ever, as the test time progressed, larger yeast particles (green circles in coli (E. Coli). As shown in Fig. 10, compared with the control plate, the
Fig. 9a, Fig. S4) could be formed through the agglomeration of yeast number of E. coli colonies on the plate treated by the PANI membrane
particles in the feed, which made an increased affinity of yeasts to displayed no distinct changes. However, the plate treated by GO/S-PANI

Fig. 10. Photographs of E. coli colonies on the agar plate: the control plate (a), PANI membrane (b), and GO/S-PANI membrane (c).

11
B. Li et al. Journal of Membrane Science 640 (2021) 119844

membrane was significantly decreased, indicating antibacterial ability Appendix A. Supplementary data
was enhanced by the GO/S-PANI membrane. The antibacterial rate
could be used to analyze the antibacterial activity of membranes Supplementary data to this article can be found online at https://doi.
quantitatively. The rate reached 92.1% for the GO/S-PANI membrane, org/10.1016/j.memsci.2021.119844.
which implied that the membrane had excellent antibacterial ability.
The mechanism of antibacterial activity of the GO/S-PANI mem­ References
brane could be explained as follows: (1) After the direct contact with GO
sheets, the destruction of cell structures could happen due to the shear [1] M.A. Shannon, P.W. Bohn, M. Elimelech, J.G. Georgiadis, B.J. Mariñas, A.
M. Mayes, Science and technology for water purification in the coming decades,
stress of GO sheets, leading to irreversible damage of bacterial cells [24]. Nature 452 (2008) 301–310.
(2) As the electrostatic adherence between Cu2+ and negatively charged [2] H.-C. Flemming, Microbial biofouling: unsolved problems, insufficient approaches,
bacteria, the bacteria wall might be destroyed and leaked out the con­ and possible solutions, in: Biofilm Highlights, Springer, 2011, pp. 81–109.
[3] A. Ronen, W. Duan, I. Wheeldon, S.L. Walker, D. Jassby, Microbial attachment
tents in bacteria, which finally led the bacteria to die, thereby the inhibition through low voltage electrochemical reactions on electrically conducting
formed GO/Cu2+ on the GO/S-PANI membrane could greatly increase membranes, Environ. Sci. Technol. 49 (2015) 12741–12750.
the antibacterial activity than that of the pure GO [46,72]. (3) As the [4] G. Chen, Electrochemical technologies in wastewater treatment, Separ. Purif.
Technol. 38 (2004) 11–41.
GO/S-PANI membrane was doped by sulfuric acid, the acid dopant on [5] P. Formoso, E. Pantuso, G. Filpo, F. Nicoletta, Electro-conductive membranes for
the PANI chains could contact the E. Coli, resulting in bacterial death permeation enhancement and fouling mitigation: a short review, Membranes 7
[73]. Overall, the GO/S-PANI membrane can be considered as a po­ (2017) 39.
[6] M. Sun, X. Wang, L.R. Winter, Y. Zhao, W. Ma, T. Hedtke, J.-H. Kim, M. Elimelech,
tential material for antibacterial.
Electrified Membranes for Water Treatment Applications, ACS ES&T Engineering,
2021.
4. Conclusion [7] J. Sun, C. Hu, B. Wu, J. Qu, Fouling mitigation of a graphene hydrogel membrane
electrode by electrical repulsion and in situ self-cleaning in an electro-membrane
reactor, Chem. Eng. J. 393 (2020) 124817.
Through electrochemical strategy, the graphene oxide/sulfuric acid [8] L. Liu, F. Zhao, J. Liu, F. Yang, Preparation of highly conductive cathodic
doped polyaniline (GO/S-PANI) membrane was readily fabricated. membrane with graphene (oxide)/PPy and the membrane antifouling property in
Copper ions were both the driving agent for electrochemical membrane- filtrating yeast suspensions in EMBR, J. Membr. Sci. 437 (2013) 99–107.
