Surface functionalization of graphene oxide using amino silane magnetic nanocomposite for Chromium (VI) removal and bacterial treatment

Amino silane magnetic nanocomposite decorated on graphene oxide (GO-Fe3O4-APTES) was successfully prepared by organic transformation reaction followed by co-precipitation method. The nanocomposite was characterised by using FT-IR, XRD, FE-SEM, TEM, EDS mapping, VSM, Raman spectroscopy, BET surface area analyzer, Zeta potential and UV-visible spectrophotometer. From TEM results we observed that 8 nm sized particles successfully modified on GO surface. The surface area of GO-Fe3O4-APTES was 57.9 m2 g−1. The magnetic Saturation value of GO-Fe3O4-APTES was 30.6 emu g−1 and the S-like magnetization of all the samples shows super paramagnetic in nature. Due to magnetic nature adsorbent, it could be easily separated from aqueous solution. GO-Fe3O4-APTES material was highly selective for Chromium (VI) removal from aqueous solution. About 91% of Chromium (VI) was removed at pH 3, 160 rpm of shaking speed, 0.3 g l−1 of adsorbent dose and 10 h of contact time. The adsorption process of Chromium (VI) on GO-Fe3O4-APTES follows Pseudo-second-order kinetic and Langmuir isotherm model because of high coefficient of determination value (R2 = 0.99). The maximum adsorption capacity (qm) of GO-Fe3O4- APTES was observed at 60.53 mg g−1. The synthesized material was desorbed with 0.5 M NaOH and recycled up to five cycles. After five cycles, the removal efficiency of Chromium (VI) possesses high efficacy towards GO-Fe3O4-APTES. Mechanistically, adsorption of Chromium (VI) follows strong electrostatic attraction between adsorbate and adsorbent. GO-Fe3O4-APTES has potential adsorbent for the adsorption of Chromium (VI) in waste water treatment. Furthermore, the GO-Fe3O4-APTES were tested for antibacterial properties against gram negative (Escherichia coli) and gram positive (Bacillus subtilis) bacterial strain. The synthesized material responds positively towards antibacterial activity.


Introduction
The water contamination throughout the world occurs by toxic heavy metals (Cr, As, Pb, Hg Ni, and Cd) above their permissible limit are a serious problem for living organism [1][2][3]. Among various toxic metal ions, Chromium (Cr) is one of the highly toxic heavy metal and often discharges to water from wood preservatives, electroplating, textile industries, metal refining and plant producing industrial inorganic chemicals and pigments [4][5][6]. A very low concentration of chromium creates serious health problem like stomach cancer, bronchial asthma, kidney damage, anaemia, liver damage and hepatotoxicity in human [7,8]. Chromium exist in water both Chromium (III) and Chromium (VI) states, Chromium (VI) is highly contaminant because of 2. Materials and methods

Preparation of Iron oxide nanoparticles (Fe 3 O 4 )
Iron oxide (Fe 3 O 4 )was prepared using previously reported chemical co-precipitation method with some modification [70]. Briefly, 4 g of anhydrous FeCl 2 and 12 g of anhydrous FeCl 3 were dissolved in 50 ml of 0.1 M HCl solution. Further, the mixture was added slowly to 500 ml of 1.5 M NH 3 solution until reaches pH 11, after wards, stirred for 2 h at the temperature 40°C. A black precipitate of Fe 3 O 4 magnetic nanoparticle was formed after stirring for 2 h, which was collected by centrifugation and washed three times with distilled water and two times with ethanol, and then dried at 60°C.

Preparation of Fe 3 O 4 -APTES (FA)
The APTES modified Fe 3 O 4 magnetic nanoparticles (FA) was synthesized based on our previously established method with small modification [71]. Firstly a solution mixture was prepared using 1 g of Fe 3 O 4 and 100 ml of ethanol in a round bottom flask. In order to disperse the Fe 3 O 4 nanoparticles the solution was ultrasonicated for 30 min. Then the resulting dispersion was bubbled with argon gas for 30 min, and then added 1 ml of APTES by a syringe under mechanical stirring. The reaction was then maintained at room temperature for 24 h. The reaction was incubated at room temperature for 24 h. Finally, the obtained solid product was collected with the help of a magnet and repeatedly washed with ethanol. The obtained FA nanocomposite was dried in a muffle furnace at 60°C.

