Differences in Nanostructure and Hydrophobicity of Cicada (Cryptotympana atrata) Forewing Surface with the Distribution of Precipitation

Although the cicada wing has a variety of functions and the nanostructure and surface properties of many species have been extensively investigated, there are no reports investigating diversity of nanostructures and wetting properties within a single species collected at locations with different rainfall conditions. In this study, the hydrophobicity and nanostructure dimensions of the forewing surface of Cryptotympana atrata were measured, based on specimens collected from 12 distributions with varying precipitation averages in China and Japan. The relationships among hydrophobicity, nanostructures, and precipitation were analyzed, and the adaption of hydrophobic nanostructures under different wet environments is discussed. The precipitation of locations in the years the samples of C. atrata were collected only has an effect on the diameter and spacing of wing surface nanostructure, and the multiple years of precipitation may have an influence on the basic diameter and spacing, as well as the height of protrusions. The rougher the wing surface, the stronger the hydrophobicity which was observed from samples taken where the rainfall conditions of the collection years are high. To our knowledge, this is one special example providing evidence of hydrophobic nanostructures found on a biological surface of a single species which shows adaption for specific wet environments.


Introduction
Through several billion years of evolution, biological surfaces have diversified into various functional structures. To float on the surface of water, for example, the upper side of the lotus leaf has a hierarchical structure and wax layer [1] which is superhydrophobic and self-cleaning. In contrast, the margins of the leaves have a much larger contact angle (CA) hysteresis with water owing to the existence of flat folds [2]. Geckos on the other hand have evolved micro/nanostructures on their feet promoting high adhesion combining a synergistic effect of the gecko's muscles [3,4]. The skin of the gecko has also proven to be a multifunctional, hierarchically structured biomaterial capable of various functions, including self-cleaning, low adhesion, superhydrophobicity, antibacterial activity, biocompatibility, and antiwetting [5]. To blend in with natural environments, many biological surfaces also incorporate specific colors (e.g., in the form of spots and patterns [6]). Such colors can result not only from pigmentation but also from patterned structures (e.g., photonic structures [7] and particularly the green color [8]).
There are numerous other examples in nature demonstrating a variety of functions and specific structures. This includes the collection of water from contrasting patterning on the surface of a desert beetle [9], small channels on a cactus [10], high adhesional structure on many mussel species for contacts with substrates on the seashore to prevent tidal and wave removal [11], structures present on the wings of various insects to provide the mechanical and structural support for flight and superhydrophobicity [12][13][14], and antiadhesional structure of soil fragments on dung beetles [15]. Generally, the varying combinations of the micro/ nanostructures and/or the specific chemical components promote or enhance the desired interaction with these naturally occurring contacting surfaces. Thus, examination of these natural templates has practical implications and is often the impetus for many researchers studying such structures and associated mechanistic processes for specific functionality [16].
Many of the features illustrated above are concerned with the interaction of the natural surface with water either to promote or more commonly resist contact. This antiwetting behavior will often depend on the habit and environment of the organism. For example, water repellency of plant leaves is more common in herbaceous species than those in trees and in subtropical regions. Wetlands also appear to have more species with water-repellent leaves when compared to other regions [1]. In addition to plants, flying insects often demonstrate wetting and more often an antiwetting behavior on various body parts (especially wings). Some termite species have finely structured hairs and additional micro/ nanostructure-like waxes to aid in flying during rainy periods, while other species have adopted less hydrophobic surfaces for flight when little precipitation is likely to occur [17,18].
One of the most interesting flying insects studied in recent times is the cicada. Studies have shown that the structure on the wings of these insects is often multifunctional. The wide variety of small-scale structures [19] on these insects has shown a variety of properties/applications including variable wettability, antireflection [20][21][22][23], selfcleaning [24], cell growth platforms [25,26], material strength properties [27], thermoisolative feature [28], and biomimetic fabrication of the nanostructures [29][30][31][32][33]. As well, it has been shown in a number of previous studies that some cicada structures demonstrate control of the interaction with solid bodies (e.g., natural biological contaminants such as pollens as well as hydrocarbons and silica materials) [21,34]. Such studies, as well as observation of the resistance to environmental decomposition of certain regions of the samples (e.g., wings) during collection [22], have led to the investigation of other biological materials such as various bacteria where the cicada membranes demonstrated an antimicrobial effect [26]. We have found previously that there are large differences in the hydrophobicity of wing surfaces among cicadas [18,22,35]. Some species show very weak hydrophobicity with a small water CA, while others exhibit superhydrophobic interactions. The hydrophobicity generally has a positive correlation with the diameter and height of the structure but a negative correlation with the general spacing of features. As well, stronger hydrophobic wing surfaces often contain higher contents of wax [18,36], while functional groups with higher surface energies can effectively decrease the hydrophobicity of the cicada wing [37]. The hydrophobicity also influences the adhesional force of droplets on the wings with hydrophilic wings showing significantly higher adhesion forces [35,37].
In this study, we have investigated a widely distributed species of cicada (Cryptotympana atrata Fabricius, 1775) collected from various locations demonstrating contrasting wetting (rain) conditions. We have measured differences in the wing nanostructure dimensions and hydrophobicity of forewing surfaces of the same species of cicada found at different locations. The relationship of the wing nanostructure differences and the precipitation (rain) conditions is tested by statistical analyses to explore the hypotheses of evolutionary patterns in relation to nanostructures and wetting environments.