[9] B.P. Chaplin, The prospect of electrochemical technologies advancing worldwide
formation and the stabilizer for GO layers. The conductivity of the GO/S- water treatment, Acc. Chem. Res. 52 (2019) 596–604.
PANI membrane (55.6 S m− 1) was three orders of magnitude higher than [10] C. Thamaraiselvan, A. Ronen, S. Lerman, M. Balaish, Y. Ein-Eli, C.G. Dosoretz, Low
that of the PANI membrane (0.019 S m− 1). Affected by the hydrophilic voltage electric potential as a driving force to hinder biofouling in self-supporting
carbon nanotube membranes, Water Res. 129 (2018) 143–153.
GO layer, the GO/S-PANI membrane became highly hydrophilic with a [11] T.X. Huong Le, L.F. Dumée, S. Lacour, M. Rivallin, Z. Yi, L. Kong, M. Bechelany,
water contact angle of 37.1◦ . The surface zeta potential decreased from M. Cretin, Hybrid graphene-decorated metal hollow fibre membrane reactors for
− 35.4 ± 1.4 mV (PANI membrane) to − 52.9 ± 5.1 mV (GO/S-PANI efficient electro-Fenton - filtration co-processes, J. Membr. Sci. 587 (2019) 117182.
[12] K. Wang, L. Xu, K. Li, L. Liu, Y. Zhang, J. Wang, Development of polyaniline
membrane), providing an increased membrane resistance to negatively conductive membrane for electrically enhanced membrane fouling mitigation,
charged yeasts (Zeta potential: − 12.9 mV). The pore structure of the J. Membr. Sci. 570–571 (2019) 371–379.
developed membrane was more favorable for rejecting yeast pollutants [13] L. Liu, J. Liu, B. Gao, F. Yang, S. Chellam, Fouling reductions in a membrane
bioreactor using an intermittent electric field and cathodic membrane modified by
(2–5 μm). Under low-intensity electric fields, the GO/S-PANI membrane
vapor phase polymerized pyrrole, J. Membr. Sci. 394–395 (2012) 202–208.
showed better antifouling performance and preferable water perme­ [14] K.G. Zhou, K.S. Vasu, C.T. Cherian, M. Neek-Amal, J.C. Zhang, H. Ghorbanfekr-
ability than the PANI membrane. For the electrofiltration process, the Kalashami, K. Huang, O.P. Marshall, V.G. Kravets, J. Abraham, Y. Su, A.
GO/S-PANI membrane demonstrated greater stability than without N. Grigorenko, A. Pratt, A.K. Geim, F.M. Peeters, K.S. Novoselov, R.R. Nair,
Electrically controlled water permeation through graphene oxide membranes,
electric fields. The improved antibacterial rate was shown by the GO/S- Nature 559 (2018) 236–240.
PANI membrane, which can be against E. Coli up to 92.1%. This study [15] Q. Zhang, P. Arribas, E.M. Remillard, M.C. García-Payo, M. Khayet, C.D. Vecitis,
provides a strategy for facile preparation of the GO/S-PANI conductive Interlaced CNT electrodes for bacterial fouling reduction of microfiltration
membranes, Environ. Sci. Technol. 51 (2017) 9176–9183.
membrane with application potential in electrofiltration and antibac­ [16] N. Li, L. Liu, F. Yang, Highly conductive graphene/PANi-phytic acid modified
terial fields. cathodic filter membrane and its antifouling property in EMBR in neutral
conditions, Desalination 338 (2014) 10–16.
[17] A. Ronen, S.L. Walker, D. Jassby, Electroconductive and electroresponsive
CRediT authorship contribution statement membranes for water treatment, Rev. Chem. Eng. 32 (2016) 533–550.
