Fabrication and Characterization of Sulfonated Graphene Oxide-Doped Polymeric Membranes with Improved Anti-Biofouling Behavior

In this study, cellulose acetate (CA) was blended with sulfonated graphene oxide (SGO) nanomaterials to endow a nanocomposite membrane for wastewater treatment with improved hydrophilicity and anti-biofouling behavior. The phase inversion method was employed for membrane fabrication using tetrahydrofuran (THF) as the solvent. The characteristics of CA-SGO-doped membranes were investigated through thermal analysis, contact angle, SEM, FTIR, and anti-biofouling property. Results indicated that anti-biofouling property and hydrophilicity of CA-SGO nanocomposite membranes were enhanced with addition of hydrophilic SGO nanomaterials in comparison to pristine CA membrane. FTIR analysis confirmed the successful decoration of SGO groups on CA membrane surface while revealing its morphological properties through SEM analysis. Thermal analysis performed using DSC confirmed the increase in thermal stability of CA-SGO membranes with addition of SGO content than pure CA membrane.


Introduction
Membrane technology has undergone advances in various fields of application like wastewater treatment, water purification, food, seawater desalination, and medicine, which is attributed to its advantageous features like simplicity, high efficacy, eco-friendly nature, insignificant chemical utilization, and cost effectiveness [1]. In previous years, a great focus has been on the development of polymeric materials with high performance which must possess good hydrophilicity, high permeability, and excellent separation [2]. However, the main obstacle which economically and from technical point of view restricts membrane performance is its "biofouling".
Biofouling is caused by the deposition, attachment, and proliferation of biological foulants (e.g., proteins and bacterial cells) present in feed water on the surface of a membrane, resulting in biofilm formation. Biofouling assists in the concentration polarization (i.e., the ratio among solute concentration on the surface of membranes and inside bulk solution) of nutrients on the membrane surface [3], which consequently blocks the membrane pores, causes tremendous reduction in salt rejection, permeates flux, and increases transmembrane pressure, thus requiring further energy for filtration [4,5]. Therefore, the key focus in context of membrane modification is fabrication of anti-biofouling membranes [6].

Preparation of Graphene Oxide (GO)
Synthesis of Graphene oxide was done using Hummer's method. The procedure involves the following steps: The graphite powder (6 g) was added along with 200 mL sulfuric acid and 2.5 g sodium nitrate into flask with constant stirring. In the next step, 30 g of KMnO 4 (Potassium Permanganate) was then slowly added into flask containing suspension for temperature maintenance up to 5 • C and kept in an ice bath with continuous agitation for 3 h. The ice bath was then removed, and the temperature of suspension increased for 45 min up to 35 • C. Distilled water (200 mL) then subsequently added into suspension with rise in temperature for 20 min up to 98 • C. The color of suspension turned brown. Distilled water (250 mL) was added for suspension dilution followed by the addition of 20 mL of H 2 O 2 (10%) for reduction of unreacted manganese dioxide and permanganate into soluble manganese sulfate. During this process, the suspension turned bright yellow followed by centrifugation of suspension which resulted in a yellow brownish paste. The suspension was washed twice using 95% alcohol and three times with deionized water, yielding graphitic oxide. At last, we sonicated graphitic oxide to generate separate layers known as graphene oxide. The resulting GO placed in oven for drying at 60 • C for 12 h [25].
Sulfonation of GO was achieved using Diazonium salt of Sulfanilic acid. Steps involved in Diazonium salt preparation are as follows: (a) First, sulfanilic acid was dissolved in solution of NaOH (10%) using water bath and little bit heating. (b) Then, the heated solution was allowed to cool at room temperature followed by addition of sodium nitrite (NaNO 2 ). (c) The solution was then placed in an ice bath under continuous stirring followed by addition of 10 mL concentrated HCl and 10 mL cooled water until the formation of white solution. (d) Afterwards, 1 g of the GO suspension was added to 100 mL of distilled water using sonication, which was then added into the above solution of diazonium salt for 4 h with addition of few drops of hydrazine to ensure a bit reduction of GO surface. (e) The suspension so obtained using centrifugation was then washed several times with distilled water (f) Last, the sulfonated graphene oxide (SGO) was obtained [26].

