An omniphobic graphene oxide separation membrane, preparation method and use

Abstract

The application discloses a novel omniphobic graphene oxide (GO) membrane, which is prepared by plasma treatment of GO membrane, followed by fluorosilane grafting. The plasma treatment not only generates in-plane nanopores in GO nanosheets to provide more water channels, but also adds active sites for fluorosilane grafting to build a robust omniphobic membrane surface. The generated omniphobic membrane exhibits superior water flux and salt rejection rate in a direct contact membrane distillation (DCMD) process.

Description

Technical Field

[0001] This invention relates to a fully hydrophobic graphene oxide separation membrane, its preparation method, and its applications, belonging to the field of membrane separation technology. Background Technology

[0002] Membrane distillation (MD) is an emerging separation technology that combines membrane separation and evaporation. In the MD process, vapor transport through a porous hydrophobic membrane is driven by the temperature difference across the membrane. Compared with traditional membrane desalination processes, MD has the advantage of high salt rejection rate, and the water flux is almost unaffected by the salt content of the feedstock. Fully hydrophobic membranes have been shown to effectively alleviate membrane wetting caused by low surface energy contaminants [1-3]. Existing fully hydrophobic membranes are mainly based on hydrophobic polymers, such as PTFE and PVDF, which are also the most studied membrane materials in the MD process. In recent years, new membrane materials have been developed to improve the water flux, hydrophobic stability and antifouling performance of MD membranes. Among them, graphene family two-dimensional materials show great potential because of their unique atomic thickness, ultrathin films, frictionless carbon surfaces that enable ultrafast water transport, and anisotropic thermal conductivity that facilitates heat utilization. In the prior art, a new type of hydrophobic GO-based membrane was prepared by embedding SiO2 nanoparticles and grafting hexadecyltrimethoxysilane (HDTMS) [4], which showed potential desalination performance only in the vacuum MD process. Overall, current graphene-based films exhibit moderate MD desalination performance, and their enormous potential has not yet been fully realized. Furthermore, graphene-dominated films have not yet achieved a truly fully hydrophobic surface.

[0003] References

[0004] 1.Lu,KJ;Chen,Y.;Chung,TS,Design of omniphobic interfaces formembrane distillation-A review.Water Res 2019,162,64-77.

[0005] 2. Zhang, W.; Wang, Z.; Li, B., Omniphobic membrane with nest-like re-entrant structure via electrospraying strategy for robust membranedistillation. Journal of Membrane Science 2021, 640, 119824.

[0006] 3. Li, C.; Li, X.; Du,

[0007] 4.Mao, Y.; Huang, Q.; Meng, B.; Zhou, K.; Liu, G.; Gugliuzza, A.; Drioli, E.; Jin, W., Roughness-enhanced hydrophobic graphene oxide membrane for waterdesalination via membrane distillation. Journal of Membrane Science 2020, 611, 118364. Summary of the Invention

[0008] The technical problem this invention aims to solve is the wetting phenomenon caused by low surface energy contaminants in the feed of membrane distillation processes. This invention prepares a novel fully hydrophobic graphene oxide (GO) membrane by plasma treatment followed by fluoroalkyl grafting. The plasma treatment not only generates in-plane nanopores in the GO nanosheets to provide more water channels but also adds active sites for the fluoroalkyl grafting, thus constructing a robust fully hydrophobic membrane surface. The resulting fully hydrophobic membrane exhibits superior water flux and salt rejection in direct contact membrane distillation processes.

[0009] The technical solution is:

[0010] A fully hydrophobic graphene oxide separation membrane includes a base membrane and a separation layer composed of reduced graphene oxide (rGO) nanosheets loaded on the base membrane. The surface of the reduced graphene oxide nanosheets is modified with fluorinated silane. The separation layer has nanopores with an average size of less than 5 nm distributed in it. The water droplet contact angle of the separation layer is 100-170°.

[0011] The fluorinated silane is selected from 1H, 1H, 2H, 2N perfluorododecyltriethoxysilane.

[0012] The base film is obtained by electrospinning and is made of PVDF.

[0013] The preparation method of the fully hydrophobic graphene oxide separation membrane includes the following steps:

[0014] Step 1: Prepare a dispersion of graphene oxide nanosheets;

[0015] Step 2: The dispersion is loaded onto the surface of the base film using vacuum suction, followed by drying.

[0016] Step 3: Perform plasma modification treatment on the dried membrane;

[0017] Step 4: Immerse the plasma-modified membrane in a fluorosilane-containing solution for grafting reaction, followed by thermal reduction treatment.