Species and Specimens.
The studied species is C. atrata (Fabricius, 1775) (Hemiptera: Cicadidae) which is widely distributed in China, North Korea, Northern Laos, and Japan. Twelve specimens were collected from Beijing, Hebei, Tianjin, Jiangsu, Zhejiang, Sichuan, Jiangxi, Fujian, and Guangxi in China and Osaka in Japan (Table 1). Among these, three specimens are from Beijing, and the other nine are from other respective locations. All samples were airdried and preserved at the National Zoological Museum, Institute of Zoology, Chinese Academy of Sciences.

Sample Preparation.
The outermost region of the forewing of C. atrata was used in this study as defined in Figure 1(a) (circled region). The forewings of the 12 specimens were washed by flowing deionized water with the end parts excised from the individual samples for further experiments.

Contact Angle
Measurement. The hydrophobicity of cicada forewings was measured by placing small water droplets on the surface. The larger the CA, the stronger the hydrophobicity is. If the CA is larger than 90°, the surface is defined as hydrophobic; if it is smaller than 90°, it is defined as hydrophilic. The end parts of the forewings were fixed on a glass slide using double-sided adhesive tape. CA measurements were carried out using a Dataphysics Contact Angle System (OCA 2.0, Germany). The water drop volume was 3 μL. Each sample was measured at 10 sites, with the average value and standard deviation then calculated (Table 1).

Nanostructure Observation and Parameter Calculation.
For observing nanostructures of the wing surfaces, the individual samples were fixed on the sample stage using a double-sided adhesive tape. The top surface of the forewing was coated with a thin layer of gold (~10 nm) using ion sputtering (KYKY SBC-12, Beijing, China). Three sites of each sample were selected to observe the basal diameter (d), basic spacing (s), and height (h) of protrusions as illustrated in Figures 1(b) and 1(c) under an environmental scanning electron microscope (Quanta 200 FEG, FEI, Eindhoven, Netherlands). These parameters were measured using Photoshop (version 12.0) software, by utilising the software ruler function and correlating the sizes with the SEM scale bars. Each parameter was measured 20 times to calculate the average value and standard deviation (Table 1).
For the large CA hysteresis (see inserts in Figures 2(a) and 2(c)), the roughness factor (γ) of the wing surfaces, and the actual cicada structure shape as shown diagrammatically in Figures 1(b) and 1(c) calculated here utilize the structural geometry of a cylinder, which describes a very basic, first principle analysis of the Wenzel model [38]: where γ is the roughness factor (the ratio of actual area to geometry projected area of surface), θ 0 is the CA on a smooth surface of the same material, and θ w is the apparent CA on a rough surface. Thus, the roughness factor (γ) of the wing surfaces was calculated using the following, with the results shown in Table 1: where d, s, and h are the basal diameter, basal spacing, and height of protrusions, respectively.

