Solar water evaporation assisted by photothermal membranes is considered to be one of the sustainable and cost-effective strategies for pure water generation and wastewater treatment. In this work, a self-assembled reduced graphene oxide (rGO) film has been prepared and proposed for direct solar thermal desalination. The morphology, structure, absorbance and desalination performance of the rGO film are explored. It is found that rGO film with optimized microstructure delivers an evaporation rate of 0.87 kg m−2 h−1 with solar thermal conversion efficiency of 46% under 1 sun illumination. Moreover, the evaporation rate of rGO film remains at 0.86 kg/m2·h−1 after ten times recycling, demonstrating the superior reusability.
Nowadays, the shortage of fresh water resources challenges the development of society (Oki & Kanae 2006; Schiermeier 2014). On our planet, the total ocean water storage is 1.322 billion km3, accounting for 96.53% of the total global water. Therefore, it is a very promising task to alleviate the shortage of fresh water resources by desalinating seawater (Li et al. 2015; Zhang et al. 2015; Gao et al. 2017). Among various desalination technologies, solar vapor distillation (SVD) shows many obvious advantages and can directly harvest solar energy to purify water by transferring heat to trigger evaporation (Neumann et al. 2013, 2015; Lou et al. 2016). In this regard, various materials serving as both solar absorbent and evaporation generator, such as carbons (Ghasemi et al. 2014; Liu et al. 2015a, 2015b, 2017a, 2017b), plasmonic metals (Liu et al. 2017a, 2017b; Wang et al. 2017; Zhou et al. 2017) and thermal conversion ceramics (Zhou et al. 2016; Ren et al. 2017) have been proposed to improve the conversion efficiency. However, SVD still suffers from low conversion efficiency due to inefficient utilization of converted heat. According to SVD, heat can be utilized in three ways, which are classified as water heating, parasitic thermal loss and water vaporization. Among them, only water vaporization contributes to high-efficiency SVD. In this case, materials subjected to SVD with high efficiency should have the following properties (Shang & Deng 2016): (1) a wide absorption band as well as high absorption value, (2) high heat usage, (3) hydrophilicity.
Graphene (Sani et al. 2018; Šest et al. 2018) with high thermal conductivity, broadband absorption and good flexibility is very suitable for SVD. Previous studies have been conducted to explore graphene and hybrid film for high efficiency SVD. Ren et al. (2017) designed SVD film by growing graphene on nickel foam through plasma-enhanced chemical vapor deposition, showing an evaporation rate of 1.3 kg m−2 h−1 with an efficiency of 93.4%. Importantly, it was found that the overflow of sunlight and the absorption of light could be controlled by adjusting the deposition angle of the graphene sheets. Zhang et al. (2017) prepared orientated three-dimensional graphene SVD film by employing directional freeze-drying technology (Zhang et al. 2017). The evaporation rate was measured as high as 1.62 kg m−2 h−1 with an efficiency of 86.5%. Latterly, Wang et al. (2018) obtained graphene and carbon nanotube composite by using vacuum filtration. The composite revealed an evaporation rate of 1.2 kg m−2 h−1 and efficiency of 80.4%. Thus, from the aspect of technique, a multilayer graphene film can be simply synthesized through vacuum filtration (Homaeigohar & Elbahri 2017; Yang et al. 2017; Boretti et al. 2018). Nevertheless, due to the two-dimensional flake structure, graphene would stack again during the filtration (Su et al. 2014; Liu et al. 2015a, 2015b). With the increase of graphene concentration, the graphene sheet gradually blocks the basement channel and thus the cascading nanochannel network will be blocked, leading to incomplete heat usage. Therefore, it is necessary to optimize the vacuum-filtration process. For this reason, the graphene SVD film was prepared by employing a vacuum filtration technique and the microstructure was examined and optimized by controlling the filtration concentration. Moreover, the morphology, structure, absorbance and SVD performance of graphene film has been explored.
