Magnetic Graphene-Based Sheets for Bacteria Capture and Destruction Using a High-Frequency Magnetic Field.

Magnetic reduced graphene oxide (MRGO) sheets were prepared by embedding Fe3O4 nanoparticles on polyvinylpyrrolidone (PVP) and poly(diallyldimethylammonium chloride) (PDDA)-modified graphene oxide (GO) sheets for bacteria capture and destruction under a high-frequency magnetic field (HFMF). The characteristics of MRGO sheets were evaluated systematically by transmission electron microscopy (TEM), scanning electron microscopy (SEM), zeta potential measurement, X-ray diffraction (XRD), vibrating sample magnetometry (VSM), and X-ray photoelectron spectroscopy (XPS). TEM observation revealed that magnetic nanoparticles (8–10 nm) were dispersed on MRGO sheets. VSM measurements confirmed the superparamagnetic characteristics of the MRGO sheets. Under HFMF exposure, the temperature of MRGO sheets increased from 25 to 42 °C. Furthermore, we investigated the capability of MRGO sheets to capture and destroy bacteria (Staphylococcus aureus). The results show that MRGO sheets could capture bacteria and kill them through an HFMF, showing a great potential in magnetic separation and antibacterial application.


Introduction
Accelerating research on carbon nanostructures led to the discovery of graphene. Structurally, graphene consists of a flat monolayer of carbon atoms tightly packed into a two-dimensional (2D) honeycomb lattice. Graphene is the basic fundamental structure for the derivative graphitic materials of all other dimensionalities including fullerenes, nanotubes, and graphite [1], and is also the thinnest and strongest material. Its charge carriers exhibit giant intrinsic mobility, have zero effective mass, and can travel for micrometers without scattering at room temperature [2]. Currently, graphene has

Preparation of Magnetic Reduced Graphene Oxide (MRGO)
GO was prepared via modified Hummers method using graphite as the precursor material followed by the oxidation processes [37]. Surface modification of GO was conducted through the direct mixing with PVP and PDDA, simultaneously. Herein, PVP (average molecular weight 10,000 Da) was used to improve the aqueous dispersibility of GO [38]. Briefly, PVP (10 mg/mL) was added into the GO solution (1 mg/mL) as the dispersant. Furthermore, the above PVP-modified GO solution was mixed with 10 mg/mL of PDDA (the stock concentration: 0.2 g/mL) solution and 0.625 M of KCl in the 20 mL of deionized water. The GO-PDDA solution was then stirred for 10 min at 90 • C and then stirred continuously for 12 h at 25 • C. The color change of the GO-PDDA solution confirms the reduction process of GO (brown) transformed to GO-PDDA (black) [37].
FeCl 2 ·4H 2 O and FeCl 3 ·6H 2 O were used as the initial precursors for the co-precipitation of Fe 3 O 4 on the GO surface. Briefly, FeCl 2 ·4H 2 O (2.5 mmol) and FeCl 3 ·6H 2 O (5 mmol), with a ratio of 1:2, were briefly mixed homogeneously at room temperature [39]. The GO solution (1 mg/mL) was mixed with the previous solution and then the temperature was increased to 80 • C. Furthermore, ammonia solution (33%) was added dropwise into the solution to initiate the coprecipitation of Fe 3 O 4 onto the GO sheets. The final precipitation was denoted as MRGO ( Figure 1a).
FeCl2.4H2O and FeCl3.6H2O were used as the initial precursors for the co-precipitation of Fe3O4 on the GO surface. Briefly, FeCl2.4H2O (2.5 mmol) and FeCl3.6H2O (5 mmol), with a ratio of 1:2, were briefly mixed homogeneously at room temperature [39]. The GO solution (1 mg/mL) was mixed with the previous solution and then the temperature was increased to 80 °C . Furthermore, ammonia solution (33%) was added dropwise into the solution to initiate the coprecipitation of Fe3O4 onto the GO sheets. The final precipitation was denoted as MRGO ( Figure 1a).