Preparation of graphene oxide (GO)
Graphene oxide (GO) was prepared according to reported work via modified hummers method [72]. In a typical synthesis, 1 g of graphite powder was suspended in 25 ml of H 2 SO 4 (98%) and ultrasonicated for 20 min. After which, 100 mg of NaNO 3 was added to the above solution and stirred for 34 h at room temperature. The obtained mixture was kept in an ice bath to cool down 5°C followed by the slow addition of 3 g of KMnO 4 with constant stirring, by keeping in mind the temperature should not rise above 20°C. The stirring was continued for 4 h at same condition. Then the ice bath was removed and kept the reacting mixture at room temperature and 250 ml distilled water was added drop wise. The stirring was continued for another 45 min and 50 ml of warmed distilled water to terminate the reaction 5 ml of H 2 O 2 (30%) was added drop wise to the contents and stirred for 12 h. A bright yellow colour indicates the complete oxidation of graphite to GO. The yellow suspension was centrifuged and washed three times with HCl (10%) to remove the metal ions and then several times with distilled water to maintain the pH 7. Then the solution was dried at 60°C and grind to get desire GO powder.

Preparation of GO-Fe 3 O 4 -APTES
Firstly 300 mg of GO powder was dispersed in 20 ml of DMF (as a catalyst) in 250 ml capacity of round bottom flask followed by the addition of 0.25 ml of SOCl 2 and the reaction was left to proceed for 24 h with constant stirring at 70°C. SOCl 2 is a strong reducing agent, it can convert less reactive -COOH group of GO into much reactive -COCl group. Solvents like DMF are suitable catalyst to speed up the reaction. Then 600 mg of Fe 3 O 4 -APTES was added to the reacting mixture followed by addition of 10ml of DMF and 0.38 ml of Et 3 N. The temperature of the following mixture was kept upto 130°C, then stirred and refluxed for next 72 h. In the following method, -COCl group react with -NH 2 group of Fe 3 O 4 -APTES to form amide bond. The solid residue was separated by centrifugation followed by washing using distilled water. The residue was dried at 80°C for 12 h to form desired product. The detailed synthetic mechanism was represented in scheme 1.

Adsorption experiment
All the batch adsorption experiments were carried out in 200 ml capacity of polypropylene bottles by considering various concentration of Chromium solution. For the adsorption studies,0.3 g of GO-Fe 3 O 4 -APTES was added in 20 mg l −1 of Chromium solution at pH 3. The reaction bottles were shaken using an incubator shaker (RC 5100) at shaking rate 200 rpm at room temperature. After 10 h, the adsorbent was separated out by Whatman-42 (mm) filter paper. The remaining concentration of Chromium was analysed by UV-Visible spectrophotometer using 1,5-diphenyl carbazide (DPC) method at the wave length of 540 nm [73]. The impact of various parameters like pH (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12), adsorbent dosage (0.15-0.35 g), time (2-10 h) and room temperature (25°C)influencing the removal of Chromium ion were examined separately by keeping the others constant, so as to optimize the adsorption process. The adsorbed amount of chromium ion and the removal efficiency (%) was determined using the subsequent equations.
From the above equation, q e shows the adsorbed amount of Chromium (VI). C 0 is the initial concentration of chromium (VI) (mgl −1 ), C e is the final concentration of chromium ion (mgl −1 ), V is the volume of Chromium (VI) solution and W is the mass of adsorbent. All experiments were repeated for 3 times. 2.8. Bacterial culture E. coli and B. subtilis were grown in Luria-Bertini (LB) broth. For colony counting purpose LB agar was taken and plates were prepared in Petridishes.

Antibacterial activity of GO-Fe 3 O 4 -APTES
Assessment of antibacterial activity of GO-Fe 3 O 4 -APTES was checked against gramnegative bacteria, E. coli and gram positive bacteria B. subtilis. For the evaluation of antibacterial activity both the bacteria were grown in LB broth and an appropriate concentration of bacteria was taken for the antimicrobial assay. The bacteria were inoculated on the LB agar plate and the antimicrobial activity of GO-Fe 3 O 4 -APTES was checked by using the disc diffusion method [74]. 2.11. Detection of ROS production Assessment of ROS generation by bacteria was check by using the 2,7-Dichlorofluorescin diacetate (DCFDA) dye. It is a peroxynitrile indicator which confirms the generation of different reactive oxygen species (nitric oxide or hydrogen peroxide) [75]. Bacterial cells treated with and without GO-Fe 3 O 4 -APTES were exposed to DCFDA dye (1 μM) and then its fluorescence intensity was measured at emission wavelength 529 nm with an excitation of 495 nm by using the fluorescence spectrophotometer.

