Graphene Oxide/Fe-Based Composite Pre-Polymerized Coagulants: Synthesis, Characterization, and Potential Application in Water Treatment

This study presents for the first time the synthesis and characterization of GO (graphene oxide), PFSiC (polyferric silicate chloride), and hybrid GO-PFSiC derivatives, aiming to enhance synergistically the performance of coagulation, when applied for the treatment of water. The structure and the morphology of composite GO-PFSiC coagulants were studied in detail by the application of FTIR, XRD, and SEM characterization techniques. Furthermore, the proposed coagulants were applied for the treatment of simulated turbid surface water. The effects of the reagent’s dosage, pH value, and experimental/operational conditions on the coagulation efficiency, applied mainly for the removal of turbidity, were examined. The results, obtained from the FTIR and XRD measurements, showed the presence of a bond between the PFSiC and the GO surface, indicating that the PFSiC particles are distributed uniformly on the surface of graphene, which was also confirmed by the SEM images. Especially, the composite compound GO-PFSiC1.5-15-0.5 presents the most uniform distribution of iron on the surface of graphene oxide and exhibits the optimum coagulation efficiency, while it significantly reduces the turbidity for doses above 3–5 mg/L, i.e., achieving the respective legislation limit as proposed by WHO. Specifically, at the alkaline pH values (>7.9), the removal of turbidity reaches 96%. Consequently, the results of this study render these materials as potential coagulant agents for further research and applications, aiming to also achieve the co-removal of other water components.


Introduction
The rapid development of nanomaterials, which present the advantages of larger surface areas and more activated functionalized sites, brings an alternative way for the preparation of novel specific reagents, dedicated to enhanced water purification [1,2]. Carbon-based nanomaterials are the most known representatives in this category, due to their numerous unique advantages, and considerable number of existing studies [3][4][5][6]. The most promising materials for treating and delivering pure water are considered those that incorporate nanoscale features and tailorable chemical properties [7].
Graphene [8,9], a two-dimensional (2-D) material, has attracted increasing interest in the field of composite materials, because of its excellent thermal [10], electrical [11], and mechanical [12] properties. This unique 2-D plane structure, coupled with an extremely high surface area, makes graphene an ideal support material. Multifunctional hybrid materials that can take advantage of the extraordinary C 2020, 6, 44 3 of 13

Preparation of Graphene Oxide (GO)
The graphite flakes, used for the preparation of graphene oxide, were purchased from Sigma Aldrich (USA). Graphite oxide (GrO) was prepared by using the modified Hummer's method [33], as improved by Debnath et al. (2014) [34]. In brief, graphite flakes (5.0 g) and NaNO 3 (2.5 g) were mixed with 120 mL of H 2 SO 4 (95%) in a 500-mL flask. The mixture was stirred for 30 min in an ice bath. Afterwards, 15.0 g KMnO 4 was added into the mixture under vigorous stirring, keeping the reaction temperature lower than 20 • C. The mixture was further stirred at room temperature overnight. Thereafter, another 15.0 g of KMnO 4 were added into the previous mixture and stirred at room temperature for another 2 h, aiming to oxidize totally any remaining graphite. On addition of the deionized water into this mixture, a yellowish paste was obtained, and the temperature was increased to about 98 • C with effervescence. This diluted suspension was then stirred for another 4 h and 50 mL of H 2 O 2 (30%) were added afterwards to the mixture. As the reaction progressed, the color of the pasty mixture gradually turned into light brownish. The mixture was subsequently washed with 100 mL of 5% HCl and 30% H 2 O 2 (five times) to purify it from any residuals of MnO 2 and sulfates. Finally, a solid was obtained after drying at 60 • C under vacuum.
Graphene oxide (GO) was prepared from the solid graphite oxide by sonicating the solid in double distilled water, aiming to convert it into a stable colloidal suspension. Then, the colloidal graphene oxide (GO) was precipitated by using 0.1 M NaOH solution and the solid precipitates were subsequently filtered through a 0.45-mm nylon membrane filter to separate the liquid phase and finally, they were subjected to drying at 60 • C under vacuum and the GO was obtained as grey powder.