Characterization
The structure and morphology of MRGO were characterized by TEM (TEM-7650, Hitachi, Chiyoda-ku, Japan) at an acceleration voltage of 100 kV. The binding energy characteristics of GO and MRGO were investigated using X-ray photoelectron spectroscopy (XPS) (ESCALAB 250, Thermo VG Scientific, West Sussex, UK), equipped with Mg Kα at 1253.6 eV at the anode. X-ray diffraction (XRD) was conducted using a D 2 Phaser BRUKER X-ray powder diffractometer to evaluate the phase of the material by scanning dried powder in the 2θ range of 20-80 • with Cu K α (1.5406 Å) radiation. The zeta potential of the graphene-based material before and after polymeric modification was determined through electrophoretic mobility measurements (Horiba Instrument). Vibrating sample magnetometry (VSM) (Lakeshore model 7400) was used to evaluate the magnetization of MRGO as a function of the magnetic field (Oe) at room temperature.
The bacteria capture by MRGO was observed using scanning electron microscopy (SEM) (JSM 5510, JEOL, Tokyo, Japan). Figure 1b shows the schematic representation of the bacteria capturing experiments by the permanent magnetic field. The bacteria capturing capability of MRGO was evaluated against S. aureus (ATCC 43894). Briefly, frozen preserved stock was thawed at room temperature, 0.1 mL was then pipetted and streaked into a quadrant on a nutrient agar plate. Nutrient agar was composed of 6 g agar bacteriological (Agar No. 1) LP0011 powder mixed with 10 g DifcoTM LB (Luria-Bertani) Broth, Miller powder and diluted with 400 mL deionized (DI) water, and cultured at 37 • C for 18-24 h to allow the formation of colonies. Afterward, a single colony was scraped with a loop and swabbed onto a 15 • -slant medium (nutrient agar) which was then incubated at 37 • C for 18-24 h. After 18-24 h of culturing, at least 2 mL of NaCl solution (1 wt%) was added and stirred with a vortex mixer for 30 s. After mixing, 1 mL of the solution was added into 9 mL of nutrient broth (5 g Difco™ LB Broth, Miller powder/200 mL DI-water) and mixed with a vortex mixer, then 1 mL of the first nutrient broth solution was mixed into 9 mL of new nutrient broth. This process was repeated four times in a row. An aliquot of 1 mL of nutrient broth containing bacteria (1 × 10 5 colony forming unit (CFU)/mL) was added to a flask containing 1 mL MRGO, followed by shaking with a rotary shaker at 250 rpm for 20 min. The separation process was conducted by exposing the solution to an external magnetic field for 10 min. After 18-24 h of culturing, the surviving bacteria were counted.

