Preparation of Hydrophobic Octadecylphosphonic Acid-Coated Magnetite Nanoparticles for the Demulsification of n-Hexane-in-Water Nanoemulsions

To design more environmentally friendly, economical, and efficient demulsifiers for oily wastewater treatment, hydrophobic octadecylphosphonic acid (ODPA)-modified Fe3O4 nanoparticles (referred to as Fe3O4@ODPA) were prepared by condensation of hydroxyl groups between ODPA and Fe3O4 nanoparticles using the co-precipitation method. The prepared magnetite nanoparticles were characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscope (SEM), Fourier transform infrared (FTIR) spectroscopy, and thermogravimetric/differential thermogravimetric (TG/DTG) analysis. The water contact angles (θW) of Fe3O4@ODPA nanoparticles were more than 120°, indicating hydrophobic nature, and the diameter of the obtained spherical-shaped magnetite nanoparticles was 12–15 nm. The ODPA coating amount (AO) (coating weight per gram Fe3O4) and specific surface area (SO) of Fe3O4@ODPA were 0.124–0.144 g·g−1 and 78.65–91.01 m2·g−1, respectively. To evaluate the demulsification ability, stability, and reusability, the magnetite nanoparticles were used to demulsify an n-hexane-in-water nanoemulsion. The effects of the magnetite nanoparticle dosage (CS), pH value of nanoemulsion, and NaCl or CaCl2 electrolytes on the demulsification efficiency (RO) were investigated. The RO of Fe3O4@ODPA samples was found to be higher than that of bare Fe3O4 samples (S0, ST, and SN) under all CS values. With the increase in CS, the RO of Fe3O4@ODPA samples initially increased and then approached equilibrium value at Cs = 80.0 g·L−1. A maximum RO of ~93% was achieved at CS = 100.0 g·L−1 for the Fe3O4@ODPA sample S2. The pH and two electrolytes had a minor effect on RO. The Fe3O4@ODPA nanoparticles maintained high RO even after being reused for demulsification 11 times. This indicates that the hydrophobic Fe3O4@ODPA samples can be used as an effective magnetite demulsifer for oil-in-water nanoemulsions.

Notably, hydrophobic octadecylphosphonic acid (ODPA) coatings were prepared on oxidized copper mesh for self-cleaning and oil/water separation and on titanium dioxide for anti-fouling biomedical applications [40,41]. In addition, ODPA coating has been used to modify the surface of Ti6Al4V alloys to improve the corrosion resistance [42]. The hydroxyl groups on the surface of TiO2 and Cu(OH)2 facilitate the coating of ODPA on the metal surface [40,41]. This provides a possibility to design a new magnetite demulsifier by using ODPA to modify Fe3O4 nanoparticles with ample hydroxyl groups under appropriate conditions. A possible schematic representing the interaction between ODPA and the Fe3O4 surface is shown in Figure 1.  In this study, ODPA-modified Fe 3 O 4 nanoparticles (Fe 3 O 4 @ODPA) are synthesized using co-precipitation with or without cetyltrimethylammonium bromide (CTAB). The prepared nanoparticles are characterized using X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy (SEM), Fourier transform infrared (FTIR) spectroscopy, and thermogravimetric/differential thermogravimetric (TG/DTG) analysis. The demulsification ability and recyclability of Fe 3 O 4 @ODPA nanoparticles are investigated by using them to demulsify an n-hexane-in-water (O/W) nanoemulsion under an applied magnetite field. Furthermore, the effects of demulsifier dosage, pH, and electrolytes such as NaCl or CaCl 2 on the removal efficiency (R O ) of oil from the nanoemulsion are examined.

Materials
Analytical-grade ammonium hydroxide (25-28 wt% NH 3 in water), n-hexane, ethanol, HCl, NaOH (granular), and oil-soluble Sudan III were purchased from Damao Chemical Reagent Factory, Tianjin, China. Analytical-grade reagents, including FeCl 3 ·6H 2 O, FeSO 4 ·7H 2 O, CTAB, NaCl, and CaCl 2 , and chemically pure Tween 60, were purchased from Sinopharm Chemical Reagent Co., Shanghai, China. Analytical-grade ODPA was purchased from Shanghai Bide Pharmatech, Shanghai, China. All of the reagents were directly used as received without further purification. The deionized water used in this study was obtained from a Hitech-Kflow water purification system from Hitech, Shenzhen, China.

