2.3. Solar Steam Generation Performance
Firstly, the performance of solar steam generation using pure water as analyte is evaluated to optimize the product composition. The impacts of aerogel density, loading amount and composition of the photothermal materials are respectively studied.
The density of an aerogel mainly affects saturation water content of the solar steam generator. As seen from Figure 2 a, the density of solar steam generator becomes larger with increased mass ratio of Janus nanofibers to dispersion medium in freeze drying process, while the saturation water content is decreased. Figure 2b shows the water evaporation performances of the solar steam generators. As the density increases, the solar steam generators display increased first and then decreased water evaporation rates. The increase in water evaporation rates can be attributed to alleviative thermal dissipation from evaporation surface to bulk water caused by decreased water contents in the solar steam generators. The thermal conductivities of CA and PVB are determined to be 0.17 W m-2K-2 and 0.23 W m-2K-2, respectively, which are substantially lower than that of water (0.59 W m-2 K-2). Hence, decreased water content in the solar steam generator is beneficial to reduce heat dissipation and increase the temperature of evaporation surface. As seen in Figure 2c, the temperature (after one-hour irradiation) of evaporation surface increases with increased density of the solar steam generator, whereas that of the bulk water is reduced. However, the increased density of solar steam generator simultaneously results in increased resistance of water transfer, which is negative to water evaporation. Therefore, appropriate aerogel density is an important basis to ensure high-efficient water evaporation.
Secondly, optimized loading amount of photothermal materials is explored, as shown in Figure 2d. It can be seen that the solar steam generator with the photothermal materials of 0.016 g·cm-2 exhibits the best performance. Insufficient photothermal materials cannot produce enough heat to sustain rapid water evaporation. On the contrary, because the pores in the photothermal materials are small (at the nanoscale), excessive photothermal materials leads to too high resistance of water transfer and thus decreases water evaporation rate.
Thirdly, the dosages of raw materials for preparing photothermal materials are optimized. To start with, the concentration of dopamine is fixed (y=2.4), while the dosage of TEOS is varied (x=0.1, 0.2, 0.3, 0.4, 0.5), and relevant SSG results are shown in Figure 2e. With the increased dosage of TEOS (i.e. enlarged content of SiO2NPs), the water evaporation rates reveal an increasing trend at the beginning and then decrease when x>0.3. To explain this phenomenon, at first, the surface temperatures of the solar steam generators are measured. As seen from Figure 2f, with the gradual increase of SiO2NPs content, the surface temperatures of the solar steam generators show a falling trend, which can be ascribed to the fact that the SiO2 NPs, although have been modified with dark-colored PDA, possess relatively lower photothermal conversion performance than CNTs. The light absorption spectra of the solar steam generators shown in Figure 2g can prove the degradation of light absorption performance with increased SiO2 NPs content. Therefore, from the perspective of surface temperature, the performance enhancement of the products within x=0.1 to 0.3 cannot be well explained. It has been known that there are three kinds of water in a material: bound water (BW), intermediate water (IW) and free water (FW).[42]Among them, BW is the water molecule in direct contact with the hydrophilic materials. Because strong chemical bonds are formed between BW and the materials, BW is difficult to evaporate. The FW is relatively far from the adsorption materials, and its property is consistent with the bulk water. The IW is sandwiched between the BW and FW. Due to the interaction between the adjacent BW and adsorption materials, the hydrogen bonds between IW and BW are weakened, leading to the lowest energy required for the evaporation of IW. That is, IW has the lowest evaporation enthalpy among the three kinds of water.[43] In this study, SiO2 NPs can fill in the gaps among CNTs and thus regulate the water state in the photothermal materials. Figure 2h visually depicts the increased proportion of IW and reduced proportion of FW when more SiO2 NPs are introduced. In addition, the variation of BW can be ignored because BW is in monomolecular type and its proportion is very low among the three kinds of water. It has been reported that the ratio of IW to FW in a material can be quantitatively measured by Raman spectroscopy, and Gaussian function can be used to fit the peaks at 3233 cm-1, 3401 cm-1, 3514 cm-1 and 3630 cm-1.[44] The peaks at 3233 cm-1 and 3401 cm-1 correspond to the in-phase and out-of-phase vibration modes of O-H bonds in water that forms four hydrogen bonds with surrounding water molecules, which represent the existence of FW. The peaks at 3514 cm-1and 3630 cm-1 are assigned to the symmetric and asymmetric stretching of O-H bonds in the weakened hydrogen bonds, which are the characteristic peaks of IW that is relatively weakly affected by hydrogen bond. The ratio of the fitted peak areas of the two kinds of water is the molar ratio of the two kinds of water. The Raman spectra of the hydrated photothermal materials are given in Figure S4 . According to the calculation of fitted peak areas, the molar ratios of IW to BW in these samples (x=0.1, 0.2, 0.3, 0.4, 0.5) are 0.38, 0.52, 0.67, 0.71, 0.73, respectively, which further demonstrates increased proportion of IW when more SiO2 NPs are introduced. However, excessive SiO2 NPs (x=0.4 and 0.5) cannot substantially increase the proportion of IW, but make the light absorption performance of the photothermal materials continuously degrade, as revealed in Figure 2g, resulting in the decreased water evaporation rate.