Characterisation
In order to study the presence of functional groups in GO-Fe 3 O 4 -APTES, FTIR measurement was carried out.   Techniques such as FE-SEM and TEM were used to explore the surface morphology of the synthesized material. Figure 3(a) shows the FE-SEM image of the Fe 3 O 4 -APTES. It is observed that the shape of Fe 3 O 4 -APTES was nano sized spherical shape and homogeneously distributed [92][93][94]. The FE-SEM image of GO (figure 3(b)) shows the layer type structure [95][96][97]. The Raman spectroscopy was carried out to study the structural configuration of GO and GO-Fe 3 O 4 -APTES during the process of adsorption and the data was shown in figure 6. The Raman spectrum of GO shows two strong peaks one is D band at 1353 cm −1 due to the disorder of GO arising from imperfection linked with vacancies, amorphous carbon species and grain boundaries and other is G band at 1607 cm −1 which specifies the E 2g phenomenon of Sp 2 hybridised carbon in a 2-dimensional hexagonal lattice [98]. After the modification of Fe 3 O 4 -APTES on the surface of GO, the intensity of D and G band became higher than pristine GO. In case of GO-Fe 3 O 4 -APTES, it is observed that two strong bands at 1350 cm −1 and 1588 cm −1 are found. The D and G band of GO-Fe 3 O 4 -APTES was slightly shifted by 3 cm −1 and 19 cm −1 as compared to pristine GO. This Raman shift of D and G band for GO-Fe 3 O 4 -APTES sheet demonstrates that the charge transfer occurs between the sheets of GO andFe 3 O 4 -APTES. It exhibit a strong interaction between GO and Fe 3 O 4 -APTES nanocomposite. The intensity ratio I D /I G of GO and GO-Fe 3 O 4 -APTES was calculated to be 1.00 and 1.12, respectively. However, the intensity ratio I D /I G of GO-Fe 3 O 4 -APTES was higher than pure GO, which was due to the defects arises by the interaction between Fe 3 O 4 -APTES and GO [99,100].
Vibrating sample magnetometer (VSM) analysis was carried out to know the magnetic properties of prepared The N 2 adsorption-desorption isotherm is carried out to calculate the specific surface area of synthesized material. Figure 8 Figure 9(a) shows the variation of initial Chromium (VI) concentration from 5 to 70 mg l −1 at constant parameters such as adsorbent dosage (0.3 g), pH 3, room temperature (25°C), contact time (10 h) and shaking speed (160 rpm). From this plot we observe the Chromium (VI) removal efficiency nearly same from 5 mg l −1 to 20 mg l −1 . After 20 mg l −1 the removal efficiency starts to decrease. This occurs due to that at lower concentrations of Chromium (VI), the ratio of the initial number of Chromium (IV) ions to the obtainable surface area of the adsorbent is high. Although at higher concentrations of Chromium (VI), the remaining sites of adsorption become lower and then the percentage of removal efficiency of Chromium (VI) decreases which depends on the initial concentration of Chromium (VI). Thus 20 mg l −1 of Chromium (VI) concentration was taken as the optimum concentrtion for further experiment.

Effect of contact time on Chromium (VI) adsorption
Among all the parameters contact time is one of the most important factor which affect the adsorption capacity of the adsorbent. This is shown in figure 9(c). We varied the contact time from 1 to 25 h by keeping other parameters constant (pH 3,20 mg l −1 Chromium (VI) concentration, adsorbent dose 0.3 g, room temperature (25°C) and shaking speed 160 rpm). In this plot we observedthat1 to 10 h of contact time the removal efficiency of Chromium (VI) gradually increases. The maximum adsorption of Chromium (VI) occurred at in 10 h to 25 h of contact time indicating 91% of removal efficiency. It occurs due to the availability of maximum numbers of unoccupied surface sites for adsorption process [103]. Thus 10 h was selected as optimum contact time for further experiments.