Precipitation Data and Statistical
Methods. The wide distributions of the C. atrata cicada species were calibrated by positive latitude and longitude, which reflect the gradient of precipitation from north to south. Rain distribution data for each month of the individual year between 1950 and 2000 and their annual average were obtained from the WorldClim data website (http://www.worldclim.org/ current) with a solution of 30′ (~1 × 1 km 2 ) and extracted using the PSDS 2.0 software [39]. The annual precipitation of the different locations was then calculated based on the average precipitation of each month from the year 1950 to 2000 (Table S1). The two samples distributed in Japan (collected in 1932) and Beijing North (collected in 2010)  are excluded from the collecting data because they were not in the specified range. Precipitation data for the specific year of C. atrata sample collection at the respective locations were downloaded from the China Meteorological Data Sharing Service System website (http://cdc.nmic.cn/home.do), with the exception of Jiangsu as it had no precipitation record in 1951 and Tianjin which did not contain information regarding the collection time (Table S2).
Because many variables do not conform to the normal distribution, Spearman's correlation of the SPSS software (version 18.0) was used for statistical analysis of significant correlations between precipitation, nanostructure, CAs, and the specific year of sample collection (the data are shown in Tables 2 and 3), where R is the correlation coefficient and P is the significant level, ( * P < 0 05 and * * P < 0 01), meaning that the two variables have significant correlation. Spearman's analysis shows that precipitation of an individual month or annual precipitation has no correlation with the positive longitude of sample distributions of C. atrata (Tables S1 and S2) but significantly correlates to the positive latitude with the exception of precipitation over several months as indicated in Table S1 (P < 0 05) and Table S2 (P < 0 05). This result is in line with climatic characteristics of China, where precipitation increases when latitude reduces from the north to the south [40]. With the exception of precipitation in July and August (Table S1), January and July to October (Table S2) have no relation to latitude, mainly due to these months showing little differences in precipitation among these various locations.

Relationship between Hydrophobicity and Nanostructure.
Observation of the different wing samples (Figure 1(a)) using electron microscopy showed the structure of the C. atrata wing surface to be cylindrical in shape, with the approximately spherical apex and significantly enlarged base of the structure (Figures 1(b), 1(c), 2(a), and 2(c)). The diameter (d) and height (h) of protrusions on the surfaces of cicada wings showed large differences among specimens (diameters (d) of 80-130 nm and heights (h) of 253-464 nm), but the spacing (s) of protrusions showed minimal differences in the range of 67-92 nm. The roughness of wing surfaces, varying from 3.74 to 7.44 nm, indicates large differences among the samples. Consequently, the CAs also varied from 96.1 to 137.9.1° (Table 1), but the CA hysteresis was large, even when the cicada wings were inverted upside down, the water drops still adhered on the surfaces (as in the inserts in Figures 2(a) and 2(c)). Such variations within a species are not usual for biological samples.
Correlation analysis of Spearman's correlation indicated that CA at the tips of the forewings was positively correlated to protrusion height (P < 0 05) (Figure 3(a)) and calculated roughness (P < 0 05) (Figure 3(b)), especially when the CA is greater than 120°, as indicated by the purple shading   Feb.

Nov.
Dec.  Figure 3. The CA however has no correlation with diameter and spacing between protrusions on the forewing (Figure 3(a)). These results indicate that rougher surfaces have higher hydrophobicity. Samples distributed in Hebei and Fujian (Table 1), as highlighted in blue in Figure 3, have lower CAs of 96.1°and 100.8°, respectively. This is due to inhomogeneous structures resulting from the quasi-ordered arrangement of protrusions on some areas of the wing surfaces (Figure 2(d)) compared to other strictly regular structures (Figure 2(b)).

