The nano-photonic structures on the wings of three Papilionidae butterflies, Papilio blumei, Papilio ulysses and Papilio peranthus, were investigated. It was observed that the photonic structure is multi-layer with alternate air and cuticle layers forming one-dimensional photonic crystal. The multi-layer structures of the three butterflies differ subtly but are sufficient to account for the differences in their iridescence. The subtleness is more obvious in their polarized reflection results. We performed the simulation of polarized reflection using characteristic matrix method with parameters obtained from SEM images of butterfly wing scales’ cross-section. The simulated reflection spectra are matched with the experimental spectra to derive the effective refractive index of the air lamina in the butterfly wing scales. It shows that through varying the optical thickness and periodicities in air/cuticle bilayer stacks, the iridescent color of these three Papilionidae butterflies appear different. The result lays the foundation for mimicking the photonic structures of these butterflies.
© Optical Society of America
In nature, many insects, animals and plants make use of their nano-structure to give optical effect such as iridescence, light detection [1–3]. In fact, these nano-structures consist of one-dimensional to three dimensional photonic crystal structures [4,5]. Many air-borne insects, such as butterflies and dragonflies, have these so-called photonic nano-structures. The structures of these butterflies’ scales have finer, nano-dimension structures that are responsible for controlling how the incident light reflected back out. This includes not only the color or wavelengths of light being reflected but also the polarization of the light. In fact, there were reports on studying the iridescence origin of the Morpho butterfly family [6–8]. Apart from the Morpho butterfly species, Papilionidae butterfly family also exhibit colorful wings that possess these nano-structures and they provide the attractive iridescence. There have been studies on the photonic structures on some of the Papilio (P) butterflies [9,10] showing that the color is due to specula reflection of incident light. However, to the best of our knowledge, a systematic study on the variation among the species within a butterfly family on Papilionidae has not been done. A systematic study can provide an indication on how bio-diversity can arise within the same family due to the influence the difference in their respective living environment. In this work, we studied three butterfly species of the Papilionidae family; P. blumei, P. ulysses, P. peranthus, on their optical property and relating it to their physical structure of the wings. Modeling was made on their dispersion property and the model provided support on how the butterfly wing structure disperses the color of the light. The result shows the structural variation among the three species giving rise to different color appearance of their wings.
2. Results and discussion
The color and appearance of the three Papilio species are quite different; P. peranthus (Fig. 1(a)) exhibits yellowish green iridescence, P. blumei (Fig. 1(b)) exhibits green iridescence and P. ulysses (Fig. 1(c)) exhibit bluish green iridescence at normal viewing angle. However, thecolor of the wings appears blue shift at oblique viewing angle. For example, the bluish green iridescence of P. ulysses turns into blue iridescence when the butterfly is viewed from steep angles (Fig. 1(d)). The nano-structure of the wings was examined using Scanning Electron Microscope (SEM); most parts of the wings are composed of scales and they have two clear shapes; one scale has smooth edge whereas the other has serrated edge at the tip of the scale (Fig. 2). From the reflection microscopic photo of black and green regions of the wings, we deduce that the smooth scale is responsible for the green iridescence and the serrated scales do not reflect light i.e. black in appearance (Fig. 2 inset).
Using focus ion beam, we sliced one of the smooth edge scales from each species and looked at its cross-section. Figure 3 shows the cross-section of P. blumei scale; it has 7 layers of cuticle alternate with the so-called air laminae which are in fact composed of cuticle pillars within large air pockets. These air pockets allow the air laminae to have refractive index close to that of air. Hence, the 7-layer scale is effectively a Bragg reflector. Cross-section of P. ulysses and P. peranthus smooth scales were also examined and they all have the same multi-layer structure except that the number of cuticle layer is different; for P. ulysses 7 cuticle layers were observed and for P. peranthus there are 8 cuticle layers. From the SEM cross-section image, we can also obtain the average thicknesses of cuticle and air laminae of the scales.