Effect of pH on Chromium (VI) adsorption
The effect of pH on the adsorption process is an important parameter and shown in figure 9(d). We varied the pH 3 to 10 by keeping other parameters constant such as room adsorbent dose (0.3 g), Initial Chromium (VI) concentration 20 mg l −1 ), temperature (25°C), contact time (10 h)and shaking speed (160 rpm). It was observed  that at pH 3 the removal efficiency was found to be 91%. After that it significantly decreases up to 34% at pH 10. In acidic medium Chromium (VI) exist as HCrO 4 − , Cr 2 O 7 2− and it is found in the form of CrO 4 2− above pH 6. Now it is important to find the surface charge of GO-Fe 3 O 4 -APTES at different pH. Zeta-potential is the best technique to confirm the surface charge of the material. From Zeta-potential (figure S1 is available online at stacks.iop.org/NANOX/1/010062/mmedia) measurement we found that 8.2 is isoelectric point. The GO-Fe 3 O 4 -APTES has positive charge at pH<8.2, so it favours electrostatic attraction [104]. Furthermore, at pH 3 maximum positive charge was observed corresponding to strong electrostatic attraction between highly positive charged material and negatively charged chromium ion. Above pH 3, the positive charge decreases so the removal efficiency decreases. Similarly, pH>8.2 the adsorbent surface becomes negatively charge and hence the material repels negative chromium ion ref.

Effect of shaking speed on Chromium (VI) adsorption
Shaking speed is one of the most important factors which affect the adsorption capacity of the adsorbent. To determine the effect of agitation speed, 80 to 160 rpm of agitation speed were set to check the adsorption efficiency. Figure S2 shows that increasing the speed of agitation from 80 to 160 rpm, the percentage of Chromium (VI) removal efficiency also increases. At 160 rpm of agitation speed maximum adsorption of Chromium (VI) occurs. This is caused due to increase in the intra-particle diffusion and film diffusivity [106].

Adsorption kinetics
To understand the mechanism of adsorption process well on contact time, we investigate adsorption kinetics. Three models were introduce to simulate the predictable data such as Pseudo first order kinetic theory, Pseudo second order kinetic theory and Intra particle diffusion model. The Pseudo first order kinetic equation is expressed as follows [93,94,107].

( )
Here K 2 = Pseudo second order rate constant. The values of q , e K 2 and R 2 were calculated from the slope and intercept of the plot t q vs t and the values are listed in table 1.
The intra particle diffusion model is expressed by using the following equation.
Here k p is the intra-particle diffusion constant (mg g −1 min 0.5 ) and C is the boundary layer thickness constant (mg g −1 ) . The values of k p , C and R 2 were calculated from the plot q t versus t 0.5 and the values are listed in table 1. Figure 10 shows the linear form of Pseudo-first-order, Pseudo-second-order and intra-particle diffusion model for Chromium (VI) adsorption. The computed result which obtained from three models is listed in table 1 comparing to the R 2 value, pseudo-second-order kinetics is well fitted with the Chromium (VI) adsorption [110]. Therefore, this result shows chemisorption in between adsorbent and adsorbate [111]. Intraparticle diffusion model is the best model to identify the adsorption diffusion mechanism. According to this model, if the line passing through origin, then the adsorption process is controlled by intra-particle diffusion, while if the data exhibit multi linear plot but does not passing through origin, then more steps involved the adsorption process [112,113]. Figure 10(c) shows three straight lines which indicate more than one steps are involved in the adsorption mechanism. The first straight line ascribe to outer surface adsorption that means Chromium (VI) diffuses through the solution to the outer adsorbent surface. The middle line corresponds to the gradual adsorption reflecting intra-particle diffusion as the rate-limiting step. The final plateau relates out the equilibrium stage and surface adsorption, since the diffusion mechanism starts to slow down and level out [114,115]. From the above results, it could be inferred that the diffusion mechanism was involved in the multi adsorption process.