[18] A. Sarihan, S. Shahid, J. Shen, I. Amura, D.A. Patterson, E.A.C. Emanuelsson,
Bojun Li: Conceptualization, Methodology, Formal analysis, Visu­ Exploiting the electrical conductivity of poly-acid doped polyaniline membranes
with enhanced durability for organic solvent nanofiltration, J. Membr. Sci. 579
alization, Investigation, Writing – original draft. Wenjing Tang: Inves­ (2019) 11–21.
tigation, Formal analysis, Validation, Software. De Sun: Project [19] E.L. Subtil, J. Gonçalves, H.G. Lemos, E.C. Venancio, J.C. Mierzwa, J. dos Santos de
administration, Conceptualization, Funding acquisition, Resources. Souza, W. Alves, P. Le-Clech, Preparation and characterization of a new composite
conductive polyethersulfone membrane using polyaniline (PANI) and reduced
Bingbing Li: Supervision, Writing – review & editing. Yanxia Ge:
graphene oxide (rGO), Chem. Eng. J. 390 (2020) 124612.
Validation. Xin Ye: Software. Wei Fang: Visualization. [20] L.L. Xu, S. Shahid, D.A. Patterson, E.A.C. Emanuelsson, Flexible electro-responsive
in-situ polymer acid doped polyaniline membranes for permeation enhancement
and membrane fouling removal, J. Membr. Sci. 578 (2019) 263–272.
Declaration of competing interest
[21] I.F. Amura, S. Shahid, A. Sarihan, J. Shen, D.A. Patterson, E.A.C. Emanuelsson,
Fabrication of self-doped sulfonated polyaniline membranes with enhanced
The authors declared that they have no conflicts of interest to this antifouling ability and improved solvent resistance, J. Membr. Sci. 620 (2021)
work. 117712.
[22] H. Alhweij, E.A.C. Emanuelsson, S. Shahid, J. Wenk, Simplified in-situ tailoring of
We declare that we do not have any commercial or associative in­ cross-linked self-doped sulfonated polyaniline (S-PANI) membranes for
terest that represents a conflict of interest in connection with the work nanofiltration applications, J. Membr. Sci. 637 (2021) 119654.
submitted. [23] H. Wang, W. Wang, L. Wang, B. Zhao, Z. Zhang, X. Xia, H. Yang, Y. Xue, N. Chang,
Enhancement of hydrophilicity and the resistance for irreversible fouling of
polysulfone (PSF) membrane immobilized with graphene oxide (GO) through
Acknowledgment chloromethylated and quaternized reaction, Chem. Eng. J. 334 (2018) 2068–2078.
[24] S. Liu, T.H. Zeng, M. Hofmann, E. Burcombe, J. Wei, R. Jiang, J. Kong, Y. Chen,
Antibacterial activity of graphite, graphite oxide, graphene oxide, and reduced
This work was supported by the Chinese National Key R&D Plan graphene oxide: membrane and oxidative stress, ACS Nano 5 (2011) 6971–6980.
[grant numbers 2018YFB0504600, 2018YFB0504603]; the Foundation [25] B. Shi, H. Wu, J. Shen, L. Cao, X. He, Y. Ma, Y. Li, J. Li, M. Xu, X. Mao, M. Qiu,
of Jilin Educational Committee, Jilin, China [grant number H. Geng, P. Yang, Z. Jiang, Control of edge/in-plane interactions toward robust,
JJKH20191318KJ].

12
B. Li et al. Journal of Membrane Science 640 (2021) 119844

highly proton conductive graphene oxide membranes, ACS Nano 13 (2019) [49] S. Eigler, S. Grimm, F. Hof, A. Hirsch, Graphene oxide: a stable carbon framework
10366–10375. for functionalization, J. Mater. Chem. A 1 (2013) 11559–11562.
[26] L. Chen, G. Shi, J. Shen, B. Peng, B. Zhang, Y. Wang, F. Bian, J. Wang, D. Li, [50] J. Tang, X. Jing, B. Wang, F. Wang, Infrared spectra of soluble polyaniline, Synth.
