Anti-fouling copper hydroxide nanowires (CHNs)-graphene oxide (GO) nanocomposites membrane was fabricated by a vacuum-assisted filtration self-assembly process. CHNs were covered on the surface and inserted into the interlayers of the GO nanosheets to form the rough surface and nanostructure channels. The membrane with water contact angles (CAs) of 53° and oil CAs of 155° exhibited superior stability, hydrophilicity, underwater superoleophobicity and ultralow oil adhesion, and hence it could separate the oil-water emulsion with a high efficiency of >99%. This membrane showed the combined advantages of high oil rejection rate and ultralow membrane fouling, making it promising for practical oil-water emulsion separation applications.
The increasing release of industrial oily wastewater as well as frequent crude oil leakage is seriously contaminating the environment. Oil contaminants in water can be divided into free oil (150 μm), dispersed oil (20–150 μm) and emulsified oil (<20 μm) according to the size of oil droplets. Among them, emulsified oil remains the most difficult to clean up or recycle from water via traditional methods because of its micro- and nano-sized oil droplets and stability (Xue et al. 2014; Fan et al. 2015). Significant attention has been paid to membrane technology to break emulsions due to its high-selectivity, small footprint, and low energy cost. Recently, a series of superwetting materials have been fabricated by designing surfaces in combination with surface chemistry and roughness (Yu et al. 2015; Li et al. 2017b). Our group reported a modified superhydrophilic and underwater superoleophobic PVDF membrane for oil-water separation (Liu et al. 2016a). However, successful application of membranes is still limited, and the major challenges are the instability of the materials, membrane fouling and low water fluxes. Therefore, there is an urgent need to prepare novel membranes, that can maintain high stability and treat emulsified water without oil fouling effectively.
Graphene oxide (GO) is a two-dimensional material with oxygenated graphene sheets bearing carboxyl, hydroxyl, and epoxide functional groups (Dikin et al. 2007; Krishnamoorthy et al. 2013). GO nanosheets offer an extraordinary potential in enhancing the flux or antifouling performance of different membranes. Our previous studies have shown that GO coatings have properties in oil-water separation (Liu et al. 2016b). However, GO-modified membranes made by filtration are unstable in water because of their extremely hydrophilicity (Dreyer et al. 2009). Hence, to enlarge the application fields of GO-based membranes, the nanochannel structures within GO-based membranes must be delicately regulated and optimally integrated with functional surfaces/interfaces by nanomaterials (Zhao et al. 2016). In some researches, copper hydroxide nanowires have been used for water treatment (Huang et al. 2013; Zhang et al. 2013). CHNs have an appropriate length-diameter ratio. In addition, the CHNs are cross-overlapping and properly deposited into a three-dimensional network structure, which facilitates the construction of nanochannels of composite materials.
In this study, we report on a GO based membrane with the surface and internal channel structure regulated by CHNs, endowing composite membranes with good stability, high separation efficiency, as well as superior anti-fouling properties. The preparation process is simple blending, followed by vacuum filtration, and can be easily applied to a large surface area. The composite membranes can separate the oil-water emulsion with high efficiency of more than 99%.
MATERIAL AND METHODS
Synthesis of GO and CHNs nanomaterials
GO is derived from natural graphite ﬂakes through the modified Hummers oxidation method (Hummers & Offeman 1958). CHNs were prepared by the chemical co-precipitation method. Briefly, 0.998 g of CuSO4·5H2O was added to 20 mL of deionized water and stirred for 15 min. After that, 30 mL of aqueous ammonia (0.15 M) was added to the solution and allowed to stand for 15 min. Then, 2.88 mL of sodium hydroxide solution (2.5 M) was added and allowed to stand for 15 min. Finally, the solution was washed with deionized water, dried and ground into powder.
FABRICATION OF GO AND CHNS-GO MODIFIED MEMBRANES
Twenty μg of GO and 80 μg of CHNs were dispersed into 50 mL of water, followed by ultrasonication for 30 min. The resulting CHNs-GO dispersions were filtrated by vacuum on mixed cellulose ester (MCE) membrane (pore size 220 nm). The as-prepared CHNs-GO modified membranes were dried at 50 °C overnight. GO-modified membranes were prepared in the same way without the addition of CHNs.