Preparation of Fe 3 O 4 @ODPA Nanoparticles
A previously reported [3,43] modified chemical co-precipitation method was used to prepare the Fe 3 O 4 @ODPA nanoparticles. The preparation process of Fe 3 O 4 nanoparticles can be found in our previous report [3]. Briefly, about 0.057 mol·L −1 of FeSO 4 ·7H 2 O and 0.123 mol·L −1 of FeCl 3 ·6H 2 O were dissolved in 350 mL deionized water under N 2 flow. The above solution was heated to 80 • C under vigorous stirring. Next, 20 mL ammonium hydroxide was added to the solution. A black magnetite suspension (Fe 3 O 4 nanoparticle suspension) was obtained, which was vigorously stirred for 30 min. Subsequently, nearly 0.50 g solid ODPA or alkaline ODPA solution (0.50 g ODPA dissolved in 50 mL alkaline solution) was added to the black magnetite suspension in the presence or absence of CTAB, and then the black suspension was continuously heated at 80 • C for 2 h. Next, the suspension was naturally cooled to room temperature, and the black magnetite nanoparticles were collected using a magnet, which were thoroughly washed with ethanol and deionized water in sequence. The obtained magnetite nanoparticles were dried in a vacuum oven at 60 • C to a constant weight, and Fe 3 O 4 @ODPA samples S1, S1T, and S2 were obtained.
Bare Fe 3 O 4 nanoparticles (S0, ST, and SN) were also synthesized for comparison. S0 and ST were prepared in the absence and presence of CTAB without adding ODPA using the same process as that for S1 and S1T, respectively. A bare Fe 3 O 4 nanoparticle sample (SN) was prepared using the same process as that for S2 without adding ODPA. The diameter and water contact angle (θ W ) of S0 were measured to be 11.3 nm and 29 • at pH = 6.3 [3].
The preparation conditions and characterization of the prepared magnetite nanoparticles are shown in Table 1.  Table 1 is the amount of ODPA added in the preparation process of Fe 3 O 4 @ODPA nanoparticles.

Preparation of n-Hexane-in-Water Nanoemulsions
To obtain the parent n-hexane-in-water nanoemulsion, Tween 60, n-hexane (dyed using Sudan III in oil phase), and deionized water were mixed in a mass ratio of 1:1:8 and stirred using a DS-1 homogenizer (Shanghai Specimen and Model Factory, Shanghai, China) at 10,000 rpm for 10 min. The parent nanoemulsion was diluted with an oil content of 25.0 g·L −1 and a mean droplet size of 311 nm (Figure 2), measured using a Zetasizer Nano S90 Mastersizer (Malvern Instruments Co., Malvern, UK), which was then used in the demulsification tests.

Preparation of n-Hexane-in-Water Nanoemulsions
To obtain the parent n-hexane-in-water nanoemulsion, Tween 60, n-hexane (dyed using Sudan Ⅲ in oil phase), and deionized water were mixed in a mass ratio of 1:1:8 and stirred using a DS-1 homogenizer (Shanghai Specimen and Model Factory, Shanghai, China) at 10,000 rpm for 10 min. The parent nanoemulsion was diluted with an oil content of 25.0 g·L −1 and a mean droplet size of 311 nm (Figure 2), measured using a Zetasizer Nano S90 Mastersizer (Malvern Instruments Co., Malvern, UK), which was then used in the demulsification tests. To investigate the effects of the pH value of the emulsion on the demulsification efficiency (RO), nanoemulsions were prepared using deionized water, whose pH value was previously adjusted by NaOH or HCl. To investigate the effects of NaCl or CaCl2 in the emulsion on RO, nanoemulsions were prepared using deionized water containing NaCl or CaCl2. The concentrations (CSalt) of NaCl or CaCl2 in deionized water were both 0.30 g·L −1 .