Magnetic Inductive Heating (Hyperthermia) Experiment by HFMF
Magnetic field-induced heat generation of MRGO was generated through HFMF exposure with a frequency of 15 KHz. Briefly, MRGO suspensions were subjected to the center of a copper coil in the HFMF apparatus for 20 min. The temperature increases were recorded every minute using an alcohol thermometer.  Figure 2b shows that the round-like shaped iron oxide (Fe 3 O 4 ) nanoparticles with a size of around 8 to 10 nm were embedded onto the surface of GO-based sheets. GO provides the host surface due to its high-surface area for the crystallization of Fe 3 O 4 nanoparticles, which generated MRGO [21]. Moreover, the oxygen-based functional groups on GO can act as the nucleation sites for the coprecipitation of Fe 2+ /Fe 3+ [21]. As shown in Figure 2b, the wrinkle structure of GO sheets disappear after the crystallization of Fe 3 O 4 nanoparticles, which further confirms the reduction process of GO into magnetic reduced graphene oxide (MRGO). This is in accordance with the previous investigation by Bo Hu et al. [26]. Figure 2c indicates the selected area electron diffraction (SAED) pattern of GO, showing the hexagonal structure and the plane (1010) and (1210). Nanomaterials 2020, 10, x FOR PEER REVIEW 6 of 14 disappear after the crystallization of Fe3O4 nanoparticles, which further confirms the reduction process of GO into magnetic reduced graphene oxide (MRGO). This is in accordance with the previous investigation by Bo Hu et al. [26]. Figure Figure 3 shows the X-ray photoelectron spectra (XPS) of GO and PVP/PDDA-modified GO (reduced graphene oxide, RGO). The C-1s spectra of XPS showed that the graphite contained spectra of C-C at 284.01 eV and C-OH at 284.85 eV, respectively ( Figure 3a). Further, the XPS spectra of GO (C-1s) characteristics peaks were found at C-C (285. 16 Figure 3 shows the X-ray photoelectron spectra (XPS) of GO and PVP/PDDA-modified GO (reduced graphene oxide, RGO). The C-1s spectra of XPS showed that the graphite contained spectra of C-C at 284.01 eV and C-OH at 284.85 eV, respectively ( Figure 3a). Further, the XPS spectra of GO (C-1s) characteristics peaks were found at C-C (285. 16    The XRD pattern of MRGO shows a similar pattern to the pattern of Fe3O4 nanoparticles, indicating the formation of Fe3O4 nanoparticles on the graphene surface. Figure 4b shows the measured saturation magnetization of MRGO and Fe3O4 nanoparticles by VSM. The magnetization of MRGO was around 20 emu/g, which was strong enough to separate the magnetic graphene from the aqueous solution. The decreasing saturation magnetization value of MRGO in comparison to Fe3O4 nanoparticles (around 60 emu/g) could be due to the interference of the graphene-based material in the magnetic field exposure. As shown in Figure 4b, the fielddependent magnetization curves are reversible, which confirms that MRGO and Fe3O4 nanoparticles were superparamagnetic materials with zero coercivity and zero remanence. MRGO could disperse well in the aqueous solution in the absence of magnetic field exposure, which was generated due to the presence of an oxygen-based functional group on the surface of MRGO. Further, MRGO could be collected in the presence of magnetic field exposure (Figure 4b) for the magnetic separation of the microbes. Figure 4c shows the zeta potential value of each component at pH 7. The zeta potential of Fe3O4 nanoparticles is −40 mV. This result might be caused by a number of hydroxyl groups (Fe-OH) [40]. GO exhibited negative charged of −60 mV suggesting the successful oxidation process of graphite through the formation of the negatively charged moiety (hydroxyl, epoxy and carboxyl) on the surface of GO. Surface modification of GO using PVP altered the zeta potential value to −70 mV. This may be caused by PVP adsorbing onto the surface of the GO. A positive zeta potential value (+42 mV)   Figure 4b shows the measured saturation magnetization of MRGO and Fe 3 O 4 nanoparticles by VSM. The magnetization of MRGO was around 20 emu/g, which was strong enough to separate the magnetic graphene from the aqueous solution. The decreasing saturation magnetization value of MRGO in comparison to Fe 3 O 4 nanoparticles (around 60 emu/g) could be due to the interference of the graphene-based material in the magnetic field exposure. As shown in Figure 4b, the field-dependent magnetization curves are reversible, which confirms that MRGO and Fe 3 O 4 nanoparticles were superparamagnetic materials with zero coercivity and zero remanence. MRGO could disperse well in the aqueous solution in the absence of magnetic field exposure, which was generated due to the presence of an oxygen-based functional group on the surface of MRGO. Further, MRGO could be collected in the presence of magnetic field exposure (Figure 4b) for the magnetic separation of the microbes. Nanomaterials 2020, 10, x FOR PEER REVIEW 8 of 14