Preparation of PFSiC
Composite poly-ferric-silicate chloride (PFSiC) coagulants were produced at room temperature (22 o C), according to a modified procedure of Zouboulis and Moussas (2008) [25] and Tolkou et al. (2015) [27] under various experimental conditions. The initial solutions used for the preparation of pre-polymerized coagulants were 0.5 M FeCl 3 ·6H 2 O (Merck) and 0.5 M NaOH (Merck, as the added alkaline agent). The 0.5 M polysilicic acid solution (pSi) was prepared according to the respective literature [26]: 1.74 mL of pSi solution were added to 20 mL of 0.5 M FeCl 3 ·6H 2 O solution, under vigorous stirring (0.3 mL/min, 20 rpm), and subsequently, 30 mL of alkaline solution were added slowly under magnetic stirring (0.1 mL/min, 70 rpm) in the mixture at the temperature of 30 • C. The produced composite pre-polymerized material, termed as PFSiC 1.5-15 to indicate the proportions of the aforementioned components, was left under stirring to mature for about 1 h and then diluted with water to the final concentration of 0.1 M (relevant to Fe). The main physico-chemical properties of this coagulant agent (used in the subsequent experiments) were pH value 1.9, turbidity 65 NTU, and conductivity 29 ms/cm.

Preparation of Composite GO-PFSiC Material
The Fe-containing graphene oxide composite materials were prepared after the impregnation of prepared graphene oxide with the pre-polymerized Fe-based coagulant (PFSiC), followed by thermal treatment at 85 • C. For a typical impregnation procedure, taking, e.g., GO-PFSiC 1.5-15-1 as a typical case with a molar ratio of [GO]/[Fe] = 1100 mg of GO were dispersed in 100 mL of water (1 g/L). The appropriate amount of aquatic solution of PFSiC 1.5-15 , i.e., 100 mg Fe (as 17.3 mL of 0.1 M coagulant solution) were added. Because of the highly acidic pH value of PFSiC 1.5-15 , the precipitation of ferric (hydro)oxide species [35] was avoided. After 30 min of stirring, the dispersion was sonicated for 30 min. It was then filtrated, washed with deionized water, and finally, it was dried in the air. A weighed amount of dried impregnated graphene oxide composite was placed in a furnace and heated up overnight to 85 • C. Two more relevant materials, but with different molar ratios, were also synthesized by following the same procedure, i.e.,  Table 1, while the weight ratio of Fe to GO was increased stepwise in order to define the most efficient coagulant. Small portions of coagulant powders were obtained after drying the aquatic solution of PFSiC in the oven (~40 • C). GO and GO-PFSiC composites were already in powdered form and used to observe the morphology of respective products, by employing a ZEISS EVO 50 scanning microscope, at EHT = 10 kV. The SEM sample was prepared by placing a drop of dilute ethanol dispersion of the prepared composites onto a copper plate attached to an aluminum sample holder and the solvent was allowed to evaporate at room temperature.

FT-Infrared Spectroscopy (FTIR)
FTIR spectra of the prepared composite products were recorded in the range of 4000−400 cm −1 , by using a Nicolet 380 FTIR Spectrometer-Thermo Scientific spectrophotometer to allow the observation of surface species and the possible polymeric interactions between the components.

X-ray Diffraction Spectroscopy (XRD)
The X-ray diffraction spectroscopy (XRD) is an important tool for the identification of crystalline compounds and it was applied to identify the possibly formed crystalline phases. The Malvern Panalytical Xpert instrument (Cu-K radiation) was used and the solid samples were analyzed in the range of 20 • -80 • 2θ with a scan rate of 1 • /min.