Demulsification Tests of n-Hexane-in-Water Nanoemulsion
To evaluate the demulsification ability of Fe3O4@ODPA for n-hexane-in-water nanoemulsion, the residual oil content (Ce) of the nanoemulsion (light red transparent liquid) was measured after settling it on a hand magnet. The designed amount of magnetite nanoparticles was thoroughly mixed with 20 mL freshly prepared nanoemulsion in a 50 mL glass vial, where the dosage (CS) of magnetite nanoparticles was 0.0-100.0 g·L −1 . The vials with the prepared mixtures were shaken in a THZ-82 thermostatic water bath shaker (Wuhan Grey Mo Lai Detection Equipment Co., Wuhan, China) under a speed of 240 cy-cles·min −1 at 25 °C for 3 h. The demulsification kinetic tests indicated that shaking for 3 h was enough to achieve demulsification equilibrium (seen Figure S1 in Supporting Information). Then, the magnetite nanoparticles and the adhered organics were separated by To investigate the effects of the pH value of the emulsion on the demulsification efficiency (R O ), nanoemulsions were prepared using deionized water, whose pH value was previously adjusted by NaOH or HCl. To investigate the effects of NaCl or CaCl 2 in the emulsion on R O , nanoemulsions were prepared using deionized water containing NaCl or CaCl 2 . The concentrations (C Salt ) of NaCl or CaCl 2 in deionized water were both 0.30 g·L −1 .

Demulsification Tests of n-Hexane-in-Water Nanoemulsion
To evaluate the demulsification ability of Fe 3 O 4 @ODPA for n-hexane-in-water nanoemulsion, the residual oil content (C e ) of the nanoemulsion (light red transparent liquid) was measured after settling it on a hand magnet. The designed amount of magnetite nanoparticles was thoroughly mixed with 20 mL freshly prepared nanoemulsion in a 50 mL glass vial, where the dosage (C S ) of magnetite nanoparticles was 0.0-100.0 g·L −1 . The vials with the prepared mixtures were shaken in a THZ-82 thermostatic water bath shaker (Wuhan Grey Mo Lai Detection Equipment Co., Wuhan, China) under a speed of 240 cycles·min −1 at 25 • C for 3 h. The demulsification kinetic tests indicated that shaking for 3 h was enough to achieve demulsification equilibrium (seen Figure S1 in Supporting Information). Then, the magnetite nanoparticles and the adhered organics were separated by a 3000 Gs NdFeB magnet (Zibo Dry Magnetite Industry Science and Technology Co., Zibo, China). The absorbance was measured at 400 nm using a 754 ultraviolet-visible (UV-Vis) spectrometer (Shanghai Precision Scientific Instrument Co., Shanghai, China), which was then utilized to determine the residual oil content though a standard curve obtained from a series of standard nanoemulsions with different oil contents [4]. Finally, the demulsification efficiency (R O ) could be calculated as follows: where C 0 (g·L −1 ) and C e (g·L −1 ) are the initial and residual oil contents of the nanoemulsions, respectively. The reported R O values are the average values obtained from three parallel tests. For comparison, blank tests were conducted on nanoemulsions without the demulsifier. The relative error of the demulsification tests was <6.0%.

Recycling Tests
To evaluate the reusability and stability of Fe 3 O 4 @ODPA, Fe 3 O 4 @ODPA sample S2 was collected by an additional magnet after the demulsification test. Subsequently, it was sequentially washed with dichloromethane, ethanol, and water and then dried in a vacuum oven at 60 • C. This recycling test has been used in our reports [3,4]. The recycled Fe 3 O 4 @ODPA nanoparticles were applied in the subsequent demulsification tests 11 times to monitor any reduction in demulsification efficiency (R O ).