Physicochemical Characterizations
of MRGO was achieved after surface modification of PVP/GO by PDDA. PDDA could be tethered onto graphene sheets to develop a net positive charge due to the anchoring of nitrogen-based moiety on PDDA [41]. Eventually, the immobilization of Fe3O4 onto the PDDA/PVP/GO altered the net zeta potential value to −65 mV, compared with the pristine Fe3O4 nanoparticles (−40 mV). Thus, highly negatively charged MRGO also proved the high stability of MRGO due to the highly electrostatic repulsion among nanoparticles, which will prevent the aggregation.  Figure 5a shows the optical images of bacteria capturing activity toward the supernatant of S. aureus solution with various concentrations of MRGO. The supernatant of bacteria colony decrease indicates that more bacteria was captured. The results show that higher concentrations of MRGO will capture more bacteria. The optimal concentration of MRGO sheets is higher than 2.4 mg/mL, which is enough to collect all of bacteria (10 5 CFU/mL) (Figure 5b). These results might be due to the higher density of functional groups and surface area of MRGO that could enhance the potential contact and interaction with S. aureus, followed by the settlement of bacteria on the surface of graphene-based materials [30]. From the point of view of antibacterial activity, graphene-based materials could damage or break the cell walls and cell membranes of bacteria through the interaction between the sharp edge of graphene-based materials and membranes of bacteria, causing the stress of the bacteria membrane [22,27,42]. This interaction could induce the leakage of RNA and transfer of charge [26]. Previous investigation proposed that the appropriate charge transfer between the bacteria and the more sharpened edges of the graphene-based materials could enhance the antibacterial activity of the   [40]. GO exhibited negative charged of −60 mV suggesting the successful oxidation process of graphite through the formation of the negatively charged moiety (hydroxyl, epoxy and carboxyl) on the surface of GO. Surface modification of GO using PVP altered the zeta potential value to −70 mV. This may be caused by PVP adsorbing onto the surface of the GO. A positive zeta potential value (+42 mV) of MRGO was achieved after surface modification of PVP/GO by PDDA. PDDA could be tethered onto graphene sheets to develop a net positive charge due to the anchoring of nitrogen-based moiety on PDDA [41]. Eventually, the immobilization of Fe 3 O 4 onto the PDDA/PVP/GO altered the net zeta potential value to −65 mV, compared with the pristine Fe 3 O 4 nanoparticles (−40 mV). Thus, highly negatively charged MRGO also proved the high stability of MRGO due to the highly electrostatic repulsion among nanoparticles, which will prevent the aggregation. Figure 5a shows the optical images of bacteria capturing activity toward the supernatant of S. aureus solution with various concentrations of MRGO. The supernatant of bacteria colony decrease indicates that more bacteria was captured. The results show that higher concentrations of MRGO will capture more bacteria. The optimal concentration of MRGO sheets is higher than 2.4 mg/mL, which is enough to collect all of bacteria (10 5 CFU/mL) (Figure 5b). These results might be due to the higher density of functional groups and surface area of MRGO that could enhance the potential contact and interaction with S. aureus, followed by the settlement of bacteria on the surface of graphene-based materials [30]. From the point of view of antibacterial activity, graphene-based materials could damage or break the cell walls and cell membranes of bacteria through the interaction between the sharp edge of graphene-based materials and membranes of bacteria, causing the stress of the bacteria membrane [22,27,42]. This interaction could induce the leakage of RNA and transfer of charge [26]. Previous investigation proposed that the appropriate charge transfer between the bacteria and the more sharpened edges of the graphene-based materials could enhance the antibacterial activity of the graphene-based materials [30]. Another possible mechanism is the isolation of S. aureus from the surrounding medium during the incubation with the MRGO; therefore, blocking the consumption of nutrients for proliferation leads to the cells' death. In Figure 6, SEM images show that the morphology of MRGO sheets and bacteria (S. aureus) were captured in the surface of MRGO. A large surface to volume ratio of MRGO provided a space to capture bacteria for the interaction between the MRGO and the cell walls of bacteria. The bacteria were embedded onto the surface of MRGO. On the other hand, magnetic exposure could separate directly the captured bacteria on MRGO in the external magnetic field. Nanomaterials 2020, 10, x FOR PEER REVIEW 9 of 14 graphene-based materials [30]. Another possible mechanism is the isolation of S. aureus from the surrounding medium during the incubation with the MRGO; therefore, blocking the consumption of nutrients for proliferation leads to the cells' death. In Figure 6, SEM images show that the morphology of MRGO sheets and bacteria (S. aureus) were captured in the surface of MRGO. A large surface to volume ratio of MRGO provided a space to capture bacteria for the interaction between the MRGO and the cell walls of bacteria. The bacteria were embedded onto the surface of MRGO. On the other hand, magnetic exposure could separate directly the captured bacteria on MRGO in the external magnetic field.