The impact of PDA is evaluated while the ratio of CNTs and SiO2 NPs is fixed (x=0.3). Here, the PDA mainly acts as an adhesive in the photothermal materials, and also contributes to photothermal conversion due to its dark color. In order to study the adhesive performance of the PDA, the solar steam generators with different PDA contents were completely immersed in water for 1 h and then naturally dried. The mass changes of the samples before and after soaking are given in Figure 2i. When PDA is not enough (y=0.8 and 1.6), some photothermal materials fall off the solar steam generators and disperse into the water, and obvious mass losses of the samples are found. When sufficient PDA is introduced (y≥2.4), the photothermal materials can be stable on the solar steam generators. In spite of this, excess PDA content is still detrimental because the light absorption performance is impaired and the water evaporation rate is decreased, as indicated in Figure 2i-l.
Based on the above, the optimized parameters for preparing the solar steam generator are as follows: the mass ratio of nanofibers to water is 0.0500:1 in the dispersion medium for freeze drying; the dosages of TEOS (x) and dopamine (y) are respectively 0.3 and 2.4 for preparing the photothermal materials; the loading amounts of the photothermal materials is 0.016 g·cm-2. The optimized solar steam generator can provide a state-of-the-art water evaporation rate (2.944 kg m-2 h-1) under 1 sun irradiation. Next, the impact of light irradiation intensity on water evaporation rate and energy efficiency is investigated. Light irradiation intensity is one of the important factors affecting the water evaporation rate. In order to further explore the water evaporation performance of the optimized solar steam generator under different light irradiation intensities, 0.5, 1, 1.5 and 2 suns are respectively adopted. As shown in Figure 3 a and b, water evaporation rates of 1.486 kg·m-2·h-1, 2.944 kg·m-2·h-1, 4.469 kg·m-2·h-1 and 5.899 kg·m-2·h-1 are achieved when the irradiation intensities are 0.5, 1, 1.5 and 2 suns, while the energy efficiencies are 88.74 %, 91.05 %, 91.94 % and 92.83 %, respectively. In most reports on solar steam generators, energy efficiency generally increases with the increase of irradiation intensity.[10,32,45] Of the solar steam generator prepared in this work, the evaporation surface temperature significantly rises with increased irradiation intensity, whereas the temperature of bulk water varies little, which indicates excellent thermal insulation performance of the prepared solar steam generator. The experimental results show that the product possesses superior SSG performance under different irradiation intensities, suggesting its application potential in the natural environment where sunlight intensity changes frequently and the conditions with concentrated sunlight by using optical concentrators such as convex lens and heliostats.
The stability of solar steam generators is also a key factor affecting their practical application. Hence, the real-time water content, continuous working stability and reusability of the prepared solar steam generator are studied. The test method of real-time water content is as following: the solar steam generator is taken out from bulk water every 20 min during SSG process, the water drops on the surface of solar steam generator are wiped off by using filter papers, and then the weight of the waterlogged solar steam generator is measured and compared with that of the dry solar steam generator. Continuous working stability is evaluated by ten-hour uninterrupted irradiation, and the mass changes of water are recorded per hour to calculate hourly water evaporation rates. The reusability is studied by twenty-cycle SSG processes. Continuous ten-hour irradiation is employed for each cycle, and the solar steam generator is dried after each cycle.
The test results are shown in Figure 4 a-c. The water content of the solar steam generator is basically constant in SSG process, indicating continuous and stable water pumping. The water evaporation rates remain steady within both ten-hour continuous SSG process and twenty -cycle reusing. These results manifest that the product possesses excellent stability and has great application potential.