Z. Qian, G. Xu, G. Liu, J. Zeng, L. Zhang, Y. Yang, G. Zhou, M. Wu, W. Jin, J. Li, Met. 24 (1988) 231–238.
H. Fang, Ion sieving in graphene oxide membranes via cationic control of interlayer [51] G. Wang, W. Xing, S. Zhuo, The production of polyaniline/graphene hybrids for use
spacing, Nature 550 (2017) 380–383. as a counter electrode in dye-sensitized solar cells, Electrochim. Acta 66 (2012)
[27] L. Jianhua, A. Junwei, Z. Yecheng, M. Yuxiao, L. Mengliu, Y. Mei, L. Songmei, 151–157.
Preparation of an amide group-connected graphene–polyaniline nanofiber hybrid [52] I. Šeděnková, M. Trchová, J. Stejskal, Thermal degradation of polyaniline films
and its application in supercapacitors, ACS Appl. Mater. Interfaces 4 (2012) prepared in solutions of strong and weak acids and in water – FTIR and Raman
2870–2876. spectroscopic studies, Polym. Degrad. Stabil. 93 (2008) 2147–2157.
[28] R.-X. Wang, L.-F. Huang, X.-Y. Tian, Understanding the protonation of polyaniline [53] X. Feng, N. Chen, J. Zhou, Y. Li, Z. Huang, L. Zhang, Y. Ma, L. Wang, X. Yan, Facile
and polyaniline–graphene interaction, J. Phys. Chem. C 116 (2012) 13120–13126. synthesis of shape-controlled graphene–polyaniline composites for high
[29] Q. Zhang, Y. Li, Y. Feng, W. Feng, Electropolymerization of graphene oxide/ performance supercapacitor electrode materials, New J. Chem. 39 (2015)
polyaniline composite for high-performance supercapacitor, Electrochim. Acta 90 2261–2268.
(2013) 95–100. [54] Y. Furukawa, F. Ueda, Y. Hyodo, I. Harada, T. Nakajima, T. Kawagoe, Vibrational
[30] T. Rezazadeh, N. Dalali, N. Sehati, Investigation of adsorption performance of spectra and structure of polyaniline, Macromolecules 21 (1988) 1297–1305.
graphene oxide/polyaniline reinforced hollow fiber membrane for [55] Z. Wu, S. Zhu, X. Dong, Y. Yao, Y. Guo, S. Gu, Z. Zhou, A facile method to graphene
preconcentration of Ivermectin in some environmental samples, Spectrochim. Acta oxide/polyaniline nanocomposite with sandwich-like structure for enhanced
A 204 (2018) 409–415. electrical properties of humidity detection, Anal. Chim. Acta 1080 (2019)
[31] H. Kweon, C.-W. Lin, M.M. Faruque Hasan, R. Kaner, G.N. Sant, Highly permeable 178–188.
polyaniline–graphene oxide nanocomposite membranes for CO2 separations, ACS [56] H. Wang, Q. Hao, X. Yang, L. Lu, X. Wang, Graphene oxide doped polyaniline for
Appl. Polym. Mater 1 (2019) 3233–3241. supercapacitors, Electrochem. Commun. 11 (2009) 1158–1161.
[32] V. Shalini, M. Navaneethan, S. Harish, J. Archana, S. Ponnusamy, H. Ikeda, [57] P. Pachfule, D. Shinde, M. Majumder, Q. Xu, Fabrication of carbon nanorods and
Y. Hayakawa, Design and fabrication of PANI/GO nanocomposite for enhanced graphene nanoribbons from a metal–organic framework, Nat. Chem. 8 (2016)
room-temperature thermoelectric application, Appl. Surf. Sci. 493 (2019) 718–724.