Characterization of Magnetite Nanoparticles
The morphology of magnetite nanoparticles was characterized using TEM (JEM-F200 microscope, Jeol Corporation, Tokyo, Japan) and SEM (ZEISS sigma500, Zeiss, Jena, Germany). The crystal structures of the samples were examined by powder XRD (D8 advance model diffractometer, Bruker Co., Karlsruhe, Germany) with Cu K α radiation (λ = 0.154184 nm) at 40 kV and 40 mA [3,4]. The chemical bonds in the samples were identified using FTIR spectroscopy (FTIR-660 + 610 spectrometer, Agilent Technologies Co., Palo Alto, USA) by the KBr wafers method. The ODPA coating mount on the Fe 3 O 4 surface was measured by TG/DTG analysis (SDT-Q-600 thermal analysis system, TA Instruments Co., New Castle, DE, USA) [3]. In this analysis, the magnetite nanoparticles (~25 mg) were heated from room temperature to 800 • C at 10 • C·min −1 in an N 2 atmosphere. The conventional drop shape method was used to measure the water contact angles (θ W ) [3,44]. Firstly, a circular disc of magnetite nanoparticles with a thickness of approximately 2 mm was fabricated by a Shimadzu press at 10 MPa. Secondly, a droplet of deionized water (~3 µL) from a syringe was carefully dropped on the surface of the circular disc, and its image was taken after reaching equilibrium (~2 min) using a Datapyhsics OCA 20 contact angle goniometer (Kruss, Germany). The θ W value was the average value measured at three different locations on the disc surface.
The specific surface area (S a ) of the magnetite nanoparticles was determined by N 2 adsorption-desorption isotherm measurements (Quadrasorb SI-MP system, Quantachrome Instruments, Boynton Beac, Florida, USA), as shown in Table 1.

Shape and Size of Magnetite Nanoparticles
The TEM and SEM images of magnetite nanoparticle shape are shown in Figures 3 and 4, respectively. It is clear that all the magnetite nanoparticles had a spherical shape. Using the TEM images, the average diameter (D) of the prepared magnetite nanoparticles was measured to be 12.5-14.1 nm (see Table 1).

FTIR Spectroscopy
The FTIR spectra of solid ODPA and magnetite nanoparticles are shown in Figure 6. In the ODPA spectrum, the two sharp bands at 2922 and 2852 cm −1 were attributed to the

FTIR Spectroscopy
The FTIR spectra of solid ODPA and magnetite nanoparticles are shown in Figure 6. In the ODPA spectrum, the two sharp bands at 2922 and 2852 cm −1 were attributed to the

FTIR Spectroscopy
The FTIR spectra of solid ODPA and magnetite nanoparticles are shown in Figure 6. In the ODPA spectrum, the two sharp bands at 2922 and 2852 cm −1 were attributed to the characteristic asymmetric and symmetric C-H stretching vibrations, respectively [3,45,47,48]. The bands at 1471 cm −1 and 716 cm −1 were attributed to the ν(CH 2 ) and ν(P-C) vibrations, respectively [49]. The absorption peaks at 1228 and 947 cm −1 were assigned to ν as (P=O) and ν as (P-OH), and the bands at 1076 and 1005 cm −1 were assigned to the symmetric ν s (PO 3 ) and asymmetric ν as (PO 3 ) stretching vibrations, respectively [40,50,51]. The spectra of bare ST and SN samples showed peaks at 3442 and 1630 cm −1 , which were attributed to O-H vibrations. This indicates that abundant hydroxyl groups were present on the surface of the Fe 3 O 4 nanoparticle, proving the rationality of the scheme shown in Figure 1. The bands at 630 and 591 cm −1 were assigned to Fe-O lattice vibrations [3,26,27,52,53]. The peaks at 2922 and 2852 cm −1 , which correspond to Fe 3 O 4 [3], were absent in the FTIR spectrum of ST, indicating that CTAB was not bonded with Fe 3 O 4 .
The spectra of the Fe 3 O 4 @ODPA samples (S1, S1T, and S2) exhibited the characteristic ODPA bands at 2922 and 2852 cm −1 and bare ST and SN bands at 3442, 1630, 633, and 591 cm −1 . The weakened peak intensities at 3442 and 1630 cm −1 in the Fe 3 O 4 @ODPA samples were attributed to the fact that some hydroxyl groups covering the surface were replaced by ODPA. ODPA can modify the Fe 3 O 4 surface by monodentate, bidentate, or tridentate bonding [40,50]. The absence of peaks at 1228 and 947 cm −1 corresponding to ν as (P=O) and ν as (P-OH) vibrations indicates that ODPA was bound to the surface of Fe 3 O 4 nanoparticles by tridentate bonding [51,54]. This inference is consistent with the design scheme shown in Figure 1.
The spectra of the Fe3O4@ODPA samples (S1, S1T, and S2) exhibited the characteristic ODPA bands at 2922 and 2852 cm −1 and bare ST and SN bands at 3442, 1630, 633, and 591 cm −1 . The weakened peak intensities at 3442 and 1630 cm −1 in the Fe3O4@ODPA samples were attributed to the fact that some hydroxyl groups covering the surface were replaced by ODPA. ODPA can modify the Fe3O4 surface by monodentate, bidentate, or tridentate bonding [40,50]. The absence of peaks at 1228 and 947 cm −1 corresponding to νas (P=O) and νas (P-OH) vibrations indicates that ODPA was bound to the surface of Fe3O4 nanoparticles by tridentate bonding [51,54]. This inference is consistent with the design scheme shown in Figure 1.