Coagulation Experiments Performed by Jar-Tests
Convenient jar-test experiments were applied for the examination of coagulants' efficiency, by using a jar-test apparatus (Aqualytic), equipped with six paddles, employing 1-L glass beakers. The simulated contaminated surface water (1 L) was prepared by using tap water of Thessaloniki (N. Greece) and clay (kaolin) suspension of an initial concentration of clay suspended particles of 10 mg/L. The pH was properly adjusted by using HCl or NaOH solutions of appropriate concentrations (1-0.01 M) and measured by a Metrohm Herisau pH meter. The properties of water samples are presented in Table 2, while the conditions used in the jar-test experimental runs are shown in Table 3.  After the coagulation treatment, samples were collected from the supernatant of each beaker and were analyzed for the determination of turbidity (NTU), by using a HACH RATIO/XR Turbidimeter. In order to evaluate the synergistic effect of GO and PFSiC in water treatment, residual turbidity values, obtained by the application of combined materials, were compared with those obtained with standalone GO and PFSiC reagents.
For clay suspension, the residual turbidity percentage (RT, %) was expressed as: where T and T o are the turbidities of the treated and raw water, respectively.

Characterization
The characterization of the new composite materials by using FTIR measurements was carried out to confirm the bonding or types of functional groups of molecules that are being present in the GO and can be impregnated into the new GO-PFSiC materials. FTIR analysis was performed at wavenumbers 400-4000 cm −1 , where in this range the characteristic absorption of functional group vibrations, as presented by the GO and the composites, can be identified. In Figure 1, the FTIR spectra can be shown, comparing the different examined materials, i.e., GO, PFSiC 1.5-15 , GO-PFSiC 1.5-15-1 , GO-PFSiC 1.5-15-0.5 , and GO-PFSiC 1.5-15-0.3 . In the FTIR spectra, primary hydroxyl C-OH (1380 cm −1 ) groups, epoxy C-O groups (1186 cm −1 ), and carbonyl C=O (1820 cm −1 ) groups can be detected in the GO spectrum, which reveals that large numbers of oxygen-containing groups have been introduced/interact into/with the GO structure [36,37]; noting also that there is an absence of absorption at the wavenumber 1725 cm −1 , which matches with the characteristics of C=C aromatic stretching vibration, indicating that GO is completely oxidized [38].  The band at 1038 cm −1 , which was found only in the spectra of pre-polymerized coagulants, can be assigned to the asymmetric Si-O stretching vibrations of Si-O-Si bonds, which is indicative of the high polymerization degree of silica. In addition, two characteristic peaks exist at 1100 and 1080 cm −1 that could be attributed to the asymmetric stretching vibration of Fe-O-Fe [30] and Si-O-Fe bonds [39], respectively. Moreover, in the spectra of PFSiC 1.5-15 and also of the composites GO-PFSiC, it can be observed that there is a peak at the wavenumber 3279.75 cm −1 , which indicates the characteristic absorption of stretching vibrations from the -OH group. The wide peaks in these vibration ranges are characteristic of -OH groups that undergo hydrogen bond formation. The absorption at the wave number 1623.36 cm −1 expresses the characteristic of vibration stretching from the C=C group. FTIR analysis can also provide a strong clue to the formation of the co-ordination bond C-O-Fe between the GO sheets and the iron particles. This can be noticed from the presence of the characteristic absorption bands of the stretching vibrations of Fe-O in the PFSiC material that were found at the wavenumber 617.19 cm −1 , which confirms the success of PFSiC loading/interactions with the GO sheets. The difference that can be observed between the FTIR spectrum of GO and the GO-PFSiC materials is that there is a weakening of the absorption intensity at the wave number around 1580 cm −1 , regarding the composite GO-PFSiC, as compared to simple GO, which at the wave number 1580 cm −1 is considered as an absorption characteristic of the carboxyl group. This shows that there has been a reduction in the quantity of carboxyl groups, because they are playing role as a ligand that can interacts with the Fe atoms of PFSiC [38,40,41].