Adsorption isotherm
The adsorption isotherm models such asLangmuir, Freundlich, Temkin, Dubinin-Radushkevic (D-R) and Elovich were selected to define the adsorption process. Langmuir isotherm is used for assuming the adsorption of Chromium (VI) on homogeneous planes by monolayer adsorption. The mathematical expression for Langmuir isotherm is written as follows.  The value of R L which is the dimensionless constant can also be considered to know the favourability or unfavourability of the process of adsorption. R L can be evaluated by using the following equation.
The calculated value of R L was found to be less than 1 (<1). This designates that the adsorption process of Chromium (VI) was favourable for this isotherm.
Freundlich isotherm is utilized for assuming the adsorption of Chromium (VI) on heterogeneous planes by multilayer adsorption. This isotherm model can be expressed as follows, Here q e is the adsorbed amount of Chromium (VI) per unit weight of adsorbent (mg g −1 ) at equilibrium, n is the density of adsorption. c e is the concentration of Chromium (VI) in solution at equilibrium time (mg l −1 ). The values of K F (Freundlich constant), n and R 2 was determine by using the slope and intercept of the plot q c ln versus ln . Here, R is the gas constant and T is the absolute temperature The value of E can be computed by using the formula, All isotherm models are demonstrated graphically in figure 11 and the isotherm parameters are listed in table 2. Among five isotherm models, Langmuir isotherm model (R 2 =0.99) is more appropriate as compare to R 2 value. The Langmuir isotherm model agree to the formation of monolayer adsorption in between Chromium  [91]. From Freundlich isotherm model, it was observed that the value of 1/n is 0.169, which is less than 1. It Shows favorable adsorption of adsorbate and adsorbent [116]. Temkin isotherm model is favourable for adsorption of Chromium (VI), because it shows smaller value of Temkin constant (B 1 =1.56) [117].

Influence of co-existing ion
The studies on the Chromium (VI) removal were observed in the presence of different ions such as Sulphate, Phosphate, carbonate, bicarbonate, fluoride, nitrate and chloride, which is shown in figure 12. The adsorption procedure was carried out in presence of these ions keeping other parameters constant i.e. adsorbent dosage 0.3g, initial Chromium (VI) concentration 20 mg l −1 , pH of the solution is 3, shaking speed 160 rpm and an optimum time 10 h. About 20 ml of each anion having a concentration of 20 mg l −1 was added to the polyethylene bottle and the adsorption efficiency was measured. Carbonate, fluoride and phosphate had more impact whereas nitrate, chloride, bicarbonate and sulphate had little impact on the Chromium (VI) adsorption. The more changes were observed due to change in solution pH caused by the anions. The adsorption capacity of Chromium (VI) increased in the order of chloride>nitrate>sulphate>bicarbonate>phosphate>fluoride>carbonate.

Reusability of adsorbent
The primary purpose of reusability is to recover the depleted material. This study is a very chief parameter to study the regeneration or effectiveness of the adsorbent. We have noticed that at lower pH maximum adsorption of Chromium (VI) occurs. Hence for the reusability study higher pH values were needed. Desorption of Chromium (VI) was conducted by washing Chromium (VI) with distilled water and various concentrations of Na 2 CO 3 , NaHCO 3 and NaOH. The desorption efficiency of H 2 O, Na 2 CO 3 and NaOH was noticed to be 1%, 83%, 74% and 92% respectively ( figure S3(a)). Hence for the desorption process 0.5 M of NaOH solution was used. The plot shows desorption of Chromium (VI) having different pH conditions. Some distilled water was used in this experiment to remove undesirable ions present on the surface of the adsorbent ( figure S3(b)). From the plot it was noticed that after 5 cycles, adsorption efficiency decreases up to 51%. This is shown that the reusability of the material was highly efficient.

Adsorption mechanism
The FT-IR peak of after adsorption of Chromium (VI) on GO-Fe 3 O 4 -APTES material was shown in figure S4.
Comparing to the FT-IR data of before and after adsorption of chromium (VI), the N-H bending vibration was shifted from 1570 cm −1 to 1577 cm −1 and -CO-NHwas shifted from 1650 cm −1 to 1664 cm −1 with high intense peak which attributes the bonding between the nitrogen and chromium. The presence of one new peak at 943 cm −1 was ascribing to stretching of Cr-O in CrO 7 2− groups [74]. These changes in the FT-IR spectrum after adsorption shows chromium (VI) successfully adsorb the synthesized material. The surface morphology of  Figure S5(c) shows the EDS mapping of the material after adsorption of chromium (VI). It was clearly seen the chromium adsorbed uniformly on the surface of GO-Fe 3 O 4 -APTES. Based on the above results, the possible mechanism for Chromium (VI) was the protonated amine groups and hydroxyl groups of GO-Fe 3 O 4 -APTES by electrostatic interaction (Shown in scheme S1).