Fabrication of SGO-Cellulose Acetate Membranes
The preparation of the SGO-blended CA membrane was done using non-solventinduced phase inversion (NIPS) [27]. The compositions of casting solutions of CA, THF as solvent, and the desired concentration of SGO nanomaterials are given in Table 1. In the beginning, the precise SGO nanoparticle amount was poured into THF solvent and dispersed for 30 min using a sonicator (ultrasonic water bath) for improvement in homogeneity. On other side, CA was added into the THF solution and under continuous stirring was poured with the dispersion of SGO at room temperature. For better dispersion of SGO, nanomaterials into CA matrix and elimination of air bubbles the solution was kept for 24 h. The casting solution was then ready to be casted using casting glass plate with thickness of (150 µm) and casting knife, and the film so obtained was then submerged into coagulation bath containing distilled water for almost 2-3 min. Afterwards, the prepared film was detached from casting plate by washing three times using pure water. The membrane so obtained was allowed to dry at room temperature and then stored for further use (Figure 1). film was detached from casting plate by washing three times using pure water. The membrane so obtained was allowed to dry at room temperature and then stored for further use ( Figure 1).

Characterization of GO and SGO Nanoparticles and CA-Composite Membranes
FTIR spectra for GO and SGO composite membranes were taken using a Perkin Elmer FTIR spectrometer (Waltham, MA, USA). The morphological features (i.e., top surface and membrane microstructures) of the prepared membranes were characterized using SEM (FEI, Quanta FEG 450, Thermo Scientific, Waltham, MA, USA). The crystalline structure of prepared SGO doped CA membranes was characterized through X-ray diffraction (PANalytical X'pert, Malvern Panalytical, Malvern, UK) technique, respectively.
Thermal properties of SGO doped composite membranes were determined using differential scanning calorimetry (DSC-250, TA instruments, New Castle, DE, USA) at a 10 °C/min heating rate under nitrogen flow of 30 mL/min. Ten milligram samples were sealed in an aluminum pan and scanned over a range of 25 °C to 500 °C [28].

Characterization of GO and SGO Nanoparticles and CA-Composite Membranes
FTIR spectra for GO and SGO composite membranes were taken using a Perkin Elmer FTIR spectrometer (Waltham, MA, USA). The morphological features (i.e., top surface and membrane microstructures) of the prepared membranes were characterized using SEM (FEI, Quanta FEG 450, Thermo Scientific, Waltham, MA, USA). The crystalline structure of prepared SGO doped CA membranes was characterized through X-ray diffraction (PANalytical X'pert, Malvern Panalytical, Malvern, UK) technique, respectively.
Thermal properties of SGO doped composite membranes were determined using differential scanning calorimetry (DSC-250, TA instruments, New Castle, DE, USA) at a 10 • C/min heating rate under nitrogen flow of 30 mL/min. Ten milligram samples were sealed in an aluminum pan and scanned over a range of 25 • C to 500 • C [28].

Measurement of Water Contact Angle, Pure Water Flux, Tensile Strength
The contact angle was measured through Attension Theta Tensiometer using the sessile-drop method. The amount of water used was 5 µL. In this method, a droplet of distilled water was dropped on the membrane surface and then the contact angle of the droplet with surface was calculated. Five random locations of each membrane were selected for each measurement of the contact angle to minimize the investigation error and then the average was reported. Pure water flux of membrane was checked using a cross-filtration apparatus and mechanical properties like tensile strength of fabricated membrane were evaluated using Instron tensile testing machine. Particularly, for each type of membrane the specimens were stretched in unidirection at the rate of 5 mm/min [29]. The pure water flux of SGO membranes was determined using dead end stirred filtration system having membrane active area of~18.5 cm 2 . The filtration cell was mounted to a reservoir containing deionized water and compressed via nitrogen gas. First, the membrane was compacted for 30 min using filtering water at 2 bar (gauge), and then filtration tests were performed at 1 bar (gauge). Permeated water was then collected and weighed out using digital balance. The following formula was employed for the calculation of pure water flux Jw (L m −2 h −1 ) [29]: where V is the permeate volume (L), A is the effective area of membrane (m 2 ) and ∆t is the time of filtration (h).