[0018] The concentration of graphene oxide nanosheets in the dispersion is 1×10⁻⁶. -4 ~5×10 -3 The dispersion, by weight, is obtained by mixing water and ethanol.

[0019] The plasma treatment time is 5-20 seconds.

[0020] The concentration of the fluorinated silane in the solvent is 0.5-5 wt%, and the grafting reaction time is 5-30 h.

[0021] The heat reduction treatment time is 0.1-5 hours, and the temperature is 120-160℃.

[0022] The above-mentioned fully hydrophobic graphene oxide separation membrane is used in membrane distillation and dehydration.

[0023] Beneficial effects

[0024] This invention proposes a novel all-hydrophobic graphene oxide (GO) membrane for direct contact membrane distillation (DCMD). The novel all-hydrophobic membrane design mitigates wetting caused by low surface energy contaminants in the feed during membrane distillation. Plasma treatment of the surface of stacked GO layers creates in-plane nanopores within the GO nanosheets, providing more transport channels for water. Simultaneously, plasma treatment introduces more oxygen-containing groups to promote subsequent fluoroalkyl grafting, creating a robust all-hydrophobic surface. Therefore, this plasma-assisted preparation strategy simultaneously improves the water flux, salt rejection, and anti-wetting properties of the graphene-dominated membrane. Attached Figure Description

[0025] Figure 1 : Schematic diagram of the fabrication of a fully hydrophobic GO membrane, (a) assembling GO stacks; (b) plasma treatment to generate in-plane nanopores and adding oxygen-containing groups on GO nanosheets; (c) silane coating to graft fluoroalkyl chains onto the surface of GO nanosheets.

[0026] Figure 2SEM images of (a) the surface and (b) the cross-section of the rGO-P-FAS film; TEM images of the original rGO and (d) rGO-P-FAS nanosheets after thermal reduction.

[0027] Figure 3 : SEM images of the surface of (a) rGO and (b) rGO-FAS films.

[0028] Figure 4 : (a) TEM images of pristine rGO and (b) rGO-P-FAS nanosheets after thermal reduction.

[0029] Figure 5 FTIR spectra of pristine GO, GO-P, rGO-P-FAS membranes and FAS.

[0030] Figure 6 XPS spectra of (a) rGO membrane and (b) rGO-P-FAS membrane.

[0031] Figure 7 EDX images of (a) rGO-FAS membrane and (b) rGO-P-FAS membrane.

[0032] Figure 8 (a) Water contact angle of GO membrane surfaces with different plasma treatment times; (b) Contact angle of rGO, rGO-FAS and rGO-P-FAS membranes with different liquids.

[0033] Figure 9 Water contact angle of pure GO membrane after 5 seconds of plasma treatment.

[0034] Figure 10 SEM image of rGO-P-FAS membrane after 10 seconds of plasma treatment.

[0035] Figure 11 DCMD separation performance of rGO-P-FAS membranes with different plasma treatment times (60℃, 35 g·L⁻¹) -1 (NaCl + 0.2 mMSDS mixed solution). Detailed Implementation

[0036] Example 1: Preparation of a fully hydrophobic GO membrane

[0037] A PVDF solution was prepared by dissolving 10 wt% PVDF in a mixed solvent of DMF and acetone (4:1 wt%) by stirring for 12 hours. The resulting PVDF solution was then loaded into the syringe of an electrospinning apparatus to manufacture an electrospun PVDF membrane.

[0038] 0.25 mg, 0.5 mg, and 1 mg of GO nanosheets were uniformly dispersed into a solution consisting of 50 g of deionized water and 120 g of ethanol by sonication for 2 hours. The nanosheets were then deposited onto an electrospun polyvinylidene fluoride (PVDF) substrate by vacuum filtration to obtain pure GO films.

[0039] The pure GO film, dried at room temperature, was exposed to plasma and treated in air at a power of 26 W and a frequency of 8.4 kHz for 5-20 s to generate in-plane nanopores, which were named GO-P.

[0040] Subsequently, the membrane was immediately immersed in a hexane solution containing 2% v / v 1H,1H,2H,2N perfluorododecyltriethoxysilane (fluorinated alkylsilane, FAS) for 24 hours. The FAS reacts with the hydroxyl groups on the membrane surface via hydrolysis, resulting in the membrane covalently bonded with fluoroalkyl chains. The membrane was then heat-treated at 140°C for 1 hour to remove any oxygen-containing groups that had not yet reacted with the fluorinated alkylsilane through thermal reduction. The GO reduced by this thermal treatment is called reduced GO (rGO). Therefore, this membrane is labeled as rGO-P-FAS (…). Figure 1 ).