Relationship of Precipitation with Nanostructure and Life
Cycle. Data analysis shows that nanostructural parameters on wing surfaces have significant correlation with precipitations of several months and annual precipitation for the years 1950 to 2000, as per WorldClim data ( Table 2). As shown in Figure 4, the diameter of the protrusions shows a positive correlation with the precipitation in October (Figure 4(a)), the spacing displays a negative relationship with three more months from September to December and annual precipitation (Figures 4(b)-4(f)), and the height of protrusions also shows a positive correlation with the precipitation of October (Figure 4(g)) and annual rainfall (Figure 4(h)) (P < 0 05), respectively. Meanwhile, using information obtained from the Chinese Meteorological data (Table 3) for the specific month and year of sample collection, the diameter of nanostructure parameters on the wing surfaces does seem to have a significant correlation (P < 0 01) with the precipitation in June (Figure 5(a)), November ( Figure 5(b)), and annually ( Figure 5(c)). The correlation between rain and spacing also seems significant overall months (Figures 5(d)-5(i)), excluding January, March to May, July, August, and October. The height of protrusions however shows no correlation with the precipitation of any month or annual precipitation over the respective collection dates (Table 3). All the three structure parameters including the diameter, spacing, and height of protrusions have no correlation with the specific year of sample collection (the analysis data in the last column of Tables 2 and 3), and this illustrates that the statistical analysis of the precipitation with nanostructure is not affected by the specific year.
The results presented in Tables 2 and 3 show that the precipitation of multiple years between 1950 and 2000 (World-Clim data (Table S1)), and the specific year (China Meteorological data (Table S2)) of the respective locations of the samples of C. atrata seems to both have an effect on the diameter and spacing of wing surface nanostructure. Precipitation levels of multiple years appear to have a positive correlation with the height of protrusions and illustrate that the wetting conditions may influence the height of protrusions.
Generally, mature nymphs start to unearth in early June and become adults after ecdysis [41]. The nanoscale protrusions on the wing surface are presumably developed when a nymph becomes an adult cicada (akin to the formation of wing scales when a pupa changes into a butterfly [42]). The precipitation level in June in the specific year of sample collection may have the greatest influence on the formation of the diameter and spacing of protrusions when mature nymphs are undergoing the process of ecdysis. In addition, the cicada pupae usually emerge after living underground for several years; hence, precipitation levels would differ from year to year. As a result, it is unlikely that there would be any direct correlation with a single month's precipitation. Therefore, the diameter, spacing, and height of the wing surface nanostructure could be influenced by precipitation over multiple years, especially in October, when conditions are right for the egg to turn into a nymph, with precipitation levels also having an influence on the development of nanostructure. Similarly, the precipitation a month prior to and two months following October, heavily influencing the development of nanostructures, with the spacing of protrusions (Tables 2  and 3) is influenced by the precipitation during those four or five months. This means that the height of protrusions is possibly a consequence of the accumulated effect of multiple years of rainfall since it has no correlation with the precipitation levels in any month of the specific year the samples were collected. The diameter and spacing are not only the effect of evolution, but is also influenced by the rainfall in that specific year which positively correlates with the trends of precipitation during the year when the samples were collected and multiple years. Table 2, it can be seen that the precipitation of an individual month and annual rainfall of multiple years shows no correlation with the CA of wing surfaces denoting the hydrophobicity with a great correlation coefficient (P = 0 171-0.626). From Table 3, while not the specific year of sample collection (P > 0 05), the precipitation during July of the specific year is negatively correlated to the hydrophobic nature of the cicada wing (P < 0 01) and has no correlation with the height of nanostructures. The hydrophobicity of cicada wing surfaces is highly correlated with the height of protrusions as shown here and previously [19,35,37]. This illustrates the height of protrusions is just a correlative but not a determinant factor.

Relationship of Precipitation with Hydrophobicity. From
In wetter conditions (like in the south of China), the bigger the diameter, the smaller the spacing and the greater the height of nanostructure; these three parameters jointly make the wing surface exhibit a higher hydrophobicity than that in drier regions (like in the north of China). Presumably, the samples from wetter areas, with corresponding nanostructures showing a high hydrophobicity, are an evolutionary trait to facilitate a quick removal of water from the wings which would maximize efficiency for movement and flight. In dry conditions, where the rainfall is low and the negative impact of rainfall would be small on flying or movement of insects, the nanostructure of the forewing on the samples had a weak hydrophobicity. In addition, the hydrophobicity of cicada wing surfaces is also dependent on another aspect besides microstructure, such as chemical components, because surface chemistry is also important in determining hydrophobicity of a solid surface in combination with the microstructure [43,44].

Conclusions
In summary, the nanostructure parameters can be seen to significantly affect the hydrophobic properties of cicada wing surfaces. Our study, even though of a small data set, shows significant differences in nanostructure correlated with precipitation at their respective locations. The height of protrusions, roughness of the wing surface, and CA are all related to the precipitation where the species are distributed. The influence of precipitation at the locations during the collection years on the cicada nanostructure diameter and spacing illustrates these two parameters are instantaneously changeable and ambulatory. Conversely, precipitation at locations where the samples were collected over a long period (e.g., 1950-2000) effected the height of nanostructures showing this parameter is constant and evolutionary. Furthermore, the evolution of these nano-and microstructures on a biosurface may be a common phenomenon and worthy of further investigation among populations within diversified taxa. A larger data set of samples collected in the future may corroborate and demonstrate more relationships and subtle differences of these parameters. Given the growing interest in biomimetic material/property development and the knowledge gained in this study, could help in the future design of novel functional biomimetic materials.

Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this article.