Antibacterial Testing
Antibacterial testing was conducted using disc diffusion method. The inoculum of bacteria was dispersed homogeneously using nutrient agar on sterile petri dish with the help of sterile cotton swab. In this study, Escherichia coli pathogenic bacteria was used. For testing, each membrane punched in the shape of disks was transferred into a petri dish containing culture of grown bacteria. Afterwards, these plates were placed for incubation at 36 • C ± 1 • C for 24 h, under aerobic conditions. Then, the inhibition zone was measured at the end against each sample [25].

Results and Discussion
The analysis of sample using FTIR was done for the determination of functional groups. Graphene oxide and SGO were analyzed using this spectroscopic technique. The FTIR spectra given in Figure 2 give the comparison among sulfonated graphene oxide (SGO) and GO. GO shows its characteristic peaks at 3000-3389 cm −1 corresponding to hydroxyl (-OH) group and peak at 1731 cm −1 to carbonyl (C=O) groups, whereas at 1210, 1035 cm −1 , 1361 cm −1 is related to stretching vibrations of epoxy (C-O), alkoxy (C-O) groups, and C=C aromatic bonds (un-oxidized sp 2 ), respectively [26]. The peak obtained at 1150 cm −1 corresponds to group, i.e., of sulfonic acid [-SO 3 H], whereas it is absent in GO spectrum. The peak at 1733 cm −1 in spectrum is decreased as in GO, which showed that in process of sulfonation the graphite oxide is reduced partially. The sharp peaks obtained at 1204 cm −1 and 1382 cm −1 reveal asymmetric and symmetric S=O stretching modes, confirming the successful substitution of sulfonic groups upon the GO surface. The peaks obtained close to 823 cm −1 and 875 cm −1 are related to stretching modes of S-C and S-O bonds, respectively, signifying the covalent existence of -SO 3 H on the surface of reduced GO. The reduction in broad peak at 3369 cm −1 suggested the partial reduction of sulfonated GO [30].
The FTIR spectra of pristine CA and CA-SGO nanocomposite membranes at various concentrations (SGO-1 = 0.02%; SGO-2 = 0.06%; SGO-3 = 0.1%; SGO-4 = 0.14%) are shown in Figure 3. Characteristic peaks exhibited by pure CA membrane are at 1037, 1224, 1369, and 1745 cm −1 which corresponds to C-O cm −1 , C-O-C cm −1 , C-CH 3 cm −1 , C=O cm −1 , and O-H cm −1 stretching modes of vibration. The IR absorptions frequency of pristine CA membrane is almost identical to CA-SGO nanocomposite membranes. However, the characteristic peaks exhibited by SGO at 823 and 896 cm −1 related to stretching mode of vibration for S-O and S-C, respectively. The presence of peak at 1739 cm −1 in spectrum for all SGO membranes showed that in process of sulfonation the graphite oxide is reduced partially.     The FTIR spectra of pristine CA and CA-SGO nanocomposite membranes at various concentrations (SGO-1 = 0.02%; SGO-2 = 0.06%; SGO-3 = 0.1%; SGO-4 = 0.14%) are shown in Figure 3. Characteristic peaks exhibited by pure CA membrane are at 1037, 1224, 1369, and 1745 cm -1 which corresponds to C-O cm -1 , C-O-C cm -1 , C-CH3 cm -1 , C=O cm -1 , and O-H cm -1 stretching modes of vibration. The IR absorptions frequency of pristine CA membrane is almost identical to CA-SGO nanocomposite membranes. However, the characteristic peaks exhibited by SGO at 823 and 896 cm -1 related to stretching mode of vibration for S-O and S-C, respectively. The presence of peak at 1739 cm -1 in spectrum for all SGO membranes showed that in process of sulfonation the graphite oxide is reduced partially.  Figure 4a-d depicts the surface morphological features of CA and composite membranes, respectively. We provide a SEM image of pure CA membrane given in Figure 4a with homogeneous surface, whereas Figure 4b-d represents cellulose acetate composite membranes with varying concentrations of SGO. SGO-1 shows membrane with less antibiofouling behavior due to less loading of SGO particles. However, with the increase in concentration of SGO within CA membrane, the anti-biofouling behavior of the corresponding membranes enhances due to reduction in fouling layer on the membrane surface (Figure 4c,d). Furthermore, the SGO-2 and SGO-3 membranes show a smooth surface with fewer particles spreading, meaning that the addition of SGO nanofiller did not crack membrane surface at this concentration. On further increase in concentration, particles start aggregating on membrane surface at intervals which reflect incomplete