Hyperthermia Experiment and Bacteria Capturing and Removing Activity Assisted by HFMF Exposure
Finally, HFMF was conducted to investigate the magnetic inductive heating of MRGO and to evaluate its capability for killing bacteria. Inductive heating is thermal energy induced from the hysteresis loss of ferrites and is dependent on the type of re-magnetization process in the HFMF [31]. Figure 7a shows the results of the inductive heating ability of GO and MRGO. GO cannot heat up during the exposure to HFMF and the temperature remained around 25 °C (room temperature). Meanwhile, MRGO could increase in temperature up to ~42 °C, after the exposure to HFMF for 20 min. After HFMF was applied in MRGO, it could significantly decrease the colony number of bacteria, in comparison to that in GO (Figure 7b). The bacteria decrease from 10 5 CFU/mL (control) to ~10 2 CFU/mL (MRGO) after exposure to HFMF for 20 min, and only 0.1% of bacteria can survive. The calculated capability of MRGO to kill bacteria under HFMF is about 99.9% (Figure 7c). These results confirmed that the temperature increase in the MRGO was closely related to the incorporation of magnetic nanoparticles in the MRGO (Figure 7d). The generation of localized heat could disrupt the surrounding microbes [43].

Hyperthermia Experiment and Bacteria Capturing and Removing Activity Assisted by HFMF Exposure
Finally, HFMF was conducted to investigate the magnetic inductive heating of MRGO and to evaluate its capability for killing bacteria. Inductive heating is thermal energy induced from the hysteresis loss of ferrites and is dependent on the type of re-magnetization process in the HFMF [31]. Figure 7a shows the results of the inductive heating ability of GO and MRGO. GO cannot heat up during the exposure to HFMF and the temperature remained around 25 • C (room temperature). Meanwhile, MRGO could increase in temperature up to~42 • C, after the exposure to HFMF for 20 min. After HFMF was applied in MRGO, it could significantly decrease the colony number of bacteria, in comparison to that in GO (Figure 7b). The bacteria decrease from 10 5 CFU/mL (control) to~10 2 CFU/mL (MRGO) after exposure to HFMF for 20 min, and only 0.1% of bacteria can survive. The calculated capability of MRGO to kill bacteria under HFMF is about 99.9% (Figure 7c). These results confirmed that the temperature increase in the MRGO was closely related to the incorporation of magnetic nanoparticles in the MRGO (Figure 7d). The generation of localized heat could disrupt the surrounding microbes [43].

Conclusions
We developed an MRGO nanosheet through surface modification of GO using PDDA and coprecipitation of Fe3O4 nanoparticles onto the graphene sheets for capturing and separating microbes. TEM observation confirmed the formation of MRGO sheets, and Fe3O4 nanoparticles were able to disperse onto the graphene sheets. MRGO sheets could capture bacteria efficiently. When the content of MRGO reaches 2.4 mg/mL, the bacteria (1 × 10 5 CFU/mL) can be fully captured by MRGO.

Conclusions
We developed an MRGO nanosheet through surface modification of GO using PDDA and coprecipitation of Fe 3 O 4 nanoparticles onto the graphene sheets for capturing and separating microbes. TEM observation confirmed the formation of MRGO sheets, and Fe 3 O 4 nanoparticles were able to disperse onto the graphene sheets. MRGO sheets could capture bacteria efficiently. When the content of MRGO reaches 2.4 mg/mL, the bacteria (1 × 10 5 CFU/mL) can be fully captured by MRGO. Furthermore, the hyperthermia effect, due to the localized heating, could be generated by MRGO through HFMF application and could kill the bacteria. These novel MRGO sheets have the potential to be developed for capturing, separating, and disrupting microbes.