[0041] Compare with Example 1

[0042] As a control, compared to Example 1, a GO film directly grafted with FAS and subjected to heat treatment without plasma treatment was labeled as rGO-FAS.

[0043] The GO nanosheets used for membrane fabrication had a lateral dimension of approximately 700 nm and a thickness of approximately 1 nm, as analyzed by AFM. The high aspect ratio facilitated the deposition of GO nanosheets on a porous support and their assembly into GO stacks. A porous support with an average pore size of approximately 650 nm was prepared by electrospinning PVDF nanofibers. SEM characterization confirmed that uniform deposition of GO nanosheets on the PVDF nanofiber support was clearly observable. Figure 2 (Region a). The nanofiber structure on the film surface is obvious, which is beneficial to improving the surface roughness of the GO film. By controlling the deposition amount of GO nanosheets, the thickness of the GO film layer was optimized to approximately 200 nm. Figure 2 (Region b) is used to balance the elimination of defects and the reduction of transmission resistance.

[0044] No changes in film morphology were observed under SEM after plasma and / or FAS coating treatment. No visible pores were found on the surface of the rGO-P-FAS or rGO-FAS films, consistent with the original GO films. Figure 3 Unlike traditional hydrophobic porous membranes, GO membranes prepared by stacking nanosheets primarily transport water vapor through their interlayer channels while simultaneously intercepting liquid water and salt ions. However, as characterized by ACTEM, plasma treatment can generate in-plane nanopores in GO nanosheets. Figure 2 Regions c and d show the thermally reduced GO nanosheets before and after 5s plasma treatment, respectively. The gray areas represent graphene regions, and the bright white areas represent functional group regions. Figure 4 The irregular light gray areas highlighted by the circle represent nanopores. TEM results show that plasma treatment resulted in the formation of more nanopores with an average size of less than 5 nm, which can provide more transport channels for water to pass through the GO membrane. Based on this, the present invention also provides a method for generating nanopores on GO nanosheets.

[0045] Chemical surface property characterization

[0046] FTIR characterization is used to study the chemical properties of pure GO, GO-P, rGO-P-FAS membranes, and FAS reagents, such as... Figure 4 As shown. Typical GO peaks are present in the spectra of GO and GO-P films, including hydroxyl peaks (3389 cm⁻¹). -1 ,-OH), carboxyl group (1761cm) -1 C=O; 1453cm -1 C-OH), aromatic ring (1625cm) -1 Compared to GO, the GO-P film showed an increased peak intensity of -OH, indicating that plasma treatment added hydroxyl groups to the film surface. More hydroxyl groups mean more active sites for subsequent fluorosilane grafting. The spectrum of the rGO-P-FAS film exhibited Si-O-Si asymmetric vibrations (1087 cm⁻¹). -1 ) and CF tensile vibration (1240cm) -1 -CF3; 1190cm -2 The peaks (-CF2-) correspond to the characteristic peaks of FAS, indicating that FAS has been introduced into the GO film. Further XPS analysis was then used to analyze the chemical bonding of the film. Figure 6 As shown, for the pure GO film, four typical GO peaks were detected: CC (284.8 eV), CO (3287.26 eV), C=O (288.76 eV), and OC=O (289.46 eV). Deconvolution of the C1s spectrum of the rGO-P-FAS film revealed two additional peaks at 291.7 and 293.8 eV, belonging to the CF2-CF2 and -CF2-CF3 groups, respectively. XPS analysis also confirmed the successful grafting of fluoroalkyl chains onto the film surface. Furthermore, EDX scanning revealed the presence of Si elements on the rGO-FAS and rGO-P-FAS surfaces. Figure 7 The distribution of dots in the image is shown. The number of yellow dots on the rGO-P-FAS membrane surface is significantly greater than that on the rGO-FAS surface, indicating that plasma treatment increases the grafting density of fluoroalkyl chains on the membrane surface.

[0047] Effect of plasma treatment time on the surface hydrophilicity of GO membranes

[0048] As shown in Figure 8 a, as the plasma treatment time is extended to 15 s, the water contact angle on the surface of the GO membrane gradually decreases from 42.5° to 16.3°. Within the measurement period of 250 s, the water contact angle of the GO membrane treated with 5 s plasma continuously decreases from 35.4° to 23.8°( Figure 9 ). Finally, after 20 seconds of plasma treatment, the surface of the GO membrane can be wetted by water immediately. The significant enhancement of surface hydrophilicity can be attributed to the introduction of more hydroxyl groups on the GO nanosheets, which facilitates the condensation reaction between C-OH on GO and Si-OH of FAS to improve the grafting density of fluoroalkyl grafting. Although the membrane surface becomes more hydrophilic with the increase of plasma treatment time, excessive plasma treatment introduces large-sized defects in the membrane( Figure 10 ).