1350–1360. [58] M. Hassan, K.R. Reddy, E. Haque, S.N. Faisal, S. Ghasemi, A.I. Minett, V.G. Gomes,
[33] W. Jr, R. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc. 80 (1958). Hierarchical assembly of graphene/polyaniline nanostructures to synthesize free-
[34] R. Muzyka, M. Kwoka, Ł. Smędowski, N. Díez, G. Gryglewicz, Oxidation of graphite standing supercapacitor electrode, Compos. Sci. Technol. 98 (2014) 1–8.
by different modified Hummers methods, N. Carbon Mater. 32 (2017) 15–20. [59] D. Shao, G. Hou, J. Li, T. Wen, X. Ren, X. Wang, PANI/GO as a super adsorbent for
[35] S. Lee, J. Oh, R.S. Ruoff, S. Park, Residual acetone produces explosives during the the selective adsorption of uranium(VI), Chem. Eng. J. 255 (2014) 604–612.
production of graphite oxide, Carbon 50 (2012) 1442–1444. [60] K. Urbas, M. Aleksandrzak, M. Jedrzejczak, M. Jedrzejczak, R. Rakoczy, X. Chen,
[36] J. Cao, P. He, M.A. Mohammed, X. Zhao, R.J. Young, B. Derby, I.A. Kinloch, R.A. E. Mijowska, Chemical and magnetic functionalization of graphene oxide as a route
W. Dryfe, Two-step electrochemical intercalation and oxidation of graphite for the to enhance its biocompatibility, Nanoscale Res. Lett. 9 (2014) 656.
mass production of graphene oxide, J. Am. Chem. Soc. 139 (2017) 17446–17456. [61] E.A. Sanches, J.M.S. da Silva, J.M. de O. Ferreira, J.C. Soares, A.L. dos Santos,
[37] S. Pei, Q. Wei, K. Huang, H.-M. Cheng, W. Ren, Green synthesis of graphene oxide G. Trovati, E.G.R. Fernandes, Y.P. Mascarenhas, Nanostructured Polyaniline
by seconds timescale water electrolytic oxidation, Nat. Commun. 9 (2018). Emeraldine-base form (EB-PANI): a structural investigation for different
[38] S. Fang, Y. Lin, Y.H. Hu, Recent advances in green, safe, and fast production of neutralization times, J. Mol. Struct. 1074 (2014) 732–737.
graphene oxide via electrochemical approaches, ACS Sustain. Chem. Eng. 7 (2019) [62] X. Wang, L. Miao, C. Liu, J. Gao, Y. Chen, Thermoelectric enhancement of
12671–12681. polyaniline grafting from graphene oxide, Mater. Sci. Forum 847 (2016) 153–160.
[39] B. Mi, Graphene oxide membranes for ionic and molecular sieving, Science 343 [63] J. Kang, J. Sheng, J. Xie, H. Ye, J. Chen, X.-Z. Fu, G. Du, R. Sun, C.-P. Wong,
(2014) 740. Tubular Cu(OH)2 arrays decorated with nanothorny Co–Ni bimetallic carbonate
[40] W. Li, L. Zhang, X. Zhang, M. Zhang, T. Liu, S. Chen, Atomic insight into water and hydroxide supported on Cu foam: a 3D hierarchical core–shell efficient
ion transport in 2D interlayer nanochannels of graphene oxide membranes: electrocatalyst for the oxygen evolution reaction, J. Mater. Chem. A 6 (2018)
implication for desalination, J. Membr. Sci. 596 (2020) 117744. 10064–10073.
[41] B. Li, D. Sun, B. Li, W. Tang, P. Ren, J. Yu, J. Zhang, One-step electrochemically [64] M.U. Shahid, N.M. Mohamed, A.S. Muhsan, R. Bashiri, A.E. Shamsudin, S.N.
prepared graphene/polyaniline conductive filter membrane for permeation A. Zaine, Few-layer graphene supported polyaniline (PANI) film as a transparent
enhancement by fouling mitigation, Langmuir 36 (2020) 2209–2222. counter electrode for dye-sensitized solar cells, Diam. Relat. Mater. 94 (2019)
[42] Y. Roichman, G.I. Titelman, M.S. Silverstein, A. Siegmann, M. Narkis, Polyaniline 242–251.
synthesis: influence of powder morphology on conductivity of solution cast blends [65] S. Park, R.S. Ruoff, Chemical methods for the production of graphenes, Nat.
with polystyrene, Synth. Met. 98 (1999) 201–209. Nanotechnol. 4 (2009) 217–224.