Antibacterial activity
Bactericidal activity of the synthesized material is dependent on the concentration of material and concentration of bacterial cell suspension [101]. Our study includes the initial bacterial load of10 6    subtilis ( figure 13). Antibacterial activity of GO-Fe 3 O 4 -APTES is dependent on the concentration and time of exposure with the bacteria. Assessment of bacterial growth inhibition by colony count method has suggested that the GO-Fe 3 O 4 -APTES shows the good antibacterial activity at 160 μg ml −1 and its activity decreases gradually with decreasing concentration of the material ( figure 14). Growth kinetics study of both the bacteria in the presence as well as in the absence ofGO-Fe 3 O 4 -APTES suggested that the significant growth inhibition observed in treated one in comparison to control ( figure 15). The proposed mechanism of antibacterial activity is supported by an increase in reactive oxygen species (ROS) generation at higher concentration 160 μg ml −1 and it gradually decreases with decreasing concentration of the prepared material. To quantify the ROS generation the bacteria was treated with DCFDA which react with the ROS and produces the green fluorescence which is measured by the Fluorimeter ( figure 16). Higherthe fluorescence intensity higher is the amount of ROS generated from the bacteria in presence of GO-Fe 3 O 4 -APTES. ROS generated from the bacteria get mixed with the culture media which shows the fluorescence in presence of DCFDA. Bacteria show the ROS production in stress condition which could be observed in the control batch which is not treated with GO-Fe 3 O 4 -APTES. FESEM study also reveals the change in membrane integrity of bacteria with alteration in its morphology (figures 17 and 18).
Both gram positiveand gram negative bacteria shows variation in the toxicity with respect to GO-Fe 3 O 4 -APTES bactericidal activity. In case of gram negative bacteria (E. coli) peptidoglycan layer is protected by an outer layer composed of lipopolysaccharide which helps to protect the bacteria from the chemical exposure [128]. Thus the bacterial death was less in E. coli in comparison to gram positive bacteria (B. subtilis). Direct contact of graphene material with the bacteria [59] increases oxidative stress [129] as the main mechanism responsible for bacterial growth inhibition and including this the iron oxide itself causes the increased oxidative    stress [130]. Increased oxidative stress is also responsible for the release of hydroxyl radical which bind to the carbonyl group of peptide linkage in the bacterial cell membrane which distorts the structure of cell membrane [131] ( figure 18(b)). It is also reported that the bacterial cell membrane gets ruptured when it comes in contact with GO [132]. The GO binds to the water molecule by the help of carbonyl group and free radical sites and thus form the colloidal solution which enhances the easy accessibility of material to interact with the bacteria [133,134].

Conclusion
Fe 3 O 4 -APTES was successfully fabricated on GO through organic transformation reaction followed by coprecipitation method. The functionalization, formation, morphology of the material, magnetic properties and surface area were characterized by FTIR, XRD, FE-SEM, TEM, HRTEM, Raman and BET technique. The synthesized GO-Fe 3 O 4 -APTES was identified as an efficient adsorbent for removal of hexavalent Chromium. Experimental results revealed that the removal efficiency was pH dependent and higher removal efficiency occurs at pH 3. Pseudo second order kinetics model was best fit for the adsorption process and shows chemisorptions'. Langmuir isotherm is best fit for Chromium (VI) adsorption on GO-Fe 3 O 4 -APTES with an adsorption capacity of 60.5 mg g −1 at room temperature (25°C). It has shown that the coexisting ions had no significant impact on adsorption efficiency. The adsorbed chromium could be effectively washed from the adsorbent in to the solution using 0.5 M of NaOH. It can be concluded that GO-Fe 3 O 4 -APTES material has got good reusable ability. GO-Fe 3 O 4 -APTES has shown inhibitory effect on the growth of E. coli and B. subtilis.