The morphology and structure of GO and of the relevant composites were further examined by obtaining representative SEM images of these materials. As typical examples, the SEM images of GO, PFSiC, and GO-PFSiC materials are shown in Figure 2. Both the pristine GO and the prepared impregnated GO-PFSiC materials were found to exhibit a typically wrinkled sheet-like structure, as it can be observed, indicating that iron was homogeneously distributed onto the GO sheets. Especially, the composite GO-PFSiC 1.5-15-0.5 , as illustrated in Figure 2d, presents the most uniform distribution of iron on the surface of graphene oxide. The morphology and structure of GO and of the relevant composites were further examined by obtaining representative SEM images of these materials. As typical examples, the SEM images of GO, PFSiC, and GO-PFSiC materials are shown in Figure 2. Both the pristine GO and the prepared impregnated GO-PFSiC materials were found to exhibit a typically wrinkled sheet-like structure, as it can be observed, indicating that iron was homogeneously distributed onto the GO sheets. Especially, the composite GO-PFSiC1.5-15-0.5, as illustrated in Figure 2d, presents the most uniform distribution of iron on the surface of graphene oxide.   Figure 3b-e can be detected, which are the characteristic peaks of cubic Fe [42], indicating that the respective peaks of PFSIC1.5-15 are formed in the composites and that the PFSiC particles are distributed uniformly on the surface of graphene, which was also confirmed by the SEM images ( Figure 2) [43].
It should be noted, however, that the intense and characteristic peak at 45° belongs to the sample carrier during the analysis by the XRD instrument and not to another compound contained in the sample [24]. In addition, the sharp peak at 32° is observed only in Figure 3b, which corresponds to PFSiC1.5-15, and belongs to NaCl that is produced during the preparation of PFSiC1.5-15 solution and it disappeared in Figure 3c-e, which present the XRD patterns of solid composite GO-PFSiC hybrids. However, it is observed that with the increase of the Fe/GO ratio (see Figure 3b to Figure 3e, respectively), the intensity of the peak at 32° gradually decreases, indicating that at a higher [Fe]/[GO] ratio, amorphous ferric hydroxide materials are intercalated between the GO sheets and a homogeneous amorphous composite is formed [13].
In addition, as shown in Figure 3a, GO material presents two characteristic peaks at 25.58° and 38.8°, corresponding to the partially oxidized part that shows few layers of graphene, and to natural graphite, which undergoes imperfect oxidation, respectively. The presence of a bond between the PFSiC and the GO surface can be seen from the shift of the GO peak to 30.58°, which is caused by the reduction of some graphene oxides to graphene, due to the precipitation reaction of Fe ion [38,44]. The results obtained from the XRD analysis are presented in Figure 3 for the cases of GO (Figure 3a  The results obtained from the XRD analysis are presented in Figure 3 for the cases of GO (Figure  3a), PFSIC1.5-15 (Figure 3b), and GO-PFSiC hybrids (Figure 3c-e). From the XRD patterns, similar diffraction peaks at 35.6°, 43.3°, 53.6°, 57.3°, and 62.8° in Figure 3b-e can be detected, which are the characteristic peaks of cubic Fe [42], indicating that the respective peaks of PFSIC1.5-15 are formed in the composites and that the PFSiC particles are distributed uniformly on the surface of graphene, which was also confirmed by the SEM images ( Figure 2) [43].
It should be noted, however, that the intense and characteristic peak at 45° belongs to the sample carrier during the analysis by the XRD instrument and not to another compound contained in the sample [24]. In addition, the sharp peak at 32° is observed only in Figure 3b, which corresponds to PFSiC1.5-15, and belongs to NaCl that is produced during the preparation of PFSiC1.5-15 solution and it disappeared in Figure 3c-e, which present the XRD patterns of solid composite GO-PFSiC hybrids. However, it is observed that with the increase of the Fe/GO ratio (see Figure 3b to Figure 3e, respectively), the intensity of the peak at 32° gradually decreases, indicating that at a higher [Fe]/[GO] ratio, amorphous ferric hydroxide materials are intercalated between the GO sheets and a homogeneous amorphous composite is formed [13].
In addition, as shown in Figure 3a, GO material presents two characteristic peaks at 25.58° and 38.8°, corresponding to the partially oxidized part that shows few layers of graphene, and to natural graphite, which undergoes imperfect oxidation, respectively. The presence of a bond between the PFSiC and the GO surface can be seen from the shift of the GO peak to 30.58°, which is caused by the reduction of some graphene oxides to graphene, due to the precipitation reaction of Fe ion [38,44].