[0049] The surface free energy and wettability of the membranes were evaluated by liquid-solid contact angle measurement. Different typical liquids were used, including water, ethanol, diiodomethane, and 35g·L containing 0.2mM SDS -1 NaCl solution (the feed solution for DCMD), and the contact angles on the membrane surface were measured. As shown in Figure 8 region b, compared with the rGO and rGO FAS membranes, the rGO-P-FAS membrane has higher contact angles for all liquids (i.e., water, 113.3±1.2°; ethanol, 71.6±0.6°; diiodomethane, 99.9±1.5°; NaCl solution containing 0.2mM SDS, 106.3±0.4°). The results confirmed that omniphobic rGO membranes were successfully prepared. According to the Owens-Wendt two-liquid geometric mean method, the contact angles of water (polar liquid) and diiodomethane (non-polar liquid) were used to determine the surface free energy of the membranes. As shown in Table 1, the order of surface free energy from low to high is rGO-P-FAS < rGO-FAS < rGO membrane (note that all membranes were thermally reduced at 140°C for 1 hour).

[0050] Table 1. Surface free energy of different rGO membranes.

[0051]

[0052] Since FAS reduces the surface free energy of the membrane, even though the surface morphologies of rGO, rGO-FAS and rGO-P-FAS membranes are almost identical, the rGO membrane with FAS coating has higher omniphobicity. The lowest surface free energy (9.54mJ·m -2This indicates that plasma treatment prior to FAS coating provides more efficient grafting of fluoroalkyl chains onto the membrane surface than direct FAS coating. Furthermore, the resulting fully hydrophobic rGO-P-FAS membrane exhibits better performance against a feed solution (containing 0.2 mM SDS in a 35 g·L⁻¹ solution). -1 NaCl solution exhibits excellent anti-wetting properties.

[0053] DCMD Desalination Test Method: The desalination performance and anti-wetting performance of the membrane were evaluated using a direct contact membrane distillation (DCMD) process. The membrane module was in countercurrent / crossflow mode, with an effective membrane area of ​​4.9 cm². 2 Feed (35g·L) -1 The NaCl solution was maintained at 50-80℃ using an oil bath, while the permeate side (deionized water) was maintained at 20℃ using a cryostat. In the DCMD wetting test, 35 g / L solutions containing 0.2 mM SDS were used. -1 NaCl solution and deionized water were used as feed and permeate solutions. Then, 0.05 mM SDS was added to the feed solution every two hours until the SDS concentration reached 0.4 mM to investigate the effect of SDS concentration on membrane performance. In the MD fouling experiment, a concentration of 30 mg / L was used. -1 35 g·L of humic acid -1 NaCl solution was used as the feed. The membrane's permeation flux and salt rejection rate were calculated from the change rate of the permeate mass and salinity, respectively.

[0054] The effect of different plasma treatment times on DCMD performance was tested. Similarly, the water flux and salt rejection rate of these membranes were compared after 1 hour of testing. Figure 11 As shown, the membrane flux increases with increasing plasma treatment time, proving that plasma treatment can indeed increase the number of transport channels to improve water flux. However, the salt rejection rate decreases significantly with prolonged plasma treatment time because longer plasma treatment times lead to large-sized defects on the membrane surface. Therefore, this experiment selected 5-10 seconds as the optimal plasma treatment time.

[0055] Stability of rGO-P-FAS membrane

[0056] The DCMD desalination performance of rGO membranes under different preparation conditions was compared. After 70 minutes of DCMD operation (35 g·L⁻¹), the performance was compared. - 1 A NaCl solution containing 0.2 mM SDS at 60 °C was used to wet the pristine rGO membrane, which underwent thermal reduction at 140 °C. This was reflected in the rapid decrease in flux and salt rejection. After approximately 12 hours of testing, the water flux of the rGO-FAS membrane dropped to approximately 10 kg·m³. -2 ·h -1The salt rejection rate decreased slowly. This suggests that the typical wetting phenomenon may be caused by the low / weak grafting density of fluoroalkyl chains on the GO nanosheets. Furthermore, the water flux of the rGO-P-FAS membranes treated for 5 s and 10 s (labeled rGO-P-5s-FAS and RG0-P-10s-FAS, respectively) was significantly higher than that of the rGO and rGO-FAS membranes, indicating that the nanopores generated by plasma treatment provide additional channels for water vapor transport while effectively retaining liquid.