particles dispersion within polymeric material and low anti-biofouling behavior [31]. membranes with varying concentrations of SGO. SGO-1 shows membrane with less antibiofouling behavior due to less loading of SGO particles. However, with the increase in concentration of SGO within CA membrane, the anti-biofouling behavior of the corresponding membranes enhances due to reduction in fouling layer on the membrane surface (Figure 4c,d). Furthermore, the SGO-2 and SGO-3 membranes show a smooth surface with fewer particles spreading, meaning that the addition of SGO nanofiller did not crack membrane surface at this concentration. On further increase in concentration, particles start aggregating on membrane surface at intervals which reflect incomplete particles dispersion within polymeric material and low anti-biofouling behavior [31].  Figure 5a,b shows the morphological features of SGO with particles sizes determined. The SEM analysis indicates the thickness in the range of 11 to 14 nm for prepared SGO nanoparticles. Moreover, particle size was also calculated through Scherrer calculator feature, which measured particle size of 6-8 nm thus confirms the fabrication of sulfonated graphene oxide [32].  Figure 5a,b shows the morphological features of SGO with particles sizes determined. The SEM analysis indicates the thickness in the range of 11 to 14 nm for prepared SGO nanoparticles. Moreover, particle size was also calculated through Scherrer calculator feature, which measured particle size of 6-8 nm thus confirms the fabrication of sulfonated graphene oxide [32]. The crystallinity of CA-SGO nanocomposite membranes, determined by XRD analysis, reflects the halo amorphous structure as shown in Figure 6. An amorphous halo exits within the range of 15 to 30 • . The existence of broader peaks at 2 theta values of 19.57 and 20.18 corresponds to GO and SGO, respectively. It suggests that all the membranes exist in an amorphous structure. Similar patterns are reported in another work conducted in [3]. The crystallinity of CA-SGO nanocomposite membranes, determined by XRD analysis, reflects the halo amorphous structure as shown in Figure 6. An amorphous halo exits within the range of 15 to 30°. The existence of broader peaks at 2 theta values of 19.57 and 20.18 corresponds to GO and SGO, respectively. It suggests that all the membranes exist in an amorphous structure. Similar patterns are reported in another work conducted in [3]. Contact angle measurement of membranes is widely used for characterizing the membrane surface hydrophilicity. One of the most significant characteristics of membranes is their hydrophilicity because it can influence the anti-biofouling ability and flux of membranes [12]. The highest contact angle of 70° is shown by CA membrane. After adding hydrophilic SGO nanofillers, a decrease in contact angle of composite membranes is observed which suggested better hydrophilicity and improved affinity of water for CA-SGO composite membranes (Figure 7). Furthermore, all CA-SGO membranes exhibited Contact angle measurement of membranes is widely used for characterizing the membrane surface hydrophilicity. One of the most significant characteristics of membranes is their hydrophilicity because it can influence the anti-biofouling ability and flux of membranes [12]. The highest contact angle of 70 • is shown by CA membrane. After adding hydrophilic SGO nanofillers, a decrease in contact angle of composite membranes is observed which suggested better hydrophilicity and improved affinity of water for CA-SGO composite membranes (Figure 7). Furthermore, all CA-SGO membranes exhibited higher hydrophilicity than the CA-GO (SGO-0 membrane). This is attributed to the introduction of water holding sulfonic acid -SO 3 H, -COOH, and O-H moieties both on the surface and within the matrix of respective hybrid membranes. These functionalities have the ability to absorb water molecules quickly through iondipole and dipole-dipole interactions. Consequently, an increase in hydrophilic character of CA composite membranes will help in promoting anti-biofouling performance and These functionalities have the ability to absorb water molecules quickly through iondipole and dipole-dipole interactions. Consequently, an increase in hydrophilic character of CA composite membranes will help in promoting anti-biofouling performance and membrane permeability [21,33]. Figure 8 shows