Determination of Optimum Coagulation Experimental Conditions
In Figure 4, the results of the experiments, as applied for the determination of optimum coagulation conditions during jar-test experimental runs, are presented. For this reason, GO and PFSiC1.5-15 were compared, regarding the removal of turbidity and expressed as the residual turbidity (NTU), (a) in different sedimentation times, and (b) in different rapid mixing period times (Table 3),  It should be noted, however, that the intense and characteristic peak at 45 • belongs to the sample carrier during the analysis by the XRD instrument and not to another compound contained in the sample [24]. In addition, the sharp peak at 32 • is observed only in Figure 3b, which corresponds to PFSiC 1.5-15 , and belongs to NaCl that is produced during the preparation of PFSiC 1.5-15 solution and it disappeared in Figure 3c-e, which present the XRD patterns of solid composite GO-PFSiC hybrids. However, it is observed that with the increase of the Fe/GO ratio (see Figure 3b-e, respectively), the intensity of the peak at 32 • gradually decreases, indicating that at a higher [Fe]/[GO] ratio, amorphous ferric hydroxide materials are intercalated between the GO sheets and a homogeneous amorphous composite is formed [13].
In addition, as shown in Figure 3a, GO material presents two characteristic peaks at 25.58 • and 38.8 • , corresponding to the partially oxidized part that shows few layers of graphene, and to natural graphite, which undergoes imperfect oxidation, respectively. The presence of a bond between the PFSiC and the GO surface can be seen from the shift of the GO peak to 30.58 • , which is caused by the reduction of some graphene oxides to graphene, due to the precipitation reaction of Fe ion [38,44].

Determination of Optimum Coagulation Experimental Conditions
In Figure 4, the results of the experiments, as applied for the determination of optimum coagulation conditions during jar-test experimental runs, are presented. For this reason, GO and PFSiC 1.5-15 were compared, regarding the removal of turbidity and expressed as the residual turbidity (NTU), (a) in different sedimentation times, and (b) in different rapid mixing period times (Table 3), based on previous research tests [27], by using 10 mg/L as an indicative dosage.

Determination of the Optimum Dosage and Type of Material
The coagulant dosage is considered to be among the most critical parameters that present a significant effect on the efficiency of the coagulation process. All prepared coagulants in this study were compared in terms of turbidity removal ( Figure 5) (expressed as NTU units), in order to obtain the optimum type of coagulant and subsequently, to determine the optimal dosage of the examined materials by treating simulated surface water samples, noting that the respective initial values of this water were 17.2 NTU and pH 7.6 ± 0.2 ( Table 2).
According to the obtained results, it can be observed that the application of composite/hybrid materials shows a better coagulation efficiency than the pure graphene oxide (GO) or the simple PFSiC1.5-15 coagulant. Particularly, the composite GO-PFSiC1.5-15-0.5 with the ratio [GO]/[Fe] = 0.5 was found to significantly reduce the turbidity for dosages above 3-5 mg/L, i.e., much lower than the relevant data found in the respective literature, e.g., >24 mg/L [23]. In addition, the turbidity removal was increased, as the dosage of GO-PFSiC1.5-15-0.5 was increased, reaching 93% (~1 NTU) for a dosage of 30 mg/L. According to the recommendations from the World Health Organization [32], turbidity must be lower than 5 NTU before the water can be adequately sanitized for human consumption and Particularly, when studying the different applied sedimentation times (Figure 4a), the solutions were stirred rapidly at 160 rpm for 2 min, followed by slow stirring at 40 rpm for 10 min, and then, the suspension was rested for 30-60 min without any stirring to allow the separation by sedimentation of the produced flocs. As noticed in Figure 4a, the optimum sedimentation time was found to be 50 min, where the system reaches a dynamic equilibrium, indicated by the relatively constant value of turbidity after 50 min of sedimentation.
Then, maintaining the optimal settling time at 50 min, the different durations of the initial rapid mixing period (Figure 4b) were tested at 160 rpm for 1, 1.5, 2, and 2.5 min, followed by the slow stirring at 40 rpm for 10 and 50 min of sedimentation. According to Figure 4b, the optimum time in this case was 2 min of rapid mixing. After that, the residual turbidity (NTU) remains rather constant. Therefore, the jar-test protocol selected for the optimization of coagulation performance was rapidly mixed at C 2020, 6, 44 9 of 13 160 rpm for 2 min, slowly stirred at 40 rpm for 10 min, and finally, sedimentation was performed for 50 min.