Reagents and materials
The hydrophilic polyvinylidene fluoride membrane (PVDF, diameter = 0.22 μm) was purchased from Haiyan New Oriental Plasticizing Science and Technology Co., Ltd. Crystalline flake graphite, sodium nitrate (NaNO3), sulfuric acid (H2SO4), potassium permanganate (KMnO4), hydrogen peroxide (H2O2) solution, nitric acid (HNO3), sodium chloride (NaCl), sodium hydroxide (NaOH) and hydrazine hydrate (N2H4·H2O) were purchased from Sinopharm Chemical Reagent Co., Ltd. All the reagents were analytical grade and used as received without further purification. De-ionized water (DI) was used in all the experiments.
Synthesis of graphene oxide
Basically, graphene oxide (GO) was prepared according to a modified Hummers' method (Li et al. 2014). Quantities of 2 g flake graphite and 1 g NaNO3 were initially put into the ice water bath. Then, 50 mL H2SO4 was added slowly under magnetron stirring. After 15 minutes reaction, 5 g KMnO4 was added and kept stirred in the ice bath for 2 h. Subsequently, the mixture was naturally cooled down to 40 °C and kept for 2 h. Then 200 mL DI was added into the solution and stirred for 40 minutes. Finally, a few drops of H2O2 were added. To remove excessive acid, the GO was washed by centrifugation until the pH equalled 7.
Preparation and characterization of graphene film
The schematic diagram of preparing the graphene film for SVD is shown in Figure 1. A certain amount of GO powder was dissolved into 4 mL DI and treated with ultrasonics for 2 hours to form a homogeneous solution. Then the solution was filtrated on the PVDF film by vacuum filtration. Subsequently, the GO-SVD film was sent into a sealed container together with N2H4·H2O, followed by increasing the temperature to 60 °C in a water bath for 2 h. For simplicity, rGO films with concentrations of 0.003125, 0.00625, 0.009375, 0.0125 and 0.2 mg/mL were defined as rGO-SVD-1, rGO-SVD-2, rGO-SVD-3, rGO-SVD-4 and rGO-SVD-5, respectively.
The morphology of rGO-SVD film was identified with a scanning electron microscope (SEM, JSF7500F). The absorbance was measured with an ultraviolet-visible-near-infrared diffuse reflectance tester (Lambda750). Information associated with functional groups was collected with a WQF 520A Fourier transform infrared (FTIR) spectrometer. The temperature and IR images were captured by an E8 FLIR.
It is shown in Figure 1 that the salty water was placed in a glass container with an inner diameter of 43 mm. Simulated sunlight was provided by a xenon lamp (CEL-PE300 L-3A, China). The SVD film was floated flat on the water. The simulated sun illuminated the film perpendicularly from the top. The glass container was placed on an analytical balance connected to a computer to monitor the evaporation of the water in real time.
RESULTS AND DISCUSSION
The SEM image of the rGO is shown in Figure 2(a). It reveals large areal transparent thin layers with flat surfaces and lateral sizes mostly larger than 10 μm. The SEM images of various rGO-SVD films are illustrated in Figure 2(b)–2(f), corresponding to rGO-SVD-1 to -5. Obviously, the morphology of rGO-SVD film is quite variable. Specifically, there are many tiny pores distributed on the rGO-SVD films, which can be observed from Figure 2(b)–2(d). However, the rGO-SVD film became smoother and the pores vanished when the rGO concentration increased to 0.0125 mg/mL, in terms of the rGO-SVD-4 film. On the other hand, cracks could be captured as indicated by arrows shown in Figure 2(e), implying weak mechanical performance. Moreover, it is obtained from Figure 2(f) that the rGO-SVD-5 film exhibits smooth surfaces and typical wrinkles caused by the stacked GO boundaries. In addition, the insets of Figure 2(b)–2(f) clearly show that the diameter of pores presenting on the rGO-SVD films is reduced from ∼1 μm to ∼220 nm, associated with rGO-SVD-1 to -3. It should be noted that the thicknesses of rGO-SVD-3 and rGO-SVD-5 films were measured for comparison, which were 10 and 200 nm, respectively. Thus, the rGO-SVD-3 film is favored for penetrating water during thermal evaporation.