TG/DTG Analysis
The TG/DTG curves of magnetite nanoparticles are shown in Figure 7. All the samples exhibited a weight loss peak below < 120 °C, which was ascribed to the desorption of physically adsorbed water. The bare Fe3O4 samples (ST and SN) showed a weight loss of 2.39 wt% and 3.84 wt% in the range of 120 °C to 600 °C, which was attributed to the condensation of surface hydroxyl groups [3,55]. The Fe3O4@ODPA samples (S1, S1T, and S2) exhibited two weight loss stages in the range of 120 °C to 800 °C. The first weight loss stage occurred at 120-600 °C due to the loss of the ODPA coated on the surface of magnetite nanoparticles [3,43,55,56]. In this stage, the weight loss occurred in two steps, as shown in the DTG curves in Figure 7. Two derivative peaks were observed at 202-331 °C and 437-492 °C, which were attributed to the breaking of different kinds of chemical bonds between -PO3 2− and Fe. The other weight loss stage appeared above 650 °C with a derivative peak at 708-770 °C, which may have been caused by the reduction of Fe3O4 by

TG/DTG Analysis
The TG/DTG curves of magnetite nanoparticles are shown in Figure 7. All the samples exhibited a weight loss peak below < 120 • C, which was ascribed to the desorption of physically adsorbed water. The bare Fe 3 O 4 samples (ST and SN) showed a weight loss of 2.39 wt% and 3.84 wt% in the range of 120 • C to 600 • C, which was attributed to the condensation of surface hydroxyl groups [3,55]. The Fe 3 O 4 @ODPA samples (S1, S1T, and S2) exhibited two weight loss stages in the range of 120 • C to 800 • C. The first weight loss stage occurred at 120-600 • C due to the loss of the ODPA coated on the surface of magnetite nanoparticles [3,43,55,56]. In this stage, the weight loss occurred in two steps, as shown in the DTG curves in Figure 7. Two derivative peaks were observed at 202-331 • C and 437-492 • C, which were attributed to the breaking of different kinds of chemical bonds between -PO 3 2− and Fe. The other weight loss stage appeared above 650 • C with a derivative peak at 708-770 • C, which may have been caused by the reduction of Fe 3 O 4 by the reducing substances produced during the ODPA degradation process. This result is consistent with the existing literature [3,43,45,56,57].
The ODPA coating amount (A O ) (g) on per gram Fe 3 O 4 for the Fe 3 O 4 @ODPA samples was calculated using the net weight loss of ODPA from the TG data, and the results are shown in Table 1. The A O values of S1 and S2 were 0.127 and 0.124 g·g −1 . Thus, the average A O value for S1 and S2 was approximately 0.126. However, the A O value of S1T was 0.144 g·g −1 , indicating that a slightly higher amount of ODPA was coated on the Fe 3 O 4 surface when CTAB participated in the synthesis process.