An unavoidable problem with solar steam generators in practical outdoor applications is the flip of the generators due to the effects of wind, water wave and etc . In order to accommodate this situation, the prepared cubic solar steam generator is fully covered with photothermal materials on its six surfaces, so that its all surfaces can be used for water evaporation, as depicted in Figure 4d. The performance recovery of the solar steam generator after flipping is investigated, and the specific experimental methods are as follows: in a sixty-minute SSG experiment, the solar steam generator is randomly flipped every 10 min to switch the evaporation surface. The temperature of evaporation surface (Figure 4e) and water mass loss are recorded after flipping. In order to meticulously investigate the effect of flipping on water evaporation rate, the concept of “interval mean evaporation rate” is proposed. As seen from the red dots in Figure 4f, every dot is also calculated by Formula S1, and the irradiation time (T ) and initial mass of water are reset at the initial point of each interval. The results reveal that temperature of evaporation surface can quickly rise to more than 90 % of the maximum temperature within 2 min, and concurrently, the interval mean evaporation rate is recovered to near maximum. The black line in Figure 4f presents the water mass change (2.740 kg·m-2 h-1 in total) over the course of sixty-minute SSG experiment with quintuplicate flipping of the solar steam generator. It can be seen that the flipping of the solar steam generator does not have a great influence on the total water evaporation rate, meaning that the prepared solar steam generator is adequate for dealing with dynamic water environment.
In order to clarify the superiority of using hydrophile/hydrophobe amphipathic Janus nanofibers aerogels (CA//PVB-based product for short) as host materials for solar steam generators, completely hydrophilic CA nanofibers aerogel (CA-based product), hydrophobic PVB nanofibers aerogel (PVB-based product) and CA-PVB blended nanofibers aerogel (CA-PVB-based product) are also respectively fabricated and coated with photothermal materials through the same freeze drying and air spraying processes as those for preparing amphipathic Janus nanofibers aerogel. The PVB-based product cannot be used as a solar steam generator at all due to its inability to pump water. The CA-based product can pump water but cannot float on water surface. Therefore, a scaffold has to be used to immobilize the CA-based product on the water surface. CA-PVB-based product also can float on water surface by itself. The water evaporation performances of CA//PVB-based product, CA-PVB-based product and CA-based product as well as the water without using any solar steam generators (blank water), are compared. The results of water evaporation rates are shown in Figure 5 a. The CA//PVB-based product exhibits much higher water evaporation rate than CA-PVB-based product and CA-based product (2.944 kg m-2 h-1 vs . 0.771 kg m-2 h-1 and 2.125 kg m-2 h-1), and the evaporation rate of blank water is only 0.414 kg m-2h-1. Compared with CA//PVB-based product, CA-PVB-based product shows a much lower water evaporation performance, even though they have the same composition in raw materials. Through carefully observing the evaporation surface of CA-PVB-based product during SSG process, it can be found that almost no water exists on the evaporation surface, proving a poor water-pumping ability of CA-PVB-based product. As illustrated in Figure 5b, the CA components of CA//PVB Janus nanofibers facilitate rapid upward pumping of water molecules due to the closely situated hydrophilic groups on CA. As for CA-PVB blended nanofibers, the presence of PVB molecular chains results in an increased distance between hydrophilic groups, leading to a much weaker water-pumping ability. Thus, a simple blending of hydrophilic and hydrophobic materials is not applicable for fabricating solar steam generators. The relatively lower water evaporation performance of the CA-based product can be attributed to its excessive internal water content (~ 0.95 g cm-3), which aggravates heat dissipation, as revealed in Figure 5c and d. In addition, it is also found that the CA-based product exhibits poor reusability, as shown in Figure S5 . In order to explain this result, the physical photos of CA//PVB Janus nanofibers aerogel, CA nanofibers aerogel, CA//PVB-based product and CA-based product before and after water soaking are provided in Figure S6 . After 10 cycles, the photothermal materials are seriously peeled off the CA-based product. This phenomenon is attributed to the swelling of CA nanofibers aerogel in water. After water soaking, CA-based product experiences a noticeable expansion in volume, which creates internal stress between photothermal materials and the swollen CA nanofibers aerogel and thus leads to the detachment of photothermal materials. As for the CA//PVB-based product, the hydrophobic PVB components suppresses the volume expansion of the CA components, and the entire aerogel retains almost the same morphology before and after being placed in water. Thus, photothermal materials can be stable on CA//PVB Janus nanofibers aerogel during SSG process, thereby achieving excellent reusability. From the above results, it is obvious that the hydrophile/hydrophobe amphipathic Janus nanofibers aerogels are superior for preparing solar steam generators in view of their high performance, reusability and ease of use.