Determination of the Optimum Dosage and Type of Material
The coagulant dosage is considered to be among the most critical parameters that present a significant effect on the efficiency of the coagulation process. All prepared coagulants in this study were compared in terms of turbidity removal ( Figure 5) (expressed as NTU units), in order to obtain the optimum type of coagulant and subsequently, to determine the optimal dosage of the examined materials by treating simulated surface water samples, noting that the respective initial values of this water were 17.2 NTU and pH 7.6 ± 0.2 ( Table 2). significant effect on the efficiency of the coagulation process. All prepared coagulants in this study were compared in terms of turbidity removal ( Figure 5) (expressed as NTU units), in order to obtain the optimum type of coagulant and subsequently, to determine the optimal dosage of the examined materials by treating simulated surface water samples, noting that the respective initial values of this water were 17.2 NTU and pH 7.6 ± 0.2 ( Table 2).
According to the obtained results, it can be observed that the application of composite/hybrid materials shows a better coagulation efficiency than the pure graphene oxide (GO) or the simple PFSiC1.5-15 coagulant. Particularly, the composite GO-PFSiC1.5-15-0.5 with the ratio [GO]/[Fe] = 0.5 was found to significantly reduce the turbidity for dosages above 3-5 mg/L, i.e., much lower than the relevant data found in the respective literature, e.g., >24 mg/L [23]. In addition, the turbidity removal was increased, as the dosage of GO-PFSiC1.5-15-0.5 was increased, reaching 93% (~1 NTU) for a dosage of 30 mg/L. According to the recommendations from the World Health Organization [32], turbidity must be lower than 5 NTU before the water can be adequately sanitized for human consumption and should be ideally below than 1 NTU. It is worth noting that the optimum GO-PFSiC1.5-15-0.5 material, as illustrated in Figure 2d, presents the most uniform distribution of iron on the surface of graphene oxide. According to the obtained results, it can be observed that the application of composite/hybrid materials shows a better coagulation efficiency than the pure graphene oxide (GO) or the simple PFSiC 1.5-15 coagulant. Particularly, the composite GO-PFSiC 1.5-15-0.5 with the ratio [GO]/[Fe] = 0.5 was found to significantly reduce the turbidity for dosages above 3-5 mg/L, i.e., much lower than the relevant data found in the respective literature, e.g., >24 mg/L [23]. In addition, the turbidity removal was increased, as the dosage of GO-PFSiC 1.5-15-0.5 was increased, reaching 93% (~1 NTU) for a dosage of 30 mg/L. According to the recommendations from the World Health Organization [32], turbidity must be lower than 5 NTU before the water can be adequately sanitized for human consumption and should be ideally below than 1 NTU. It is worth noting that the optimum GO-PFSiC 1.5-15-0.5 material, as illustrated in Figure 2d, presents the most uniform distribution of iron on the surface of graphene oxide.

Determination of Optimum pH
The effect of pH on the coagulation ability of optimal GO-PFSiC 1.5-15-0.5 material, as obtained from the previous experiments, was examined in the pH range 4-9, by using 10 mg/L of GO-PFSiC 1.5-15-0.5 . The removal of turbidity, expressed as percentage (Figure 6a) and as residual values (NTU) (Figure 6b), is presented in Figure 6. According to the results, it is found that the treated water by the composite material GO-PFSiC 1.5-15-0.5 can follow the respective legislation (by WHO) limits with an increase of pH values over 8. In particular, at more alkaline pH values (>7.9), the removal of turbidity reaches 96%, by applying only the 10 mg/L dosage, while in pH 6, where very small flocks rather slowly formed and barely settled, the relative percentage decreases significantly (68%). However, at pH 4-5, the turbidity removal efficiency was highly improved (up to 75-85%), in accordance with the relevant literature [23]. The original pH value of the examined simulated surface water (at pH 7.6 ± 0.2) was in the optimum pH range for the treatment of turbid water samples, in agreement with Figure 6. In strong alkaline media, the formation of flocks was very quick and began even from the first minute of high-speed agitation.
composite material GO-PFSiC1.5-15-0.5 can follow the respective legislation (by WHO) limits with an increase of pH values over 8. In particular, at more alkaline pH values (>7.9), the removal of turbidity reaches 96%, by applying only the 10 mg/L dosage, while in pH 6, where very small flocks rather slowly formed and barely settled, the relative percentage decreases significantly (68%). However, at pH 4-5, the turbidity removal efficiency was highly improved (up to 75-85%), in accordance with the relevant literature [23]. The original pH value of the examined simulated surface water (at pH 7.6 ± 0.2) was in the optimum pH range for the treatment of turbid water samples, in agreement with Figure 6. In strong alkaline media, the formation of flocks was very quick and began even from the first minute of high-speed agitation.