[0057] Subsequently, continuous DCMD tests were performed on the rGO-P-FAS membranes with different plasma treatment times. Although the initial flux of the rGO-P-10s-FAS membrane was higher than that of the rGO-P-5s-FAS membrane, after 30 hours of testing, both flux and salt rejection rate began to decrease (35 g·L⁻¹). -1 NaCl solution containing 0.2 mM SDS at 60 °C. However, the DCMD test time of the fully hydrophobic rGO-P-5s-FAS membrane was significantly improved, exhibiting high and stable flux and salt rejection even after more than 450 hours during the DCMD process (35 L of solution containing 0.2 or 0.4 mM SDS). -1 (NaCl solution). During the first 370 hours of the experiment, the feed solution was 35 g / L containing 0.2 mM MSDS. - 1 NaCl solution, membrane water flux of ~35 kg·m -2 ·h -1 The salt rejection rate was 99.99%. In subsequent tests, the SDS concentration in the feed solution was increased to 0.4 mM, and then the membrane flux was slowly reduced to approximately kg·m⁻²·h⁻¹, with the salt rejection rate decreasing slightly to 99.9%. However, the MD process remained stable during subsequent 80-hour continuous testing. The significant enhancement in the stable DCMD operating time of the rGO-P-FAS membrane can be attributed to the plasma treatment introducing more active sites to generate a high density and more fluoroalkyl grafts on the GO membrane surface. For comparison, we summarize the latest fully hydrophobic membranes for DCMD processes with feed solutions containing low surface energy contaminants. These fully hydrophobic membranes are primarily based on hydrophobic polymers such as PTFE and PVDF. Most membranes were tested for stability performance for only about 10 hours.

[0058] One of the main indicators for measuring membrane wettability is the liquid inlet pressure (LEP), which is affected by the material surface energy, the surface tension of the feed solution, and the membrane pore size and geometry. LEP is defined as:

[0059]

[0060] Where γ Lθ is the surface tension of the liquid, θ is the apparent contact angle (CA) between the feed solution and the membrane surface, and r max is the maximum pore size of the membrane, and B is the pore geometry coefficient (B = 1.0 for circular pores). The presence of low-level surfactants such as SDS reduces the surface tension of the feed solution. LEP is linearly proportional to surface tension, therefore surfactants in the feed solution facilitate wetting. Consequently, existing membranes struggle to withstand prolonged DCMD processes. In contrast, the rGO-P-FAS membrane exhibits superior stability and water flux during DCMD. The rGO-P-FAS membrane surface is defect-free. Water vapor transport through this membrane is primarily facilitated by interlayer channels in the GO nanosheets and in-plane nanopores generated by plasma treatment. These pores are significantly smaller than those in PVDF or PTFE membranes. With the same feed solution, smaller pores result in higher LEP, making the membrane more difficult to wet. Due to ultrafast water transport and reduced energy consumption caused by anisotropic thermal conductivity, the rGO-P-FAS membrane achieves relatively high water flux while maintaining a stable high salt rejection rate.

Claims

1. The application of a fully hydrophobic graphene oxide separation membrane in direct contact membrane distillation and desalination, characterized in that, The fully hydrophobic graphene oxide separation membrane includes a base membrane and a separation layer composed of reduced graphene oxide nanosheets loaded on the base membrane. The surface of the reduced graphene oxide nanosheets is modified with fluorinated silane. The separation layer has nanopores with an average size of less than 5 nm. The water droplet contact angle of the separation layer is 100-170°. The fluorinated silane is 1H,1H,2H,2H perfluorododecyltriethoxysilane; The preparation method of the fully hydrophobic graphene oxide separation membrane includes the following steps: Step 1: Prepare a dispersion of graphene oxide nanosheets; Step 2: The dispersion is loaded onto the surface of the base film using vacuum suction, followed by drying. Step 3: Perform plasma modification treatment on the dried membrane; Step 4: Immerse the plasma-modified membrane in a fluorosilane-containing solution for grafting reaction, followed by thermal reduction treatment. The plasma treatment time is 5-20 seconds; The concentration of the fluorinated silane in the solvent is 0.5-5 wt%; The grafting reaction time is 5-30 hours.

2. The application according to claim 1, characterized in that, The base film is obtained by electrospinning and is made of PVDF.

3. The application according to claim 1, characterized in that, The concentration of graphene oxide nanosheets in the dispersion is 1×10⁻⁶. -4 ~5×10 -3 The dispersion, by weight, is obtained by mixing water and ethanol.

4. The application according to claim 1, characterized in that, The heat reduction treatment time is 0.1-5 hours, and the temperature is 120-160℃.

Images