the effect of GO and SGO nanofillers on water flux of cellulose acetate blended membranes. The results suggested that CA blended membranes with GO and SGO led to an increase in water flux as compared to pure CA membrane. The water flux for pristine CA membrane was 50 Lm −2 h −1 /bar. Addition of SGO (SGO-1, SGO-2, SGO-3, and SGO-4) nanofillers in the casting solution of CA improved the visible changes in the pure water flux. Among the CA-SGO blended membranes, the water flux of SGO-3 reached a peak value of 152 Lm −2 h −1 /bar in comparison to pure CA membrane. This indicates that the presence of additional sulfonic groups (SO 3 H) upon graphene oxide (GO) holds a thick layer of water and hence increases the water flux. The -SO 3 H group anchored within SGO is a strong H-bonding group rather than the -OH/-COOH groups present on GO. Additionally, SGO nanofillers showed a decrease in contact angle which can positively act in promoting the permeability of water [21]. Moreover, the slight decrease in water flux is shown by high SGO-4 (1.4 wt %) content mainly due to pore blockage because of nanoparticles agglomeration at highest concentration of embedded nanoparticles within fabricated membranes. The antibacterial activity of fabricated CA-based SGO-doped membranes was evaluated via agar well diffusion method through measurement of inhibition zone. Our chosen testing microbe, Escherichia coli, was inoculated into the plate containing agar medium through spreading over whole space. Afterwards, five to six holes of 6 to 8 mm diameter were aseptically punched with a tip or sterile cork borer into a plate containing agar medium followed by introduction of antimicrobial extract solution and control with highest antibacterial activity (Ampicillin) using microliter syringe into wells which already contained E.coli as a testing microbe. The agar plates were then incubated under suitable conditions depending on the test microorganism. This causes diffusion of antimicrobial agent into agar medium which inhibits the tested microbial strain growth. The inhibition zone was developed against each well by antimicrobial extract. From Figure 9 and Figure 10, it is revealed that antibacterial activity is increased with increased content of SGO within CA membrane. The surface of CA-SGO composite membrane exhibited negative charge due to hydrophilic SGO functional groups like sulfonic acid and hydroxyl groups which creates electrostatic repulsion among microbe and membrane [14]. The antibacterial activity of fabricated CA-based SGO-doped membranes was evaluated via agar well diffusion method through measurement of inhibition zone. Our chosen testing microbe, Escherichia coli, was inoculated into the plate containing agar medium through spreading over whole space. Afterwards, five to six holes of 6 to 8 mm diameter were aseptically punched with a tip or sterile cork borer into a plate containing agar medium followed by introduction of antimicrobial extract solution and control with highest antibacterial activity (Ampicillin) using microliter syringe into wells which already contained E.coli as a testing microbe. The agar plates were then incubated under suitable conditions depending on the test microorganism. This causes diffusion of antimicrobial agent into agar medium which inhibits the tested microbial strain growth. The inhibition zone was developed against each well by antimicrobial extract. From Figures 9 and 10, it is revealed that antibacterial activity is increased with increased content of SGO within CA membrane. The surface of CA-SGO composite membrane exhibited negative charge due to hydrophilic SGO functional groups like sulfonic acid and hydroxyl groups which creates electrostatic repulsion among microbe and membrane [14]. DSC curves of prepared CA/SGO membranes with different concentrations in this study are given in Figure 11. The Tg (Glass transition temperature) is generally used for interpretation of membrane structure while employing the thermal analysis upon membrane.
It is reported that the Tg value of pure CA membrane is 55 °C [34] without any additives. The endothermic peaks shift to region of higher temperature (100-170 °C) with addition of nanofillers at higher concentration. All membranes, i.e., SGO-0, SGO-1, SGO-2, SGO-3, and SGO-4, possess a Tg value greater than 60 °C of pristine CA membrane, which  DSC curves of prepared CA/SGO membranes with different concentrations in th study are given in Figure 11. The Tg (Glass transition temperature) is generally used fo interpretation of membrane structure while employing the thermal analysis upon mem brane.