The morphology of each membrane surface was observed by scanning electron microscopy (SEM, S-4800, Hitachi, Japan). X-ray diffraction (XRD) was performed on an X-ray powder diffraction (Dmax-2500, Rigaku, Japan). Contact angles (CAs) were measured on a CA goniometer (CM20, KSV, Finland). The oil concentration was measured by UV spectrophotometry (UV-2450, Shimadzu, Japan). The particle size of the emulsions was measured by optical microscopy (EX20, Sunny, China).
OIL–WATER SEPARATION TEST
The oil-water emulsion was prepared by mixing 100 mg of diesel oil, 100 mL of deionized water and 10 mg of sodium dodecyl sulfate (SDS) by ultrasonication for 30 min. The emulsion was stable for at least 1 week without obvious stratification.
RESULTS AND DISCUSSION
Characterization of membranes
The self-assembly of CHNs-GO modified membranes via vacuum filtration is schematically depicted in Figure 1. CHNs were covered on the surface and inserted into the interlayers of the GO nanosheets to form the rough surface and nanostructure channels. The as-prepared CHNs-GO modified membrane is flexible for application in different structures. The epoxy, hydroxyl, carbonyl, and carboxyl groups on the GO nanosheets acted as anchoring sites for the attachment (Chen et al. 2010). Therefore, CHNs with plentiful hydroxyl groups could be intercalated into adjacent GO nanosheets via hydrogen bond interaction (Wang et al. 2013), and GO nanosheets were assembled into laminate structures through π-π stacking and cation cross-linking. These forces increased the stability of the membrane. Under the same oscillation condition, slight peeling of the GO layer was observed on the GO-modified membrane while no visible structural destruction was observed on the CHNs-GO modified membrane. This indicated the better stability of CHNs-GO modified membranes due to intercalation.
SEM images of the original MCE, GO modified membrane and CHNs-GO modified membrane are shown in Figure 2. The original MCE had a smooth surface with an average porous diameter of approximately 0.22 μm (Figure 2(a)). Nanocomposites were attached to the membrane after filtration (Figure 2(b)), the nanowires were evenly distributed without agglomeration. The GO-modified membrane was quite smooth and possessed wrinkled corrugations (Figure 2(c)), and the CHNs-GO modified membrane showed greater roughness due to the coverage and intercalation of the nanowires (Figure 2(d)).
The XRD patterns of GO, CHNs, and CHNs-GO are shown in Figure 3. The XRD pattern indicated the nanowire was composed of Cu(OH)2 crystals, which was in agreement with the values in the standard card (JCPDS No. 80–0656). The characteristic peak corresponding to the (001) diffraction plane reflection of GO appeared at around 2θ = 9.7°. It could be seen that the interlayer distance increased from 0.91 nm (2θ = 9.7°) to 1.03 nm (2θ = 8.6°). It implied the successful intercalation of CHNs into the interlayers of GO nanosheets (Zhao et al. 2016). The characteristic peak of GO in composite materials was very weak, which was mainly caused by the following two reasons: (1) the low quantity and relatively low intensity of GO in the sample; (2) CHNs were covered on the surface and inserted into the interlayers of the GO nanosheets (Dubale et al. 2014; Ghayeb et al. 2017).
WETTABILITY BEHAVIOR OF MEMBRANES
Some studies have demonstrated that the special wettability is governed by both channel structures and surface properties (Li et al. 2016; Zhao et al. 2016). For the first term, compared with the surface of the GO-modified membrane, the CHNs-GO modified membrane exhibited hierarchical roughness, enhancing the potential underwater superoleophobicity. On the other hand, the architecture of the CHNs contributed hierarchical nanostructures and endowed membranes with a robust hydration layer at the water/membrane interfaces. Both water CAs in air and oil CAs under water were measured to evaluate the wetting behavior of the membranes. The original MCE membrane displayed oleophobicity with oil CAs of 60° (Figure 4(a)), and the GO modified membrane had water CAs of 67° and oil CAs of 150° (Figure 4(b)). The CHNs-GO modified membrane had water CAs of 53° and oil CAs of 155° (Figure 4(c)), indicating its better hydrophilicity in air and underwater superoleophobic character. In addition, the oil CAs of the CHNs-GO modified membrane was much larger than the antifouling hydrolyzed polyacrylonitrile/GO membrane, which had an oil CA of 120° for highly effective separation of oil-water emulsion (Zhang et al. 2017).