Effect of Demulsifier Dosage on the Demulsification Efficiency
The magnetite nanoparticles were used as a demulsifier for n-hexane-in-water nanoemulsion. Figure 8 shows the influence of demulsifier dosage (C S ) on the demulsification efficiency (R O ) at PH = 6.9 and 25 • C. As C S increased, the R O of Fe 3 O 4 @ODPA nanoparticles (S1, S2, and S1T) initially increased and tended to stabilize at C S~8 0.0 g·L −1 . The R O values of all Fe 3 O 4 @ODPA samples were found to be higher than those of the bare magnetite samples (S0, ST, and SN). The order of R O for the magnetite nanoparticles was S2 > S1T > S1 > bare Fe 3  the reducing substances produced during the ODPA degradation process. This result is consistent with the existing literature [3,43,45,56,57]. The ODPA coating amount (AO) (g) on per gram Fe3O4 for the Fe3O4@ODPA samples was calculated using the net weight loss of ODPA from the TG data, and the results are shown in Table 1. The AO values of S1 and S2 were 0.127 and 0.124 g·g −1 . Thus, the average AO value for S1 and S2 was approximately 0.126. However, the AO value of S1T was 0.144 g·g −1 , indicating that a slightly higher amount of ODPA was coated on the Fe3O4 surface when CTAB participated in the synthesis process.

Effect of Demulsifier Dosage on the Demulsification Efficiency
The magnetite nanoparticles were used as a demulsifier for n-hexane-in-water nanoemulsion. Figure 8 shows the influence of demulsifier dosage (CS) on the demulsification efficiency (RO) at PH = 6.9 and 25 °С. As CS increased, the RO of Fe3O4@ODPA nanoparticles (S1, S2, and S1T) initially increased and tended to stabilize at CS ~ 80.0 g·L −1 . The RO values of all Fe3O4@ODPA samples were found to be higher than those of the bare magnetite samples (S0, ST, and SN). The order of RO for the magnetite nanoparticles was S2 > S1T > S1 > bare Fe3O4 (SN, ST, and S0). The RO values of the three bare Fe3O4 nanoparticles were very close to each other, nearly 60%, indicating that the bare Fe3O4 nanoparticles had some interfacial activity. A maximum RO of ~93% was observed for the sample S2, verifying the potential of Fe3O4@ODPA nanoparticles in oil-water separation.
The specific surface areas (Sa) of determination were determined to further analyze the demulsification performance. The Sa values of Fe3O4@ODPA samples (S1, S1T, and S2) were 78.7, 87.0, and 87.0 m 2 ·g −1 , respectively, and less than those of the bare Fe3O4 samples (ST and SN) of 142.9 and 119.0 m 2 ·g −1 . The results do not seem to be related to demulsification performance. The relationship between specific surface area and demulsification ability needs to be studied further. It has been reported that the amount of coating modifier on the Fe3O4 surface and wettability can affect demulsification [3,58,59]. Therefore, the water contact angles of magnetite samples (pH 6.9) were also measured, and the results are shown in Figure 9. It can be seen that the θW values of the bare Fe3O4 samples (ST and SN) were 15° and 19°, respectively, indicating a hydrophilic nature. This is consistent with the hydrophilic Fe3O4 surface containing abundant hydroxyl groups. After coating ODPA, the interfacial activity of the magnetite nanoparticles increased, causing an increase in their demulsification ability.
This suggests that samples S1, S1T, and S2 contained ODPA-coated structures, and the higher θW of S2 may have been the reason for its better demulsification performance. According to the above results, the bare Fe3O4 nanoparticles had a similar size, similar shape, approximately similar θW (<90°), and approximately similar demulsifying ability.
This indicates that the properties of the bare Fe3O4 nanoparticles were unaffected by the addition of CTAB or alkaline aqueous solution in the absence of ODPA during the prepa- The specific surface areas (S a ) of determination were determined to further analyze the demulsification performance. The S a values of Fe 3 O 4 @ODPA samples (S1, S1T, and S2) were 78.7, 87.0, and 87.0 m 2 ·g −1 , respectively, and less than those of the bare Fe 3 O 4 samples (ST and SN) of 142.9 and 119.0 m 2 ·g −1 . The results do not seem to be related to demulsification performance. The relationship between specific surface area and demulsification ability needs to be studied further.
It has been reported that the amount of coating modifier on the Fe 3 O 4 surface and wettability can affect demulsification [3,58,59]. Therefore, the water contact angles of magnetite samples (pH 6.9) were also measured, and the results are shown in Figure 9. It can be seen that the θ W values of the bare Fe 3 O 4 samples (ST and SN) were 15 • and 19 • , respectively, indicating a hydrophilic nature. This is consistent with the hydrophilic Fe 3 O 4 surface containing abundant hydroxyl groups. After coating ODPA, the interfacial activity of the magnetite nanoparticles increased, causing an increase in their demulsification ability. The θ W values of Fe 3 O 4 @ODPA samples (S1, S1T, and S2) were 124 • , 125 • , and 133 • , respectively, demonstrating that they were hydrophobic due to the ODPA coating.
tively, indicating a hydrophilic nature. This is consistent with the hydrophilic Fe3O4 surface containing abundant hydroxyl groups. After coating ODPA, the interfacial activity of the magnetite nanoparticles increased, causing an increase in their demulsification ability.
This suggests that samples S1, S1T, and S2 contained ODPA-coated structures, and the higher θW of S2 may have been the reason for its better demulsification performance. According to the above results, the bare Fe3O4 nanoparticles had a similar size, similar shape, approximately similar θW (<90°), and approximately similar demulsifying ability.
This indicates that the properties of the bare Fe3O4 nanoparticles were unaffected by the addition of CTAB or alkaline aqueous solution in the absence of ODPA during the preparation process.  This suggests that samples S1, S1T, and S2 contained ODPA-coated structures, and the higher θ W of S2 may have been the reason for its better demulsification performance. According to the above results, the bare Fe 3 O 4 nanoparticles had a similar size, similar shape, approximately similar θ W (<90 • ), and approximately similar demulsifying ability. This indicates that the properties of the bare Fe 3 O 4 nanoparticles were unaffected by the addition of CTAB or alkaline aqueous solution in the absence of ODPA during the preparation process.