It is reported that the Tg value of pure CA membrane is 55 °C [34] without any add tives. The endothermic peaks shift to region of higher temperature (100-170 °C) with ad dition of nanofillers at higher concentration. All membranes, i.e., SGO-0, SGO-1, SGO-SGO-3, and SGO-4, possess a Tg value greater than 60 °C of pristine CA membrane, whic DSC curves of prepared CA/SGO membranes with different concentrations in this study are given in Figure 11. The Tg (Glass transition temperature) is generally used for interpretation of membrane structure while employing the thermal analysis upon membrane.
confirms the enhanced thermal stability of composite membranes. A broad endothermic peak is shown generally by all membranes. This difference is attributed to differences in contents of SGO plus the density of packing among various membranes. The increase in Tg value is related to the cross-linking density of the membrane, meaning the chains of polymer are more stiff and compact [35,36]. A tensile test was used for evaluating mechanical properties of CA-SGO nanocomposite membranes. An improved in tensile strength is shown in all membranes (SGO-0, SGO-1, SGO-2, and SGO-3) as compared to pure CA membrane. As indicated from ( Figure  12) due to high surface area of SGO nanoparticles an interface was obtained with CA matrix due to their better dispersion and increase in cross-linking due to SO3H groups presence which enhances the mechanical properties of resulting membranes. However, further increases in concentration of nanofillers, as in SGO-4, might result in decrease in tensile strength, which is related to particles aggregation on selective points of membrane during membrane fabrication. It is reported that the Tg value of pure CA membrane is 55 • C [34] without any additives. The endothermic peaks shift to region of higher temperature (100-170 • C) with addition of nanofillers at higher concentration. All membranes, i.e., SGO-0, SGO-1, SGO-2, SGO-3, and SGO-4, possess a Tg value greater than 60 • C of pristine CA membrane, which confirms the enhanced thermal stability of composite membranes. A broad endothermic peak is shown generally by all membranes. This difference is attributed to differences in contents of SGO plus the density of packing among various membranes. The increase in Tg value is related to the cross-linking density of the membrane, meaning the chains of polymer are more stiff and compact [35,36].
A tensile test was used for evaluating mechanical properties of CA-SGO nanocomposite membranes. An improved in tensile strength is shown in all membranes (SGO-0, SGO-1, SGO-2, and SGO-3) as compared to pure CA membrane. As indicated from ( Figure 12) due to high surface area of SGO nanoparticles an interface was obtained with CA matrix due to their better dispersion and increase in cross-linking due to SO 3 H groups presence which enhances the mechanical properties of resulting membranes. However, further increases in concentration of nanofillers, as in SGO-4, might result in decrease in tensile strength, which is related to particles aggregation on selective points of membrane during membrane fabrication.

Conclusions
In summary, CA membranes doped with SGO additives were fabricated through

Conclusions
In summary, CA membranes doped with SGO additives were fabricated through phase inversion method. The influence of SGO nanoparticles has been studied with main focus on the hydrophilicity, morphology, permeability, mechanical strength, and anti-biofouling behavior of CA-SGO nanocomposite membranes. With the addition of SGO nanoparticles within CA polymer matrix, an enhancement in water uptake level was observed with reduction in contact angle. SGO-3 showed maximum water flux with reduced contact angle than pristine CA membrane. The anti-biofouling behavior was enhanced through hydrophilic SGO additives which provide resistance against microbes. An increase in thermal stability was also attributed to SGO content, which efficiently reduces Tg value as compared to pure CA membrane. Based on such comprehensive results, SGO additives can be regarded as potent inorganic additives for application in wastewater treatment using various polymeric membranes.