A series of GO and CHNs-GO-modified membranes were employed to separate oil-in-water emulsion by a vacuum driven filtration system at 0.05 MPa. The membranes were fixed in a dead-end filtration cell; each membrane was first precompacted at a specific operating pressure with ultrapure water until it reached a stable permeate flux, and then an oil-in-water emulsion was poured into the filtration cell. When the emulsion was added, and a pressure applied, water permeated through the membrane while oil was retained above the membrane, showing that the emulsion was separated successfully. As shown in Figure 5, a complete separation was achieved for the emulsion. The rejection ratios of emulsions were more than 99%, as determined by UV-vis spectrophotometer. The water and emulsion fluxes of the membranes were tested to evaluate their permeation performance (Figure 6). The original MCE membrane has a water flux of 380 L m−2 h−1 bar−1, but would be clogged when filtering no more than 5 ml of emulsion. The water and emulsion fluxes of the GO-modified membrane were 166 L m−2 h−1 bar−1 and 76 L m−2 h−1 bar−1, respectively, while the water and emulsion fluxes of the CHNs-GO-modified membranes were 349 L m−2 h−1 bar−1 and 278 L m−2 h−1 bar−1, respectively. The CHNs-GO-modified membranes achieve a similar water flux to the original MCE membranes and can be used for the separation of emulsions. These results were in good agreement with the underwater superoleophobic and low oil-adhesive water/membrane interface of the GO/CHNs-GO-modified membranes. The increasing flux of CHNs-GO-modified membranes was caused by the generation of nanowires intercalated channels within the interlayers of the GO nanosheets. In addition, the flux of CHNs-GO modified membranes is much higher than reduced GO/g-C3N4 composite membranes, whose flux of emulsions is around 25 L m−2 h−1 bar−1 (Li et al. 2017a).
The dynamic underwater anti-fouling performance of the CHNs-GO modified membrane was tested. Two μL of oil droplet was captured on the tip of the micro syringe to contact with the membrane and then leave on the surface. As can be seen from Figure 7, the oil droplet was compressed on the membrane surface from a spherical to an ellipsoidal shape, and then left the surface easily. It remained spherical and there were no residual grease spots on the membrane surface, indicating the ultralow oil adhesion of the surface. When the oil-water emulsion was in contact with the membrane, water could pass through because of penetration and the 3D capillary effect (Yang et al. 2015). The trapped water in the hierarchical nanostructures prevented the direct contact of oil droplets with the surface of the membrane (Li et al. 2015).
STABILITY OF CHNS-GO-MODIFIED MEMBRANES
Stability is a common and tough issue for filtration membranes during the process of oil-water separation. The stability of CHNs-GO-modified membranes was monitored by detecting the change in oil rejection during several cycles of oil-water separation. As shown in Figure 8, there was no obvious damage to the coating or changes in the oil rejection of CHNs-GO-modified membranes for the 10 cycles, indicating their good reusability and antifouling properties. The results were consistent with the dynamic underwater anti-fouling performance test. This is probably because oleophobic and most hydrophilic surfaces can achieve the properties of self-cleaning and antifouling at the solid-water-oil interface (Jung & Bhushan 2009).
In summary, we fabricated CHNs-GO modified membranes with good stability, high separation efficiency, and anti-fouling properties through the controlled assembly of CHNs into stacked GO nanosheets. These could separate the oil-water emulsion with high efficiency of >99%. The GO-based nanocomposites membrane might find practical applications in oil-water emulsion separation.
This research has been supported by Jiangsu Province Research Joint Innovation Fund-Prospective Joint Research Project (BY2014123-08), Yangzhou city policy guidance plans (Collaborative Innovation) (YZ2016198), Research Innovation Project for College Graduates of Jiangsu Province (SJLX16_0434) and Jiangsu Collaborative Innovation Center of Technology and Material of Water Treatment.