Effects of pH and Electrolytes on the Demulsification Efficiency
The influences of pH and electrolytes (NaCl or CaCl 2 ) on demulsification efficiency (R O ) were investigated to assess the demulsifying property of Fe 3 O 4 @ODPA in different aqueous environments. Figure 10 shows the effect of the pH value of the nanoemulsion on the R O of sample S2 under C S = 40.0 g·L −1 and 80.0 g·L −1 at 25 • C. It can be seen that the R O did not change significantly in the pH range of 4.8 to 10.5. This indicates that the pH value of nanoemulsion had a minor effect on the demulsification efficiency in the studied range. The result is consistent with an earlier report [3,34]. As was reported, the R O of Fe 3 O 4 @PEI@β-CD and Fe 3 O 4 @OA was reduced from 96.8% to 95.3%, and there were no obvious differences as the pH of the emulsion increased from 4 to 10, respectively [3,34]. The effect of NaCl or CaCl 2 with a concentration (C salt ) of 0.30 g·L −1 on the R O of sample S2 in the nanoemulsion at pH = 6.9 and 25 • C is shown in Figure 11. Compared with the n-hexane-in-water nanoemulsion without electrolyte (bare nanoemulsion), R O changed slightly after adding a low concentration of NaCl or CaCl 2 when C S increased from 20.0 to 80.0 g·L −1 . At C S = 80.0 g·L −1 , R O was 90.8%, 88.9%, and 91.1% for the bare nanoemulsion, nanoemulsion containing NaCl, and nanoemulsion containing CaCl 2 , respectively. This is consistent with an earlier report about the demulsification of Fe 3 O 4 @OA for cyclohexanein-water nanoemulsion [3]. consistent with an earlier report about the demulsification of Fe3O4@OA for cycloh in-water nanoemulsion [3].
Consequently, the variation in the pH value or the addition of low-concen NaCl or CaCl2 (Csalt = 0.30 g·L −1 ) hardly affected the demulsification performa Fe3O4@ODPA nanoparticles for n-hexane-in-water nanoemulsion. consistent with an earlier report about the demulsification of Fe3O4@OA for cyclohexa in-water nanoemulsion [3]. Consequently, the variation in the pH value or the addition of low-concentra NaCl or CaCl2 (Csalt = 0.30 g·L −1 ) hardly affected the demulsification performance Fe3O4@ODPA nanoparticles for n-hexane-in-water nanoemulsion. Bare namoemulsion NaCl CaCl 2 Figure 11. Influences of electrolytes (NaCl or CaCl 2 ) on R O of sample S2 at pH = 6.9 and C Salt = 0.30 g·L −1 .
Consequently, the variation in the pH value or the addition of low-concentration NaCl or CaCl 2 (C salt = 0.30 g·L −1 ) hardly affected the demulsification performance of Fe 3 O 4 @ODPA nanoparticles for n-hexane-in-water nanoemulsion.