In order to elucidate the optical process in the nano-structure of the butterfly wings, angle-resolved reflection spectroscopy measurements were taken. The sample was placed at the centre of the sub-stage mounted on a precision rotating stage that can turn ± 60° away from normal incident. The wings are orientated such that the scales on the wings with their long axis parallel to the incident plane of the reflectance measurement (X-Y plane in Fig. 5 inset). The reflection is found to be stronger in the TE (transverse electric) mode and significantly weaker in the TM (transverse magnetic) mode. Hence all the subsequent dispersion results are done in TE mode. Figure 4 shows the experimental TE polarized spectral reflectivity results of P. ulysses, P. blumei and P. peranthus. The reflectivity as function of reflecting angle shows a blue shift with increasing reflecting angle. This indicates that the optical characteristics of the smooth scale originated from the multi-layer observed in Fig. 3. The TE polarized spectral reflectance and dispersive characteristics of the three species were also investigated theoretically. We modeled the butterfly wing scales as one-dimensional photonic crystal consisting of bilayer stacks with high refractive index (cuticle lamina) interleaved with layers having lower refractive index (air lamina). We adopted the value 1.56 for the refractive index of cuticle laminae . For the air lamina, since it is not completely air-filled and has supporting cuticle structure, therefore it has an effective refractive index value between the refractive index of cuticle and air. With the physical parameters obtained from the wing scales cross-section SEM images, we used the characteristic matrix method  and perform the simulation using the software Matlab to simulate the reflection spectra from the multilayer structure in the butterfly wing scales for different viewing angles and matched them with the experimental results. This allows us to derive the effective refractive index of the air laminae in each species. The optical thickness is defined as n1d1 + n2d2 where the n is refractive index and d is the thickness of the lamina. The subscripts 1 and 2 represent the cuticle and air laminae respectively. It basically follows the Bragg reflection condition λgap = 2(n1d1 + n2d2), and that the reflection spectrum peak blue shift with smaller periodicity. Table 1 summarizes the average thicknesses and optical thicknesses of cuticle and air laminae for P. ulysses, P. blumei and P. peranthus. Detail analysis of the butterfly spectral characteristics is shown in Fig. 4. At 10 degree reflection measurement results as an example, the peak wavelength for P. ulysses, P. blumei and P. peranthus are at 460nm, 518nm and 559nm respectively. Also, the full-width-half-maximum (FWHM) of the reflectance spectra for P. ulysses, P. blumei and P. peranthus are 200nm, 174nm and 128nm respectively. In fact, the peak wavelengths and the FWHM of the reflectance spectra are closely related to the detail structures in the scales of the three species. P. ulysses has the shortest optical thickness for one air/cuticle period (250nm) giving rise to bluer color than the other two. P. blumei has longer optical thickness for the one air/cuticle period (271nm) and P. peranthus has the largest optical thickness (277nm) (Table 1). Figure 4(d), the peak wavelengths of the both experimental and simulated reflection spectra for P. ulysses scale are plotted with respect to the viewing angles. Similar results are obtained for P. blumei (Fig. 4(e)) and P. peranthus (Fig. 4(f)) and they confirmed that color dispersion is due to multi-layer of the scale causing Bragg scattering. The results show that FWHM width of the reflection spectrum is a function of the contrast between the refractive indices of the cuticle laminae and the air laminae. The larger the difference of the refractive indices between the layers, the broader is the reflection spectrum. Thus, P. peranthus which has the least contrast in the refractive indices (Table 1) has the narrowest FWHM of the reflection spectra among the three species.
From the physical measurement and the model results, we are able to correlate with the optical results explaining how the photonic nano-structures of these three species dictate their color. For the two-tone color of the P. ulysses wing, the normal view color is blue-green and the color shifts into blue with increasing viewing angle. This is due to the short periodicity of its multi-layer structure that reflects more efficiently in the green – blue region. The change in color with view angle can be shown more clearly when the color coordinates of the three species are plotted onto the CIE (Commission Internationale d’Eclairage) chart (Fig. 5) which gives the color, in coordinates of x and y, we perceive projected onto a display panel. It shows the color shifting trend of P. blumei, P. ulysses and P. peranthus. In fact, it also shows that when view at normal angle i.e. the sight is perpendicular to the wing, P. peranthus has a yellowish green color in appearance, P. blumei has a green color appearance and P. ulysses has a bluish green color appearance. The color blue shifted with increasing view angle as shown in Fig. 5. For the P. ulysses, its color changes from bluish green to blue as the viewing angle getting increasingly acute, whereas the color of P. blumei and P. peranthus changes from green and yellowish green to blue as the viewing angle increased to larger than 50 degree away from normal.
In conclusion, the iridescence characteristics of three species of the Papilionidae butterfly family were investigated. Using the parameters obtained from the SEM images of the scale cross-sections, we simulated TE polarized reflection spectra using characteristic matrix method and matched them with the experimental results. This allows us to derive the effective refractive indices of air laminae inside the scales of the three different species. From the results, we also show that the difference in their iridescence is due to the difference in optical thickness and their periodicities of the air/cuticle bilayer stacks of their respective wing scale. The P. peranthus (P. ulysses) has the least (largest) contrast in the refractive indices between its air and cuticle laminae, and has longest optical thickness for the one air/cuticle period resulting in longest reflection peak wavelength. While P. ulysses has the largest contrast in the refractive indices between its air and cuticle laminae, and has shortest optical thickness for the one air/cuticle period resulting in shortest reflection peak wavelength. The difference in their iridescence is more obvious when their color shift as function of viewing angle is mapped onto the CIE color space. This shows how the wing color changes when we see the butterflies at different angle. These differences through subtle structural changes are sufficient to mark the difference between the three Papilionidae families.
This work is supported in part by Hong Kong Ocean Park and the Papilionidae butterfly species are from by Penang Butterfly Farm, Pulau Pinang, Malaysia. Authors are grateful to Photonic Center, Hong Kong Science Park, which did the cross-sectioning of the butterfly wings.
References and links
4. J. D. Joannopoulos, R. D. Meade, and J. N. Winn, Photonic Crystal: Molding the Flow of Light (Princeton University Press, 1995), pp. 38–93.
5. K. Sakoda, Optical Properties of Photonic Crystals (Springer, 2001), pp. 1–11.
6. P. Vukusic, J. R. Sambles, C. R. Lawrence, and R. J. Wootton, “Quantified interference and diffraction in single Morpho butterfly scales,” Proc. Biol. Sci. 266(1427), 1403–1411 (1999). [CrossRef]
8. S. Kinoshita, S. Yoshioka, and K. Kawagoe, “Mechanisms of structural colour in the Morpho butterfly: cooperation of regularity and irregularity in an iridescent scale,” Proc. Biol. Sci. 269(1499), 1417–1421 (2002). [CrossRef] [PubMed]
10. H. Tada, S. E. Mann, I. N. Miaoulis, and P. Y. Wong, “Effects of a butterfly scale microstructure on the iridescent color observed at different angles,” Opt. Express 5(4), 87–92 (1999). [CrossRef] [PubMed]
11. E. Hecht, Optics (Addison Wesley, 2002), pp. 426–428.