The surface functional groups of the GO and rGO-SVD films were examined and are drawn in Figure 3(a). It reveals (Zhang et al. 2010) O–H stretching vibration ranging from 3,000 to 3,800 cm−1, C = O vibration at 1,698 cm−1, C = C vibration at 1,523 cm−1 and C–O vibration at 1,157 cm−1. Basically, the FTIR features of rGO-SVD are similar to that of GO, but intensities are all significantly weaker. In particular, the C–O bond disappeared completely in rGO-SVD, suggesting partial reduction due to the treatment of hydrazine hydrate vapor. The light absorbance of the rGO-SVD films was acquired to confirm the solar light-harvesting efficiency (Figure 3(b)). It reveals the average absorbance values of 0.631, 0.323, 0.244, 0.213, and 0.147, in accordance with rGO-SVD-5, rGO-SVD-4, rGO-SVD-3, rGO-SVD-2, and rGO-SVD-1, respectively. This is consistent with the rule that the absorbance of graphene increases with the increase of the graphene layers (Zhang et al. 2009). Figure 3(c) illustrates the surface temperature of the rGO-SVD films. It is often employed to verify the light-to-heat conversion performance of solar thermal films. The measurement was recorded under 1 sun light irradiation and the typical IR image of the rGO-SVD-3 film is shown in Figure 3(d). When all the rGO-SVD films were exposed under the solar irradiation for 60 mins at room temperature, the surface temperature sharply rose to a certain value and gradually stabilized after 60 mins. Taking rGO-SVD-3 film for example, the surface temperature reached up to 44.8 °C after 120 mins. Remarkably, the difference in surface temperature between rGO-SVD-1 and rGO-SVD-5 was over 10 °C.
The mass change of the rGO-SVD films with respect to time under 1 sun irradiation (1 kW/m2) is drawn in Figure 4(a), showing that rGO-SVD-3 owns the best desalination performance. Under the same conditions, the mass change of GO-SVD-3 was 1.0. Obviously, the rGO film is superior in application to solar thermal evaporation. To demonstrate the evaporation performance and reusability of the rGO-SVD-3 film, the solar water evaporation experiment was repeated ten times using 0.2 M NaCl (Figure 4(b)). Obviously, the evaporation rate remained at 0.86 kg/m2·h−1. Moreover, the solar conversion thermal efficiency η of rGO-SVD-1 to -5 is revealed in Figure 4(c). Basically, η varies greatly with the rGO concentration. From rGO-SVD-1 to -5, the evaporation rate is 30%, 38%, 45.5%, 42% and 38%, respectively. Figure 4(d) is a schematic of solar thermal conversion on the PVDF membrane, and rGO-SVD-1, rGO-SVD-3 and rGO-SVD-5 films. Regarding the pure PVDF membrane, water can be easily transported onto its surface via capillary force, however, the solar thermal conversion of the PVDF membrane is undesirable due to the low solar absorbance. In terms of rGO-SVD film, the PVDF membrane serves as both scaffold for supporting rGO film and water transportation channel. During evaporation, the rGO film is responsible for absorbing solar energy to vaporize the surface salty water and thereby the potable water can be collected. In this case, the surface texture of the rGO-SVD film is of great importance. Regarding rGO-SVD-1, the pores formed on rGO have larger diameter than those of PVDF pores, which implies that the water cannot reach the surface of the rGO, leading to low mass change. Further, when the rGO pores have almost the same diameter as PVDF pores (rGO-SVD-3), the water transported onto the surface of the rGO film can be fully vaporized, resulting in high solar thermal conversion efficiency. However, when the PVDF pore is completely covered by rGO (rGO-SVD-5), the heat usage decreases, resulting in low solar thermal conversion efficiency.
In conclusion, we propose a simple and effective method to self-assemble a graphene oxide (rGO) film for solar evaporation desalination (SVD). It is found that the surface structure of the rGO film can be controlled by the filtration concentration. Thus, the optimized rGO film delivers an evaporation rate of 0.87 kg/m2·h−1 under 1 sun illumination with solar thermal conversion efficiency of 46%. Furthermore, the evaporation rate of the rGO film remains at 0.86 kg/m2·h−1 after ten times cycling, demonstrating the superior reusability. It is believed that this study has important implications for the graphene-based hybrid films for high-efficiency SVD.
This work was supported by the National Natural Science Foundation of China (No. 21862016), and the Higher Educational First-rate Discipline Construction Project of Ningxia (Chemical Engineering and Technology, No. NXYLXK2017A04).