Recyclability Test of Fe 3 O 4 @ODPA
Compared with conventional chemical demulsifiers, magnetite demulsifiers can be recycled and reused to demulsify different nanoemulsions [2,3,26]. To evaluate the recyclability of hydrophobic Fe 3 O 4 @ODPA, the S2 sample was reused for demulsification 11 times. As shown in Figure 12, the sample still maintained a high R O value after 10 cycles. Furthermore, the FTIR spectra and TG curves were obtained to observe the stability of the ODPA coating layer on the Fe 3 O 4 @ODPA surface (seen Figures S2 and S3 in Supporting Information), and no remarkable change was observed over the 10 cycles. This validates the good recycling performance of the hydrophilic Fe 3 O 4 @ODPA nanoparticles.
clability of hydrophobic Fe3O4@ODPA, the S2 sample was reused for demulsification 11 times. As shown in Figure 12, the sample still maintained a high RO value after 10 cycles. Furthermore, the FTIR spectra and TG curves were obtained to observe the stability of the ODPA coating layer on the Fe3O4@ODPA surface (seen Figures S2 and S3 in Supporting Information), and no remarkable change was observed over the 10 cycles. This validates the good recycling performance of the hydrophilic Fe3O4@ODPA nanoparticles.

Conclusions
Hydrophobic and quasi-spherical Fe3O4@ODPA nanoparticles were prepared using the modified co-precipitation method. The size, AO, and SO values of the Fe3O4@ODPA nanoparticles were approximately 12-15 nm, 0.124-0.144 g·g −1 , and 78.65-91.01 m 2 ·g −1 , respectively. The θW of Fe3O4@ODPA nanoparticles was more than 120°, indicating hydrophobicity. The Fe3O4@ODPA nanoparticles exhibited efficient demulsification and excellent recyclability for n-hexane-in-water nanoemulsions under an external magnetite field. Further, the change in pH value of n-hexane-in-water nanoemulsion and the addition of a small amount of NaCl or CaCl2 in the nanoemulsion had a minor effect on the demulsification efficiency of Fe3O4@ODPA nanoparticles. Sample S2 exhibited a maximum RO of 93% and excellent recyclability, with the highest θW of 133°. Overall, the findings of the study verify that Fe3O4@ODPA nanoparticles are an excellent demulsifier and can be effectively used for oily wastewater treatment.

Conclusions
Hydrophobic and quasi-spherical Fe 3 O 4 @ODPA nanoparticles were prepared using the modified co-precipitation method. The size, A O , and S O values of the Fe 3 O 4 @ODPA nanoparticles were approximately 12-15 nm, 0.124-0.144 g·g −1 , and 78.65-91.01 m 2 ·g −1 , respectively. The θ W of Fe 3 O 4 @ODPA nanoparticles was more than 120 • , indicating hydrophobicity. The Fe 3 O 4 @ODPA nanoparticles exhibited efficient demulsification and excellent recyclability for n-hexane-in-water nanoemulsions under an external magnetite field. Further, the change in pH value of n-hexane-in-water nanoemulsion and the addition of a small amount of NaCl or CaCl 2 in the nanoemulsion had a minor effect on the demulsification efficiency of Fe 3 O 4 @ODPA nanoparticles. Sample S2 exhibited a maximum R O of 93% and excellent recyclability, with the highest θ W of 133 • . Overall, the findings of the study verify that Fe 3 O 4 @ODPA nanoparticles are an excellent demulsifier and can be effectively used for oily wastewater treatment.