[43] S.M. Krishna, S. Cheunkar, B. Lertanantawong, R. Thippeswamy, N. D H, [66] W.R. Bowen, T.A. Doneva, J.A.G. Stoton, Protein deposition during cross-flow
W. Surareungchai, B. Geetha, K. Reddy, Graphene oxide-Cu (II) composite membrane filtration: AFM studies and flux loss, Colloids Surf. B Biointerfaces 27
electrode for non-enzymatic determination of hydrogen peroxide, J. Electroanal. (2003) 103–113.
Chem. (2016) 776. [67] D. Quéré, Wetting and roughness, Annu. Rev. Mater. Res. 38 (2008) 71–99.
[44] K. Yang, B. Chen, X. Zhu, B. Xing, Aggregation, adsorption, and morphological [68] C.E. Goodyer, A.L. Bunge, Mass transfer through membranes with surface
transformation of graphene oxide in aqueous solutions containing different metal roughness, J. Membr. Sci. 409–410 (2012) 127–136.
cations, Environ. Sci. Technol. 50 (2016) 11066–11075. [69] L. Xu, S. Shahid, A.K. Holda, E.A.C. Emanuelsson, D.A. Patterson, Stimuli
[45] X.-B. Lv, R. Xie, J.-Y. Ji, Z. Liu, X.-Y. Wen, L.-Y. Liu, J.-Q. Hu, X.-J. Ju, W. Wang, L.- responsive conductive polyaniline membrane: in-filtration electrical tuneability of
Y. Chu, A novel strategy to fabricate cation-cross-linked graphene oxide membrane flux and MWCO, J. Membr. Sci. 552 (2018) 153–166.
with high aqueous stability and high separation performance, ACS Appl. Mater. [70] J. Yin, G. Zhu, B. Deng, Graphene oxide (GO) enhanced polyamide (PA) thin-film
Interfaces 12 (2020) 56269–56280. nanocomposite (TFN) membrane for water purification, Desalination 379 (2016)
[46] S. Zavareh, E. Norouzi, Impregnation of GO with Cu2+ for enhancement of aniline 93–101.
adsorption and antibacterial activity, J. Water Process Eng. 20 (2017) 160–167. [71] P. Bhol, S. Yadav, A. Altaee, M. Saxena, P.K. Misra, A.K. Samal, Graphene-based
[47] M. Chandler, A. Zydney, Effects of membrane pore geometry on fouling behavior membranes for water and wastewater treatment: a review, ACS Appl. Nano Mater.
during yeast cell microfiltration, J. Membr. Sci. 285 (2006) 334–342. 4 (2021) 3274–3293.
[48] M. Ramezanzadeh, M. Asghari, B. Ramezanzadeh, G. Bahlakeh, Fabrication of an [72] Y. Zhou, M. Xia, Y. Ye, C. Hu, Antimicrobial ability of Cu2+-montmorillonite, Appl.
efficient system for Zn ions removal from industrial wastewater based on graphene Clay Sci. 27 (2004) 215–218.
oxide nanosheets decorated with highly crystalline polyaniline nanofibers (GO- [73] N.L. Shi, X.M. Guo, H.M. Jing, J. Gong, C. Sun, K. Yang, Antibacterial effect of the
PANI): Experimental and ab initio quantum mechanics approaches, Chem. Eng. J. conducting polyaniline, J. Mater. Sci. Technol. 22 (2006) 289–290.
337 (2018) 385–397.

13