Conclusions
This study presented the application of new composite materials obtained with the interaction between GO (graphene oxide) and PFSiC (polyferric silicate chloride), i.e., an Fe-based prepolymerized coagulant agent, as a promising hybrid material to synergistically enhance water treatment. The hybrid GO-PFSiC derivatives are firstly reported in the literature.
The structure and morphology of composite GO-PFSiC coagulants were studied in detail by the application of FTIR, XRD, and SEM techniques. According to the presented results, the success of PFSiC loading on GO sheets was confirmed by the FTIR spectra at a wavenumber of 617.19 cm −1 , corresponding to the Fe-O bond in the GO-PFSiC material. The XRD analysis confirmed the characteristic peaks of PFSIC1.5-15 in the composites with GO, indicating that the PFSiC particles were distributed uniformly on the surface of graphene, especially in the composite GO-PFSiC1.5-15-0.5, which was also confirmed by the SEM images.
Furthermore, the proposed coagulants were applied for the treatment of simulated surface water contaminated with clay. The effects of the dosage, pH, and experimental/operating conditions on the coagulation efficiency, regarding the removal of turbidity, were examined, aiming to determine the optimal turbidity removal conditions. The jar-test protocol selected for the best coagulation

Conclusions
This study presented the application of new composite materials obtained with the interaction between GO (graphene oxide) and PFSiC (polyferric silicate chloride), i.e., an Fe-based pre-polymerized coagulant agent, as a promising hybrid material to synergistically enhance water treatment. The hybrid GO-PFSiC derivatives are firstly reported in the literature.
The structure and morphology of composite GO-PFSiC coagulants were studied in detail by the application of FTIR, XRD, and SEM techniques. According to the presented results, the success of PFSiC loading on GO sheets was confirmed by the FTIR spectra at a wavenumber of 617.19 cm −1 , corresponding to the Fe-O bond in the GO-PFSiC material. The XRD analysis confirmed the characteristic peaks of PFSIC 1.5-15 in the composites with GO, indicating that the PFSiC particles were distributed uniformly on the surface of graphene, especially in the composite GO-PFSiC 1.5-15-0.5 , which was also confirmed by the SEM images.
Furthermore, the proposed coagulants were applied for the treatment of simulated surface water contaminated with clay. The effects of the dosage, pH, and experimental/operating conditions on the coagulation efficiency, regarding the removal of turbidity, were examined, aiming to determine the optimal turbidity removal conditions. The jar-test protocol selected for the best coagulation performance was rapid mixing at 160 rpm for 2 min, slow stirring at 40 rpm for 10 min, and finally, sedimentation for 50 min. According to the obtained results, the composite material GO-PFSiC 1.5-15-0.5 with the ratio [GO]/[Fe] = 0.5 can significantly reduce the turbidity for dosages above 3-5 mg/L, i.e., much lower than the respective presented values in the relevant literature. The effect of pH on the coagulation ability of GO-PFSiC 1.5-15-0.5 was tested in the pH range 4-9 by using a 10 mg/L dosage. It was found that the GO-PFSiC 1.5-15-0.5-treated water can follow the stringent legislation (by WHO) limits with an increase of the application pH values. In particular, at more alkaline pH values (>7.9), the removal of turbidity can reach 96%, by applying only a 10 mg/L as dosage.
Consequently, the results of this study render these materials as potential coagulants for further research and potential applications regarding the co-removal of other water components.